8

Can anyone give a cogent, simply described explanation of why the verb BE in:

  • What is it?

... doesn't seem to be able to be contracted with the subject:

  • What's it? *

Compare the sentences above with:

  • What's this?
  • What's that?

These are perfectly fine. In fact, the contractions here should be expected in almost all examples of spoken English.

Answers with references to authoritative vetted sources would be greatly appreciated.

Araucaria - Him
  • 47,146
  • 1
    Though the question is entirely different, I would say that this answer of mine also adequately answers this question—assuming you accept that it in this function is inherently unstressed and cannot populate the mandatory predicate stress slot (unlike this, and also it in some other cases, like “That’s it!” or even “What’s ‘it’?”). – Janus Bahs Jacquet Sep 09 '14 at 23:39
  • @JanusBahsJacquet Well that's what I thought (kind of) on first consideration. Then I realised that What is the object here, therefore part of the predicate. When first considering What is it? I decided that the is is stranded because its complement has been fronted, so therefore is must be a strong form. However, consideration of What's that? seems to shoot that to pieces. The only difference I can perceive is that that, being deictic, is stressed, whereas it will usually be de-accented, because it's virtually always old information. Don't know if this would matter... – Araucaria - Him Sep 09 '14 at 23:49
  • @JanusBahsJacquet Very, very nice post btw! – Araucaria - Him Sep 09 '14 at 23:49
  • 1
    I don't think the rules that govern sentential stress slots really care whether what is subject or predicate on a deeper syntactic level—there's simply a ‘subject slot’ and a ‘predicate slot’, and the cutoff point between them is the verb (which is included in the latter). – Janus Bahs Jacquet Sep 09 '14 at 23:53
  • @JanusBahsJacquet (Must remember which comment I'm on!)... I don't quite understand why according to that answer is wouldn't need to be strong in What's that? but would need to be in What is it?. I'm not sure, but it also seems to me that if we have what as part of the predicate, and obligatorily stressed, then is shouldn't need to be stressed according to the theory as described? – Araucaria - Him Sep 10 '14 at 00:12
  • I would say that what fills the ‘subject slot’ (SS) and is it/that the ‘predicate slot’ (PS). If we accept that this usage of it is mandatorily unstressed, then the stress in the PS must fall on the verb, ’cause there's nowhere else for it to go. That, on the other hand, is mandatorily stressed and can carry the PS stress, which means the verb doesn't have to and becomes a candidate for contraction. – Janus Bahs Jacquet Sep 10 '14 at 00:18
  • @JanusBahsJacquet Well, this is what got me onto asking the question in the first place. That's what I originally proposed to my fellow grammar anaorak - colleague. Problem is that isn't in the predicate!! – Araucaria - Him Sep 10 '14 at 00:21
  • @JanusBahsJacquet ... because that is the subject and what is the complement ... (which is why I had to backtrack and then put this question up on here because my head hurt ...) – Araucaria - Him Sep 10 '14 at 00:24
  • Like I said higher up, and the reason I'm using quotes around the terms, I don't see these slots as being equivalent to syntactical entities. They're prosodic entities and don't necessarily have to correspond to syntactical entities. The default sentence type is SVO (=SP), and that's what the slots are based on. In cases like these, where you basically have OVS (or CVS), the stress slots still pivot around the verb in the same way. Wackernagel trumps syntax when stressing sentences, I think. – Janus Bahs Jacquet Sep 10 '14 at 00:26
  • 1
    Note that taking the verb as the pivot (and perhaps speaking of a pre-verbal slot PVS and a verb-complex slot VCS instead) also accounts for cases with fronted complements or objects: “Boring he is, but clever he's not”—the PVS here includes both the subject complement and the subject, whereas the VCS simply starts at the verb and covers everything that comes after, so the first sentence can't be contracted, but the second can. – Janus Bahs Jacquet Sep 10 '14 at 00:37
  • 2
    @JanusBahsJacquet Do the same test trying to contract a three syllable sentence beginning with Who is into Who’s, and you find something interesting. Normally Who’s it? is blocked, although Who’s it for? is not because the sentence now ends with a stressed word. However, if you are playing “Tag! You’re It!” then suddenly it’s allowed, since it has become a noun more than a pronoun, and so becomes stressed and valid. You can say “Who’s it?” in that context. Because possessive pronouns are stressed but but not possessive determiners, some resist ending the sentence with the p.p. its. – tchrist Sep 10 '14 at 00:51
  • @JanusBahsJacquet See also this question about the resistance that some native speakers have against using its at the end a sentence even when this *its* is a stressed possessive pronoun like mine instead of as the unstressed possessive determiner like my. In sentence-final position, even normally unstressed purely function words like is, has, are, have, can, with, in, and gain stress there — and when they do, now you are allowed to contract words like is or has that came right before them, when normally you could not. – tchrist Sep 10 '14 at 01:02
  • 1
    It's not contracted because it's final; final be doesn't contract: *I said that it isn't but he said that it's. This and that are stressed after is, but it normally isn't, so that makes it sound weird uncontracted, too. – John Lawler Sep 10 '14 at 02:11
  • @JohnLawler I think there might be something in your comment, but there seems to be a hole in it, to me at least. You can strand BE in other situations but have following stressed words, but this doesn't seem to allow the cliticisation of BE. For example: A: "Is it open?" B:"I think it is sometimes". Now that answer from B can't be "I think it's sometimes". – Araucaria - Him Oct 03 '15 at 23:20
  • It's right before a deletion site, though. Open has been deleted by conjunction reduction, and I think the same rule applies -- probly I shoulda stated it as forbidding contraction before a deletion site, but it's fairly far out of my area. – John Lawler Oct 04 '15 at 03:26
  • 2
    In Araucaria's example "I think it is sometimes", the "sometimes" is a sentence modifier that is not part of the verb phrase "it is". The last thing in the verb phrase has to be stressed. On the other hand, the "sometimes" is in the verb phrase when it is positioned between the verb and an element of the verb phrase, as in "I think it is sometimes open". That's why the former "is" cannot be contracted, but the latter one can be. – Greg Lee Oct 04 '15 at 16:34
  • To answer, the reason why you think "it doesn't seem to be able to be contracted" seems pertinent. My guess is that your answer will be something along the line of "it doesn't sound right", in which case your question posits that a matter of opinion can be resolved by a matter of fact...which may be true, but if so, the mechanism whereby it is true is unknowable. – JEL Oct 05 '15 at 22:18
  • @tchrist, Re your first comment, in the last sentence did you mean "resist ending the sentence with the p.d. its"? Since you said possessive pronouns are stressed but possessive determiners aren't, shouldn't ending a sentence with the p.d. its be found more unacceptable? – HeWhoMustBeNamed Mar 03 '20 at 09:07

3 Answers3

16

(1) The word "it" doesn't like to be stressed. (2) Normally, a sentence has its strongest stress on the last thing that can be stressed, which in a simple subject-verb-object sentence will the object, since that is the last thing.

Principles (1) and (2) interact to give the strongest stress on the verb of a sentence, in case the object is "it" -- since the stress can't go on the "it", the last eligible thing for stress is the verb. Compare "I like yoghurt" with "I like it".

(3) Stressed vowels cannot be deleted.

Putting together (1-3), we deduce that the "is" in "What is it?" will be stressed, and consequently cannot be contracted to *"What's it?", because that would require deleting the "i" of "is", which must be stressed because of the following "it".

Greg Lee
  • 17,406
  • 1
    How do you account for the contractability of "Who's it?" (commonly encountered, for example, in the game of tag)? – JEL Oct 05 '15 at 23:21
  • 2
    The rule that "it" is not stressed concerns the pronoun "it". The "it" in the game of "it" is not a pronoun, so it gets stress, and consequently preceding "is" can be contracted. – Greg Lee Oct 05 '15 at 23:28
  • So, asking "Who's it?" about someone at the door doesn't work. Reminds me of a knock-knock joke, but I take your point. The OP is asking for "authoritative vetted sources". Do you have any such that, particularly, make the distinction between between pronominal 'it' and nominal 'it'? I ask because your answer specifies "the word", not the pronoun. – JEL Oct 05 '15 at 23:41
  • My account of the stress is a simplification of the treatment in SPE, where the nuclear stress rule formalizes the tendency in English for the main stress of a phrase to come at the end. Also, Chomsky and Halle note that pronouns don't take stress. I don't think they draw the obvious implication about contraction, but of course you can't stress a vowel that isn't there. (I was thinking of myself as a vetted source.) – Greg Lee Oct 05 '15 at 23:53
  • I accept your self-vetting, but I suspect the OP is thinking more along the lines of peer-reviewed--despite the strong evidence, if not proof, that being 'peer-reviewed' is no more likely an indicator of underlying expertise than, for example, the acquisition of 'reputation' here by reason of longevity, prolificity, and popularity. – JEL Oct 06 '15 at 04:28
  • @JEL "How do you account for the contractability of "Who's it?"" <---- That's a no brainer. The word it there is a proper noun, not a pronoun. – Araucaria - Him Oct 06 '15 at 08:40
  • @Araucaria, I guess...not when we played, and not like Cousin It, but okay, different linguistic cultures I suppose. The name of the game was "Tag" for us. Maybe I misunderstood at the time. Anyway, Mr. Lee said quite clearly "the word it", not "the pronoun it"; but again, now that it's been made clear, I guess such imprecision is acceptable in this domain (where duplicates are not duplicates, obsolete is not obsolete, etc.), however questionable it may've been to start. That leaves the "doesn't seem to be able to be contracted" of your question a matter of opinion. Shibboleth? – JEL Oct 06 '15 at 09:14
  • 2
    @JEL, Araucaria, When I wrote my answer, it just didn't occur to me that I should formulate it with "the pronoun 'it'". I noticed after JEL brought it up. – Greg Lee Oct 06 '15 at 11:26
  • My answer referred to the word "it". "It" is not "this". So why shouldn't "this" be stressed? In fact, it is stressed. "This" is not even a pronoun. You seem to be imagining generalizations or alternatives to my answer, then criticizing those. So "is" is not a content word -- what does that have to do with what I said? What is this "stranding"? – Greg Lee Oct 06 '15 at 12:35
  • @GregLee I'm not criticising your answer! I'm just still ruminating over the difference between that sentence and the other two examples in the question - which is where my original confusion, interest in the question stems from. So my purpose of speaking to you here is just to pick your brains and converse with you. Stranding just means that the complement of the auxiliary has been deleted or moved to a different site. Here, it seems to me, the complement of BE is probably the word What. The only thing that I don't understand in relation to your answer is that if some particular .... – Araucaria - Him Oct 06 '15 at 13:34
  • @GregLee ... word becomes destressed we'd normally expect the nucleus to move to the previous content word, wouldn't we? So in your like example, I'd expect like to take the nucleus because it's a content words and it, as you say doesn't take stress, and in any case represents old information in the conversation. Maybe it would be better if I gave a genuine example that seems to buck your explanation (I'll delete most of the comments above and then these ones after you've read them. (PS, I won't get pinged if you don't address any comments to me!) – Araucaria - Him Oct 06 '15 at 13:46
  • @GregLee So how about the following conversation (all thats and thises are incidental in this btw): A: "That's not the one we've been looking for, it's that one with the blue top." B: "Oh, that's it! I thought it was the other one!" C: "What's it? I can't see anything". In that conversation, it is also the last word, but BE doesn't seem to become stressed. And B's comment would work, imo, if they said "Oh, that is it! ..." with the BE uncontracted. Those examples don't surprise me because I wouldn't expect BE to take stress normally, because it's a function word ... – Araucaria - Him Oct 06 '15 at 13:53
  • @GregLee And then there's C's response too. It seems to me that is isn't stranded in C's answer which is why it can be weak - and that stranding *in conjunction with your 1-3* might explain why is is stressed in the OQ. But I don't know I'm just bandying around an idea. I hadn't got anywhere till I'd read your answer! – Araucaria - Him Oct 06 '15 at 13:59
  • @GregLee *But* ... if the fact that the Complement's been fronted means that is must be stressed, then the other two examples given for comparison in my question seem to pose a problem ... namely why is is cliticizable there? [and the reason I'm asking you under your post is because I suspect you're the only person who might be able to help! - not because I'm trying to criticise your excellent answer] – Araucaria - Him Oct 06 '15 at 14:12
  • On your example conversation with A, B, C sentences, I agree that the last "it" would be stressed. I attribute that to its metalinguistic sense -- it's part of a comment about the language that was used in the earlier part of the conversation, as opposed to the ideas expressed. The convention is to use scare quotes here. What is this "it" that you refer to? – Greg Lee Oct 06 '15 at 17:12
  • Broadly speaking, in describing the stress, you are appealing to two ideas that are entirely absent from the SPE analysis, and from mine. One is the notion that stress depends on syntactic derivation (this was part of Joan Bresnan's analysis), and the other is that it has something to do with information structure -- old/new, topic/comment, or whatever. It's a good thing to pursue other ideas, but you ought to know when you're headed off in a different direction. In the SPE analysis, the stress contour depends only on the surface phonological and syntactic structure. – Greg Lee Oct 06 '15 at 17:22
  • 1
    @GregLee I guess a lot of that's also because I was trained by phoneticians and not phonologists. I'm more from a Wells background, less from a generative one ... – Araucaria - Him Oct 07 '15 at 23:31
  • 1
    Phoneticians taught you to appeal to derivations? I didn't think phoneticians even knew about derivations, or if they did, certainly didn't believe in them. – Greg Lee Oct 08 '15 at 00:37
  • Isn't saying "Who's it" wrong unless it's followed by something else, such as "Who's it concerning"? I've never heard anyone say "Who's it?" in place of "Who is it?" – SteJ Oct 09 '15 at 15:03
  • Do you know of any good references to learn more about this topic? I'd like to know more about it, but I'm very new to this. – HeWhoMustBeNamed Jan 31 '20 at 16:59
  • @MrReality, Other than what I said in previous comments, I don't know of other good references. – Greg Lee Jan 31 '20 at 20:03
  • @GregLee, is the rule that "Normally, a sentence has its strongest stress on the last thing that can be stressed", true if the object has an article? And if it is a proper noun? ... – HeWhoMustBeNamed Mar 02 '20 at 10:31
  • ... Also, re your reply to JEL, can you elaborate on why the "it" in "Who's it" in the game of tag isn't a pronoun? Does it lose any properties of a pronoun and/or gain some properties of a noun? – HeWhoMustBeNamed Mar 02 '20 at 10:32
  • @MrReality Because it refers to a role in a game rather than a preceding noun phrase in the conversation. – Greg Lee Mar 02 '20 at 11:55
  • @GregLee, what about the comment previous to that? Do you know whether the stress rule you mentioned normally also applies to the objects having an article, or that are proper nouns? – HeWhoMustBeNamed Mar 02 '20 at 12:19
  • @MrReality Yes, it does, with the qualification that later mentions of the same phrase have low stress (as do pronouns), – Greg Lee Mar 02 '20 at 12:29
9

It's a perfectly grammatical contraction, though it doesn't commonly stand alone; it's usually followed by another word.

  • What's it all about, Alfie?
  • What's It? - The Award Winning Game Where Creative Minds Think Alike!
  • Sun, sea and silver service: what’s it like crewing on a superyacht?
  • What's It? - Information Today
  • What's It Like on the Pope's Plane?
  • What's it to you, anyway?
  • *What's it do?"

The problem with your example is that in the short, stand alone phrase "What is it?" (just as with Who is it?), the emphasis is on is, not what. If there is no emphasis on is, then the phrase is simply What? If the emphasis is on it, then something for the dummy-it must be stated, as in the cases above, or, What's it like outside? (Who's it gonna be?)

edited to add: Please see @John Lawler's comment.

anongoodnurse
  • 55,278
  • It's certainly helpful! :) One of my problems is that the necessary strong - ie not-contracted - form of is is the exceptional bit. As you show, when there are following complements or adjuncts, BE, of course, is always contractible. I'm not sure about the Information Today title, not because it's not a proper title, but because I'm not sure that What's it can be used in real speech as a question in so-called "standard English". Also why obligatory emphasis on is in what is it?, but not in What's that? – Araucaria - Him Sep 10 '14 at 00:03
  • 4
    I think this answers something that is slightly, but crucially, different from the question. When you add more elements to the sentence, you change the basis for stress assignment and thereby also contraction, so though there are many instances of “what’s it […]”, those all end up being incomparable to the ‘simplex’ sentence “*What’s it?”. Now the game show and the Information Today titles, those are very interesting, because they actually do use something that isn't valid as a natural sentence. – Janus Bahs Jacquet Sep 10 '14 at 00:06
  • +1 for your previous answer ( no less for this one). But I don't think I follow with the theory that if there's no emphasis on is the phrase is simply What? If someone gives you a strange object why do you need to stress is in What is it?, but you don't in What's that?. The reasons for stressing is in both examples seem to be the same. What? on it's own doesn't seem to be an adequate response in this situation - imo. What do you think? – Araucaria - Him Sep 10 '14 at 00:08
  • 1
    It also reminds me that nouns derived from similar phrases seemingly always have contracted forms: whatsit, whatchamacallit, etc. Most of them have additional elements, so contractions aren't that odd, but whatsit doesn't. You'd really expect that to be called a whaddisit or something. – Janus Bahs Jacquet Sep 10 '14 at 00:09
  • 2
    How quickly we forget the childhood game of Tag! Tag, you’re it! Who’s it? I’m it! – tchrist Sep 10 '14 at 01:06
  • @tchrist Very true. As you say above, it's more like Who's It? in that example, so there does seem to be something requiring a stressed word after the contraction even if it's not a complement ... – Araucaria - Him Sep 10 '14 at 09:37
  • @Araucaria - Agree with the above; tchrist nailed the perfect example. You asked a very interesting question. Haven't been overwhelmed with those recently. It's been an interesting discussion! – anongoodnurse Sep 10 '14 at 09:44
  • 1
    What's it do? is not ungrammatical; it's the contraction for What does it do? The -'s (really a /z/, but devoiced to /s/ after /t/ from What) can stand for is or does (though not was, the same way -'d can stand for could or did); What's he do? is short for What does he do? – John Lawler Oct 05 '15 at 19:34
  • @JohnLawler - please forgive me, that answer was written a long time ago. You're correct, of course. I'll edit. – anongoodnurse Oct 05 '15 at 20:31
3

What is it?
What's this?
What's that?

Why are people so averse to providing context? English lives and breathes through context.

The proper way to approach this is to ask how the expressions are used, not to examine them as isolated phrases that can be poked at and dissected without any reference to real-life usage.

Example

(a) Jane points at an object lying on the table. She says, "What's that?"

The word 'that' is strongly stressed. We understand that something new has come to Jane's attention. The something is that. The verb 'is' cannot take a position of stress because that is already occupied.

(b) Jane points at an object lying on the table. She says

"What is that?"

This time, the word 'is' is strongly stressed. Jane is talking about something that is already under scrutiny by those present but now she is saying, "I see it but now I want to know what it is" Clearly we can't abbreviate 'is' in this case.

Now all we have to do is apply the same reasoning to 'it'.

(c) Someone hands an object to Jane, she enquires, "What is it?" This is similar to case (b). There is a known object and Jane wants some information about it.

(d) Now we come to the 'impossible' case. There is an object.

Jane says, "What's it?"

This should be like case (a). That is to say, the object is new to Jane. However a vital property of the word 'it' is that it must have an antecedent. Therefore Jane must already have mentioned it. So it can't be like case (a). This case doesn't happen because there can never be a context that warrants it.

Conclusion

You can only understand the reason by considering the contexts in which the expressions can be used. You cannot do it by simply reasoning about the isolated phrase—English requires context.

Unlike for 'this' and 'that', there is simply no context that allows emphasis on the word 'it'.

Community
  • 1
chasly - supports Monica
  • 37,215
  • 2
    It is not actually a "vital property" of the word "it" that it have an antecedent. For instance, the "it" at the beginning of the sentence preceding this one. For another example, take "It's raining". – Greg Lee Oct 09 '15 at 03:24
  • @GregLee - That's a very good observation. In fact, it blows the whole thing apart. I'll amend my answer. – chasly - supports Monica Oct 09 '15 at 07:36
  • Jane's "What's it?" is a nominal (or noumenal, you choose) use of 'it' with reference to the 'it' in John's "It's raining". Try this: "At the other end, the phone rang three times, and then a woman's voice said "Yeah, what's it?". – JEL Oct 09 '15 at 08:02
  • @JEL - Good points. With regard to the phone scenario - there's a deletion. The woman is effectively saying, "You rang me about something. What is it?" – chasly - supports Monica Oct 09 '15 at 08:04
  • @JEL - You're right about the nominal. I'm going to have to revert to my older answer. The impersonal 'it' in expressions like 'It's raining', is being used in a different way to the normal 'it' that requires an antecedent. Suppose you'd never seen rain before: You would not point at rain and say "What's it?" because the idiom in English is to point and say, "What's that?" If we didn't have demonstrative pronouns in English then maybe we would say, "What's it?" – chasly - supports Monica Oct 09 '15 at 08:15
  • Having taken note of comments, I think I need to do a rewrite. I'm still happy with my answer but I need to sharpen it up. In a while I'll delete this version and submit a clearer version. – chasly - supports Monica Oct 09 '15 at 08:39
  • A: It shows that dogs are more intelligent that cats? –  Feb 01 '21 at 02:17
  • B: Does it now? What’s it? –  Feb 01 '21 at 02:18
  • So is "Where's that?" correct? – Bob Nov 04 '22 at 01:15