A broadcasting correspondent from Canada writes to ask if the use of English vocabulary is shrinking - by which she means the following: 'Perhaps we're using fewer complex words in favour of more, smaller words that can be cobbled together for the same meaning? Technology is certainly pushing us to shorten words and use fewer words, as we try to fit text onto smaller and smaller screens. I also wonder if - in a multicultural society such as Canada - we're using fewer complex words in consideration of non-native English speakers? (I know my friend feels this way with her Spanish boyfriend, for instance).'
I don't know of any survey on this issue, though impressions abound. Vocabulary (by comparison with grammar and pronunciation) remains a hugely neglected topic of linguistic study. Even in child language acquisition, generalizations about vocabulary peter out as the child passes age 2 or so. It's the scale of the exercise which is so off-putting. We're talking about tens of thousands of words.
Because studies haven't been done, there is no benchmark from years ago (or however long my correspondent is thinking) to make a comparison. And any comparison would have to be very careful to match people in terms of - what? social background? age? gender? regional background? occupation? subject-matter? Which of these (and other) variables have most influence on vocabulary size and range? We all have our impressions, but how many of these relate to fact is a quite different matter.
What we do know is that people tend to underestimate vocabulary sizes. Most people think that the average size of a person's vocabulary is a few thousand words, whereas it is tens of thousand. Also, we know that a distinction would have to be made between spoken and written vocabulary. Whatever is happening on screen is going to be very different from what is happening in the air.
You have to compare like with like. I did a quick test. BBC News Online started in 1997. The oldest article I have managed to trace is for 22 October that year: a health article, 'Teenager gets gift of colour from scientists'. This consisted of 261 words and 1225 characters: 4.69 characters per word. I took another health article from BBC News Online yesterday, with a similar content of report and interview: 'Facing a very different Christmas'. That had 3471 characters and 745 words: 4.66 characters per word. Virtually identical.
That's my impression, actually. I've not noticed any particular change in word length in the contexts I've heard and read, over the past decade. Yes, sentences have got shorter (in such settings as instant messaging and texting), but one of the consequences of shorter sentences is that more reliance is placed on the selection of the words they contain, which can actually get longer. The opposite effect can be seen in dictionaries which rely on a restricted defining vocabulary, such as the Longman Dictionary of Contemporay English. If you have to create definitions but have only 2000 words to choose from, your sentences inevitably get longer.
The non-native case is a separate issue. It's well-known that native speakers adapt (accommodate) when talking to non-native speakers (eg avoiding or explaining certain idioms) or speakers of English from other parts of the world than their own. As countries become increasingly multi-ethnic, doubtless more of this kind of thing goes on. But this tells us nothing about how native speakers talk to each other.
I think the bottom line, as far as technology is concerned, is that it hasn't been around long enough for trends to be clear. The new small screens may have an effect, but it's too soon to say. For most people, the exeperience of using the Web, chat, email, and so on, is less than a decade. That's not even an eye-blink in relation to language change.