Modect Englich Cpelling Reformc

post by Gordon Seidoh Worley (gworley) · 2022-04-16T23:38:50.212Z · LW · GW · 38 comments

Contents

41 comments

Given the voting on this post suggests it's pretty controversial, I should add a note that this is mostly just for fun. I would have posted this during Good Heart Week but forgot about it, so you're getting it late. This post is more lazy afternoon pipedream than anything else.

It's no secret that English spelling is a mess. And we know the main reason: English spelling was standardized in the middle of a vowel shift, so it reflect an intermediate state of the language rather than the comparatively stable state the language is in today (yes, vowel positions vary across dialects, but these follow patterns that would mostly permit a standardized spelling if one was attempted today).

But not all of the problems are due to the vowel shift. Some of them are because English is written using the Latin alphabet which doesn't quite line up with its sound system. And so we have letters that simultaneously do too much work and not enough.

Real spelling reform for English is nearly impossible at present. It would require the coordination of a billion people across dozens of nations to have a real shot of sticking. Also, there's no mechanism by which to carry a spelling reform out since there's no standards body everyone agrees with (English dictionaries are descriptive rather than prescriptive, Anglo school systems are decentralized, etc.). Publishers and writers could try to unilaterally change spelling, but it's going to be hard to get enough people on board with learning a new spelling system to read their work. Maybe the next J.K. Rowling could pull off a spelling reform if it gradually showed up in a highly popular children's book, but that's tall order.

The most realistic option is to aim for modest spelling reforms and introduce them gradually. Get influential publishers to make small changes to their standards, let people get used to them, and then turn the crank on more reforms generation after generation until the spelling is better.

Here's an idea for the first such modest spelling reform we could make.

Let's fix the letter "c".

"C" makes two sounds, /s/ and /k/. By extension this means "s" and "c" are two letters doing the same job, and in the same way "c", "k", and to some extent "q" and "x" are also used to represent the same sounds.

Let's clean this up. Let's drop 3 letters out of these 5 and keep just 2.

I'm going to suggest we keep the letters "c" and "x". Why these two? Basically for handwriting reasons. If you're ever had to write on a backboard, "c" and "x" are easy enough. "k" requires extra strokes, "s" looks like "5", and and "q" has a bunch of fairly similar letters and I'm in favor of letters that are more easily distinguished. However, this is just my preference. I think the more realistic version is to use "s" and "k", but I'll stick with my preference for the rest of this post.

So what does this mean? "c" makes the /s/ sound, "x" makes the /k/ sound. How does spelling change?

Just replace "s" with "c" and keep using "c" in places where "c" makes the /s/ sound. "sass" becomes "cacc", "city" remains "city", etc.

Replace "k" with "x" and replace "c" with "x" where it makes the /k/ sound. So "cat" becomes "xat", "kick" becomes "xix", etc.

Replace "q" and "qu" with the digraph "xw" generally and "x" with "xc" in most places. So "queen" becomes "xween", "extra" becomes "exctra", etc.

For words with etymological spellings where the above won't work, just fix them to have phonetic spellings. So fix "xylophone" to "zylophone", etc.

As you'll notice this this still not perfect. It's not hard to find words that would still have weird spellings. But that's okay. The goal here is incremental improvement, not fixing everything at once, since I don't think that's possible. Instead, roll out a small change like this, wait 20 years, then try the next one. That way, one day, we'll have fixced our cpelling xonfucionc.

38 comments

Comments sorted by top scores.

comment by Valentine · 2022-04-17T13:32:22.998Z · LW(p) · GW(p)

I've lamented something similar. I irrelevantly desired:

  • 'c' to make the "sh" sound
  • 'x' to make the "ch" sound
  • Banish 'q'. Just replace "qu" with "kw".
  • Remove "extra" sounds from letters:
    • 's' never makes the 'z' sound. If you want a 'z' sound, use 'z'.
    • Likewise 'g' never makes the 'j' sound.

So "The quick brown fox jumps over the lazy dogs" becomes "The kwik brown foks jumps over the lazy dogz." And "Choose wisely" becomes "Xooze wizely".

(I also wondered about replacing the "th" sound with a single letter such as the banished 'q', but it usually makes everything look illegible & awful.)

But in playing with this, I realized that there's still pointless redundancy. Why not convert "choose" to "xuz", and "wisely" to "waizly"? Isn't that better, despite it looking kind of terrible?

This later evolved into wondering why we don't just universally use the international phonetic alphabet (IPA) everywhere. I mean, how are we going to deal with the 12ish vowel sounds English forces to fit into five written letters? Wouldn't the IPA just be… better? Clearer?

But it slowly dawned on me: No. No, it wouldn't.

Consider "butter". In my native dialect it's pronounced "BUH-dr". In some English dialects it's "BUH-tah". I'm not sure where it's from, but I've sometimes heard people from some region pronounce it "BA-dah".

So the same word gets an IPA spelling that's totally differently depending on which dialect you're spelling it from.

If I remember right, this was an old problem in English. It wasn't until the printing press that English spelling got standardized.

So presumably the pronunciation of "butter" in some dialect really was something like "BOO-ter" — much like in German!

And that silly "k" in front of "knight", "knife", and "knoll" really was once pronounced. We just dropped it eventually in each case (apparently because England's French-speaking conquerors couldn't handle it?).

So I think part of what we're looking at here is a tradeoff between…

  • …having spelling match pronunciation versus…
  • …using the same spelling for the same word even across different dialects.

To the extent you go with the latter strategy, spelling is basically guaranteed to make less and less sense over time as language transforms.

(Fun related fact: This is what the hat accent in French is for. At some point the French went through a centralized spelling reform but wanted to encode in the written language that some letter (usually 's') had been dropped in the past. For instance, "to be" in French is "être", which has the same etymological root as Spanish "estar".)

All of which is to say:

I super relate…

…and I think this project is doomed.

Replies from: gworley, Measure
comment by Gordon Seidoh Worley (gworley) · 2022-04-17T16:59:34.927Z · LW(p) · GW(p)

…and I think this project is doomed.

Same. Basically this is just for fun. I don't think I made that very clear originally, so added some text about it. I thought it would be obvious but I think people are taking this more seriously than I meant it.

I think there's two things that are fun to think about here. One is how would we reform English spelling in an idealized way that would work across dialects. The other is to think about "realistic" changes that you can just barely believe could happen under the right circumstances.

Replies from: Valentine
comment by Valentine · 2022-04-19T16:22:12.599Z · LW(p) · GW(p)

Ah, yep. That totally makes sense!

Sorry to sprinkle a bit much serious sauce into your play stew.

comment by Measure · 2022-04-17T16:20:46.109Z · LW(p) · GW(p)

The ch sound is a combination of t + sh, and likewise the 'j' in "just" makes a combination of d + zh (the voiced sh).

'th' has a voiced ("this") and unvoiced ("with") version that would require different letters.

For vowels, I had seven base vowel sounds (pat, pet, pit, pot, putt, put, and peat), and the others were combinations of these.

Replies from: Valentine
comment by Valentine · 2022-04-19T16:18:01.963Z · LW(p) · GW(p)

Aha! I hadn't noticed the ch = tsh thing before. Very cool! I can totally feel it in my mouth. And yeah, I can't make the sound without doing the double letter thing. Huh!

And yeah, lots of letters have this vocal/non flip. Like d is vocalized t, z is vocalized s, etc. Sometimes I wondered about having some kind of "And this is vocalized" accent mark to sort of free up letters.

Heh. To make it all easy to type, I guess one could just use capitalization for vocalization. Skipping the letters that only make sense vocalized of course (like 'n' and vowels).

then eFrithiNG luks afl.

:-D

Replies from: Measure
comment by Measure · 2022-04-19T17:31:27.440Z · LW(p) · GW(p)

Fun fact: When whispering, everything is unvoiced and you have to figure out from context which sounds are meant to be voiced.

Replies from: Rana Dexsin
comment by Rana Dexsin · 2022-04-20T05:37:36.013Z · LW(p) · GW(p)

In English, most unvoiced consonants whose voiced counterparts are in-repertoire non-allophones are aspirated and their voiced counterparts non-aspirated, which seems to make most of the difference in some brief experiments with whispering. Fricatives are the hardest and seem to rely on subtle articulatory differences instead, but this isn't a blinded test, so…

comment by tailcalled · 2022-04-17T08:30:10.058Z · LW(p) · GW(p)

c is actually an important, non-redundant letter:

Replies from: gworley
comment by Gordon Seidoh Worley (gworley) · 2022-04-17T17:03:29.139Z · LW(p) · GW(p)

Is there some summary of this? I watched the first couple minutes but the case made there was kind of weak (removing "c" would be confusing sometimes, but keeping it is also confusing in different ways and if we're changing spelling we can fix these issues) and I'm not invested in this enough to watch a 30 minute video to find out if it makes a compelling case.

Replies from: tailcalled
comment by tailcalled · 2022-04-17T18:31:01.954Z · LW(p) · GW(p)

The main thing is that the soft c sound is voiceless, whereas s is often voiced (same sound as z). I asked a linguist friend about it, and you could probably get away with changing the spelling of words that use s for a voiced sound to instead use the letter z.

Replies from: gworley
comment by Gordon Seidoh Worley (gworley) · 2022-04-17T22:32:18.541Z · LW(p) · GW(p)

Oh then I don't buy the argument of the video then. Not that there aren't real patterns here, but they vary by dialect and if consistent can be fixed by swapping to z as your friend says. C still seems redundant to me.

comment by jimv · 2022-04-17T09:15:02.476Z · LW(p) · GW(p)

"sass" becomes "cacc" ...

... "kick" becomes "xix"...

Any reason why sass would need a double c at the end, but the ck in kick just becomes one x?

Replies from: gworley
comment by Gordon Seidoh Worley (gworley) · 2022-04-17T16:49:23.715Z · LW(p) · GW(p)

Oh, I suppose not!

comment by Measure · 2022-04-17T00:46:27.152Z · LW(p) · GW(p)

exstra

Typo here

Replies from: jimv
comment by jimv · 2022-04-17T09:13:16.728Z · LW(p) · GW(p)

cepelling 

Typo here too, perhaps?

Replies from: gworley
comment by Gordon Seidoh Worley (gworley) · 2022-04-17T16:54:02.098Z · LW(p) · GW(p)

oh, yes. I'm not an expert at my proposed changes!

comment by TLW · 2022-04-17T12:47:42.479Z · LW(p) · GW(p)

"c" makes the /s/ sound, "x" makes the /k/ sound

Making c and k more consistent I understand. Replacing s->c and k->x is a much larger change. It makes it difficult for someone who knows English spelling as it is now to adapt to the change[1].

 

  1. ^

    Or at the very least, it's enough of a change that I can't quickly read words any more[2]. In particular, s is a very common suffix, but c at the end of a word is generally a /k/ sound not a /s/ sound[3], so my parser gets confused and I need to backtrack.

  2. ^

    Or worse, I can quickly read, but misread. I did not read the title as "reforms", plural. I read it as reform, singular.

  3. ^

    cynic, psychic, sac, arc, disc, etc.

comment by gjm · 2022-04-17T01:32:07.588Z · LW(p) · GW(p)

I'm not sure there's really such a thing as incremental improvement here.

That is: if you publish a book whose "only" divergence from currently-standard English orthography is that it makes (say) the change you propose here, then you are publishing a Book With Weird Spellings, and the gap between a book in ordinary English and a book in English-with-weird-spellings is much bigger than the gap between any two specific sets of weird spellings.

No publisher is ever going to say "these changes are so modest that they won't seem like a big deal to readers and reviewers so f--- it, let's give it a try".

(Maaaaaybe a much smaller change could be incremental enough, though I'm having trouble thinking of plausible examples that aren't close to "this single word should be spelt differently", which I think one could get away with. But if you want to get from current standard English orthography to something much simpler or cleaner by a series of changes that small, I think it'll take you several hundred years. Which is to say, I think not coincidentally, you're looking at a rate of change comparable to the rate of organic change in the language's spelling and vocabulary.)

Replies from: mruwnik, ChristianKl, 9eB1, RamblinDash
comment by mruwnik · 2022-04-17T18:03:10.779Z · LW(p) · GW(p)

"ph" seems like a good start, seeing as it's pretty much always an "f". 

Dwarfs -> Dwarves is an example of a word sort of being changed in order to be more consistent. Though in this case Tolkien had to continuously make sure that editors didn't "fix" it. And to be honest it sort of introduced an extra spelling, rather than fixing the current one... 

comment by ChristianKl · 2022-04-17T22:44:31.992Z · LW(p) · GW(p)

For most normal printed books the effort would be prohibitive. On the other hand, ebooks don't have similar problems. It's easy to both publish the ebook in old spelling and in the new spelling. The same goes for websites where you can automatically parse between the spellings. 

When it comes to actual physical books it would be possible to print a special edition of the book in the new spelling with a limited print run and sell it for more money as a collector's item.

Replies from: gjm
comment by gjm · 2022-04-19T20:19:52.969Z · LW(p) · GW(p)

Yup, true, you could pretty easily do ebook editions in pairs. But my prediction for that is that the only people who would buy the new-cpelling edition are the ones who are already on board with cpeling riform, and you can't change a language just by selling reformed-spelling books to people who already think spelling should be reformed: you need to change the minds of the people as a whole.

Maaaybe you could bring them in by spreading the idea that New Spelling Is High Status, but to me that feels unlikely to work. I think that if you want to do that you need some body like the Académie Française officially pushing the new spelling, and even then it's going to be difficult: I think everyone in France ignores the AF when it says unpopular things like "don't use le weekend". It's just really difficult to change a language from the top down.

comment by 9eB1 · 2022-04-17T03:11:05.585Z · LW(p) · GW(p)

Cormac McCarthy successfully published a fiction book with no capitalization (and some reduction in other punctuation), The Road, to critical acclaim. Of course, it didn't lead to a groundswell of novels without capitalization.

Replies from: gjm
comment by gjm · 2022-04-17T13:49:46.025Z · LW(p) · GW(p)

I haven't read The Road, but: isn't this a case like (e.g.) Riddley Walker or A Clockwork Orange where the language is deliberately nonstandard, and meant to be perceived as nonstandard, as a sign that we're looking at a world radically changed from the present one?

Replies from: 9eB1
comment by 9eB1 · 2022-04-17T19:51:58.315Z · LW(p) · GW(p)

Yes, it is intended to be nonstandard, and underscores the misery of the world, as if to say they things have gotten so hopeless they can't even muster up the energy to add quotation marks.

Replies from: gjm
comment by gjm · 2022-04-17T20:03:58.959Z · LW(p) · GW(p)

Then (to make explicit something I left implicit before) I don't think it says anything about what success a publisher would have trying to use currently-nonstandard orthography just because they think it's better.

(Note that even if some nonstandard orthography would be better if everyone used it, it may still be much worse for most readers now because it's not what they're used to, and readers will likely not be happy to have their reading made more difficult because a publisher is on a crusade to improve the English language. And, accordingly, publishers won't do that because they like selling books.)

Replies from: philh
comment by philh · 2022-04-26T18:27:58.296Z · LW(p) · GW(p)

My understanding is that parent commenter is wrong, and The Road was like that just because Cormac McCarthy generally writes novels like that. https://en.wikipedia.org/wiki/Cormac_McCarthy

But The Road is the only one of his that I've read, partly because of that, so to say more I'd need to do more research than a quick look at Wikipedia.

comment by RamblinDash · 2022-04-26T19:55:00.209Z · LW(p) · GW(p)

A single-word change that you can observe in the wild is through -> thru. It appears that way on road signs and I've started to see it creep into informal and even semi-formal writing, e.g. internal-only business email. Unfortunately, the related change throughout -> thruout just looks bizarre, though I guess the original is kinda a bizarre word too.

comment by Jiro · 2022-04-18T00:23:34.561Z · LW(p) · GW(p)

Words have different pronunciations in different dialects. Any attempt to "reform" spelling has to take this into consideration.

comment by Gunnar_Zarncke · 2022-04-17T08:50:38.641Z · LW(p) · GW(p)

Typo in "exstra".

Replies from: gworley
comment by Shmi (shminux) · 2022-04-16T23:59:58.444Z · LW(p) · GW(p)

i'd say, use Spanish spelling rules for English words, and any sounds that do not have a Spanish equivalent should be phased out.

Replies from: gworley
comment by Gordon Seidoh Worley (gworley) · 2022-04-17T16:56:13.555Z · LW(p) · GW(p)

I don't think there's any way to make that work because flattening English vowels down to Spanish vowels would result in way too many homophones.

Replies from: shminux
comment by Shmi (shminux) · 2022-04-17T18:46:57.793Z · LW(p) · GW(p)

Nothing's wrong with cat and cut being spelled the same way, right? Can always double-t one of them to distinguish from the other.

comment by TAG · 2022-04-18T19:36:20.873Z · LW(p) · GW(p)
  1. So you want X to have a sound it has in no other language...so that the very international words "taxi" and "kiosk" have to be respelt "taksi" and "xiosx" ?

  2. Why not do something with "s"? Indeed why insist on one-letter-per-sound, in addition one-sound-per-letter?

  3. Why the lack of interest in vowels?

  4. Why the lack of interest in -gh and -ough? "("Cough", "hiccough", "through", "plough", thorough", etc)

  5. Why the lack of interest in dialects?

Replies from: gworley
comment by Gordon Seidoh Worley (gworley) · 2022-04-18T20:30:15.598Z · LW(p) · GW(p)

I don't see 1 as a real problem. Plenty of orthographies use letters to make sounds they don't in any other language. It's just confusing because of the status quo, but it's also kind of confusing that English "i" doesn't make the /i/ sound most of the time.

For 2, this is a bit of an idiosyncratic thing, but "s" and "k" kind of suck to write, especially on a blackboard, especially if you're writing math. "s" looks like "5". "k" quickly becomes "lc". I want to remove letters that are ambiguous when you write them. I agree that "s" and "k" are the easier options given they already make the intended sounds.

For 3-4, it's not that I'm not interested, it's that I think these are too hard to take on as a first fix. English vowels are complicated so I chose to just ignore them, even where there's low hanging fruit that others have been actively picking (like replacing "through" with "thru", "doughnut" with "donut", etc.).

For 5, I'm actually quite interested in dialects! In particular, I want the spelling the hold across dialects, and am sensitive to changes that would make the spelling more favorable to a particular dialect. Luckily most all dialect differences in pronunciation in English are about vowel placement, so by ignoring vowels I can put that problem off for later. I'd like to maintain the single spelling system (with minor local variations, if necessary) for all dialects, and that means having the spelling make sense for most dialects, e.g. by preserving differences that exist in some dialects and not in others (e.g. don't merge the spelling of "cot" and "caught" or "pen" and "pin").

Replies from: philh, TAG
comment by philh · 2022-04-26T18:32:25.988Z · LW(p) · GW(p)

“s” looks like “5″. “k” quickly becomes “lc”.

Though I had a professor who somehow managed to make "x" looked like "n", so maybe we just need to eliminate handwriting entirely.

comment by TAG · 2022-04-20T15:59:50.224Z · LW(p) · GW(p)

Plenty of orthographies use letters to make sounds they don’t in any other language.

Yes but that's a bad thing, so you don't want more of it. English has plenty of orthographic inconsistencies , but the point of spelling reform is to reduce the number, not create new ones.

comment by Measure · 2022-04-17T00:49:13.297Z · LW(p) · GW(p)

I once worked on something similar. In my version I kept 'S' for the s sound, and 'C' made the sh sound. I think I used 'X' for one of the vowels.

comment by lalaithion · 2022-04-26T20:02:26.149Z · LW(p) · GW(p)
comment by TLW · 2022-04-17T12:00:34.923Z · LW(p) · GW(p)
comment by jimv · 2022-04-17T09:14:45.569Z · LW(p) · GW(p)