Tuesday, 24 May 2016

The ones I could see

I'm reading The City and the City by China Mi√©ville at the moment. I'm only about 60 pages in but I already really like it, partly for its clever ideas which are currently just emerging, but also for its languages. Mi√©ville has a linguistics background and this shows in the plot of some books (Embassytown, for example) and the writing in all of them: he's creative and clever and does really unusual things with words. Not that you need to have a linguistics background to be able to do that, but it seems like he really knows what he's doing with language. I don't know, I know nothing about literature, so I can only waffle in an uninformed way on that point. But his sentence structure is so clever sometimes, and in this book the characters are from somewhere in central Europe, so the names are all made up but completely believable, and things like that. I think there was even some zeugma at one point and I really love zeugma. 

[Very slight spoilers follow, maybe? Not sure, I haven't read enough of it myself to know.] 

One thing that made me stop and admire it today was a clever use of could. There are parts of the city that the main character can't see. The sentence that caught my eye was describing some railway arches. He says (my bolding):
Not all of them were foreign at their bases. The ones I could see contained little shops and squats decorated in art graffiti.
At this point in the story, the reader doesn't know what the concept is here. Is it that he was able to see the arches, or that he was permitted to see them? This latter idea is what other events have hinted towards, but so far it could be either. 

This is because can and could (present and past tense respectively) can mean two different things. The first meaning, where he was able to see the arches, illustrates logical modality, or the possibility of him seeing them. The second, where he was permitted to see them, illustrates deontic modality: what is possible within the rules. This is an absolutely lovely use of this ambiguity which allows him not to reveal too much of this conceptual device this early in the story. I'm pretty sure he knew exactly what he was doing when he wrote it that way. 

This ambiguity is also, by the way, why you get smartarses responding to a question like 'Can I go to the toilet?' with 'I don't know, can you?'. 

Sunday, 22 May 2016

Gretchen McCulloch on 'Getting linguistics out of the ivory tower'

Gretchen McCulloch, of All Things Linguistic fame, is very excitingly in the UK for a bit and managed to fit in a talk at our place. I livetweeted it and have storified the tweets. The hashtag search wouldn't work for some reason, so I located a couple more but will have missed some tweets - sorry.

Tuesday, 10 May 2016

Grammar! It's the bestest!

Schoolchildren in this country are currently mired in SATs. These are tests designed to assess how well a school (not individual children) is doing, but have been being roundly criticised because they're stupid and pointless. Michael Rosen in particular has been very vocal on Twitter about the SPaG (spelling, punctuation and grammar) one, which I believe is happening today.

It does seem a stupid test, and many of the examples he has picked out do contain confusing or apparently pointless things. The main criticism many people seem to have is that young children don't need to know this stuff, that it's too hard, and that it doesn't take creativity or expression into account, and that grammar puts kids off language. This facebook post is (I think) an analogy which makes exactly that point - that focussing too much on the mechanics ruins the fun of it. Lynne Murphy has written a good post pointing out that learning grammar is good and useful and helps you to know more about language. I would go further, though, and make the case that grammar is absolutely bloody brilliant.

Spelling and punctuation are a bit dull. The types of grammar that kids are having to learn is a bit dull: they're essentially labelling parts of speech. But real grammar, the kind that I spend all day every day thinking about, the grammar that I chose to study for 8 years and then make my career, is fantastically and endlessly interesting.

How can you not be fascinated by the fact that words might not really exist, that adjectives occur in a particular order (small green apples, not green small apples), that the rules of be deletion in AAVE are precisely the same as the rules of be contraction in British English, that you can have a cheeky Nando's but not a cheeky salad, that speakers can innovate constructions like because noun and the exact same damn thing happens in Finnish and French and who knows what other languages, that through reanalysis and tiny shifts Latin became the romance languages, that languages all over the world are wonderfully diverse but equally astonishingly similar... hell, even the most basic fact about syntax, that it is a hierarchically-structured system, is still amazing to me and something that most people don't even realise.

And even more than this, most of what I've just said is controversial to some people. We don't know the answers. Language, and I take grammar to be central to language (some would disagree but I'm happy to be biased), is inextricably bound up with our humanity and we don't even know how it works. We are still finding out. Isn't that exciting?

Don't teach 8-year-olds about subordinating conjunctions. It probably will put them off writing stories for fun. And don't ban 'slang'. That'll make them scared to speak. Once again, I make my call for all teachers to study linguistics and then teach everyone grammar - but fun grammar. Learn a foreign language and see how it's like English, or different from English, and wonder why. Look at Beowulf and Chaucer and marvel at how far English has come, and what happened to that verb-second word order. And then come and study linguistics.

Friday, 6 May 2016

Acknowledgement text tokens

The other day, I was having a conversation with someone over social media. Bear with me, because I'm about to stray into conversation analysis and this is so far out of my area it might as well be cell biology.

Conversations are prototypically turn-taking exercises, with A saying something, B responding, A responding to that, and so on. But obviously this isn't always (or even usually) the way - we interrupt each other, talk over each other, don't finish our sentences, and so on. Another thing we often do is talk for an extended length of time because we have a long story to tell or whatever. When we do this, the other person typically nods, makes encouraging noises, smiles, and generally lets the speaker know they're still listening. But in a text conversation you can't do this. If you try and do it, the little symbol that shows you're typing can put off your interlocutor because they think you're saying something meaningful, and they might stop telling their story. But if you don't, might they think you've gone off somewhere or fallen asleep or lost interest? We need a button that just sends an 'mh-hmh' symbol to show we are still there and paying attention. These are called 'acknowledgement tokens', and we need a text version.

Friday, 29 April 2016

Antidisestablishmentarianism is a very long word

There's a really great post at Merriam-Webster about why the word antidisestablishmentarianism isn't in the dictionary. Basically, it's because it isn't a word in common usage. This raises interesting questions about the nature of lexicography, what 'in common usage' means, meta-linguistic mention vs. use, and compositionality of meaning.

The word clearly does have a meaning. Merriam-Webster say it's this:
opposition to depriving a legally established state church of its status
I thought it was something like the movement against the separation of church and state, but maybe that's the same thing - I'm not at all clear about what that actually means. But the point is that the meaning arises directly as a sum of its parts: it's compositional. Well, this isn't strictly true: there is some idiomatic meaning to do with the church and the law as well. But the length of it comes from attaching affixes with strictly compositional meanings. When you have compositionality, you can theoretically create longer and longer words, up to the limit of your cognitive capacity. I could add a morpheme and create antidisestablishmentarianismation, for instance. Some long words in quite common use aren't in the dictionary simply because they're made by standard suffixation processes, and that suffix is in the dictionary so you can work out the meaning for yourself.

So antidisestablishmentariansim is a real word, in the sense that people recognise it and can understand its meaning and it's made with proper word-formation processes. But for M-W, it's not a word in common usage. They have only three quotations for this word which use the meaning they give above, and dictionaries rely on written uses of words. This is why they can seem slow: words only get in when they've achieved plenty of use in print materials.

They do have plenty of quotations of the word with reference to it being a long word, however. It is in the OED, and their quotations refer to this:
1984   T. Augarde Oxf. Guide Word Games xxvi. 216   The longest words that most people know are antidisestablishmentarianism..and supercalifragilisticexpialidocious.
This is metalinguistic mention of the word, not an actual use of the word. We're stepping outside language and talking about the word itself, not using the word to say something. This is sort of difficult to get your head round because we have to use language to talk about language, a long-recognised philosophical problem in linguistics.

Interestingly, M-W say at the end of their piece that it might be considered a viable entry 'simply because it's a well-known word'. The meaning is not well-known, however, so it would be an entry whose definition read something like 'famous long word', and the definition as a secondary bit of information. That would be quite cool and very meta.

Irrelevant postscript: There's a really stupid joke that goes like this: 'Antidisestablishmentarianism is a very long word. How do you spell it?' and the answer is, of course, 'I, T'. I sometimes try that out on my students and some of them groan but a lot of them simply don't understand the joke. Maybe it's the way I tell 'em. 

Monday, 18 April 2016

You and your family's best interests

There's been a leaflet sent round lately about the EU referendum happening in the UK. The government has sent this leaflet to all households, setting out the case for remaining in the European Union. Here's how it puts it:
The Government believes it is in you and your family's best interests that the UK remains in the European Union.
A friend mentioned on facebook that he was disappointed that despite the £9m spent on this leaflet, it contained this grammatical error. I begged to differ: this is stylistic variation, not a mistake.

He argued that it should be your and your family's best interests, as both conjuncts should be possessive. This is right, as you should be able to leave either out and it still be grammatical.

But there's two complicating factors here. The first is that the possessive your is kind of already a combination of you plus 's, so perhaps the 's is redundant. I don't actually think this is the case, because I think there's another reason why it's OK to say you. 

It's to do with the nature of 's. This is what we call a clitic, which means that it's phonologically dependent (has to attach to) another word, but is not as tightly linked to the word as a suffix like the plural s. While the plural suffix can only attach to countable nouns, the possessive can attach to a much wider range of things. The only requirement is that the noun it refers to be within the phrase it attaches to. This means that we can have phrases like the following, where the possessive attaches to something other than the possessor, and sometimes not even a noun:
The woman with the long hair's dog
The guy I was talking to's friend
The girl dressed in blue's mother
It's a little more complicated with coordination, of course, as we have to have the 's referring to both conjuncts. But I think that's OK. Language Log have discussed this before, and examples like this are all right:
I and my friend's work (a bit clumsy in my opinion, but not bad)
Me and my friend's work (perfectly well-formed)
So, unusually, I'm with the government on this one, both in terms of their grammar in in staying in the EU.

Friday, 8 April 2016

(At) home

It's a well-known fact that home has no preposition when it occurs with go:
I went (*to) home
(The asterisk inside the brackets means that it's ungrammatical if you include to.)

And it has to have one if it's an adverbial phrase (optional extra information about the event):
I worked *(at) home today
(The asterisk outside the brackets means it's ungrammatical without at.)

There are some verbs where the preposition is optional, such as stay:
I stayed (at) home.
I think there might be some regional variation on that one, though I'm not sure.

But when it's with be, omitting or including the preposition gives a meaning difference. I ran a twitter poll to make sure I wasn't alone in this, and found overwhelming agreement with my judgements. In a context in which I've been for a night out and want to tell my friend that I've arrived back at my house safely, I would say I'm home. If my friend had rung my and wanted to know where I was, I would say I'm at home. I could use either in either context, but both I and those who responded to my twitter poll felt that the distinction above was right. So that preposition at being pronounced has a kind of locative meaning - location in a place - while omitting it has some sort of directional meaning - movement to (or arrival at) a place.