Thanks to Arok for letting me share his recent Facebook rant about the relationship between writing, sound and meaning. Arok writes:
The best tools for learning spoken language (besides face-to-face interaction) involve video and/or audio files so people can actually hear the language. Spoken language equates sound with meaning. That is what people need to learn when speaking a language – the equation of sound and meaning. Anything else (such as writing) can be a distraction. Similarly, sign languages equate visual hand shapes and gestures with meaning. Trying to spell words out in sign language is a “long-cut” which is far less effective than using the meaningful gestures/signs themselves. The same can be said for written languages.
The problem with a writing system that tries to equate symbols to sound before getting to meaning is that it is distracting and takes the long way around. If it is only a tool for sound, it can never be standardized because people (speaking many variable dialects, all of which must be respected) will continue to sound things out every time they see a word (especially when that word is written differently every time anyone writes it), and will never develop actual literacy in the language because no single form will ever come to be recognized for the meaning it represents, only an approximation of sound. The best way to teach sound is with sound, not silent symbols on a page.
In order to help with literacy on the other hand, the best writing system equates complex symbols (consistent spelling of words) with meaning and has far less to do with sound. So when fluent speakers see a word spelled, they should know what it means instantly (and already know how it sounds in their dialect) without having to sound it out. Fluent English readers know what “knee” means without bothering to think about how the ‘k’ sound contributes to the pronunciation of the word only to discover it doesn’t contribute at all and then have to reject it every time. Fluent English readers know what ‘tomato’ means (whether they pronounce it [toMAYto] or [toMAHto], etc.) and don’t need to sound it out every time (or argue about the spelling when we do pronounce it differently). The one spelling works no matter the pronunciation. That’s why the English spelling system (rather than the English symbols) is so powerful – because everyone agrees on that link between spelling and meaning.
In essence, what I am saying is that spelling shouldn’t be used to teach sounds because if it is, then it can’t do the work that spelling is good for. Sounds should be used to teach sounds and these should be learned before spelling. In turn, sign language gestures and consistent written spelling should be used to teach communication in the ways they were designed for. With all the tools at our disposal now, we should be teaching sound with sound, leaving the (consistent) spelling in the background until such time that learners know enough of the spoken language that they can just learn spelling as the secondary means of communication that it is meant to be.