Thursday 30 July 2009

On prosiopesis

A correspondent writes to ask for an explanation of prosiopesis and whether this can be used in relation to writing.

The term was introduced by Otto Jesperson in his Philosophy of Grammar. Here's his definition: 'the speaker begins to articulate, or thinks he begins to articulate, but produces no audible sound (either for want of expiration, or because he does not put his vocal chords in the right position) till one or two syllables after the beginning of what he intended to say'. He gives the example of 'Morning' for 'Good morning'. Another example would be 'Kyu' for 'Thank you'.

This is plainly a phonetic definition, and it could only apply to writing in cases where the same type of communicative pressure applies. I suppose one could adapt the definition as follows: 'the writer begins to type, or thinks he begins to type, but produces no graphic form (either for want of energy, or because he does not put his fimgers in the right position) till one or two syllables after the beginning of what he intended to type'.

I do this often, when typing on screen, but you don't see the results in print because typing allows revision in a way that speech does not. However, in styles of writing which simulate spontaneous speech, I think we can see the same sort of process in operation. Looking back over some instant messaging logs, for example, I can see several examples. My daughter sent me one the other day which began 'Morning'. And any genre of spontaneous written electronic communication (chat, social networking, twitter, etc) is likely to display such things.

The notion becomes more obviously applicable to writing if we replace Jesperson's phonetic definition by one in terms of syntactic or semantic processing. When someone says 'Coming out tonight?' or 'Looks like rain', these days people talk more in syntactic terms, such as 'elision', 'pro-drop', and suchlike.

Monday 27 July 2009

On aren't I

A correspondent writes to ask how aren't I? became acceptable usage. 'For first person questions it is easy enough to say, for instance, "I am right, am I not?" So why would it have developed as alternate usage?'

The history is a bit obscure, but it seems to be this. The verb forms of English have long existed in two styles - formal and informal. Alongside I am going we have informal I'm going. Alongside, formal are you not (earlier are not you) we have aren't you. And so on.

The first person followed this pattern. We find both am I not and amn't I - the latter usage still the colloquial norm today in Irish English and some Scots. But there's a pronunciation problem - the sequence of /m/ and /n/ is awkward, and it was a natural development to simplify the consonant cluster. The final /t/ made it more likely that the simplification would go to /ant/ rather than /amt/, and this is what we find in 18th century texts, where it appears as an't. The OED has an earliest citation for 1799, but I'm sure much earlier instances will turn up in due course.

The pronunciation of the /a/ vowel probably varied in length - sometimes short, sometimes long ('ahnt'). That would have made it sound exactly the same as the other forms in the paradigm (aren't you / we / they) - bearing in mind that the /r/ after the vowel would not have been sounded in the newly emerging Received Pronunciation around 1800. So, if the first person sounded like the other persons, it would have been very natural for people to start spelling the word in the same way as the others. It's an example of orthographic analogy. Aren't I became the standard form in British English, and an't I (very popular in the 1800s) gradually fell out of use. It's widely used in US English too, but some Americans dislike it, finding it genteel.

As soon as aren't I became the norm, it lost its colloquial status. So, if people could say and write aren't I in formal situations, what could they say in informal situations? The stage was set for the emergence of a further alternative: ain't, which originally didn't have the nonstandard resonance that it has today, being widely used as a colloquialism among upper-class as well as lower-class speakers. It was probably the frequent use of this form in the literary representation of lower-class speech (especially in Dickens) that eventually turned educated people against it. Fowler tried to resuscitate it, in his 1926 Dictionary of Modern English Usage, describing ain't as 'a natural contraction... supplying a real want', but his view had no influence.

Saturday 11 July 2009

On being typical(ly)

A correspondent writes from Germany to say that he is often corrected for saying typical German (for typisch Deutsch). He has been advised to say a typical German or typically German, but he feels that his version is all right. What do I think?

The use of the article isn't relevant here, as that depends on whether the noun is countable or not.

That's a typical English tree
That's a typically English tree.

That's typical English weather.
That's typically English weather.

These examples display a very slight difference in meaning: typical means simply 'characteristic of', whereas the adverbial force of typically highlights the way of behaving. (The difference is more marked with some other pairs, such as basic/basically, happy/happily.)

There's no usage issue here. An issue arises only when English (or other such nouns) is made the head of the noun phrase. Normally we wouldn't find two adjectives in predicative position without modification. If we start with That's tasty home-made cake, we wouldn't normally say (in a single intonation contour) That's tasty home-made, but something like That's tasty and home-made.

But this has happened with the type of example which motivated this post, where we find both:

That's typically English.
That's typical English.

Typical has taken on an adverbial role, and this is what makes some people uncomfortable. They like adjectives to stay adjectives, so they object when people say It's looking good, Drive slow, and suchlike. It's part of the prescriptive tradition in English.

But the fact is that both constructions are common. The present-day usage has probably been reinforced by frequency. Constructions such as typical English/German are actually three times as common as those with typically, as a quick Google search will confirm.

So we now have a pair of sentences which mean the same thing. And when this happens, a stylistic difference is bound to emerge. Typical is more informal than typically. I can imagine a curator in an art gallery stopping in front of a picture and saying That's typically Dutch - less likely, That's typical Dutch. But I wouldn't be at all surprised to hear the latter from a group of people chatting about the picture.

We can get a stronger sense of the informality if we change the example. That's typically Rembrandt is the sort of thing one would say about a picture. That's typical Rembrandt might be heard after someone told a story about his naughty behaviour. So I think my German correspondent needs to look to the context before deciding whether to say or write typical German or not.

Tuesday 7 July 2009

On a child's view of English

A colleague writes from Switzerland with a nice child language story. Her German-speaking 6-year-old daughter picked up English very quickly on a visit to New Zealand, and has been keeping it up since. At one point her husband pretended he needed help with his English, so the little girl read a story to him, and he asked her questions about spelling and suchlike, which she tried her best to answer (e.g. why now and know sound so differently, despite being similar in spelling). In the end, she said Weisst Du, Papa, Englisch ist so schön klanglich. ('You know, dad, English is nice and soundly.') The adverb is a nonce formation in both languages.

This story made me think: I don't have many examples to hand of the reactions of young children to the languages they're in the process of learning. Descriptive statements like this one are especially rare. It would be nice to have a few more. Anyone recall any?