Choice of contractions in English

[alice]

In sequences of three words where the middle word can contract with either of the other two, is it known if there are any factors which govern which contraction is actually used? As an example, consider "you aren't" versus "you're not"; the first seems more natural to me, but the second isn't particularly unnatural.

[Salmoneus]

In sequences of three words where the middle word can contract with either of the other two, is it known if there are any factors which govern which contraction is actually used? As an example, consider "you aren't" versus "you're not"; the first seems more natural to me, but the second isn't particularly unnatural.

Dialect and context.

Different dialects obviously prefer different contraction patterns. But also, the context, and the desire to place emphasis, has a role. So, I'd usually say "you're not listening" - but if someone said "I AM listening", obviously I'd reply "you AREn't listening".

[Linguoboy]

Whereas when the negated sentence has imperative force, "not" seems more natural:

"I AM going to the party!"
"You're NOT going to the party and that's final!"

This has got to be a very well-studied area. Have you checked Pullum and Huddleston?

[alice]

This has got to be a very well-studied area. Have you checked Pullum and Huddleston?

What is that?

[Travis B.]

In my dialect it is normal to use -n't whenever possible except for with am, must, shall, or ought, unless negation is being specifically emphasized or a higher register is being used.

[Kuchigakatai]

This has got to be a very well-studied area. Have you checked Pullum and Huddleston?

What is that?

He's talking about The Cambridge Grammar of the English Language (2002) by Rodney Huddleston and Geoffrey Pullum (of Language Log fame) (eds.). It's an English grammar 1800+ pages long, a wonder of a book.

[alice]

This has got to be a very well-studied area. Have you checked Pullum and Huddleston?

What is that?

He's talking about The Cambridge Grammar of the English Language (2002) by Rodney Huddleston and Geoffrey Pullum (of Language Log fame) (eds.). It's an English grammar 1800+ pages long, a wonder of a book.

It costs at least £200. What does he think?

I know Pullum more as the co-author of the Phonetic Symbol Guide.

[Linguoboy]

It costs at least £200. What does he think?

Sorry, I was remembering a time before the latest round of Tory authority when Britain still had public libraries.

If I have a chance, I'll check our library's copy later today.

[din]

According to Carter and McCarthy's 2006 Cambridge grammar of English:

With the verb be [...], there is a choice of contracted negative form in the present tense [...].

In informal spoken contexts, when the subject is a pronoun, the preference is overwhelmingly for the forms with 's not/'re not. When the subject is a lexical noun phrase, the choice is more open, but with a strong preference for the isn't/aren't forms:

So they would argue that:
[*] She's not coming
Is more common than:
[*] She isn't coming

While on the other hand:
[*] This bolt isn't big enough
Is more common than:
[*] This bolt's not big enough

[Travis B.]

That actually sounds right for me at least, with personal pronouns receiving -'m/-'s/-'re not but nouns and other pronouns taking isn't/aren't by default in everyday speech.

[zompist]

Doing some quick Google searches, I think they're right on the pronouns, wrong on the nouns.

E.g. "you aren't" - 55m
"you're not" - 325m

Other pronouns are similar, though it's not as stark with he/she.

But: "not good" - 82m
"isn't good" - 19m

And similarly with some other common adjectives.

[akam chinjir]

What nouns did you test? I got 130K for "my cat isn't" but just 6.3K for "my cat's not." (I figured the internets talk a lot about cats.)

[Dē Graut Bʉr]

His post implies that he didn't test any nouns at all, but instead tested some common adjectives, presumably under the assumption that a phrase like "isn't good" would always be preceded by a noun, which obviously isn't true.

[zompist]

Yeah, I was doing that in about 5 minutes. Taking another 5, with some common nouns, X's not is less common than X isn't, as the website says. But X is not is far more common (like, by two orders of magnitude) than either.

[Frislander]

I think I tend towards X's not rather than X isn't, which I think is typical of the far north of England iirc (similarly "don't" is "divven't" in Geordie and nearby varieties).

Also related, but with contraction of "will" it seems I contract it more contexts than many people, because I keep getting red underlines when I write things like "I wonder if anyone'll respond" or "it seems no-one'll do it" or even "this man'll do it", which are absolutely perfectly grammatical for me and people in my area.

[Linguoboy]

Also related, but with contraction of "will" it seems I contract it more contexts than many people, because I keep getting red underlines when I write things like "I wonder if anyone'll respond" or "it seems no-one'll do it" or even "this man'll do it", which are absolutely perfectly grammatical for me and people in my area.

Y'all'll fit right in around here.

[Travis B.]

Also related, but with contraction of "will" it seems I contract it more contexts than many people, because I keep getting red underlines when I write things like "I wonder if anyone'll respond" or "it seems no-one'll do it" or even "this man'll do it", which are absolutely perfectly grammatical for me and people in my area.

For me, and from what it seems, most people here, will will contract to -'ll in almost all cases except when emphasized or when negated.

[mèþru]

Contracting will to 'll is way more common in speech than in writing.