1

The Cambridge Grammar of the English Language (Page 385) has this to say about "cardinal numerals":

7.6 Cardinal numerals

The cardinal numerals are primarily determinatives but they have a secondary use in which they inflect for number and hence belong in the noun category: They set off in threes/enrolled in their hundreds. In practice, only low or round numerals are used in this way. 24

...

24Numerals are often used metalinguistically, as the names of symbols: They added a ‘3’ before all the Brisbane telephone numbers.

What's the part of speech of the following numbers in CGEL?

Now, these orbital images tell me that the hostiles' numbers have gone from a few hundred to well over 2,000 in one day. <Avatar (2009)>

Since they don't inflect for number, they don't seem to belong in the noun category. Nor do they look like determinatives. The only choice left is to say that they are being used metalinguistically, but this doesn't seem right either.

tchrist
  • 134,759
JK2
  • 6,553
  • 2
    Fused determiner heads. – BillJ Dec 13 '23 at 08:29
  • 1
    There's an argument that numbers form a distinct grammatical category. Are you specifically interested in what CGEL says or will you accept alternative theories? – Stuart F Dec 13 '23 at 11:03
  • @StuartF I'm primarily interested in CGEL's approach, but other approaches might be necessary if CGEL can't explain what these numbers are. – JK2 Dec 13 '23 at 11:10
  • You ask a lot of questions like this. Why did you not like the other answers? – tchrist Dec 13 '23 at 12:01
  • @tchrist I don't think I've asked "a lot of questions like this". This question is clearly different from my other questions. – JK2 Dec 13 '23 at 12:08
  • You've asked lots of fused-head questions. I don't know what you're looking for here, so you'd best precisely enumerate for us the allowable set of "parts of speech" you'll accept which people are expected to choose from in answering you. Are you trying to force people to pick one of Donatus’s storied eight: nomen, pronomen, verbum, adverbium, participium, coniunctio, praepositio, interiectio? If those eight ᴀʀᴇ ɴᴏᴛ the exact allowed set without additions or subtractions, you'll need to tell us just which ones you'll permit. – tchrist Dec 13 '23 at 12:16
  • @tchrist Just because I have previously asked "fused-head questions" doesn't mean those previous questions are the same as or even similar to the current question. And I've clearly asked how CGEL would treat the numbers in question. So hopefully those who are familiar with CGEL would answer it without me having to explain the basics of parts of speech in CGEL. – JK2 Dec 13 '23 at 12:33
  • This is the kind of issue I have with CGEL: Numerals are often used metalinguistically, as the names of symbols:"They added a ‘3’ before all the Brisbane telephone numbers". Really? The number 3 is a symbol? – Lambie Dec 13 '23 at 14:26

3 Answers3

2

In CGEL terminology, cardinal numerals—like 2,000 (or two thousand) as used in your sentence—are in the lexical category (part of speech)* of determinative.

With the exceptions noted in your quote (and which don’t apply to your sentence), CGEL says:

We will therefore treat all numerals as belonging to the lexical category determinative, on a par with other unmodified determinatives . . . (p. 385)

Your determinative appears as part of a determiner in a fused-head noun phrase. Under Types of determiner (pp. 355–356), we see:

. . . determinatives can . . . combine the function of determiner with that of head in the fused-head construction, as in [4vi–viii]:
. . . [4] vi He gave ten copies to me and [–six–] to the others. [simple fused head] . . .

In a simple fused-head noun phrase, “a head can be added that is recoverable from an antecedent” (p. 333). In your sentence, the recoverable antecedent is not the noun numbers, but it might be hostiles—as in two thousand hostiles (that is, if you don’t mind a possessive antecedent like hostiles’; if you do mind, look for a suitable antecedent earlier in the discourse).

To put together well over two thousand, see “Minor Determiners PP” (p. 355).

Summary: 2000 is a determinative.

*The parts of speech according to CGEL are: noun, verb, adjective, adverb, preposition, determinative, subordinator, coordinator, interjection.

Tinfoil Hat
  • 17,008
  • In the Avatar example, you say "the recoverable antecedent is not the noun numbers, but it might be hostiles —as in two thousand hostiles." If that's the case, why can't we replace 2000 with 2000 hostiles? _...the hostiles' numbers have gone from a few hundred hostiles* to well over 2,000 hostiles in one day._ In CGEL's example, on the other hand, you can add copies: He gave ten copies to me and six copies to the others. – JK2 Dec 23 '23 at 04:24
  • You can replace 2000 with 2000 hostiles. Hostiles is a noun here, meaning hostile entities. Compare e.g. 2000 natives, where natives stands for native people. – Tinfoil Hat Dec 23 '23 at 04:41
  • Well, I can't seem to add hostiles in that sentence, because we're talking about "numbers" not "hostiles". How about this example then? 13 is an unlucky number. Do you think 13 here is also a fused-head noun phrase? – JK2 Dec 23 '23 at 04:52
  • Numbers here means “a multitude of people”... Their multitude of people has gone to 2000 people... – Tinfoil Hat Dec 23 '23 at 05:48
  • 13, used “metalinguistically,” is in the noun category. – Tinfoil Hat Dec 23 '23 at 15:25
  • Then, how about this example? Two and three is five. Here, are the numbers either determinitives or used metalinguistically? – JK2 Dec 24 '23 at 00:22
  • Also, I don't quite understand your claim that numbers in the Avatar example means “a multitude of people”. In the example, the hostiles' numbers means "the number of hostiles" , not "a great number of hostiles". Strictly speaking, therefore, number should have been singular: ...that the hostiles' number has gone from...to.... – JK2 Dec 24 '23 at 00:26
  • It means the numbers of the hostiles — not the number of hostiles. And “a multitude of people” is literally the dictionary definition (OED). And looking at the script, your antecedent is horde: ”a large unorganized group of individuals.” I think you’re trying to do something like this: The number is well over a number of 2,000. But that’s not what’s happening. – Tinfoil Hat Dec 24 '23 at 04:24
  • You're referring to the prior sentence: There's an aboriginal horde out there massing for an attack, where an aboriginal horde means the same thing as the hostiles in the quoted sentence. But the sentence's subject is not the hostiles but the hostiles' numbers. And the use of plural numbers here is unfortunate because he's not talking about two or more numbers. He's talking about a single number that is changing from a few hundred to well over 2,000. – JK2 Dec 24 '23 at 05:58
  • Fused heads, where mobile phones is the recoverable noun: But the proliferation of mobile phones shows what is possible. Within ten years, their numbers have gone from zero to nearly 1 billion in both India and China. 1. Not fused heads, where the noun nesting turtles is explicit: The Hawksbill turtle is one of the seven species of marine turtles and is found in nearshore tropical and subtropical waters of the Atlantic, the Indian, and the Pacific Ocean. Their numbers are believed to be between 20,000 and 23,000 nesting turtles... 2 – Tinfoil Hat Dec 24 '23 at 16:07
  • In the first example, the plural numbers is used because both the numbers in India and China are being discussed. Similarly, turtles live in different regions, so the plural numbers is used. But in the Avatar example, there is no reason to use the plural numbers. – JK2 Dec 26 '23 at 02:04
  • For idiomatic English best results, please study the word numbers as it pertains to multitudes or masses. If you need your numbers to be nouns, try, for example: The deli counter numbers go from 1 to 99. – Tinfoil Hat Dec 26 '23 at 03:56
  • So are you saying that it'd be wrong to use the singular number in the Avatar example? Also, in Two and three is five, are two, three, and five either determinitives or used metalinguistically? – JK2 Dec 26 '23 at 05:57
  • It wouldn’t be wrong; it just wouldn’t be what comes naturally when talking about... (4) numbers plural : a numerous group : many. In two and three is five, I suppose those could be nouns or fused-head determiners, depending on the context. – Tinfoil Hat Dec 26 '23 at 17:57
  • In M-W's definition (4), its example is: numbers died on the way, where numbers cannot be singular. Note in this example that numbers denotes living things because of the predicate is died on the way, whereas in the Avatar example, the predicate have gone from a few hundred to well over 2,000 in one day doesn't require the subject to be living things. – JK2 Dec 27 '23 at 00:51
  • Why can't it be singular? A number died on the way. (And we’re not talking about poor noun 2000 perishing.) Your entire argument seems to be based on the premise that numbers is being used incorrectly for number in your Avatar example. But it’s not. – Tinfoil Hat Dec 27 '23 at 01:43
  • But "a number" is an indefinite NP whereas the hostiles' numbers is a definite NP. And the meaning of this NP in the avatar example is not "a number of hostiles" but simply their number itself. You can't say this: _Now, these orbital images tell me that a number of hostiles* have gone from a few hundred to well over 2,000 in one day_. – JK2 Dec 27 '23 at 03:45
  • In your example, numbers is correct, and as such, it renders 2000 a fused-head. Bye. – Tinfoil Hat Dec 27 '23 at 03:59
0

In the example CGEL give to illustrate the nounal usage, few would substitute numerals except perhaps in very informal contexts, say in notemaking. Though 'a few hundred' does not lend itself to numeralisation, the use of 2 000 and the obvious parallel

  • ... the hostiles' numbers have gone from about 300 to well over 2,000 in one day

argues for non-nounal usage, in spite of the presence of 'numbers'.

On the other hand,

  • ... the hostiles' numbers have gone from the low hundreds to the low thousands in one day

argues for nounness.

This seems a hybrid usage; 'fused determiner head' seems a reasonable term.

Edwin Ashworth
  • 83,272
  • In your first example, are you saying that 300 and 2,000 are 'fused determiner heads'? If so, what's the common noun that has been fused with 300 and 2,000? Hostiles, numbers, or something else? – JK2 Dec 13 '23 at 20:09
  • Wikipedia describes the usage: << Determiners may bear two functions at one time. Usually this is a fusion of determinative and head in an NP where no head noun exists. >> – Edwin Ashworth Dec 13 '23 at 22:48
  • Your wiki quote is taken from CGEL. And the subsequent sentence in the quote is "In the clause many would disagree, the determiner many is the fused determinative-head in the NP that functions as the subject." In this example, many is a fusion of a determiner many and a head noun people. Then, in your example what head nouns are the determiners 300 and 2,000 fused with? – JK2 Dec 14 '23 at 01:14
  • 1
    @JK2 Fused determiner-heads don't contain some sort of invisible noun that's been "fused" with the determiner. Instead, the constituent acting as the determiner is also acting as the head of the NP; it's called "fused" because these two functions, normally separate, are being fulfilled by the same element. – alphabet Dec 14 '23 at 06:45
  • (And while it is true that you can often replace a fused determiner-head with that same determiner + a head noun, that doesn't imply anything about the syntactic structure of the fused determiner-head itself.) – alphabet Dec 14 '23 at 06:49
  • @alphabet Please give me some examples of Fused determiner-heads that don't contain some sort of invisible noun that's been "fused" with the determiner. – JK2 Dec 14 '23 at 06:57
  • @JK2 No fused determiner-heads work that way. – alphabet Dec 14 '23 at 13:27
  • Aarts has written an overview of approaches to classifying word usages where the POS is arguable {see 'How can one prove a word is a noun?}. One is to see a word as a fusion of two POS's. I can't duck this with say 'galore' (quantifier [post-nominal] and adjective [splendid]}. Another is the gradience model (I prefer this with say 'Brown's deftly painting his daughter was a joy to watch'). Another is to say 'It seems 54% noun, 46% verb: it's a noun.' I eschew this. – Edwin Ashworth Dec 14 '23 at 14:54
  • @alphabet IF no fused determiner-heads work that way, then you should be able to produce some examples of fused determiner-heads that don't contain some sort of invisible head noun. But apparently, you can't. So, I think you have proved yourself that your statement is not correct. – JK2 Dec 15 '23 at 01:42
  • @JK2 What do you mean "fused determiner-heads that contain some sort of invisible head noun"? That's a contradiction in terms, since by definition fused determiner-head NPs don't contain a head noun outside of the determiner. – alphabet Dec 15 '23 at 01:57
  • Of interest, perhaps: https://linguistics.stackexchange.com/questions/46800/is-it-possible-to-have-a-determiner-without-a-complement – Tinfoil Hat Dec 15 '23 at 02:44
  • (Let me put it this way: if there were an "invisible head noun" in a fused determiner-head, then the determiner wouldn't be the head, which would mean that it wasn't a fused determiner-head.) – alphabet Dec 15 '23 at 06:34
-2

Now, these orbital images tell me that the hostiles' numbers have gone from a few hundred to well over 2,000 in one day. <Avatar (2009)>

This gives us:

Now, these orbital images tell me that the hostiles' numbers have gone from a few hundred hostiles (in a day) to well over 2,000 hostiles in one day. <Avatar (2009)>

It seems from this that "hostiles" is being used as a common noun derived from the substantive - and not a very natural one. (I have no qualifications in Avatarese.)

Let's take a more orthodox example:

Now, these orbital images tell me that the deaf**'s** numbers have gone from a few hundred deaf (in a day) to well over 2,000 deaf in one day.

We can parallel this with

Now, these orbital images tell me that the enemy's numbers have gone from a few hundred enemy (in a day) to well over 2,000 enemy in one day.

If you consider "hundred" to be an adjective - and adjectives are not inflected - then all that has happened is that "deaf people/hostiles" has been omitted a couple of times - as is normal in informal speech.

Greybeard
  • 41,737
  • 1
    The classification of numerals (spelled out or symbols) as adjectives is in many's opinion untenable. There are even good reasons not to class numerals as quantifiers, to which they are much more closely related (both describe the relation of the noun's referent to the environment rather than give inherent details about the referent). – Edwin Ashworth Dec 13 '23 at 22:57