When the contraction can't be used



Accidentally I've come up with a sentence where the contraction cannot be used, here it is:

*I can't tell you how excited I'm.

Obviously, any other sentence with similar structure (i.e. having a contraction of the verb be at the end of the sentence) falls prey to this as well.

I was wondering if there are other cases where the corresponding contraction (not necessarily of the verb be) cannot be used?


Posted 2015-02-10T08:44:47.673

Reputation: 1 626


One theory holds that you can't contract away a stressed syllable. I'm not sure that's the whole story, but you can find it discussed at “Is there some rule against ending a sentence with the contraction ‘it’s’?”.

– Ben Kovitz – 2015-02-10T11:07:18.110

@BenKovitz The rule is quite subtle. ITbasically dictates that an auxiliary can't have a weak form when it is stranded. Only weak forms can contract with pronouns. It is often wrongly said that the auxiliary will then be stressed. This is sometimes incidentally true, but has nothing to do with stranding. We can see this from examples like Who's coming along? - John is! where is cannot contract but is not stressed. Compare that to Is John coming? - He is Where is can't contract but definitely is stressed! – Araucaria - Not here any more. – 2015-02-10T14:53:45.753

@Araucaria I seem to remember you and StoneyB discussing this in another question. Do you remember which question? (I can't seem to find it.) BTW, nice counterexample, "John is," and nice explanation. The "stress" theory just smelled wrong! – Ben Kovitz – 2015-02-10T14:58:12.983

@BenKovitz No, I've been trying to track it down too! – Araucaria - Not here any more. – 2015-02-10T15:03:34.967


user132181, please wait a day or two before accepting an answer. This question may take some time to figure out, and it might attract many good answers (and edits as people improve their answers). For more about why waiting a day or two is generally a good idea, please see “Not so fast! (When should I accept my answer?)”.

– Ben Kovitz – 2015-02-10T15:23:55.360



When we use an auxiliary verb without any other verbs or complements, we say that that auxiliary verb has been STRANDED. This is usually because we know the listener can userstand what other parts of the verb phrase are missing:

  • Can you open it?
  • Yes, I can! [means "Yes, I can open it"]

Sometimes it is because the complements of the verb have been preposed. This is normally because they are part of a wh- phrase that has moved to the front of the sentence. This is what has happened in the Original Poster's example, where the wh- phrase has moved to the front of the exclamative clause:

  • I can't tell you how excited I am [ this excited.]

When an auxiliary is stranded it must have its strong form. That means it must have a full vowel. Often, but not always, it will be stressed. We can't contract pronouns with auxiliary verbs if the auxiliary has a strong form.

  • Are you listening?
  • *Yes, I'm (X)
  • Yes, I am (grammatical)

Edit note: As noted by Ben Kovitz, the verb have doesn't always seem to have a strong form when stranded after a modal verb: /hi kʊd əv/. I don't know why this is, though. And I've never found anyone who does!

Hope this is helpful!

Araucaria - Not here any more.

Posted 2015-02-10T08:44:47.673

Reputation: 25 536

@BenKovitz I suspect that it's phonetic. "I could've" isn't as different, phonetically, from "I could have" as other contractions. It's pronounced essentially identically to "I could have" when "have" is de-stressed. In other words, we may be writing "I could've" when the speaker may actually be saying "I could have." – phoog – 2016-02-03T16:24:49.490

@BenKovitz Does this ELU answer answer your question adequately? I believe it does, because the mandatory stress slot it is talking about can be filled by "could" here, leaving "have" free to be contracted.

– HeWhoMustBeNamed – 2020-05-07T13:20:00.037

@Araucaria: With reference to you saying that you haven't found an explanation for why have is able to be contracted in "I could've.", do you find anything wrong with the the theory detailed in the answer I linked to in my previous comment, according to which there's nothing odd about "I could've."? Indeed, as far as I can see, though different from the theory you've presented here in only factoring in the surface phonological and syntactic structures as opposed to the deeper structure concepts such as stranding, it's a very good theory. (It also seems to have a different . . . – HeWhoMustBeNamed – 2020-05-07T13:52:38.583

... definition of "stress": as opposed to what you say in your comment above, it would say is in "John is!" is stressed.)

– HeWhoMustBeNamed – 2020-05-07T13:52:55.607

2What verb is missing in the OP's example? – Ben Kovitz – 2015-02-10T15:08:17.660

4How about "I could've" (as a complete sentence)? – Ben Kovitz – 2015-02-10T15:13:36.087

2I think so. Now I'm still wondering what's going on, but apparently it's a mystery even to linguists. – Ben Kovitz – 2015-02-10T15:26:31.413


You never want to end a sentence with a contraction unless it's a contracted form of not. Otherwise it sounds like the sentence is unfinished because contractions are often used to shorten auxillary verbs - or it some cases it could sound/look like you are using possessive case.

It is what it's (wrong)

It is what it isn't. (ok)

He told me that he wasn't. (ok)

He told me that he's. (wrong)

Did you talk to him? Yeah, I told him I won't. (ok)

You did this? No, I didn't. (ok)

Who's the winner today? You're. (wrong)

Have you talked to him? I haven't (ok).

You're lying - have you talked to him? I've. (wrong)

Would you do that? I'd. (wrong)

Come on, really, would you do that? You're right, I wouldn't (ok).


Posted 2015-02-10T08:44:47.673

Reputation: 31 841