On making language accessible.

I realise that most days I am not exactly a poster child for the joys of speech pathology, but if there’s one area of my job I really enjoy, it’s Augmentative and Alternative Communication (AAC).

AAC is all about making language accessible in order to help a person communicate better. When speech isn't an option - for example, if a person has a physical disability that makes their speech unclear; or difficulty with putting their thoughts together, or with language itself - and keeping in mind that spoken language is already a symbolic representation of ideas - then AAC solutions often look to further represent symbolically an already symbolic system. This can involve using symbols for input, or output, or both.

That sounds complicated, but the idea is to make it easier, right?

So for example if I can't think of the Welsh word for 'cat' or 'milk' or 'Brain Injury Rehabilitation Trust', I can look these up in a dictionary - I'm using a symbolic representation of the word as input to jog my memory and help me speak.

Or as another example if I go to a really loud concert and lose my voice and can only use text messages to get my point across, I'm using a symbolic representation - again, writing - as my output.

Symbols make things easier.

Kind of.

Essentially the problem for AAC users is that the twenty-six letters of the alphabet is a pretty excellent way of representing every word in the English language, ever, but if you can’t spell (and many of my clients are functionally illiterate), then writing is just not a great way of symbolising language. It doesn't make it easier. It doesn't make it accessible.


The Alphabet. Brilliant for writing, not so great if you're illiterate. (source: www.tyf.com)

This is where the great logistical nightmare and emotional joyride (I’m not joking) of developing a pictorial system for language begins.

You could represent each word with a picture, but who has room to store and capacity to learn a million symbols for the English language?

Okay then, a million is a bit over the top; following the 80/20 (or is the 95/5?) rule of language, you could encode 3000 English words into symbols and be able to express most things you want to say. That’s a much more manageable number. In a physical book of 20 pictures per page, that’s about 75 pages, double sided.

If you categorise your pictures, then you speed up the process of trying to find the word you want (as long as you’re capable of using categories). But you have to be thoughtful with this approach. Take a word like “apple”- do you categorise it under ‘food’, or ‘things for home’, or ‘things for school lunch time’ or ‘types of operating system’? Potentially your categories are going to blow out to more than the number of words you started with.

There are more efficient systems, though. Minspeak, for example, doesn’t use scrapbook like categories for words. Instead, it allows each symbol to stand for multiple ideas, and then the particular sequence of symbols determines the meaning produced.

This isn’t an exact example from a device that uses Minspeak but it gives you the idea:

The apple symbol can mean lots of things. When you combine apple with the symbol for noun, you get “apple” ; when you combine it with the symbol for verb you get “eat”; and when you combine it with the symbol for adjective you get “hungry”.

Three words starting from one symbol.

Add a bed symbol and you get:

• bed + noun = “bed”
• bed + verb = “sleep”
• bed + adjective = “tired”

An extra three words and you’ve only added one symbol.

Another example of symbol combination, in Minspeak. The paintbrush symbolises "adjective", I think.

A multi-meaning iconic system like Minspeak reduces the number of symbols you need to generate the largest possible set of ideas. It’s pretty brilliant. I’m in love with it, can you tell?

Unfortunately Minspeak can have – or be perceived to have – quite a steep learning curve, and I wouldn’t often use it with my clients who have particularly limited cognition. It’s much easier when a picture only symbolises one thing, right?

The question then is, if my client has capacity to use only a limited number of words (it may be cognitive capacity to understand; it may be physical capacity to access a device without fatiguing; whatever) – what words are going to be the most important, the most useful for them to say?

No really. If you could only use 100 words to communicate, what would they be? If you could only use 50? 20? 10?

This is a bugger of a question because often I find the most important words for a client are also the ones they’re already pretty good at communicating. Yes, no. Coffee, tea, food. Toilet. Go away. Unless someone has absolutely no way of talking, making sounds, or signing, these high-frequency, high-context words are surprisingly easy to understand.

So how do you organise the billowing mass of low-frequency, low-context words that a client needs to say but has doesn’t have capacity to sort through? To be honest, I’m not sure yet; a combination of a few approaches, I suspect, and something that I keep working on.

so close, so far

After having spent four and a half years at university (bachelor of science with honours in linguistics thank you very much) telling anyone who would listen and anyone else as well that I would not, could not, was not ever going to be a speech pathologist - I found myself a few months after graduation, applying to do a masters of speech pathology.

There was a good reason for this change of heart. I'd been working with a little autistic kid who was having a bit of difficulty with language and in the structure of the programme that I was given to follow, I was basically teaching him English - first pronouns, then nouns, then present tense verbs, until one very exciting day when I looked at the updated programme and saw that we were going to be starting on past tense verbs.

I don't think the kid cared much to be honest but I thought it was probably the best thing ever. It meant that he was going to be able to tell me not just what he was doing in that moment, but what he had done every moment before that. Like this whole world of conversation - heck, this whole world full stop - was going to open up for us. Very exciting.

So then I thought, hey, yeah, I want to do this for a proper living. Not really knowing what speech pathology was - yes, I decried an entire profession without knowing very much about it and now you should know to expect that sort of thing from me - I decided that being a speech pathologist was probably the best way for me to get to do this language teaching thing, like, all the time. For a living and stuff.

The thing about speech pathology is - well, it's a lot of things. It's one of the most broad ranging and varied and misunderstood jobs I know (doesn't everyone think they're the most misunderstood?) - but it really is at least broad ranging, from stuttering and accent reduction to articulation to language to voice to swallowing to literacy to sensory stuff and as far beyond that as people will stretch your job description.

But the one thing it kind of really isn't, is language teaching.

I can help a kid acquire language. I can help an adult regain their language. But what I was doing with the autistic kid, sitting with him and giving him words and showing him how to use them and stuff - teaching language - I just don't really get to do that as a speechie.

And maybe there's something in that. I mean, there is. Language educators could take a couple of pages from the old speechie guide book and instead of just teaching languages in this sort of provisional manner, they could be supporting people to work out how to get language themselves. There's a lot to be said for that approach and I'll put it on my list of things to blog about.

In the meantime, though, at least for me in my situation, it feels like I'm so close but so far from doing that thing that got me going with that little kid. So here we go, then; this is what the blog is all about, with a few meanderings along the way. Me, our loveable protagonist, getting back - no, going forward - to language education.