what am i doing again

One thread I'm trying to follow up at the moment is the overlap between spoken language and written language. People who are literate have a different perspective on spoken language compared to, eg, people in cultures where literacy doesn't feature. One significant characteristic of people who are literate is that they conceptualise spoken language as if it was written - you hear speech as consisting of words and sentences and letters. But this isn't actually a very realistic idea of spoken language, as you can tell by listening to someone speaking a foreign language - it all flows out in a continuous stream, and you can't tell where one word ends and the next begins. You can also tell by recording speech and graphing, say, loudness/intensity (in decibels) against time - the gaps in the waveform won't necessarily correspond to the boundaries between words or sentences.

One of the major issues in linguistics, particularly developmental linguistics, is how people manage to end up perceiving the continuous stream of spoken language as if it consisted of units - like sentences, words, and smaller segments like syllables and phonemes. And one of the major factors, it turns out, is people's process of learning to read and write. That's what makes people so good at knowing the difference between a "t" and a "p", even though they're acoustically pretty similar. (It's also what makes it so hard for people to believe that there's a "t" in pencil and a "k" in taxi.)

So given the influence that written language has on people's (conceptualisation of) spoken language, what can you make of the argument that people who have difficulty in learning to read and spell must have problems in their spoken language skills? That's the predominant explanation for the difficulties experienced by the roughly 5% of the population who have dyslexia - their literacy difficulties must be caused by a linguistic impairment. It's backed up by countless studies which unanimously report that people with dyslexia aren't as good as non-dyslexics at counting the number of sounds in a word (eg three sounds in t-o-p; four sounds in s-t-o-p), or judging whether light rhymes with might. This is called the phonological deficit in dyslexia.

The difficulty with those kinds of study is that you could equally argue that it's only if people are good at spelling that they'd think of top as having three sounds in the first place (rather than, say, five, as in t-h-o-p-h, or two, as in t-op). So if someone gives you an unexpected answer to the question of how many sounds are in a word, you can't tell whether they got it "wrong" because of a difficulty processing spoken language, or just because they're not too competent at using alphabetic letters.

In order to investigate whether the phonological deficit in dyslexia is really phonological (really comes from a spoken-language deficit) or whether it's a re-statement in different words of what we knew already (that people with dyslexia have difficulty with reading and writing), we basically need to look at areas of spoken language that don't have a counterpart in written language. And this, dear chaps, is where the toy factories come in. But I've gone on too long already, so will have to break here for the night.

What a cliff hanger eh!