Spelling Reform and the Reason It's Impossible (1998) | Hacker News Reader

Spelling Reform and the Reason It's Impossible (1998)

xibalba.demon.co.uk

34 pointsreitanqild24 comments

It is not impossible! The only thing that seems to be impossible is to get people here to actually consider a technical solution to spelling reform [1]. I have posted this and it has sunk without a trace [2]. To not try is a defeatist attitude - what is the point of technology if we don’t use it to solve problems that people think are impossible.

[1] http://www.cutspel.com

[2] https://news.ycombinator.com/item?id=8158579

Edit. Sorry for the tone, but I am feeling rather annoyed that a post saying that something is impossible gets more attention than a post that demonstrates the impossible is possible.

To be fair, the linked post is largely ridiculing the arguments that orthographic reform is impossible and the author would probably agree with you that it isn't.

As for the browser extension for alternative spelling: I think it's brilliant that someone's actually done it (having had a similar idea a couple of years ago and done absolutely nothing). Unfortunately interesting links often sink without trace on HN; there's a fair bit of luck and timing involved in getting traction on here...

Unfortunately, I also dislike Cut Speling orthography (I prefer Valerie Yule's Interspel or even more basic "spelling without traps", which manages to simultaneously be more radically phonetic and look closer to the original English). That's probably the biggest problem English language reform faces: the advocates end up like the Peoples's Front of Judea versus the Judean People's Front...

For the sake of actually justifying the Cutspel dislike

- cutting "schwa" sounds is mostly unnecessary, not least because they often sound in most dialects much more like they're written in English than the Cutspel alternative anyway. (e.g. Most speakers of English pronounce the last syllable of "lemon" much more closely to the original spelling than to cutspel's "lemn", which appears like the "mn" sound in Autumn, omniscient, or damnation.) Brevity for brevity's sake is a bad thing.

- similarly the optional rendering "-cion", "sion" suffixes as shn is enforcing consistency for the sake of consistency, since "cion" is very simply and consistently pronounced throughout English, and also makes the words more intelligible to vast numbers of speakers of Latin languages, not to mention vast numbers of other languages that use loan words like "television" in their everyday speech (I'm glad you appear to have left that part alone)

- Cut Speling changes a very high proportion of English words, including those which are very common, whose spelling is not at all confusing, or both. This makes parsing it far more difficult than it need be for an English speaker

and Interspel

- it includes a short list of very common words which are excepted from the rules like 'the' and 'are'). This makes the reading experience much more acceptable for people familiar with written English, who are used to the words, and requires relatively little memorisation for language learner (and also hypothetically, for people familiar with Interspel to cope with standard English)

- sparing use of a single diacritic mark offer a reasonable pronunciation guide without changing word forms much for native English speakers

That said, spelling reform to avoid confusion could probably start with your surname (and mine)! Bet you always have to spell it out, repeat the "double 'l' and double 't'" and still get people write it down wrong half the time"...

Thanks for the thoughtful post. I agree we could use other reforms other than cut spelling. I have posted the source code on GitHub and anyone can modify it to any spelling reform you like. While I agree that cut spelling has its problems, I do like it is simple and that it results in shorter words. Something like cutspel is the start of the journey, not the end :)

Cutspel has a lot of problems.

> "...but the “ll” in holly is kept"

I count the same number of "L" sounds on "holly" as there are in "holy", namely, one.

> Cut unstressed vowels before l,m,n,r

So is it "betr" or "bettr"? And does that mean "better" or "bettor"?

Why is "ho" "who" and not "hoe"?

Why is "preferences" "prefrnces" instead of "prefrncs"?

Does "knight" become "night" or "nit"?

Your example text doesn't follow the rules. It spells "onwards" as "onwards" when by the above rule it should be "onwrds".

Those are just a few examples I found on a cursory inspection.

In regards holy and holly in my accent they both have two l sounds - hol-lee and hole-lee.

CutSpel just implements a subset of cut spelling (this is for performance reasons and to avoid cutting proper nouns). This is one of the features of cut spelling is that it is not an all or nothing reform - it can be mixed with tradition spelling without issue.

> in my accent they both have two l sounds - hol-lee and hole-lee

If your spelling scheme spells words differently depending on the writer's accent that is a problem in and of itself.

But what accent is it that has a double L sound? I can't even conceive of what that would sound like. If you say two L sounds temporally adjacent to each other without an intervening pause they would be indistinguishable from one L, i.e. "hole-lee" sounds indistinguishable from "hoe-lee". (Hm, how do you cut-spell "wholly"?) Putting enough of a pause between "hole" and "lee" so allow the two L sounds to be distinguished would sound really weird. Is there some audio of this accent on the web?

My accent is Australian, but in most accents there are two ls. Cut Spelling is not my scheme - I have just used it for CutSpel.

I think you might be missing the point of cut spelling. The rules say don't cut where the cutting will cause confusion. It has a "if in doubt don't cut" rule. This one of the major reasons I used it for cutspel as I wanted to only implement a subset of cut spelling.

Agreed. I'm a fan of spelling reform, but the rules they propose are seriously flawed.

The source code is on GitHub so you are free to use it to propose any reform you like.

I agree, just look at how Internet English differs from "correct" usage.

If you build the new spellings into predictive text systems then they will become the default pretty quickly.

This is the idea. I have found that after a couple of hours using cutspel I forget that it is even on - I just read. The fact that the cut spelling forms are shorter and easier to type I find I often automatically use the cut spelling form without thought and it is my spell checker that picks up the "mistake". The browser is a powerful tool to reform spelling!

I didn't post it for its conclusion. : )

Wow. Cutspell looks interesting. Bonus because it makes everything look Scottish to me : )

Thanks. I might try posting again and see if it can some traction. There is obvious a large interest in the topic of spelling reform here so maybe it is just luck what gets picked up.

From Wikipedia (http://en.wikipedia.org/wiki/Non-native_pronunciations_of_En...): Italians learning English have a tendency to pronounce words as they are spelled.

Being Italian myself, I can't help wondering: "If not for pronunciation, what else is spelling about?".

Just for the record, my first "shock" was with English, but learning French was more painful. Really, when you see a sentence written in French you do recognize its Latin origin. But, when you hear it... What went wrong during its evolution? At least, I was happy to discover that a major problem in learning French is common to several nationalities: there are several groups of sounds which the non-francophones do not distinguish, while the francophones cannot understand how they may be confused with one another.

Orthography is more about disseminating communication in an asynchronous way than pronunciation. Over time, it crystallises into something most people can agree on and use to communicate; however, in that time, it will pick up relics that are hard to get rid of. It is also true that while it reaches a relatively stable point, it still evolves slowly (e.g., in texts from about 100 years ago, you sometimes see "cooperate" spelled "coöperate"; in really old texts, you'll see the "long s", like in "highneſs").

The main point is that writing is a human invention, while language itself is thought by most to be a natural cognitive facility, similar to face recognition, that we've evolved to survive and flourish. When a spelling was coined for a particular word, chances are that it was relatively close to its pronunciation at the time, but oral language changes faster than written tradition.

> but oral language changes faster than written tradition

Yes and no! Sure, if they evolve at different speed, oral changes faster than written. But, in order to have any difference at all, you have to have accepted them to diverge at a certain moment in the past.

Luckily :-) this was never the case in Italy. Starting at least with Romans, written has always been meant as a phonetic transcription of the language. The pronunciation rules did change during time, but the underlying correspondence principle has never been given up.

As a native English speaker who learned French, I would have to say that pronouncing written French is easier than pronouncing written English. Indeed there are sounds that are very hard to distinguish ("en dessus" versus "en dessous"), but these sounds are a function of your native language. If you are a native Spanish speaker, the sounds you find difficult will be different than those for an English speaker. However, reading French amounts to pronouncing what is written on the page. Granted it is a far cry from Italian, which I've been told has no exceptions concerning pronunciation), but there are many fewer exceptions than English.

> "If not for pronunciation, what else is spelling about?"

Of course we should be able to read words more or less just by spelling out the letters according to simple rules.

In this regard "my" language is far better. We however have to deal with "word genders" (mostly arbitrarily assigned it seems for me) and other rules that makes little sense other than from an historical perspective.

> there are several groups of sounds which the non-francophones do not distinguish, while the francophones cannot understand how they may be confused with one another.

Common problem : ) An immigrant arab who lived close to my place advertised a "water bomb" for sale in the local newspaper. It turned out to be a water pump. This was funny for us (many years ago) but the reason was obvious: he couldn't hear the difference between b and p.

> ...there are several groups of sounds which the non-francophones do not distinguish, while the francophones cannot understand how they may be confused with one another.

In linguistics, these are called allophones:

http://en.wikipedia.org/wiki/Allophone

>In summary, then… as long as people understand the ways accents vary (a body of knowledge which will clearly be one of the main influences on the system's rules, but which any Cockney already needs for communication with non‐Cockneys), there is no reason to imagine that there are any insurmountable problems here

I'm not sure how the author comes to that conclusion. Assuming, as he does, that this isn't intended to fix pronunciation, all it does is introduce a new set of inconsistent spellings.

For example, my wife pronounces "pass" and "parse" pretty much the same way, where as I pronounce "pass" in the same way as "ass", but pronounce "parse" in the same way as her. As far as I can see, there's no one consistent way to spell all of these words so that they make phonetic sense to both of us.

Or you've got "scone", should that use the same phonetic spelling as "cone" or "one"?

It's interesting to me that spelling reform almost always seem to assume that the pronunciation of the word is definitive and it's the spelling that should change. But there are words in English where the British pronunciation matches the spelling (been, in RP, for example) but the American pronunciation does not. I've always thought that any reform should leave a word alone if any major dialect of English pronounced it as it is spelled. The changes that I think would be the most helpful would be to get rid of archaic spellings such as the 'ough' mess, changing ph to f where appropriate, fixing up more of the -ice, -ise, -ize ambiguities, etc. Don't worry about things like doubled letters, that will sort itself out later.

The language I grew up with has, to a certain degree, mangaged to adapt a lot of words to our writing style, -chocolate for example will be written in the obvious way.

However every time the language committee releases new recommendations people complain loudly, blissfully unaware of our long and successful history in adapting English word to our (saner : ) writing style.

The pronounciation of words changes much faster than spelling does. People write stuff down so that it lasts. You're optimizing the wrong thing if you want to fit spelling to reflect the pronounciation. Making people say words as they are written is much easier to implement. It's still a dumb idea, but it makes much more sense.

> The pronounciation of words changes much faster than spelling does.

Please, have a look at my answer to Xophmeister (https://news.ycombinator.com/item?id=8182703).

Reply on news.ycombinator.com