Linguistics on the Bus

After my recent blogiversarial lamentations that I’ve strayed away from my original intention of blogging about linguistics and languages and focusing more on politics and indigenous affairs, I feel re-inspired to write about linguistic curios a little more. So here’s a purely matter-of-fact, non-partisan linguistic post.

On the bus this morning, I noticed a sign on the back of the driver’s little compartment, that read something along the lines of:

Drinking alcohol is prohibited on this bus
Possession of an opened container
of alcohol is prohibited on this bus

I’ll draw attention to the second point. Note that they don’t merely use an unmarked adjective ‘open’, but a perfect participle construction, a de-verbal adjective ’opened’. It is of the generalised form verb-en.

The intended meaning is clear: any container that has ever been opened (after being initially sealed, that is), as opposed to merely being temporarily ‘open’ and thus closable, is prohibited. Sydney Buses are clearly precluding any protests from drunken juvenile delinquents claiming that their container of alcohol is not in fact ‘open’, after they’ve quickly re-lidded their bottle of spumante.

‘Open’ is one of those verbs that is so telic that it doesn’t even permit its own inverse: *unopen. This is a purely linguistic constraint though, since it’s perfectly pragmatically acceptable to perform the inverse action of ‘open’, namely, to ’close’.

I don’t know if there’s a linguisticky term for these verbs, but I’m sure some of my more knowledgeable readers will know at least one.

As a test, these verbs often occur in the construction un-verb-en, but never un-verb. That is, it is possible to describe the state of something that hasn’t undergone the change of state; you can describe something as ’unopened’, but it is (linguistically) impossible to say you will unopen something (in English at least). However, these verbs often have counterparts; verbs that encode the inverse action, even if the original verb cannot be inversed.

For instance break: it is entirely appropriate to say you broke my arm and the doctor mended/fixed my arm but it is impossible to say *the doctor unbroke my arm. Moreover, break will happily allow unbroken as in unbroken lines (on a road).

Doesn’t everyone think about these things on the bus?

9 Responses to “ Linguistics on the Bus ”

Comments:

lauredhel says:
October 4th, 2007 at 8:06 pm

Interestingly enough, break also allows unbroken in reference to bones – “the doctor reassured her that her leg was unbroken”. But not, as you say, after the event: eight weeks after a fracture, her leg doesn’t become unbroken.

And you can use unbreak – but only to say that it’s an impossible thing to do! – as in the truism “Once you break it, you can’t unbreak it.”

Yes, you’re not the only one who contemplates these things. Now I’m trying to think of other verbs in this category. “Choose”? “Take”? “Show”? “Write”? “Split”? “Infect” (if the object is animate)?

john says:
October 5th, 2007 at 4:56 am

An unopened container is one that has not yet been opened. But to unopen a container, I would close a container that had already been opened. At least that’s what “unopen” suggests to me.

Jangari says:
October 5th, 2007 at 7:59 am

Or, that’s what it would mean if ‘unopen’ were not an impossible verb. It’s meaningful all the same, but you’d never hear it, unless, as Laurdhel points out, its use is some literary device. The only example I can think of in which I’d expect to hear it is something like:
I broke my wookie.
Then unbreak it.

john says:
October 6th, 2007 at 8:36 am

Yes, I mean that’s what it would mean if it wasn’t an impossible verb. I think it’s interesting that what it would mean is the opposite to what its participle form does mean. I don’t know what that means.

Jangari says:
October 6th, 2007 at 4:30 pm

Actually, I would argue that they’re consistent.

I think of the construction un-verb-en as functioning to promote the initial state of a change-of-state verb and getting rid of the target state. So a verb open represents the change from the initial state of being [closed]¹ to the target state of being [open]. unopened is a derived adjective that excludes anything other than the initial state.

This is contrasted with another construction, verb-en, like opened or broken, which does the opposite; it excludes anything apart from the target state (inasmuch as event structure is concerned).

If the regular suffix un- functions to reverse the event (a crude analysis, I’ll concede), then, and as you rightly point out, the change-of-state becomes [open] going towards [closed] when open is changed to unopen.

So now, and here’s the beef, analyse the verb as unopen and affix the regular adjective derivational suffix, that is, apply the verb-en construction, creating unopened, and you have promoted the target state of the (otherwise impossible) change-of-state verb unopen, [open], and removed the initial state [closed].

That’s all a bit awkward an analysis, but I think it shows that the hypothesised meaning of an impossible verb unopen is logically consistent with the attested meaning of unopened.

¹I’m using the convention of italics inside square brackets to represent states that aren’t analysed or deconstructed any further.

John says:
October 12th, 2007 at 7:58 am

My theoretical thinking is a bit rusty, but I think I understand you.

You’re saying that the initial state of unopen would be [closed]? un- would reverse the action but the initial and target states would remain the same.

So add -en to unopen and you exclude everything but the initial state. So unopened means [closed].

John says:
October 12th, 2007 at 8:38 am

Ok, my previous comment is a mess. But I do understand you.

Jangari says:
October 12th, 2007 at 8:47 am

Sorry, I had this mixed up:

…and you have promoted the target state of the (otherwise impossible) change-of-state verb unopen, [open], and removed the initial state [closed].

The states [open] and [closed] should be around the other way. The target state of unopen (which eventually get’s highlighted by the -en suffix) would be [closed] and the initial state (which gets omitted) would be [open].