Saturday 18 October 2008

On instruments

A correspondent writes to ask: 'What is the difference between The window was broken with a brick and The window was broken by a brick.

Well, not much, from the window's point of view, or the house-owner's. And semantically, in this example, most people would hardly notice a difference. But they are grammatically different. What's happening is that two usually distinct constructions are overlapping because they are following a passive verb.

Grammars typically use such terminology as 'means' or 'instrument' for the first, and 'agent' for the second. Note that they answer different questions. The instrumental sentence focuses on the means used:

'What was the window broken with? or 'How was the window broken'. Answer can be: With a brick, or - so as not to repeat the with - A brick.

The agent sentence focuses on who or what performed the action:

'Who broke the window? or, in this case, 'What broke the window?' Answer cannot be With a brick. It has to be A brick - or, of course A brick broke the window.

With and without are the primary markers of instrumentality. But an instrumental meaning is often expressed by a by-phrase, and that's where the overlap with an agentive meaning comes in. You can check for instrumentality by seeing if you can substitute by means of, using, or some such phrase.

They communicated by signs.
They communicated by means of signs.
They communicated using signs.
They communicated with signs.

Note that you can't do this the other way round: an agentive meaning can't be expressed by a with phrase.

They were driven to town by a bus.
They were driven to town by a farmer.
*They were driven to town with a bus.
*They were driven to town with a farmer.

The only interpretation you could have for the latter is: 'along with a bus/farmer'.

There's a useful discussion of these constructions in The Cambridge Grammar of the English Language, pp. 673ff.

On correspondents

A correspondent writes to ask 'Who are these correspondents who always introduce your posts?'

Well, like this one, I have no idea, usually. The cult of anonymity is so strong on the Internet that you can't tell from the e-name, and rarely do people explain who they are. Occasionally they do. Their email might begin 'I am a teacher of English in such-a-place'. But usually I am in the dark. Does it matter? Not really. It would help sometimes to know what the mother-tongue is of correspondents, or where they live, as sometimes their linguistic background is relevant to the linguistic point they raise. And sometimes knowing the level of expertise behind a question would help me understand where a correspondent is, as they say, 'coming from'.

But blog answers are not like email answers. In fact, as we all know, they are called 'posts' - analogous to 'posters' - intended for a general readership. So I cut out all personal details in my responses and try to generalize the point at issue. If people want to personalize their comments to a post, that's fine - as long as they remain polite. No flaming is allowed on this site!

My correspondent asks where people are from. That I can establish, by going to Google Analytics. This tells me that somewhere between 150 and 200 people come to this blog every day, and that (to take the last month as an example) they/you are from 94 countries. The average number of pages visited is 1.51 and the average time spent on the site is 1 min.21 secs. Top four countries of origin are UK, USA, Germany, and Australia.

My correspondent also asks whether I answer all questions that come in. No, life is too short. I can only respond when I'm at home, and that isn't as often as I would like, hence the occasional gaps in posting. Also, some people send in questions which are simply too long to be answered - they are mini-essays. I admire the work that has gone into them, but to answer them would be tantamount to writing a journal article, and that's not what blogs are for. In some cases, I have a sneaking suspicion that the question is from a school or college assignation, where the questioners are hoping to get the work done for them. There's nothing wrong with quoting from this blog, of course, but the quote should always be attributed. And finally, some people have unrealistic expectations: they want an answer yesterday, and get angry if I (or sometimes my office, when I'm away) send an acknowledgement explaining that it just isn't possible right now. They don't get an answer at all!

Sunday 12 October 2008

On ending inconclusive(ly)

A correspondent writes to ask about the sentence The meeting ended inconclusive, which he says he has seen in newspaper report headlines. He also points to Christoph Loreck's book Endymion and the Labyrinthian Path to Eminence in Art, which contains the sentence The foreword ends inconclusive. But 'should this be inconclusively?', he asks.

Standard English usage after end is adverbial, so it would normally be inconclusively. I've seen the adjective used in news reports too - a Google search suggests that these are mainly from South Asian publications, so it might be a feature of Indian English. Replacing an adverb by an adjective is also a common interference pattern from some other languages. A German-speaking character in the film The Third Man advises Holly Martins to 'go careful' in Vienna. Perhaps this is the source of Loreck's preference, for he is German. And the vacillation turns up in some English dialects as well, for all kinds of reasons, and these will gradually influence informal standard English. For instance, ease of articulation motivates they lived happy ever after instead of the standard they lived happily ever after.

Why has the adjectival usage arisen at all? One possibility is ellipsis: the speaker is thinking of the sentence as short for something like 'The meeting ended with everyone inconclusive'. But I think it more likely that the verb is being influenced by the very similar verb end up, which readily takes an adjective: We ended up happy. The similarity in meaning can be seen from such examples as The day started sad and ended sad and I didn't know whether to be happy because the series ended happy or to be sad because the series ended. It's all very interesting, albeit somewhat inconclusive.

Tuesday 7 October 2008

On using

A school correspondent writes to point to a change in usage. In my Discover Grammar (p. 37) I list some sentences where someone has used transitive and intransitive verbs wrongly. One of the examples is I'm using now. My correspondent's students point out that this is perfectly OK, these days, in the context of drug taking. And she asks: 'if the object is implied but ellipted, as in this case, is the verb actually still transitive?'

Transitivity is a shifting thing, indeed. That's why, in this book and elsewhere, I always talk about transitive and intransitive 'uses' of verbs. A verb has a potential to be either, and it never surprises me to see a previously transitive verb being used in an intransitive way, or vice versa. In fact, intransitive uses of use can be found from the Middle Ages, and the drug-using meaning is first recorded as long ago as 1953. The OED gives an example from William S Burroughs' novel, Junkie: 'it is practically impossible to stop using'.

I wouldn't call I'm using an ellipted transitive, because the meaning is different from a transitive use. It means 'taking drugs'. I'm using something has one of the various meanings of use, such as 'employ', 'make use of', and so on. This is typical of verbs which have both uses. Compare the contrast between the other examples on page 37 of Discover Grammar, such as The neighbours have moved their car and The neighbours have moved.

Another point to note is that ellipsis isn't a useful notion here because it's not clear what is actually being left out. Ellipsis, in the grammatical approach I follow, is explicitly identifiable from the context. For example, if I say I like apples and someone comments So do I, it's reasonable to talk about ellipsis because the 'full' form of the sentence is plainly So do I like apples. But when someone says I'm using, in isolation, just like that, it's impossible to say exactly what's been omitted.