English Spelling Reform

I’m sure almost everyone has encountered some gag related to spelling reform in English. I will not bother to reproduce any of those here as one can simply apply any reasonable search engine to the task and discover many dozens of pages discussing such schemes.

Before I go too far in this commentary on the subject, allow me to mention the wonderful article at Zompist.com that discusses the subject of English spelling. I highly recommend that anyone interested in spelling or other topics linguistic to check out that site.

Now, on to English spelling.

I’ll assume that since you are reading this that you have at least a passing familiarity with written English. If you don’t, how are you reading this? (Yes, I know there are translation gimmicks out there but they will no doubt have made a hash of this.)

In any event, it is a common complaint that spelling in English is difficult. Of course, there are basic rules which are helpful to some extent but there are exceptions to many of them and then there are exceptions to the exceptions and so on. This leads to what many have described as a complete mess.

Let’s take an example of some of the commentary that has occurred. It has been suggested at least once that under the current spelling system in English, one could write “ghoti”. Apparently that is a common word. I have to admit, however, that no matter how I squinted at it, I only managed to come up with a pronunciation that sounds like “goatee”. However, “ghoti” is alleged to be a valid spelling of “fish”. This example, however, assumes there is far more chaos to English spelling than there is. For instance, no native English speaker would even begin to think that “gh” at the beginning of a word is anything but a hard “g” sound like in “ghost”. In fact, most will not even associate the “gh” in “cough” with the “f” sound since most will have learned most of the words ending in “ough” by rote since the pronunciation varies. The “ti” runs into the same problem because it only has the “sh” type sound when it appears in specific contexts such as “action”. And even then, depending on your dialect, it may sound like “ch” rather than “sh”. The “o” may have more justification except for the fact that it is allegedly from the “o” in “women” which is something of an aberration rather than something any English speaker will recognize as a rule. Go ahead, find other words where “o” sounds like a short “i”. I can’t think of any off the top of my head. No, when I see an “o” in the middle of a stressed syllable (which it must be if there is only one in the word), it takes one of the various sounds like in “cot” or “caught” or “coat” or what have you. It does not even degenerate into a schwa, stressed or not.

Let’s leave aside the fact that English spelling is considerably more regular than the spelling reform advocates wish it to be. English has a far bigger problem for spelling than the complexity of the current system. This particular problem must be solved by any replacement system. The fact that no replacement system has gained any real traction is suggestive that this particular problem has not been solved. The problem I am referring to is, of course, the insane number of regional dialects that exist.

While these dialects are all clearly English, their pronunciation varies wildly, particularly with respect to vowels and the letter “r”. The “r” business is easily handwaved away as the difference between rhotic and non-rhotic dialects is quite regular and need not cause any substantial trouble for a writing system. Vowel sounds, on the other hand, are critical. Have you ever wondered why “cot” and “caught” are spelled differently? If you haven’t, then your dialect distinguishes between the two vowels. If you have, then your dialect doesn’t. A trivial amount of research and thought reveals many dozens of such examples between varying dialects, and that only counts the major ones. It turns out that the vowel variance between dialects, even within a single country, simply cannot be represented uniformly by any spelling system. There are too many variations. The Zompist link above talks about that problem.

So any reformed spelling system for English must have the following characteristics to be useful:

  • It must be regular and straight forward to learn.
  • It must be applicable to all dialects of English to be useful. After all, if only one region can read the particular spelling, what’s the point of it?

Let us return to the current spelling system. It certainly has irregularities but the vast majority of words do follow known patterns which can be identified by readers with little effort, to the point that even most unfamiliar words can be pronounced with a bit of thought and practice. It does, however, have its warts, particularly with respect to the vowels, but there are also some issues related to consonant sounds.

As discussed above, the vowel problem is not truly solvable. However, since English has such a complex vowel system, it was never really going to be solvable anyway without the addition of letters to the alphabet or the introduction of diacriticals. (Try enumerating all the different vowel sounds you use in common words; ignore how the vowel is written but rather listen to how the word sounds compared to others. You may be surprised how many distinct vowel sounds you find in your own dialect.) I contend that it is not the vowels that cause the trouble anyway, but the irregularities with consonants.

Consonant sounds do tend to be more consistent across dialects and where they vary, they tend vary due to vowel variance and they tend to do so in predictable, and natural ways. The consonants in English are relatively simple and, for the most part, have perfectly simple means of expression. For instance, the “k” sound or the “s” sound or the “z” sound or the “t” sound. However, there are extra ways of writing such sounds which tend to confuse the issue. As a common example, why is “ph” needed as another way to spell “f”? Why does it matter that the “ph” words come from Greek? We are not speaking Greek nor are we using the actual Greek alphabet so why keep an archaic spelling? After all, “fotograf” is as clear as “photograph”. But that one is not a particular problem for reading English, only writing it. Let’s take the letter “c” as a real example. It sounds like “s” sometimes and “k” other times. Why is that needed at all? Simply replace it with “s” or “k” as appropriate! There are many other examples of such irregularities which have crept into the language for various reasons ranging from aesthetics to laziness to spelling standardization going one way in one part of the world and another in another part.

My contention is that if we were to clean up the mess with consonants some, English spelling would become quite a lot easier. It seems to me that a great deal of the variation in spelling for various sounds exists in the consonants rather than the vowels. If there was only one way to write “f” and only one way to write “sh”, then even with vowel confusion, one might be able to at least guess what a “fosh” is. At the very least, “fosh” looks like it could be English where “ghoti” looks foreign.

There is one important further note that I feel I should add. I have to this point completely ignored the issue of foreign words coming into English. Aside from the consonant confusion we inherited from various sources, there is a much bigger issue with loan words. Almost all of the confusion with spelling in English could be eliminated if load words were respelled to follow standard English phonetic rules! Why should I need to know that “gila” is actually pronunced “heela”? It should not matter where “gila” comes from; it should be spelled according to standard English rules when it is being written as an English word. If one must represent it as its original language source, it would be more properly written gila (note that italics to denote a foreign word).  The same thing should apply regardless of which language the word or phrase came from. For instance, “maître d'”. I usually hear it pronounced in English as “maiter dee” so why do we not write it that way?

Since I have, as is usual, blathered on at great length, allow me to summarize what I believe we should do for spelling reform.

  • Keep the current alphabet and basic spelling rules for basic English words
  • Do not mess about with the vowel representations at all because any such effort is doomed fail; there is simply too much vowel variance between dialects
  • Clean up some of the more obvious extra spellings of various consonants. We can replace such things as “ph” with “f” and “c” with “k” or “s”. “x” is another candidate for complete replacement. Also, we can make all hard “g” sounds “g” and all soft “g” sounds “j”. There are other things that could be done to improve consistency among the consonants.
  • Naturalize the spelling of all foreign loan words to follow the standard rules of English spelling and phonetics where possible. The rules of English spelling are sufficiently flexible that a great many foreign sounds can be spelled without overloading existing spellings. Particular attention should be paid to consonent sounds but vowel sounds should be adjusted as well to reflect the fact that English underwent the Great Vowel Shift while other languages did not.
  • Finally, recognize that it is not possible to have a grand unified spelling system that represents all English dialects accurately. Simply accept that each dialect is going to have exceptions that speakers of that dialect will have to learn.

I hope you enjoyed my rant on spelling. As you can see, I am in favour of some reforms but not a wholesale replacement of everything. Even if none of my suggestions are taken except this one, I would be happy. When we mean the “j” sound, let us write “j”; we have the letter so why do we insist on overloading “g”? And no, “historically ‘j’ did not exist as a distinct letter” does not cut it. We have it now. Let’s use it. And no, “j” does NOT sound like “y” in English!

Okay, enough griping. I now return you to your regularly scheduled whatever it is you do.

2 thoughts on “English Spelling Reform”

  1. This is a very interesting and thought-provoking post. Here are my reactions:
    You are absolutely right that GBS’s ‘ghoti’ is complete nonsense as it doesn’t take into account letter placement within a word. I also agree that English spelling is complex but actually reasonably regular. However, I was surprised to read “I contend that it is not the vowels that cause the trouble anyway, but the irregularities with consonants.” Certainly my research with English language learners (non-native speakers) shows many more problems with vowels than consonants when it comes to spelling (encoding) although admittedly consonants may be more misleading when reading (decoding).

    As for the spelling of foreign loanwords, are you advocating changing the spelling of new ones only or also those that have already entered the English language. If the latter, how far do you go back? Actually I have recently written a post on The Spelling Blog ( http://thespellingblog.blogspot.com )about this.

    My personal take on speling reform is that it’ll either happen naturally, as for example ‘alright’ is certainly gaining ground on ‘all right’ (though I suppose then it’s not called ‘reform’) or it won’t happen at all. The spelling systems of English are so complex that if you mess with one bit you can lose etymological and lexical links with other words. You could argue for a complete overhaul of the language, but then say goodbye to literature in its original language. No thanks!

    Johanna

  2. I’ll freely admit that vowels cause more trouble for writing than reading. I confess that I do not have much experience related to non-native speakers but I do believe that a large percentage of their problems with vowels are related to prejudices from their previous language(s). More precisely, they are trained to distinguish vowels used in their native languages but English makes different distinctions. I have, in fact, experienced that problem going the other way when attempting to learn French in grade school. Vowels that are definitely different in French sound the same to me which lead to some interesting misspellings, something that would be less likely for a native French speaker who would clealy identify the difference.

    Perhaps I overstated the matter somewhat but given the significant variability in vowel pronunciation across the English speaking world, I simply cannot see any way around the vowel mess no matter what we do. It might be better to suggest concentration on consonant reform rather than vowel reform for this reason alone.

    Regarding foreign loan words, I am certainly advocating respelling any that vary significantly from the usual regular spelling rules. However, doing so wholesale would be less than helpful, especially for words that have become very common. For instance, it is not clear that respelling “tsunami” as “tsoonommee” or some such would be terribly helpful. Still, it seems that many of the older load words have already been respelled since they came before spelling standardization. Certainly any new borrowings should be respelled rather than brought across unmodified; it would, at the very least, lessen the confusion. I would certainly not object to correction of some of the more egregious departures from English phonetics like “gila” or “jalapeno” which contain consonants that do not follow anything resembling standard English practice.

    It is interesting that you bring up etymological links as that is, in my opinion, the primary driving force against respelling loan words. It is not clear to me, however, that there is any particular benefit to maintaining lexical representations of etymological links when doing so introduces confusion for spelling. Indeed, there are counter arguments to such as well, including the fact that many words acquired incorrect etymologies over the years due to coincidental similarities to other words.

    I do fully agree that a wholesale overhaul of English would not be helpful. One of the nice things about Modern English is that it has stayed relatively stable for so long. Any spelling reform must occur in a manner that does not completely render older texts undecipherable and this is the biggest problem most proposals have. The proposals make such sweeping changes that it is nearly impossible to transfer between current spelling and the proposed respelling.

    As you say, if spelling reform is going to happen, it will happen naturally as part of the normal evolution of English writing, driven by those doing the writing, much the same way the spoken language changes over time. And that is very probably for the best.

Leave a Reply

Your email address will not be published. Required fields are marked *