6

I noticed an oddity in the sentence Why don't you just do it?: Although I always thought of don't simply as of a short form of do not it seems to me as if this is not the case in this sentence. Instead, don't appears as a particle of its own, i.e. it cannot be deconstructed any more. The sentence *Why do not you just do it? sounds ungrammatical to me, but Why don't you just do it? seems fine. (I am not a native speaker.)

I have three questions:

  1. Is "don't" a particle of its own?
  2. Is there a name for this grammatical phenomenon?
  3. Are there other cases besides negated questions where don't cannot be deconstructed?
  • 3
    Don't let's worry about this too much. – Peter Shor Nov 22 '23 at 19:23
  • 1
  • I think (don't know) that *don't* has been *gramaticalized.* Full Oxford English Dictionary: To express by means of the grammatical structure; to adopt as a grammatical requirement. Usually in passive. Again, I think this can be contrasted with *lexicalize: (1) To accept into the lexicon, or vocabulary, of a language,* (2) To express (a difference that is already expressed in the grammatical structure, or could be) by means of a different lexical item. – FumbleFingers Nov 22 '23 at 19:53
  • You might ask a followup here: https://linguistics.stackexchange.com/questions/19006/analyzing-negation-with-a-syntactic-tree – TimR Nov 22 '23 at 20:28
  • Why don't you just do it?=Why do you not just do it. "don't you" is negative interrogative. – Lambie Nov 23 '23 at 15:14
  • 3
    Arguably, "don't" / "doesn't" is a negative auxiliary verb. That's not the way it's usually analyzed in English, but it works pretty much exactly like the negative verbs in Uralic languages like Finnish, Estonian and Sami. – Ilmari Karonen Nov 23 '23 at 19:32
  • 1
    Just what do you mean by a “particle”? Do you mean like the to in English infinitives, or zu in German ones? Or do you mean a clitic or another sort of bound morpheme? Are you perhaps thinking of how some German verbs have “separable particles” (trennbare Vorsilben) whose leading bit (its “prefix”, as aus‑ in ausbleiben, durch‑ in durchfahren, and her‑ in hertragen) ᴄᴀɴ and under specific circumstances ᴍᴜꜱᴛ detach to independently move elsewhither syntactically? – tchrist Nov 23 '23 at 19:40
  • Once something is analyzed as a word, it takes on a life of its own, separate from the components it was formed from. It's the same reason why the «"PIN number" is wrong» argument is wrong. – hobbs Nov 23 '23 at 19:55
  • @IlmariKaronen Indeed! From PIE both Latin and OE inherited a free negative lexeme or particle ne which could also be used proclitically to derive negated verbs but of which few vestiges now remain in today’s English or Romance. Ancient examples include: L. nolo/nescio meaning don’t want/know, < ne‑ + positives volo/scio for want/know; OE+ME nit < ne wit meaning to not know, to know not; or as in will he or nill he where nill < ne‑ + will. – tchrist Nov 23 '23 at 20:08
  • @tchrist I realize I probably used the term particle in a wrong way. What I want to get across with the term is that it is something "atomic", i.e. it cannot be deconstructed into smaller parts. Reading don't as a contraction of do not would render it as a complex, composed, non-atomic entity. I think you are right that in assuming that my use of the term draws inspiration from the separable particles in German. – Jonathan Scholbach Nov 23 '23 at 20:09
  • 2
    Yours was not a “wrong way” sensu stricto so much as one I worried readers might lack the background or context to understand. See the supercited Wikipedia article and especially here where they write: “The negative marker ‑n’t as in couldn’t etc. is typically considered a clitic that developed from the lexical item not. Linguists Arnold Zwicky and Geoffrey Pullum argue, however, that the form has the properties of an affix rather than a syntactically independent clitic.” – tchrist Nov 23 '23 at 20:22
  • 3
    Why do not you just do it? is not idiomatic. However, Why do you not just do it? might be uncommon, but to this native speaker, it is 100% idiomatic. It allows emphasis that using don't doesn't. Why do you not just do it? Why do you not just do it? – CJ Dennis Nov 23 '23 at 22:17
  • Is this simply about someone having persuaded you that 'don't' isn't just as easily used as a contraction of '… do not…' as of '…do you not…'?

    Either way, what made you think 'don't' could have any meaning 'of its own', as a particle or anything else? Are you suggesting 'don't' could be something other than a contraction

    – Robbie Goodwin Nov 24 '23 at 22:39
  • @RobbieGoodwin Well, I don't think don't is a contraction in the example sentence, as it cannot be re-expanded any more. I don't understand the rest of your question. – Jonathan Scholbach Nov 24 '23 at 23:11

1 Answers1

11

Questions like Why do you play chess? display subject auxiliary inversion; the auxiliary verb do appears before the subject you.

In a normal declarative clause, the adverb not occurs after the first auxiliary, before the next verb in the chain of verbs (or before any modifiers of that following verb phrase):

  1. You do not [play chess].
  2. You do not [just play chess].

If we have subject auxiliary inversion, the adverb not stays in this same position:

  1. Do you not [play chess]?

However, if—and only if—the adverb not is contracted with the auxiliary, in this case the verb do, it gets pulled along with it to the pre-subject position:

  1. Don't you [play chess]?

This is exactly what is happening is the Original Poster's example:

  1. Why do you not [just do it]?
  2. Why don't you [just do it].
  3. *Why do not you [just do it]? (ungrammatical)

Above we see that (7) is ungrammatical because the negative adverb not is not in its normal position before the following verb phrase. It cannot move from this position because it is not contracted with the auxiliary verb.

Some writers, for example The Cambridge Grammar of the English Language (Huddleston & Pullum, 2002), argue that don't and other negative 'contractions' are not actually contractions at all, but negative inflections of the auxiliary verb. They cite evidence such as the fact that won't is nothing like an amalgam of the words will and not, as well as the phenomenon noticed by the Original Poster in their question. Notice that although with do and don't it looks from the spelling as if we have just tacked n't onto the verb do, the vowel in don't is completely different from the vowel in do, and thus the situation is very similar to the one with will and won't.

Other environments where don't cannot deconstruct:

Unlike the situation with negative questions, where it could be argued that don't can deconstruct, but not in situ—the not must appear in a different position, not next to the auxiliary—there is one case where don't does not readily deconstruct at all. It is also a case where some people have argued that don't is actually a marker and not a verb. This is the case of negative imperatives with subjects. And this time, the phenomenon really is to do with the item do, and not to do with auxiliaries in general.

Imperatives often occur without a subject:

  1. Dance!

If we want to negate the imperative we can use either the item don't or the two words do not:

  1. Don't dance!
  2. Do not dance!

So far, so good. Now, imperatives can also come with subjects. These are usually either the second person singular pronoun you or a word like somebody, anybody, nobody and so forth:

  1. You dance!
  2. Everybody dance!

[ These are pronounced very differently from the case where we have a vocative expression like Bertha/you/everybody, where we would expect a separate prosodic unit (a separate tune) for the vocative:

  1. Bertha, dance!
  2. You, dance!
  3. Everybody! Dance!]

Now, if we want to negate an imperative clause that has an expressed subject, we need to put the word don't before the subject:

  1. Don't you dance!
  2. Don't everyone talk at once.

However, here, there doesn't seem to be any good way at all of decomposing the don't into two separate words in these cases, especially with a second person subject:

  1. *Do not you dance! (ungrammatical)
  2. *Do you not dance! (ungrammatical imperative - would be ok as question)
  • Ha! Try to repeat "wiln't" ten times and we'll see if it won't become anything like "won't" – Stef Nov 23 '23 at 12:19
  • 1
    @Stef Well, define “anything like”. To me, repeating willn’t produces something like [wɪɫʷnt] or perhaps [wɪwnt] if the /l/ is debuccalised altogether. Is that ‘anything like’ won’t? Sure. Is it the same vowel as the /oʊ/ in won’t? No. Don’t forget that the vowel quality in won’t comes more from the formerly common present form woll than from will. – Janus Bahs Jacquet Nov 23 '23 at 12:55
  • I don't know how to say "You dance!" and "Everbody dance!" except when "you" and "everybody" are separate prosodic units, so I'm not sure what you mean there. – TimR Nov 23 '23 at 15:00
  • "Sheltie, step forward. Shiba Inu, step forward. Wolfhound, step forward". They're all being addressed individually and told what to do. – TimR Nov 23 '23 at 15:10
  • @TimR Yes, and that's why they're vocatives and also why the negation of those imperatives has the don't afetr and not before the vocative phrase. – Araucaria - Him Nov 23 '23 at 15:12
  • What about "Do not dance, you!", or "Do not dance, everybody!"? That's how I would probably say it if I felt the need to include the subject in the imperative when using "do not". (And that can also be contracted to don't and still sound correct.) – GentlePurpleRain Nov 23 '23 at 20:26
  • @GentlePurpleRain Good question. Those words at the end there everybody or you are vocatives. They tell you who the sentence is addressed to. The best clue there is the preceding comma, which tells us that those words/phrases have their own intonational phrase. You can use those with an imperative with its own subject. For example "You dance, you" or "Don't you dance, you!" – Araucaria - Him Nov 23 '23 at 20:30
  • @GentlePurpleRain Zwicky and Pullum’s point about English’s ᴍᴀɴʏ auxiliaries negated via ‑n’t, including both modal and non-modal ones alike (won’t, don’t/didn’t, isn’t/aren’t/weren’t, hasn’t/hadn’t, can’t, shan’t/shouldn’t, oughtn’t, daren’t, etc) is that these behave syntactically far more like integral/first-order lexemes derived by affixation than they do ones with a syntactically independent enclitic you’d be able to freely break up and move independently, including by interposing other syntactic elements between the deconstructed pieces—as you ᴄᴀɴ ᴅᴏ in the Queen of England’s hat. – tchrist Nov 23 '23 at 21:01
  • 1
    I’m left pondering whether the negative imperative Don’t you dare dance! can ᴇᴠᴇʀ have a corresponding positive version and if so what that is—and if not, then why it cannot do so. – tchrist Nov 23 '23 at 21:11
  • @tchrist I doubt it, but some writers believe that pseudo-conditional constructions like (You do/)Do that again, (and) you're fired contain an imperative form even though the speaker doesn't want the listener to 'do that again'. I'm not sure about that though. There seems to be a clear contrast between You dare dance and I'm leaving which has no command-like force and where You dare xyz is possible and ? You dare dance or* I'm leaving, where the first conjunct needs to be a real imperative, but seems ungrammatical or at least infelicitous. Compare: Apologise, or I'm leaving*. – Araucaria - Him Nov 24 '23 at 09:47
  • Are those examples at the end of this answer actually imperatives or are they some kind of subjunctive? – Carsten S Nov 24 '23 at 10:15
  • @CarstenS Which examples? (16) and (17)? – Araucaria - Him Nov 24 '23 at 10:28
  • @tchrist Sorry forgot to explain why I doubt it, which is basically because dare there seems to be modal-like in that it's taking a bare infinitival construction as complement. And modal dare's a negative polarity item. [However, a modal with do-support is squifffy. The data for dare is all over the place]. – Araucaria - Him Nov 24 '23 at 11:08
  • 1
    @Araucaria-Him, I was thinking foremost of (12) and then maybe (11). I of course realise that such a distinction is rather pointless in English, because those forms are jdentical to the infinitive anyway. – Carsten S Nov 24 '23 at 12:27
  • 1
    @CarstenS That's right. We can't really talk about infinitival, subjunctive and imperative forms of the verb (well, one can, but it's very difficult to justify), but we can talk about subjunctive constructions and imperative constructions and so forth. One of the main ways imperatives with subjects are traditionally differentiated from subjunctives is by negation. Subjunctives take not, whereas imperatives require don't/do not, in other words they require do-support. So that would seem to confirm (11-12) as imperatives, if we take that line of thinking. – Araucaria - Him Nov 24 '23 at 12:52
  • 1
    @Araucaria-Him, thank you for the explanation, that’s interesting. – Carsten S Nov 24 '23 at 14:45