This week: The Problem with English Spelling

English, it is said, possesses the worst spelling system the world has ever known. 80% of the time you can't tell how to pronounce a word just from its spelling. When learning a new word, you have to know how to spell it and how to pronounce it, and ne'er the twain shall meet.

Why is it that one letter can have 8 different sounds, and one sound can be spelled 10 different ways? The answer is down to the fact that English is such a shabbily structured language, with no completely dependable rules. The vocabulary of English has been built up from so many other world languages that it is near impossible to spell things consistently.

Many attempts have been made to standardise the spelling of the language, with such methods as new alphabets or consistent vowel transcriptions. Most have not been readily accepted, some completely shunned. It seems that we native English-speakers are happy with our mish-mosh higgelty-piggelty hodge-podge spelling system, and do not want to (or could not be bothered) changing.

Spelling reforms are common to many European languages, such as French and German. These countries usually have some sort of governmental regulatory body in charge of keeping the language pure. The French especially so. English, being such a widespread and geographically mutated language, has been met with much opposition along the rocky road to consistency.

In this article, I will cover a selection of the spelling reform suggestions made by linguists world-wide.

The first section to be covered is the many new alphabets that have been suggested. In these new alphabets (or more appropriately orthographies), each symbol represents one phoneme and nothing more. They include Shavian, Unifon, Deseret and Pitman Shorthand. I will cover each in detail.

The other systems are based on re-spelling words to conform to a regular pattern, i.e. making only one combination of letters make represent one phoneme. These systems would continue to use the standard Latin alphabet which we use today.

The downside of all the new alphabets mentioned hereafter, is that they rely on pronunciation, which does differ (often quite considerably) from country to country due to accent and dialect. Even, especially in the USA and UK, from state to state or city to city. A Received Pronunciation speaker from London will not pronounce words the same as a Yorkshireman, nor a New Yorker the same as a Texan. Consider also a German as compared to a Japanese person. For any new orthographical changes to be made to English, a standard and correct pronunciation would have to be agreed upon, and that is definitely not going to happen easily at all. The alphabets discussed below are based mainly upon New Yorker pronunciation, making them inappropriate for a worldwide audience.

To save space, the scripts are on separate pages.
For a full outline of the script, click each link.

The Shavian Alphabet was created in 1958 by competition winner Kingsley Read, after linguist George Bernard Shaw left £500 prize money in his will for a new alphabet to be created for the English language. Shaw was a very vocal critic of the use of the inappropriate Latin alphabet for English. Although never actually having seen the alphabet created in his name, Shaw provided the impetus for its creation.

Shavian bears no resemblance to Latin, which on one hand is useful to avoid confusion, but on the other allows no correlation between existing symbol recognition or mnemonic hints.

See below the Shavian alphabet. The words given as pronunciation guides are also used as the names for the letters, as prescribed by Shaw. There are 48 separate phonemes* represented in Shavian, including separate rhotacised vowels.

Having no capital letters, Shavian utilises the so-called namer-dot before the first letter of a word to indicate proper nouns or names. Shavian has received limited support, but is one of the more common alphabets of its kind. It is highly unlikely that Shavian will become a standard script for English.

Unifon was created by John Malone in 1959. Unifon means one sound (i.e. uni-phone), therefore, it uses a system of 40 symbols to represent the 40 English phonemes* which were identified by the creator. Many of these symbols are copied directly from Latin, representing the standard sound made by that letter. Other symbols are derived from similar letters.

Unifon does not contain any lower-case letters, nor make any such distinction as Shavian's namer-dot does.

Despite the creator's best efforts, he has managed to leave out a few vital phonemes. Whatever happened to the long a in ask or the schwa in about, or the sound in hurt? These phonemes do not exist as we know them in American English pronunciation. Therefore they were left out. This makes it inappropriate for anything other than American English phonetics. Unifon is probably the least accepted new alphabet of its kind.

* In fact, there are estimated to be more than 40 phonemes in English. As seen above, Shavian utilises 48 (most of these extras being rhotacised vowels). The best estimate is approximately 43 phonemes, depending on a speaker's accent, whereby some similar phonemes may be indistiguishable or new ones utilised.