Ghoti and choughs again
« previous post | next post »
English is not the worst imaginable choice as a medium of international communication — Chinese would be worse, among a few others. But on the whole, it's seriously bad luck for the human species that English happened to hit the linguistic jackpot. The problem is not the English language itself, which I love dearly and would otherwise be happy to recommend to others. The problem is the way that English is written, which is really, really hard to learn, in comparison to most other languages with an alphabetic writing system.
According to Philip H.K. Seymour, Mikko Aro, and Jane M. Erskine, "Foundation literacy acquisition in European orthographies", British Journal of Psychology, 94:143-174, 2003, this is because children learning to read and write English face a double problem: complex syllables and opaque letter-to-sound correspondences. (They don't discuss how much of this problem is shared by adult learners, who are typically trying to master the English writing sytem at the same time as learning the spoken language, but it's logical that some of the children's difficulty would carry over over to adults.) Their abstract:
Several previous studies have suggested that basic decoding skills may develop less effectively in English than in some other European orthographies. The origins of this effect in the early (foundation) phase of reading acquisition are investigated through assessments of letter knowledge, familiar word reading, and simple nonword reading in English and 12 other orthographies. The results confirm that children from a majority of European countries become accurate and fluent in foundation level reading before the end of the first school year. There are some exceptions, notably in French, Portuguese, Danish, and, particularly, in English. The effects appear not to be attributable to differences in age of starting or letter knowledge. It is argued that fundamental linguistic differences in syllabic complexity and orthographic depth are responsible. Syllabic complexity selectively affects decoding, whereas orthographic depth affects both word reading and nonword reading. The rate of development in English is more than twice as slow as in the shallow orthographies. It is hypothesized that the deeper orthographies induce the implementation of a dual (logographic + alphabetic) foundation which takes more than twice as long to establish as the single foundation required for the learning of a shallow orthography.
Here's their conceptual ranking of European languages in terms of syllabic complexity and orthographic depth:
And here's a graph of some of their test results on first-grade children in various European countries:
Error rates (per cent) for familiar word and simple nonword reading by simple syllable language groups and complex syllable language groups.
Because of national differences in the age of entry to primary school, the children were not all in the same age range, and the "English" children (in Scotland) were the youngest of the lot, by a year or two — though the authors argue that age is not responsible for the differences seen, on the basis of regression analysis with age as the independent variable.
It should also be noted that the subjects in each country were drawn from a small number of schools, apparently in most cases from a single school; and this may increase the importance of local differences in teaching methods. In particular, the question of how to teach reading and writing in English is a very controversial one, and the choice of method can have very large effects. In this case, the representative school for the English language was in Dundee, Scotland, and the authors do note that
Another possibility is that differences in teaching methods may be responsible. In shallow orthographies, such as German, synthetic phonic methods are commonly used (Wimmer, 1993). Some commentators argue that rates of progress could be improved by using these same methods in English. However, this contention ignores the distinction between shallow and deep orthographies. In Scottish schools there is a preference for a mixed method which combines the teaching of a vocabulary of sight words with the teaching of the letters and decoding procedures (Duncan et al., 1997). These methods are well adapted for deep orthographies in which commonly occurring words contain letter structures which are inconsistent with the principles of simple grapheme–phoneme correspondence.
My impression is that terms like "mixed method" can in fact cover a very wide variety of different approaches, in which the proportion, nature and effectiveness of the "synthetic phonic" part can vary widely. So I'd like to see another study in which the same test instruments are used to compare the effectiveness of different teaching methods for English.
Seymour et al. argue that
In shallow orthographies, the tasks of familiar word recognition and decoding are based on a common set of principles (a consistent set of simple grapheme–phoneme correspondences) and may, effectively, be handled by a single process. In deep orthographies, the principles underlying word recognition and decoding are distinct. Beginning readers of English encounter numerous common words (house, father, nice, was, etc.) which contain complex graphemes, contextual variations and irregularities which are not consistent with their concurrent learning of grapheme–phoneme correspondences. To accommodate this discrepancy, word recognition (the logographic process) follows a distinctive developmental pathway.
I don't know this literature well enough to evaluate the single-process vs. dual-process claim. My own experience reading English in broad phonetic transcription suggests to me, unscientifically, that word recognition is a separate process from decoding even in that case. However, no one can rationally deny that the complex subregularities and outright irregularies of the English orthographic system pose significant problems for learners.
Their conclusion:
It may be that the rate of learning can be influenced at the margins, being further delayed by socioeconomic disadvantage (Duncan & Seymour, 2000), possibly accelerated by modifications to the teaching of phonics, and perhaps sensitive to the child’s cognitive maturity when the teaching of reading is introduced. However, there will always be a cost associated with the implementation of a dual foundation process [i.e. one that mixes logographic and phonetic processing], and this will create irreducible differences in rates of progress between learning to read in English or other deep orthographies and learning to read in a shallow orthography.
I looked up this paper because it was mentioned in The Economist "You write potato, I write ghoughpteighbteau", 8/14/2008, and Jarek Hirny wrote to ask about the claim that children take more than twice as long to learn to read English as most other European languages. This passes the test of common sense, it seems to me, and (with the caveat that the Seymour et al. study didn't attempt a systematic study of learning rates within each country), it seems to be empirically supported as well.
The Economist's article is mostly about the idea of spelling reform, and was stimulated by the recent release of test results showing that about a third of 14-year-olds in the U.K. are not reading at what is considered to be an appropriate level:
GHOTI and tchoghs may not immediately strike readers as staples of the British diet; and even those most enamoured of written English’s idiosyncrasies may wince at this tendentious rendering of “fish and chips”. Yet the spelling, easily derived from other words*, highlights the shortcomings of English orthography. This has long bamboozled foreigners and natives alike, and may underlie the national test results released on August 12th which revealed that almost a third of English 14-year-olds cannot read properly.
One solution, suggested recently by Ken Smith of the Buckinghamshire New University, is to accept the most common misspellings as variants rather than correct them. Mr Smith is too tolerant, but he is right that something needs to change. Due partly to its mixed Germanic and Latin origins, English spelling is strikingly inconsistent.
Three things have exacerbated this confusion. The Great Vowel Shift in the 15th and 16th centuries altered the pronunciation of many words but left their spelling unchanged; and as Masha Bell, an independent literacy researcher, notes, the 15th-century advent of printing presses initially staffed by non-English speakers helped to magnify the muddle. Second, misguided attempts to align English spelling with (often imagined) Latin roots (debt and debitum; island and insula) led to the introduction of superfluous “silent” letters. Third, despite interest in spelling among figures as diverse as Benjamin Franklin, Prince Philip and the Mormons, English has never, unlike Spanish, Italian and French, had a central regulatory authority capable of overseeing standardisation.
There are many conceptual, cultural, and political barriers to English spelling reform, as the Economist goes on to explain.
And I don't know anything about methods for reading intruction in the U.K., beyond what Seymour et al. wrote about "mixed methods" in Scotland. But similarly depressing statistics about effective literacy rates in the U.S. seem to me to argue less for a reform of English spelling (which is unlikely to happen, whatever your opinion about it), and more for a reform of American educational methods.
[The tests of U.K. reading abilities are reported e.g. in Graeme Paton, "Sats results: School reading standards drop", 8/12/2008, which says:
Sats results reveal almost a third of 14-year-olds are unable to read to an acceptable level – three years after starting secondary school. […]
Some 62 per cent of boys can read to the appropriate level, against 76 per cent of girls, as the gender gap widened.
I haven't yet found a detailed description of what the expected levels are, or a more extensive account of the distribution of scores. I did find numerous suggestions that the managerial competence of the various public and private agencies involved is more seriously challenged than the students' literacy is.]
Eleanor said,
August 16, 2008 @ 2:09 pm
Is it not possible that the status of English as an international business language will lead to spelling reform? I agree that it can't happen here (UK) or in the US. But it doesn't seem nearly so implausible that my German, Finnish, Spanish, and Chinese colleagues, who communicate between themselves entirely in a business-English rather different from from my own, might agree on some more transparent writing system, publish a dictionary, and simply start using it. We would then adopt it whenever we got round to it. If my Indian colleagues chose to participate too, the game would be over.
Carin said,
August 16, 2008 @ 2:40 pm
Could you please comment more (for the interested but under-informed) on the meaning of "syllabic complexity" and on why Danish is harder on both measures than the other Gmc languages?
Benjamin Zimmer said,
August 16, 2008 @ 2:45 pm
The Economist article provides this footnote: "Fish: gh as in tough, o as in women, ti as in nation (courtesy of GB Shaw). Chips: tch as in match, o as in women, gh as in hiccough." 'Twasn't Shaw, as Language Log readers know.
kamper said,
August 16, 2008 @ 2:50 pm
Has there been any research into any sort of cause-and-effect relationship between English's complexity and its dominant position? I can't imagine that being complex led it to popularity, but did complexity result from influences from languages with which English imperialism brought it into contact? I guess what I'm asking is: was it really "bad luck for the human species that English happened to hit the linguistic jackpot" or would any other language have become just as complex given the same opportunities?
On a different, less serious note, one could argue that the dominant world language being hard to learn is good for linguistic diversity in general. That is, people will learn English whether it's hard or not but they might be more likely to stick with learning a relatively less useful language if it is relatively easier to do.
Daniel B said,
August 16, 2008 @ 4:05 pm
As a person that was brought up as a bilingual, speaking both Danish and English, I would say that spelling in Danish is even harder than spelling in English.
We have words such as Sprængt, Spradebasse or Travlt. The (c)(c)(c)(v)(c)(c)(c) construction is not that unusual in Danish. In contrast to German where almost everything is written as it is said (with a few exceptions), nearly nothing is written as is it is said in Danish.
I find the dual process claim interesting and not far fetched at all.
Amy de Buitléir said,
August 16, 2008 @ 4:06 pm
Spelling reform has long been seen as impractical, but I think that technology will change that situation in a few years by minimising the impact on those who don't wish to learn the new spelling.
Although machine translation leaves much to be desired, software to change new spelling to old (or vice versa) should be easy to develop, and reliable. For example, I would expect to be able to configure my email client to automatically convert emails to my preferred system. So we could carry on an email "conversation", each using the system we're most comfortable with. The same software could be used to convert electronic documents.
Vending machines that can print and bind books on demand have already been developed and are beginning to be installed. This should make it easier and cheaper for publishers to make books available in both the old and new spelling, whichever the customer prefers. And even if the publisher only makes it available in one format, I would expect those the book vending machines to be able to automatically convert it for you. I would also expect E-book readers like the Kindle to have the conversion software built in.
Signage could still be a problem, but I think it would be a minor inconvenience. Important messages such as "high voltage" or "one way" already use icons in addition to the written word. Retailers will probably include both spellings in store signs as long as customers want it.
I'm not saying that it would be easy, but it will be far easier to reform spelling when the supporting technology is available.
Chad Nilep said,
August 16, 2008 @ 4:11 pm
@Eleanor: As wiser heads than mine (e.g. Robbins Burling) have suggested, political upheavals no less weighty than the American Revolution or China's Cultural Revolution have had only minor changes on complex orthographies. Reactions to the small differences in American versus British spelling further suggest how changes outside the "inner circle" (see <a href="http://en.wikipedia.org/wiki/Braj_Kachru" Kachru) of native speakers may be resisted.
I have my doubts that "English as an international business language" will affect spelling very much.
Gemma said,
August 16, 2008 @ 6:27 pm
Your title has "ghoti and choughs" rather than "ghoti and tchoghs" as in the article.
A couple years ago I wore some (modern) Cornish tartan to work on Tartan Day in Scotland. I was patriotically expounding on what the colours represented when I got to "red is for the beak and legs of the Cornish chough". At which point everyone fell about laughing, as to their ears I had just said "chuff", which has some euphamistic meanings you can read on urban dictionary. My protestations of "it's a bird!" and "it's not spelt like that!" fell on deaf ears. :)
[links] Link salad got up early and went hiking, and all I got was this lousy post | jlake.com said,
August 16, 2008 @ 7:19 pm
[…] Ghoti and choughs again — Language Log with some deep geekery about the inefficiency of learning English. […]
Andy J said,
August 17, 2008 @ 4:36 am
@Amy de Buitléir.
Your suggestion for how spelling reform might be accomplished seems practical, but I foresee great problems for which the word prescriptive would be totally deficient. Who will create the new spelling rules, write the software and build the implementing machinery? The big players like Microsoft, Google etc of course, and given their tendency to dominate the markets they operate in, we face the prospect of the spelling systems being copyrighted and licensable only in exchange for money. And the Cupertino phenomenon will seem but a trifling quirk by comparison with some of the output which the spelling translators will come up with. Personally, I prefer the more democratic method of change, namely the one we have had for the past couple of millennia.
JJM said,
August 17, 2008 @ 7:56 am
"Has there been any research into any sort of cause-and-effect relationship between English's complexity and its dominant position?"
The dominant position of English has absolutely nothing to do with any intrinsic qualities of the language at all. In fact, its dominance has nothing to do with language per se.
It has everything to do with the simple historical fact that, since the Battle of Waterloo, the last two global political and economic superpowers back-to-back have been English-speaking.
If, by some fluke, Finland had become the dominant world power, there's no doubt we'd all be sweating through the complexities of Finnish grammar in an effort to apply for that career-enhancing MBA in Helsinki…
James C said,
August 17, 2008 @ 9:08 am
JJM, I rather think the point was not about whether or not English had some magical quality that made nations of its speakers superpowers, but rather that the very fact of its speakers being in such a position of dominance for so long might have affected the language. Which is an interesting question, but I'm not sure how one would go about examining it.
Tristan McLeay said,
August 17, 2008 @ 10:07 am
Andy J, you're referring to the method that got us into this mess in the first place ? If we want to reform spelling, we'll need a push in that direction. An indisputable overlord can give us that push, and will probably let us make a few changes if the results are suboptimal. In fact, I'd say this would be essentially necessary ; particularly if people have the option of using traditional orthography or reformed, the reformed orthography should account for irregular dialectal differences with different spellings. The overlord can't be expected to know the intracacies of every dialect, and so the locally irregular words would be determined by the local people. All they have to do then is respond to use.
JJM said,
August 17, 2008 @ 10:11 am
Well, I'm not sure what the term "complexity" is supposed to mean. I doubt English is any more or less "complex" than any other language in relative terms.
Andy J said,
August 17, 2008 @ 10:48 am
@Tristan McLeay. It's phrases like 'indisputable overlord' which rather reinforce my comments about Microsoft and Google, and the IPR aspects of this proposal. As there is no equivalent of l'Académie Française to act as overlord, and I would venture that this is beyond the means of academics to fund in any case, the involvement of [American] big business would seem highly probable, and to me at least, that is highly undesirable. I would have thought the example of the involvement of the pharmacuetical corporations in genetics and similar areas was warning enough.
And just to be clear, I would not advocate spelling reform just to make it easier for non-natsive speakers to learn English.
language hat said,
August 17, 2008 @ 11:03 am
I entirely agree with Andy J, and the phrase "indisputable overlord" gives me hives.
Aaron Davies said,
August 17, 2008 @ 11:34 am
No, English's current world dominance has (almost) nothing to do with its inconsistencies. Most of them arose between 1000 and 400 years ago, coming from the merger of Anglo-Saxon and Norman French into Middle English, followed by the Great Vowel Shift during the late Middle Ages and the mass borrowing of hundreds or thousands of Latin words during the Renaissance and the Enlightenment. (There were strange things going on even earlier, as Old English emerged from the combination of half a dozen Low German dialects with a liberal seasoning of Norse, but I think they mostly led to simplifications–the general lack of a case or gender system, the simplified conjugation table, etc.) All the Imperial period(s) really did is add a lot of borrowed words to the vocabulary, and most of those were regularized to at least some extent during their transition from "obviously foreign" to "obviously borrowed" to "assumed to be native".
If anything, I'd say the influence usually goes the other way around–the French Academy has found its assigned task of keeping French "pure" essentially impossible, as the actual speakers of French insist on saying "le weekend" and "le software". If the French were still running the world to anything like the degree they were (or at any rate thought they were) in, say, 1675, presumably it would be us, not them, facing this dilemma. (Assuming we even cared–English has always been fairly anarchic, as famously stated by James Nicoll.)
Andy J said,
August 17, 2008 @ 11:47 am
On the topic of language dominence through empire, it seems to me that the duration of empire is an important factor. The dominance of Latin throughout the Roman Empire is evident today through the Romance languages. The dominance of Spanish and Portugese in South America, and to a lesser extent of French in the Francophone countires is in proportion to longevity of their respective empires. I would hazard a guess that Russian will not last long in those countries* which were formerly part of the Soviet Union (despite recent events) because the "empire" only lasted a comparatively short period of time (roughly a single generation).
*excluding those in the current Russian Federation, of course
Bardioc said,
August 17, 2008 @ 11:54 am
We germans think that our language is very difficult. Now, I must learn that it is English, the language we consider very easy. Odd!
It is a very blindfold wish to reform a writing system. Once you have learnt that system, it is part of you, you'll never be able to get rid of it. If it comes to reading a longer or serious text in reformed spelling, you"ll get annoyed, if there are just few changes, or the text will be totally obscure, if many changes were made to the spelling.
Advanced reading is not picking up letter by letter and produce the associated sound, as many people might think. You start out with that in elemantary school. Advanced reading is getting aware of the word shapes as a whole. That's much faster than the letter by letter picking method. Maybe English orthography exposes pupils much earlier to the need to read "word shape"-wise. This may be more difficult in the beginning, but should make them aware that in languages, there's hardly a one-to-one correspondence between letter and sound.
Before advocating a spelling reform, take a longer text you don't know and ask somebody to do the spelling changes you would like to have. Than take the reformed text and try to read it out loud. (Best in front of a huge audience.) Then, you're healed!
If you know German, than, for information over a spelling reform and the problems it causes, see here:
http://www.sprachforschung.org
http://www.vrs-ev.de
Amy de Buitléir said,
August 17, 2008 @ 12:08 pm
If spelling reform ever happens, perhaps it will be parents and teachers who push for it.
For years now I've studied Irish, which underwent a spelling reform in the sixties. After only a few months studying the language, I could correctly pronounce almost any word I saw, and correctly spell almost any word I heard, whether I knew the word or not.
I have read estimates that spelling reform could free up the equivalent of one school year for students to learn other things. Based on my experience with Irish, I can easily believe it. I wonder what I could have learned in that extra year…
Bardioc said,
August 17, 2008 @ 12:37 pm
@Amy de Buitléir, concerning her post August 16, 2008 @ 4:06 pm
"Spelling reform has long been seen as impractical, but I think that technology will change that situation in a few years by minimising the impact on those who don't wish to learn the new spelling."
Spelling reforms are very impractical, regardless of the presence of technology to turn it the other way round in either direction. You normally (if your're not blind) read by eyes, not by technology. Do you like to carry your computer around just to change the orthography on an ad on the wall. Magical eyeglasses would be better for that.
"Although machine translation leaves much to be desired, software to change new spelling to old (or vice versa) should be easy to develop, and reliable."
It should, but is it? Can it ever be? A computer is not conscious, it doesn't know about the context and therefore is never be able to do correct translations (transcriptions) to the other spelling system. (BTW, why deal with two ore more spelling systems? Very confusing!)
Why not stepping further and allowing everyone his own personal spelling?
"For example, I would expect to be able to configure my email client to automatically convert emails to my preferred system. So we could carry on an email "conversation", each using the system we're most comfortable with. The same software could be used to convert electronic documents."
If there's only one spelling, everyone is most comfortable in that single spelling, no need for expensive and unreliable technological crutches.
"Vending machines that can print and bind books on demand have already been developed and are beginning to be installed. This should make it easier and cheaper for publishers to make books available in both the old and new spelling, whichever the customer prefers."
In Germany, it's not an issue of "whichever the customer prefers". Aside form exeptions, you only get books in reformed spelling. It's not intended to have a choice.
It's easiest and cheepest with just one orthography!
"And even if the publisher only makes it available in one format, I would expect those the book vending machines to be able to automatically convert it for you. I would also expect E-book readers like the Kindle to have the conversion software built in."
What do you mean by format? The "orthographic format" (i.e. spelling)
or the actual size of the book? "Automatically" means "full of mistakes", see above!
"Signage could still be a problem, but I think it would be a minor inconvenience."
Why having a "minor inconvenience" if you can have no inconvenience at all? I wouldn't consider it a minor inconvenience, if I would end up in another place due to misspelling on signage.
"Important messages such as "high voltage" or "one way" already use icons in addition to the written word."
So adding a third, or fourth, or fifth, etc, spelling, even if it's not needed? The markers must be very big then. Very expensive for the municipal.
"Retailers will probably include both spellings in store signs as long as customers want it.
Certainly not!
"I'm not saying that it would be easy, but it will be far easier to reform spelling when the supporting technology is available."
Today, everything needs a supporting techology. Even spelling. We should use our brains for more important issues.
Spelling Reforms YOU Can Make said,
August 17, 2008 @ 1:07 pm
If we try, we can all-but-eliminate the single most annoying trigraph of the English language. The fight has already begun — "hiccup" has replaced "hiccough" on one side of the Atlantic. Similarly "donut" has triumphed over its longer alternatives. "Thru", "tho" and "enuf" can be found in writing that isn't hard-core reformist — tho they're not likely to be accepted in formal contexts. Words like "thot" and "dauter" look rather odd at first, but are unlikely to give anyone much touble.
Spelling reform comes from below. With the advent of the Web, there's much more "below" more widely visible than ever before. Slowly weed the sequence "ugh" from your own writing and others will follow suit. Eventually even formal publications will accept the simpler spellings. In the not-terribly-distant future the "ghoti" story will have to become a "foti" story — and then your grand-children can start working on the "ti" part.
Seriously.
—
Brought to you by the Committee to END CAVEMAN SPELLING.
"Ugh" means UGH !
—
(Well, semi-seriously :-)
Bardioc said,
August 17, 2008 @ 1:30 pm
If you think about your ancestry that way …
dr pepper said,
August 17, 2008 @ 2:34 pm
Personbally, i like the idea of haveing a conoical spelling for everything and let variants be treated as images, not strings, and classified as logos, you know, like Toy's R Us. Or Tony Toni Tone.
Amy de Buitléir said,
August 17, 2008 @ 2:41 pm
@Bardioc, I think that a program to automatically convert old spellings to new (or vice versa) that would be very reliable, assuming most of the spelling changes are just that — spelling changes. The program would replace one sequence of characters with another, wherver that sequence occurs. It's nothing like translation, which I concede is a difficult problem.
Of course, things get more complicated if you decide that one sequence of letters should be replaced with different sequences depending on the meaning or part of speech. If pairs of words that are homographs (like "row"=argument and "row"=propel a boat) are no longer homographs in the new system (we might change them to "row" and "roh"), that will complicate the "translation" process. It will also complicate "translation" if we introduce new homographs (deciding to spell both "bough" the same as "bow"). In those cases, the program will need some logic to figure out which meaning is intended.
I have no idea how many homographs there are in the English language, so perhaps I'm being overly optimistic. I assume the majority of homographs will remain homographs, even though their spelling might change. For example "tire" = become fatuiged and "tire"=rubber wheel might stay as they are, or they might both become "tyr", but as long as they are spelled identically, the program doesn't need to worry about which meaning is intended.
Even with those complications, though, I think it's still a much simpler process than translating from one language to another.
Andy J said,
August 17, 2008 @ 4:00 pm
@Amy de Buitléir. You're being too timid. If reform is to have the full effect of "tidying up" English then, yes, all homographs and all homophones need to be disambiguated, but don't stop there. Let's use the opportunity to tidy up difficult grammatical stumbling blocks, say by [re-]introducing full inflection, making all verbs regular in their conjugation, all nouns in their declension and provide new spellings or suffixes for nouns when used as adjectives in noun+noun constructions, standardise the prefixes which give the negative or opposite meaning to a word, and so on. The possibilities are endless. And of course let us not forget the oft-proposed opportunity to shed a few redundant letters from the alphabet. I look forward to reading Chaucer and Shakespeare in Newspeak – no wait, someone else coined that word…
For US readers, please apply an irony filter before reading.
Robert said,
August 17, 2008 @ 8:10 pm
"Spelling Reforms YOU Can Make" – Your proposal for the word thought is not appropriate for those amongst us who pronounce the words cot and caught differently. In fact, some people would think an appropriate pronunciation spelling for thought would be thort.
Stephen Jones said,
August 18, 2008 @ 1:12 am
Few of my foreign students have difficulties with English spelling, probably because they see the spelling more or less at the same time they first hear the word.
When I taught Spaniards it was their Spanish spelling I was constantly correcting, not their English spelling.
Another thing to bear in mind is that for Romance or Germanic speakers English spelling is a boon. A German speaker will find it much easier to see the correlation between 'night' and "Nacht' than between 'nite' and 'Nacht'. The same applies to all the '-ion' endings for Latin speakers.
The problem with simplifying spelling is that it makes it harder for some, because the new 'phonetic' spelling does not represent the pronunciation in their dialect.
After all how can you have a 'phonetic' spelling for a language the speakers of which have very large differences in the individual vowels (there are twenty vowel phonemes in Standard British English, less in Network English, and even less I'm told in Canadian English).
Bardioc said,
August 18, 2008 @ 9:22 am
@ Amy de Buitléir, August 17, 2008 @ 2:41 pm
"@Bardioc, I think that a program to automatically convert old spellings to new (or vice versa) that would be very reliable, assuming most of the spelling changes are just that — spelling changes. The program would replace one sequence of characters with another, wherver that sequence occurs. It's nothing like translation, which I concede is a difficult problem."
I thought about writing "transcription", which would be more the thingy you intended, or "translation", which indicates that the thingy must be seen form an higher point if you want to get better results. Mind, one single typo will mess up your simple replacement scheme and distort meaning.
"Of course, things get more complicated if you decide that one sequence of letters should be replaced with different sequences depending on the meaning or part of speech. If pairs of words that are homographs … are no longer homographs in the new system …, that will complicate the "translation" process. It will also complicate "translation" if we introduce new homographs (deciding to spell both "bough" the same as "bow"). In those cases, the program will need some logic to figure out which meaning is intended."
Do you have any clue about programming? The program would need some knowledge to figure out the meaning. How to provide that to the program? Seems to be very complicated! In Germany, we have programs (spell checkers, "Rechtschreibprüfung") to automatically correct texts in text editor programs to reformed orthography as well as to classical orthography. I never use them. They make much mistakes. (Ok, in German, orthography is not just spelling.) But I wouldn't use them even if they were error-free. I don't consider orthography as a matter of being treated by a computer.
"I have no idea how many homographs there are in the English language, so perhaps I'm being overly optimistic. I assume the majority of homographs will remain homographs, even though their spelling might change."
If you have no idea about such things, why do you advocate for a reform?
"For example "tire" = become fatuiged and "tire"=rubber wheel might stay as they are, or they might both become "tyr", but as long as they are spelled identically, the program doesn't need to worry about which meaning is intended."
So why do you want to confuse your reader about what is meant? A human being is no computer, where you just install a new program, you can't expect that somebody aquainted with the classical spelling will understand the new spelling at once.
"Even with those complications, though, I think it's still a much simpler process than translating from one language to another."
The process might be simpler, but it will yield more confusion.
Bardioc said,
August 18, 2008 @ 10:34 am
@ Amy de Buitléir, August 17, 2008 @ 12:08 pm
"If spelling reform ever happens, perhaps it will be parents and teachers who push for it."
In Germany, there's a lot of protest against the reform, especially from parents and teachers, see here:
http://www.raytec.de/rechtschreibreform/
Wir Lehrer gegen die Rechtschreibreform (We teachers against the Rechtschreibreform)
http://www.schriftdeutsch.de/herladen/DEV-Rat1.pdf
… und die Unzufriedenheit der Eltern über die an Zerstörung
grenzende Beschädigung des Kulturgutes Schriftsprache darzulegen
(… and to state the discontentedness of the parents about the close-by destruction damaging of the cultural possessions of the literary language)
http://www.schriftdeutsch.de/orth-a04.htm
Petition zur Beendigung des Rechtschreibreformprojekts (Petition for the termination of the spelling reform project)
http://www.schriftdeutsch.de/orth-a19.htm
Sprachwissenschaftliches Gutachten (Linguistic report)
http://www.schriftdeutsch.de/orth-a22.htm
Die Schülerinnen und Schüler finden sich nicht mehr zurecht. Auch kein Lehrer kann heute in strittigen Feldern noch eine unstrittig richtige Schreibweise festlegen, da verbindliche, einheitliche und fehlerfreie Wörterverzeichnisse nicht mehr verfügbar sind.
Pupils don't find their way any more. Even teachers can't appoint an undisputed spelling in disputed fields due to of the lack of binding, consistent, and error-free word lists.
In dieser Situation mutet es geradezu absurd an, daß den Schülerinnen und Schülern die einzige wirklich anerkannte, bewährte Schreibweise, nämlich die der allgemein üblichen Rechtschreibung ohne Reform-Veränderungen, die nach wie vor von der ganz überwiegenden Mehrheit der deutsch Schreibenden angewandt wird, die in Milliarden Büchern zu finden ist, in der zahlreiche Zeitungen erscheinen, die an ausländischen Schulen im Deutschunterricht gelehrt wurde und wird, in der Familienmitglieder, Freunde, Geschäftspartner, wahrscheinlich auch Sie persönlich, untereinander in Briefen und Emails kommunizieren, in der Schule als fehlerhaft angekreidet wird.
…
"For years now I've studied Irish, which underwent a spelling reform in the sixties. After only a few months studying the language, I could correctly pronounce almost any word I saw, and correctly spell almost any word I heard, whether I knew the word or not."
I think it's not justified to compare Irish with English or German.
"I have read estimates that spelling reform could free up the equivalent of one school year for students to learn other things."
Could that really be? How do they come to that conclusion? It's the old stupid agrument: For the benefit of the children …
You learn for life, not for your childhood!
"Based on my experience with Irish, I can easily believe it. I wonder what I could have learned in that extra year…"
You're probably no school kind anymore, so your situation is different.
In having learnt the classical English orthography, you've really learnt very much. You are now able to have easy access to 500 years of English literature. You've got easy access to your cultural heritage. (You could be very proud of that!) That's what you fool the children out, if a reform happens. They're then cut off their cultural heritage. Compare it to the cut-off childeren in Bolvangar in His Dark Materials.
Even just memorising unususal spellings trains your brain in memorising something. If you want to learn a foreign language, you will often encounter such situations. In English and French, I learnt spelling and pronunciation seperately. That's quite natural.
Most likely, you would've learnt nothing in that extra year but hanging aroound …
Rob Chametzky said,
August 18, 2008 @ 12:45 pm
For those with world enough and time, the book to read on language and empire(s) is Nicholas Ostler's "Empires of the word: A language history of the world" (2005). And for those with a taste for research results laced with pretty fierce convictions, Diane McGuinness on the (non)teaching of reading of English is a bracing brew.
–RC
Mark Liberman said,
August 18, 2008 @ 1:15 pm
Rob Chametzky: (recommends Nick Ostler and Diane McGuinness)
I'll second both recommendations.
Alexis said,
August 18, 2008 @ 6:35 pm
@Tristan: I can't tell if you are being serious, but:
the reformed orthography should account for irregular dialectal differences with different spellings.
Augh, is all I can say to this. Then the English-speaking world would be truly divided by its common language.
Amy de Buitléir said,
August 18, 2008 @ 7:38 pm
I guess I never thought of the arrangement of letters as being a significant repository for culture, only the words themselves. So if I have a copy of Shakespeare on my hard disk, it has no culture because it's spelled with high and low voltages instead of letters. And then if I display it on my screen, or print it, it re-acquires culture?
I'm joking, of course. I am fond of our quirky spelling, and I do think that something of value would be lost if we simplified English spelling. Something of value was lost when Old English morphed into Middle English and then Modern English; otherwise I would be able to read Beowulf without a translation. But I also think we're missing out on other things of value by having such a difficult spelling system. I would give up our quirky spelling with a heavy heart, but I would give it up.
Andy J said,
August 19, 2008 @ 2:28 am
@Amy de Buitléir. I guess I was a little rude to you in my last post so here's a go at reconcilliation. I'm a photographer so I deal in images- pretty universal, no language barriers, maybe just some small cultural biases along the way. But the minute an image enters the digital world – similar to your model for the machine transliteration of Shakespeare – all kinds of issues arise which can potentially change the image that I submit and you, the viewer, see. Compression techniques removal detail which de-compression cannot restore, color rendition in monitors and printers can utterly change the subtleties [there's a spelling which needs reform!] of the palette, and other factors can change the ratio of height to width, the contrast etc. Those who use different browsers will know that Internet Explorer renders HTML differently to Mozilla Firefox. My point is that the quirks which are in our current spelling are important for the way we "see" the language on the screen or on the page, and letting Microsoft or anyone else mediate that process is not a step I would want to comtemplate. [As a complete aside, I have a totally unresearched theory that of all the technologies, past and present, computing has the least original technical vocabulary of all, with barely any neologisms of significance, and a liberal use of poor homography – perhaps a subject for an LL post?]
Amy de Buitléir said,
August 19, 2008 @ 8:05 am
@Andy J, you raise good points. As a computer geek, I get carried away thinking about what an interesting project it would be to write a program automatically converting from one spelling system to another. But as you point out, there are philosophical and psychological issues to consider here.
Bardioc said,
August 19, 2008 @ 8:20 am
@Amy de Buitléir
You should consider the whole, not just pick out some marginal aspects which seem to support your attitude. In Old English times, only a few people were able to write and to read (yes, reading is the more difficult aspect), and there were very few books compared to nowadays.
Clearly, spelling is a cultural thing. More cultural than spoken words actually. Spoken words are spoken — and then gone. A spoken word differs slightly form the same word spoken by the same speaker uttered at a different time, maybe just seconds later. Writing lasts much longer than just speaking. Without writing (the Bible) our culture wouldn't exist.
Shakespeare was a writer, he worte his plays, and writing is another thing than just speaking. (You most likely work month or even years on a peace of literature. These wouldn't be possible without writing it down.)
Spelling preserves a former state of the language, and I consider that a good thing. (There are people interested in how proto-languages where spoken or at least written, but have no clue because of there's nothing left during the centuries.)
Ken Brown said,
August 19, 2008 @ 2:01 pm
A transparently logn psudonym said: "Words like "thot" and "dauter" look rather odd at first, but are unlikely to give anyone much touble"
"Dauter" might not, but "thot" certainly will. I had to read it three times before I was sure what it meant, and I've heard of the cot/caught merger. Most people round here have no idea that somewhere on the other side of an ocean some (but not all) English speakers have lost the vowel we think of as a "short O".
If you did mean "thought", if you wanted most English-speakers outside some parts of North America to get it at once you should have written "thawt".
Or just possibly "thort". There are many English people for whom "thought" and "fort" are homonyms.
Ann said,
August 19, 2008 @ 2:42 pm
All this talk about spelling reform. How about pronunciation reform? We just start pronouncing words the way they're spelled! Problem solved! Nothing needs to be reprinted, no committees needed…
I kid. I'm a kidder.
Joel said,
August 20, 2008 @ 11:41 am
At the financial company I work for, I regularly see bought written as "bot". I'm new to finance so I don't know how widespread this is in the industry.
Also, I disagree with Amy de Buitléir's comment [Aug 16 @ 4:06 pm] that signage would lag the electronic translations. No! They're the first to do it: Late Nite, Drive Thru, etc.
Dorienne said,
August 20, 2008 @ 10:46 pm
"This has long bamboozled foreigners and natives alike, and may underlie the national test results released on August 12th which revealed that almost a third of English 14-year-olds cannot read properly."
I am very surprised this has not been mentioned, but has anyone considered the problem with a decrease in literacy abilities might come from an increase of things text messaging and chatroom speech? I work with younger children a lot and many of them have very low reading skills because they learn to instant message and text message one another before learning formal writing skills. These days some children learn how to type on a computer before they receive any kind of English training in school. Many youngsters think it makes more sense to write "ill c u l8ter" as opposed to "I'll see you later." Could this not drive those results shown?
[(myl) In a word, no. People have been worried about students who fail to learn to read in English at least since the publication of Why Johnny Can't Read in 1955, half a century before texting and chat rooms. The research on the subject shows that these days, more texting is associated with better reading and writing skills, not worse skills (though the direction of cause and effect is uncertain).]
Spelling reform faces many, many challenges to even approach being discussable. What about how the brain simply understands the written word? Surely many people have seen this example:
"Acndrocig to a raserceh at Camgirbde Univeristy, it dosen't mttaer in waht oredr the lretets in a wrod are. The olny imtrnpaot tnihg is taht the frist and lsat lteetr be in the rihgt palce. The rset can be a ttoal mses and you can sitll raed it wuothit pelobrm. Tihs is bsceaue the huamn mnid deos not raed evrey letetr by iteslf, but the wrod as a wohle."
[(myl) This particular example is a hoary old internet hoax. But in fact quite a bit is known about how the brain processes letters in reading, and this research has largely been motivated by the question of how to teach reading (and to a lesser extent spelling reform).]
The way the brain reads a word throws an even bigger monkey wrench into the issue as even native English speakers with "good" reading skills cannot get the specifics right. For example, throughout the majority of my life, my name has been mispronounced as "Doreen." It was not until I got to college that I found people were capable of pronouncing my name correctly.
Creating even more problems would be the issue of which English-speaking country would set the standardization. Logically, it would be the UK considering it is, more or less, the originator of the language, but considering the number of large American businesses that would probably refuse to accept the new formats, special interest groups would ensure that the standardization would be geared toward American pronunciations and spellings most Americans would prefer. When the variations between UK and US English are considered, eg: towards versus toward, reform seems even farther from a reality.
I find this topic absolutely fascinating as I had never once considered the "tchoghs" issue nor the idea that someone would, or could, interpret English that way. This sounds like something my grandchildren will be discussing with their own grandchildren before any headway is made.
Bardioc said,
August 21, 2008 @ 11:18 am
Here a link to a thread discussing the German spelling reform: http://www.antimoon.com/forum/t11030.htm
Bardioc said,
August 21, 2008 @ 11:38 am
The headline of that thread isn't fair play, but manipulative! Nobody would seriously write that way, English spelling is not that way. With it, English spelling is accused for being that way without being that way. It's a striking headline, but not fair one at all!
In many languages, there are sounds produced slightly different according to the positions in the word or the adjacent sounds. Remember English t at the beginning and in the middle or the end of a word. There are orthographies where they use different forms of a letter depending on the position of that letter. This is quite normal.
But the headline suggests that in English, you can randomly select from several orthographic representations of a sound, trying to convince people about the alleged need for a reform. This is no good style for a scientific work.
[(myl) The title of this post was taken from the lede in the cited article in the Economist, whose title in turn was "You write potato, I write ghoughpteighbteau ". Neither the Economist article nor this blog post is "a scientific work" — though there are certainly scientific works these days that use humor or irony to make a point.]
Bardioc said,
August 21, 2008 @ 12:13 pm
@Dorienne, August 20, 2008 @ 10:46 pm
Josef Weizenbaum claimed that in America, after leaving middle school, most of the children are functional analphabets.
Yes, Internet, SMS, and chat rooms came later, but this doesn't say that there's no connection between that and the lacking of reading skills. If you don't read besides the need to read in school, you probably never really learn to read and to write. You must use that skill. Without practise, you will never do right. If there're other more interesting things, regardless what it is, you'll never practise or really use that skill.
There's much violence in school, much bullying. I know that by personal experince. I fell victim to bullying for many years. Nobody was interested in that. My parents didn't understand what was going on. I was insulted by the use of several monikers almost every day. There were violence, too. There were others who became victims, too. But all that was mostly ignored by the adults.
As an adult, you can get sued if you just say one insulting word. As a child, you are exposed to massive insulting and other kind violence day by day by rights for many a years and never realized as a victim but are accused of misbehaviour by your own parents, who never were interested in what has happened. You'll never get the chance to sue that terrorists called your "fellow students". This is scandalous injustice!
This are the real problems in school, not spelling!
Instead of a spelling reform, we need a moral reform! We need a reform of the ten commandments!
[(myl) I'm sorry you were bullied, and I agree that bullying is often a serious problem in schools. But this is not only irrelevant to the original post, it's even irrelevant to the discussion that the comments have drifted into. If you want to argue for moral reform, do it on your own blog, and we'll link to it when and if it connects to the topics discussed here. Further interventions in this style will simply be deleted.]
Valerie Yule said,
August 23, 2008 @ 8:21 am
II would like to take up Bardioc’s challenge, ‘Befor advocàting a spelling reform, take a longr text u dòn't kno and ask sombody tu du the spelling chànges u would like to hav. Than take the reformd text and try tu read it out loud. (Best in front of a hùge audience.) Then, u'r heald!
Here is his text reritten in ‘Spelling without traps for reading’ in which no grapheme has mor than two possibl pronunciàtions, apart from retaining 31very common irregùlar wurds plus the ending –tion, which make up so much of all running text. (Thòse 31ar not too much for beginners to lern as special silly wurds.)
“We germans think that our language is very difficult. Now, I must lern that it is English, the language we consider very easy. Odd!
It is a very blìndfòld wish to reform a rìting sistem. Once u hav lernt that sistem, it is part of u, u'll never be àbl tu get rid of it. If it comes to reading a longr or sèrius text in reformd spelling, u"ll get annoyd if there ar just few chànges, or the text will be tòtally obscùre, if mèny chànges wer made tu the spelling.
Advanced reading is not picking up letter by letter and prodùce the assòciàted sound, as meny pèpl might think. U start out with that in elementary scool. Advanced reading is getting awair of the wurd shapes as a hòl. That's much faster than the letter by letter picking method. Maybe English orthography expòses pùpils much erlier tu the need to read "wurd shàpe"-wise. This may be mor difficult in the beginning, but should make them awair that in languages, there's hardly a one-to-one correspondence between letter and sound.
‘Befor advocàting a spelling reform, take a longr text u dòn't kno and ask sombody tu du the spelling chànges u would like to hav. Than take the reformd text and try tu read it out loud. (Best in front of a hùge audience.) Then, u'r heald!
That is a sweeping spelling reform bàsed on màking English spelling keep to its rules. An alternativ, that has been coming in gradùally but could be made faster, is simply to cut out surplus letters in wurds which ar of no ùse to indicàte meaning or pronunciàtion and often mislead – as in gardian, garantee, gage. My experiments, which u can replicàte, sho that this streamlìning principl helps beginners, spellers, and lerners of English language, and mòst readers du not nòtice mòst of the chànges.
Dictionaries alredy acsept program, omelet, music, horror, and hundreds mor.
Andy J said,
August 23, 2008 @ 12:11 pm
@Valery Yule
“Maybe English orthography expòses pùpils much erlier tu the need to read "wurd shàpe"-wise. This may be mor difficult in the beginning, but should make them awair that in languages, there's hardly a one-to-one correspondence between letter and sound.
Befor advocàting a spelling reform… ”
< sarcasm filter> For aditional simplisity and uniformity, leters in blue in the above extract could be rendered as “f”, leters in green rendered as “ee” [for the present tens of read, ‘red’ for the past tens] and orl other duplicated leters which serve no fonological purpose shud be reduced to a single ocurence [beginning, correspondence, letter, spelling] but not [need, between]. Altho not a speling reform, maybe elipses cud be reduced to 2 dots.
< /sarcasm filter>
Andy J said,
August 23, 2008 @ 12:14 pm
Sorry about the HTML < and > which didn't work in the previous post, Looked all right in the preview.
Bardioc said,
August 24, 2008 @ 11:17 am
@ Valerie Yule, August 23, 2008 @ 8:21 am
Why are you insulting me? I've never written what you quoted! Quoting means, that the text is taken form the original as is was written there, even with errors. It shows the original you want to refer to. That's the purpose of quoting.
You misused the quoting "mechanism"! Doing so in a scientific text would disqualify you.
Valerie Yule said,
August 26, 2008 @ 8:30 am
Bardioc,
I apologise that the final paragraph that I wrote August 23 was not clearly distinguished from the earlier paragraphs, which I had transliterated into a streamlined English spelling from your post of August 17th. It included your challenge
“Before advocating a spelling reform, take a longer text you don't know and ask somebody to do the spelling changes you would like to have. Than take the reformed text and try to read it out loud. (Best in front of a huge audience.) Then, you're healed!”
So I did exactly that. I was taking up your suggestion, which is a good one.
I explained, ‘Here is his text re-ritten in ‘Spelling without traps for reading’.
Bardioc said,
August 26, 2008 @ 11:14 am
I'm not surprised, obviously, it is not so easy to do the transliterations Amy de Buitléir proposed on August 16, 2008 @ 4:06 pm.
As for "my challenge", did you really took an unknown text, did you really asked somebody to do the "encipherment" and did you really read it out loud in front of a huge audience? It would be better to take an encryption scheme you don't know so you would be in the situation of someone who is first exposed to such a distorted text.
If the frequency of replacements is low, you as a native speaker can easily "reconstruct" the meaning (you do that unconscisiously with typos), but a non-native will probably not. If the frequence of replacements is high, you — if you be honest — will not read much of the text, because you get annoyed for not instantly getting the meaning, isn't it?. (I didn't read your distorted text, I just found my nickname and some wired text quotetd to be me!)
Bardioc said,
August 26, 2008 @ 11:39 am
You are also prejudiced about your spelling reform idea, so if you by yourself read out that text encrypted by your own scheme, it doesn't proove anything, isn't it?
Valerie Yule "I explained, ‘Here is his text re-ritten in ‘Spelling without traps for reading’."
This headline is somewhat manipulative: it suggests that ordinary spelling has traps for reading and your text don't. That's no good scientific style. Do you actually know how reading works? I don't, but by self-observation, I know that it is not picking up one letter after the other.
Have you ever learnt to read some kind of shorthand? I'll give you good advice: try to learn the standard shorthand of your native tounge, and you will learn much about reading and writing. (I once learnt Deutsche Einheitskurzschrift and I still use it.) It is like learning to read once again. You will make the experience, that (at least for DEK) it takes you some month to be able to write it, but it takes you much longer to actually read it halfway fluently.
See that thread about German reform and DEK and how spelling helps to grasp the meaning of words and about why shorthand is much more difficult to read. It's in English with exemples in German and added translations:
http://www.antimoon.com/forum/t11030.htm
Bardioc said,
August 26, 2008 @ 12:45 pm
Have they also asked the 14-years old pupils how often they read a book?
Bardioc said,
September 1, 2008 @ 1:27 pm
Here's an interesting link about a reaction on a discussion of spelling reforms: http://www.antimoon.com/forum/t11487.htm
See also: http://shisaku.blogspot.com/2008/08/measure-of-uncommon-decency.html
There're some quotations there: Bradley, Jespersen, something about Shaw's alphabet and Benjamin Franklin. It seems that Mike Okrand, the inventor of Klingon, was not the first one who used capital letters for further sounds.
Henry Bradley began his "On the Relations between Spoken and Written Language–with special reference to English" as follows:
// from Bradley
Many of the advocates of spelling reform are in the habit of asserting, as if it were an axiom admitting of no dispute, that the sole function of writing is to represent sounds. It appears to me that this is one of those spurious truisms that are not intelligently believed by any one, but which continue to be repeated because nobody takes the trouble to consider what they really mean. I do not merely deny the truth of the pretended axiom as a description of the relations between speech and writing as they exist at the present day in English and other languages. I assert that, so far as peoples of literary culture are concerned, there never was a time when this formula would have correctly expressed the facts; and that it would still remain false, even if an accurately phonetic spelling had been in universal use for hundreds of years.
// end of Bradley
Susan said,
September 11, 2008 @ 8:08 am
On the subject of 'ghoti', and lots more on the English Language, do have a read of this pamphlet:
http://www.sounds-write.co.uk/documents/spelling_theory_and_lexicon.pdf
Language Links » Is English a difficult language to learn? said,
April 15, 2009 @ 2:26 pm
[…] language itself, which I love dearly and would otherwise be happy to recommend to others," writes Mark Liberman on the blog Language Log. "The problem is the way that English is written, […]
HansB said,
November 6, 2010 @ 11:21 am
Maybe the complexity is part of the reason why English is a dominant language, and that English cultures have been so successful? Might it not be that the complexity induces the development of more advanced neurological networks in the brain of English readers, giving them a head start in other matters too?
Nelson said,
May 19, 2011 @ 10:06 pm
No wonder the Chinese and Japanese economies are going to overtake the English and American then: they have even more complexity, which gives even more advanced brains! We should reform our spelling, but the other way from what is usually proposed: switch to a logographic language with spiral direction of writing, realistic drawings of unrelated objects for words, and color, size and thickness of the lines of logographs distinguishing between meanings. This would bring us to world domination.
On an unrelated note, spelling "tchoughs" for chips isn't fair at all, because "hiccough" isn't even one of the usual "ough" pronunciations, but is made on analogy with "cough". In Britain, do people not write "hiccup" yet?