What is the grammar (or syntax, if you will) of constructions of the form of the below sentence?

He has committed I don't know how many crimes.

In this sentence, for example, what is the grammatical role of the clause 'I don't know how many crimes'. Is it the object of the verb 'committed'? What type of clause is it?

  • The implication is that he has committed many crimes - so many that the speaker cannot guess the number of them. Sep 9, 2021 at 16:28
  • 4
    Actually, it should be punctuated like this; He has committed I-don't-know-how-many crimes. This larger-than-life character was truly amazing. Ergo, those phrases function as adjectives.
    – Lambie
    Sep 9, 2021 at 16:34
  • 1
  • 5
    @Lambie - I've never seen the expression hyphenated like that. Sep 9, 2021 at 18:46
  • 1
    I don't know how many is an extended quantifier; 36 or a few or many, many could substitute for it. The quantifier modifies (binds) the NP crimes, which is the object of committed. The odd part is the fact that an independent sentence has been used as a quantifier; but quantifiers can have subordinate clauses and quite complex structure, e.g, more than I would have expected is also a quantifier. Sep 9, 2021 at 23:18

4 Answers 4


Hyperextension of adjectives is not two or even three. They are longer phrases and sometimes even a full sentence used as a prepositioned adjective. AKA multiple hyphen compound adjective

Also dealt with here slightly differently: answer 5 by Sven Yargs

Here are two examples from an academic paper: Hawking Hyphens in Compound Modifiers

abstract: Abstract

The first principle of legal writing is surely its clarity — visible actors (unless the action matters more), uncluttered syntax, and, of course, logical structure. But the little things can matter to clarity, too — such as deliberate punctuation that signifies. In the language of law, in which compound nouns are rife, the reader can feel adrift as to where modifiers end and the noun begins. (Consider government-subsidized health flexible-spending arrangement without those hyphens.) Hyphens help. Whether an author cares to hyphenate the noun is his call; but hyphenating compound modifiers (also called phrasal adjectives, though they may include adverbs — more-abundant paperclips) follows a logic that is worth learning. This essay describes that logic. But its pitch is that legal writing, of all writing disciplines, should practice a deliberate, consistent use of such hyphens, rather than the more-relaxed practice readers see in less-formal writing (whose effects, of course, are usually also less consequential).

There are limits, of course, to how far a compound stitched with hyphens can stretch. Brian Garner dubs such hyperextension “snakelike compounds,” and suggests “rework[ing] the sentence.”32 Exceptions are compounds crafted tongue in cheek, such as Fred Rodell’s typology of footnotes: “There is the explanatory or if-you-didn’t-understand-what-I-said-in-the-text-this-may-help-you type. And there is the probative or if-you’re-from-Missouri-just-take-a-look-at-all-this type.”33

hyperextension of adjectives

So, that gives us: He has committed I-don't-know-how-many crimes. It is an example of a hyperextended adjective.

This type of hyperextension is found, yes, in comical texts but is also fairly common in speech. And therefore, in plays and scripts.

Examples I have just made up:

She is not a for-better-or-for worse woman.
They are not but-I-don't-want-to-Mommy kids.
We are definitely the hold-your-nose-and-do-it people in this kind of situation.

  • And here is one I just found online in a formal text.

The second point is that the Protestant British (God bless them!1) undermined and obstructed the Catholic enterprise here; and, more tragically, the Anglo-American Catholicism that came with the Maryland colonizers was a weak, timid, all-too-impressed-with-their-fellow-countrymen-who-were-heretics type of religion. I shall save that second point, with its too-long hyphenated adjective, for another time. Now I would like to focus on the first point, our too-secret Catholic history.


[All bolding is mine. I fully expect posters to post ones of their own under this answer. Perhaps they have ones that are more germane and funnier.]

  • you've got your upvote.
    – Centaurus
    Dec 11, 2023 at 13:17

This is a fairly well accepted construction which, however does not fit into the traditional scheme of things in the grammar of English. What can be thought about it from the knowledgeable point of view of grammarians is given a fundamental introduction in section 17.112 of the Comprehensive Grammar of the English language (Quirk et al, 1985). If one needs to refer to the structure preceding such noun phrases as "crimes" in the query, one says that the structure can be considered either as involving premodification by a sentence or that the noun phrase is the object (of know) in an embedded nominal clause. The nature of this premodification is more or less adverbial; the premodification of nouns by regular adverbs is rather rare (an away game, the then chairman, in after years, …).

(CoGEL § 17.112) Premodification by sentence

What [can be] said of adverb phrases [as pertains to premodification of noun phrases] applies to premodification by a sentence :

♦ (?) I visited his what do you call it cottage [ cf: What do you call it when a cottage has walls made from overlapping pieces of'timber? Clapboard] [user LPH: the interrogation point means that native speakers do not find this type of sentence fully acceptable.]

A few institutionalized examples retain a colloquial or slang flavour: a whodunit story is one about crime, and the nonstandard grammar or spelling are preserved as part of the ironic slang. Do-it-yourself as in a do-it-yourself job has become so often used as to pass out of the area of slang (and sometimes be reduced to DIY).

Somewhat more widely acceptable are noun phrases which can be interpreted either as having a sentence as premodifier or as being object (usually of know) in an embedded nominal clause:

♦ He asked I don't know HÒW many people. [user LPH: "how" carries nuclear stress.] [1]
♦ He asked I don't KNÒW how many people. [1a]

With either intonation, the meaning is 'He asked a relatively large number of people, though I don't know precisely how many'. The meaning is somewhat different if the sentence is reordered, enforcing a different grammatical structure:

♦ I don't know how many people he asked. ['I don't know the number of people he asked.']

For the most part, however, sentence premodifiers have an air of the outrageous and improvised. Part of a political leader's election campaign was described by a journalist as

today's meet the people (if they can find you) tour

Far more remarkable is the following quotation from a literary comment in which the sentence premodification itself has highly irregular and sophisticated punctuation to convey highly irregular coordination devices:

     His other comments ignore. . . the obvious fallacies inherent in the
       'But the poem (play, novel) was meant to be tedious/pretentious/pointless' line of critical argument.

Note: There is matter for disagreement among grammarians in the assertion that such constructions as "away game" and "then chairman" show examples of premodification by adverbs. The SOED dictionary lists "away" and "then" as adjectives and gives the example "his then wife"; this approach appears to be preferable.

  • @Lambie This is not my appreciation; I am quoting the grammar I have been using. It takes into account the American version of the language, but perhaps not as well as the British since it has been written by British grammarians. It is possible also that you base your evaluation only on AmE. Moreover, you might consider nothing more than colloquial speech, whereas those grammarians might be thinking about a more formal type of English.
    – LPH
    Sep 9, 2021 at 22:19
  • @Lambie Your source is American (Duke University).; one of the authors is (or was, I think he might be dead too) from Sweden. Note that they do not mention compound adjectives though, and keep strictly to sentences. Nevertheless, their findings being now 35 years old, they are possibly not quite up to date; things will have changed in the domain of premodification by sentences.
    – LPH
    Sep 9, 2021 at 23:01
  • You don't mean appreciation. You are using the French meaning of the word.
    – Lambie
    Dec 30, 2021 at 17:02
  • 2
    @Lambie Now, you've erased your commentary, and it is impossible to know what the matter is; you shouldn't do that if you intend to criticize and expect people to reply to you in a proper manner.
    – LPH
    Dec 30, 2021 at 17:37
  • 1
    Find it hard to follow this thread with half of the comments vanished into thin air, but I can vouch that in France appréciation is what teachers write on/about students' works when assessing them. One can write une bonne appréciation or une mauvaise appréciation as appropriate, just as in Britain one can write a good or a bad assessment.
    – None
    Dec 31, 2021 at 7:57

These types of constructions are actually covered in CGEL (Ch. 11 Content clauses and reported speech, section 5.3.4, p. 984):

A distinction needs to be drawn between the following constructions:

i: He made some mistakes, though I don’t know how many.

ii: He made I don’t know how many mistakes.

In [i] how many is a reduced interrogative clause functioning as complement to know. This differs from the constructions we have been discussing only in the reduction of the interrogative clause; it is interpreted anaphorically as “how many mistakes he made”.

In [ii], however, there is no ellipsis, and how many mistakes is not an interrogative clause. It means very much the same as "I don’t know how many mistakes he made", and its form seems in some way derivative from the latter. But clearly made is the verb of the matrix clause and the proposition that he made some mistakes is the main assertion, not backgrounded information; and there is an implicature that he made a large number of mistakes. Syntactically, I don’t know how many mistakes must be an object NP with mistakes as head; I don’t know functions as an irregular type of modifier to how.

A variant of I don’t know is God knows. These modifiers occur with most of the interrogative words (though not why) – compare They’re inviting God knows who to the reception.

  • See also @BillJ's answer, which is almost identical to mine, except that he doesn't cite CGEL and phrases it in his own words.
    – Eric
    Sep 12, 2021 at 17:16
  • CGEL does not cover this usage specifically where you have a full sentence or quasi-sentence used as an adjective. And this is not reported speech. I have provided numerous examples of this.
    – Lambie
    Sep 13, 2021 at 13:54
  • Could you explain to me why you think 'I don't know how many' is an adjective? What makes your analysis correct, and CGEL's wrong? (Apologies for the bluntness.) It may well be that CGEL's analysis is wrong, or at least not a definitive source on this matter.
    – Eric
    Sep 13, 2021 at 15:32
  • It is not that CGEL's analysis is "wrong". It is a sentence premodifier. CGEL just don't analyze it thoroughly or give other typical examples. As used in the OP's utterance, it is adjectival. If you move it to the end of the sentence, it then becomes something else as in your i). The occurrence of three-hyphen phrases is well described "in the literature"; Longer phrases or sentences are not. However, it is very, very common in speech and also used in writing. See my answer and the link to Sven Yargs on a similar question.
    – Lambie
    Sep 13, 2021 at 16:04
  • 3
    The problem here is that some people wrongly insist in using the misleading term 'adjectival' for everything that modifies a noun. "I don't know how many" may modify "mistakes", but it is not an adjective (phrase), not some kind of compound adjective.
    – BillJ
    Sep 14, 2021 at 9:38

He has committed I don't know how many crimes.

The meaning here is similar to "I don't know how many crimes he has committed", and in form seems in some way derivative from the latter.

But clearly "committed" is the verb of the matrix clause and the proposition that he committed some crimes is the main assertion, not backgrounded information; and there is a clear implicature that he committed a large number of crimes.

Syntactically, "I don't know how many crimes" must be an object NP (not a clause) with "crimes" as head. "I don't know" then functions as an irregular type of modifier to "how".

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.