A (dia)Critical Look at English Spelling

This morning my brain woke me up at 4:30am, and demanded that I couldn’t go back to sleep until I completed a task: I needed to go through the English written language and replace all of the consonant combinations with new letters and diacritical marks. Some version of this happens every time I come to Eastern Europe–I have so much fun figuring out how to spell and pronounce people’s names that I get all excited about diacritical marks, and I want to add a bunch to English to help standardize our ridiculous spelling system. I tried to ignore it, but I also wanted to sleep, so I finally gave in and gave my brain what it wanted in the hope that it would let me go back to bed. In the end it took about four hours, so I never got back to sleep at all, but on the other hand I do have this blog post, so: enjoy.

First things first: my goal here is not to produce a phonetic alphabet, partly because those already exist but mostly because they serve a different purpose–I don’t want to reproduce the exact way we speak, because people speak in vastly different ways depending on the region they come from and where they learned English. All I really want to do is get rid of our wacky system of letter combinations: the o makes a certain sound, as in bot, but sometimes we want it to sound different so we add an extra o for boot, or we add an a for boat, or we tack an e on the end for bote (which is not a real word, but bode is, and you know what I mean.) This is especially stupid where it’s inconsistent: sometimes oo makes a long sound, as in boot, and sometimes it’s a short sound, as in soot. And sometimes that same short sound is written with a u, as in put, and sometimes that u sounds totally different, as in putt. Wouldn’t this be so much easier and simpler with a bunch of diacritical marks, so that every sound has a single symbol that always makes that same sound? Of course it would. Diacritical marks are going to save us. Hooray!

(Hooray, by the way, is going to look completely different by the time we’re done.)

(Side note: I’m not trying to get rid of letter clusters in general, just the ones that we use as hacks to make the letters say different sounds. The pr in pray, for example, can totally stay, because both the p and the r are doing exactly what they’re supposed to do. The sh in show, on the other hand, forces both the s and the h to do things they don’t normally do. That’s the kind of thing we’ll be changing.)

We’ll start with consonants, because they’re easier, and the first one on the block is the h in sounds like sh and ch. The simplest change here is to steal the Slovakian system, which is to add a little arrow (called a caron or haček) over the top of the letter, so those are now Š and Č. The combination th is harder to deal with, because it makes two sounds: it’s voiced in words like this, and it’s unvoiced in words like thistle. And even though that difference is almost never phonemic (ie, it almost never changes the meaning of a word), sometimes it is: word pairs like thy and thigh are extremely rare, but they do exist, so we need a way to handle them both. I thus declare that a voiceless th shall be written as Ť, and the voiced version shall be written as Ď. Because both of those letters have risers in the lower case, they’re written as half-carons, which mostly look like accent marks: ť and ď.

Just for the sake of completeness, anytime z makes the same kind of sound, as in azure, we’re going to give it a caron as well: Ž. Why? Because I can.

(Side note: yes, I realize that the voiced partner of Č is J, but as much as I’d like to I’m not going to add a caron to the J, because in English J only has one job, so we don’t need a caron to differentiate it from anything else. If we start adding a caron to every affricate in the alphabet we’ll go insane.)

We have more consonant clusters to deal with, and one of them is super easy: ck is now just k. Sorry, c, but if we’re being honest you’re pretty unnecessary to the language as a whole. Your hard sound can be handled by k, your soft sound can be handled by s, and your only job in the consonant cluster is to preserve a vowel sound, so that, for example, the e in baker doesn’t mess with the a in backer. (Imagine of every kickstarter had bakers instead of backers–the world would be delicious but confusing.) Now that we have diacriticals to handle all our vowel sounds we don’t need to protect our vowels from the tyrannical e, so I’m afraid we’re downsizing the language and c is being let go. The only place it’ll show up, in fact, is in the Č, which is a far cry from it’s former glory but it’s better than nothing.

(Side note: yes, I realize that I just said I’m not putting a caron on the J because that’s it’s only job, and I just made that C‘s only job as well, so why am I still putting a caron on Č? Because it will be super confusing otherwise. That may seem like a silly reason, but trust me. It’s way too crazy without that caron.)

Speaking of nothing, that’s what Q and X get: they are eliminated completely. When we need the qu sound we can already make it with kw, so we don’t need a stupid vestigial letter that only functions when it’s clustered with something else. X dies for the opposite reason: it’s a self-contained cluster, without any other letters involved, so we’re killing it and just using ks. Done.

And what about ng? That’s not really an n or a g, so it needs it’s own letter. The phonetic alphabet writes that sound with a funky symbol that most keyboards can’t produce easily, so I’m nominating the Ń instead. It’s not a perfect solution, but at least we can type it. Maybe when the world’s smartphone and keyboard makers adopt my genius system we’ll come up with a better solution.

The last consonant cluster I want to talk about is wh. Sometimes the w is dropped completely, as in who, and sometimes the h is dropped completely, as in what, and sometimes the h and w are both pronounced but in reverse order, as in, um, what. Maybe what was a bad example? The problem is that some English-speaking accents pronounce the h and some don’t, which is where we get, for example, the Family Guy clip of Stewie saying “Cool Hwip” and Brian being confused by the h. This is further complicated by the fact that in some cases the difference is actually phonemic: the h changes the meaning between, for example, which and witch, or whether and weather. So how are we going to handle this in our glorious new system? By siding with the more popular pronunciation and dropping the h from our written language, though you’re still welcome to use it in your spoken language if you want. There are going to be a jillion spoken dialects no matter how we alter the written language, so we’ll do our best to keep things simple.

And as for wr? Screw wr. We’re just going to write it as an r.

Now: on to the vowels. Vowels are going to be hard because English has a shocking number of vowel sounds, and almost all of them are phonemic. The root form b-d is a great example, because you can add almost any vowel sound to those consonants and change the meaning of the word: bad, bade, bed, bead, bid, bide, bod, bode, bud, booed, bowed, Boyd, and probably a bunch more that I’m forgetting. Vowel sounds that don’t fit this particular construction, such as the short u in put, are still phonemic in other words, as we illustrated earlier by the difference between put and putt, or look and Luke, or soot and suit. Compare this to something like Spanish or Japanese, which only have five phonemic vowel sounds each, and you see one of the reasons that so many people have trouble learning English. The fact that we represent these sounds in grotesquely inconsistent ways (ie, boot and soot don’t rhyme, and for that matter neither do bowed and bowed) only makes it worse.

(Side note: as we did with the wh cluster, we’re going to be ignoring certain regional pronunciations during this streamlining process. There are accents on the American east coast where, for example, Don and Dawn sound completely different, and I in fact once knew a family who named a son and daughter Don and Dawn, respectively, and were shocked that people from other parts of the country pronounced them the same. So yes, some vowel sounds are phonemic in some dialects, but if the rest of us can get by without them so can you. Feel free to keep saying them however you want; all we’re changing here is the spelling.)

(Other side note: American English is currently undergoing a massive change in vowel pronunciation called the Northern Vowel Shift. If you live in the great lakes area, odds are good that you or your neighbors pronounce, for example, bag as beg or even baeg, with the same long vowel as bade. This is linguistically fascinating, but throws a wrench in my plans to alter spelling based on vowel sounds. I’m still going to do it, but I’m going to use classic, midwestern, pre-vowel shift vowel sounds. Remember that my goal is not to reflect spoken pronunciation, but to replace the letter combination system: I’m less concerned with how you say hat than with how we as a language turn hat into hate.)

So: we’ll keep the base letters pretty much the same as they’re pronounced in the list I showed above.
A as in bad and lash
E as in bed and felt
I as in bid and ship
O as in bod and far
U as in bud and some

This is already going to shift a bunch of our spellings, since the old system used a for the o sound all the time. Father and mother will become foďer and muďer–and already you’re seeing us start to reverse some of the major consonant and vowel shifts that differentiate modern English from our Roman and Germanic roots. And don’t worry, because foďer and muďer are going to change EVEN MORE by the time we’re done. It’s going to be awesome.

The next set of vowel sounds we change is going to be the long vowels, which I choose to represent with a line over the top, like so:
Ā as in bade and pray
Ē as in bead and need
Ī as in bide and sight
Ō as in bode and cold
Ū as in booed and rude

Next we’re going to deal with the diphthongs, which are vowels we think of as a single sound, but are actually two. We often represent this by tacking on an extra vowel or even a w, as in ouch and wow and point. Using the sounds we’ve codified above, what you’re actually saying in those words is aūč and waūw and poēnt. Sometimes a diphthong is just one letter by itself: technically the ī sound is a diphthong; say the word my really slowly and you’ll see that what you’re actually saying is moē. Even the ā is often pronounced as . But I don’t want to handle diphthongs as vowel clusters, because a simple diacritical mark is way simpler: makes sense to us in a way that moē does not. So I’m going to leave ā and ī as-is, and I’m going to write ow and oy as:
Å as in ouch and wow
Ø as in point and toy

Why am I using these specific symbols? Especially since I’m not using them to mean what they already mean in the languages that already use them? I’m doing it for expedience: I need symbols that a typical smartphone keyboard can produce, just like I did with ń, so here we are.

On the other hand, we’re going to write the short u (put, soot, look) as Ö, because that’s how a lot of languages already do it, so we’re not being completely ridiculous.

There’s one more thing that I want to do, though I’m not sure if it’s strictly necessary. We have a lot of words in English that use a kind of null vowel, such as the second syllable of social, or both syllables in curdle–you can’t really tell what those vowels are, and they certainly aren’t any of the vowel sounds we’ve already defined. The two liquid consonants, L and R, have a strong tendency to modify vowels this way, by reducing them to a kind of faceless placeholder between two other sounds, but even D does it, as in the final sound of landed. These three sounds are all different from each other–the final sound in tumble, and the final sound in tumbler, and the final sound in landed–and they’re not really e‘s either because they’re not really anything. The consonants dominate them to the point that they lose their own identity. I kind of want to eliminate them completely, which the Internet is already doing with things like tumblr, but that creates all kinds of stupid situations: if someone tumbled, I don’t want to write tumbld, and if someone is a murderer I definitely don’t want to write mrdrr. But I don’t want to write murderer either, because we’ve already defined what those u‘s and e‘s mean, and it’s not that. So I’m going to create a new vowel that just means “there’s a vowel here, but only in the most technical sense.” And that symbol shall be henceforth Ë, because it’s very similar to what we’ve already defined the Ö to mean, and it lets us use the umlaut again, which is always fun. So remember when I said foďer and muďer would get even weirder? Now they’re foďër and muďër.

(Side note: A lot of words that end in ed don’t actually give it a vowel sound at all, such as looked or leaped. We don’t use our null vowel in these cases, so if you looked before you leaped it would actually be lökt and lēpt.)

And that’s the whole thing. Our new alphabet looks like this:


See how fun this is? By which I mean:

Sē hå fun ďis iz? Wē kan rīt evrēťēń wē wont wiť dīukritikël morks! Hërā! Ińgliš māks sō muč mōr sens nå. Ī ekspekt evrēwun to rīt ďis wā frum nå on: insted uv år kluńkē ōld sistëm wār wun letër köd mēn a bunč uv difrënt ťińs dëpendiń on wič letërz it wuz nekst tū, wē hav an ajël, eksītiń sistëm in wič evrē letër olwāz māks ďu sām sånd. It’z simpël and kënsistënt. Jøn mē in ďis glōrēës revōlūšën! Rīs up ugenst ďu folēz uv år antikwātëd rītiń sistëm!

Deť! Deť!

Or mābē Ī got u litël kerēd uwā. Nō deť rēkwīrd. Just tīpiń.

3 Responses to “A (dia)Critical Look at English Spelling”

  1. Antonio says:

    Are the names in the sequel to Blue Screen going to be pronounceable? Is there going to be an influx of Icelandic characters?

  2. admin says:

    Pronounceable to who?

  3. Wayne says:

    Well, now I know what the Anglish of the 23rd century looks like.

Leave a Reply