5

In Swan's PEU (3rd Edition), an entry number 157 of 'discourse markers' reads...

There are a very large number of these 'discourse markers', and it is impossible to give a complete list in a few pages.

I'm finding it difficult to understand. Why is there a plural verb are and not is? I also note that discourse markers are defined in quotes ('...') which means it's considered as a single entity.

There are a very large number of 'X' out there does not look better to
There is a very large number of 'X' out there

Let me quote a quote here

Despite the digital age, there is a very large number of venues and spaces that are looking for plays, and many of them are looking for new plays.

Also, a result from the Google Books

What makes the fields 'classical' is the fact that there is very large number of sources contributing.

What am I missing?

StoneyB on hiatus
  • 175,127
  • 14
  • 260
  • 461
Maulik V
  • 66,059
  • 109
  • 310
  • 456
  • This reminds me of "proximity agreement". I couldn't find a good reference, but I found this: As noted in A Comprehensive Grammar of the English Language (1985), "Conflict between grammatical concord and attraction through proximity tends to increase with the distance between the noun phrase head of the subject and the verb." – Damkerng T. Aug 20 '14 at 12:47

3 Answers3

5

CGEL (The Cambridge Grammar of the English Language) discusses (at 3.3, p.349) situations where quantification is “expressed by means of a noun as head with an of PP as complement”. In these constructions terms like a lot, lots, a great deal, plenty, oodles and a number are in syntactic form the heads of the NPs in which they occur, but the semantic head is the ‘oblique’, the term which is the object of of:

a. Lots of workoblique is left to be done.
b. A lot of peopleoblique were present.

Note that in a lots is plural, but the entire NP takes a singular verb, is, because its oblique, work, is non-count, while in b a lot is singular but the entire NP takes a plural verb, were, because its oblique, people, is plural-only.

In these cases, says CGEL, “lot is number-transparent in that it allows the number of the oblique to percolate up to determine the number of the whole NP.”

In other words, semantics trumps syntax with these “number transparent quantificational nouns”: the verb agrees with the sense rather than the form.

In your sentence, consequently, the verb correctly takes the plural because its real subject is the oblique these 'discourse markers', not a very large number.

With many singular number-transparent nouns such as number, however, many people feel uncomfortable with a violation of syntactic concord. Consequently, you are just as likely to encounter a 'number-opaque' treatment. In the circumstances, neither can legitimately be labeled “incorrect”.

Note, by the way, that the phrase a very large number acting as a NP on its own, not quantifying an oblique, will take a singular verb:

A small number is expressed in words; a large number is expressed in digits; a very large number is expressed in exponential notation.

StoneyB on hiatus
  • 175,127
  • 14
  • 260
  • 461
  • What is this language coming to if we need one grammar book to understand the other... – oerkelens Aug 20 '14 at 13:10
  • @oerkelens lol. It's not "coming to" anything: it's exactly where it's always been. See Goold's Grammar of English Grammars, 1851.... Actually, I don't know Swan, but it is spoken of very highly by users I respect; and I betcha it addresses this question somewhere. – StoneyB on hiatus Aug 20 '14 at 13:13
  • The book seems to address the usage as you said indeed, though it might not explain the issue in depth. For those who have PEU, it's discussed in 526.2. – Damkerng T. Aug 20 '14 at 13:25
  • @DamkerngT. Thanks. You're one of the users mentioned in my previous comment, and it's clear that PEU will have to be my next purchase. – StoneyB on hiatus Aug 20 '14 at 13:34
  • @StoneyB Luckily, PEU is much cheaper than CGEL :-) Actually, Swan acknowledges both H&P 2002 and Quirk et al 1985 in the first pages of his book—he's distilled a lot of wisdom from those voluminous tomes into a much smaller pedagogical grammar. Occasionally you might feel he's oversimplified but overall it's quite good! And there is a little information there that isn't in either reference grammar. If nothing else, it's worth having because it's the most popular reference for ELLs. –  Aug 20 '14 at 18:32
  • This pretty much makes sense, except for the bit about the "real subject". You can completely remove the supposed "real subject" and still have a sensible sentence. Interesting perspective, though. – DCShannon Jun 12 '15 at 20:36
  • @DCShannon In that case you have a fused-head determiner: a determiner (in this case a quantifier) which swallows up its head, just as you do in "{Some / Many / Three / A few } [of our students] won scholarships." – StoneyB on hiatus Jun 12 '15 at 20:42
2

As pointed out by Lucian Sava in his answer, Google's corpus supports that both uses are often found in standard English. It is possible to analyse them as follows:

There are [NP a very large number of these 'discourse markers'], and it is impossible to give a complete list in a few pages.

The sentence above can be analysed as a complex noun phrase (NP):

a very large number of these 'discourse markers'

where the NP head is "'discourse markers'" and "a very large number of these" is functioning as a determiner.

By accepting this analysis, it is possible to understand that the verb "are" and the NP head "'discourse markers'" concord in number.

Despite the digital age, there is [NP a very large number of venues and spaces that are looking for plays, and many of them are looking for new plays].

The use of "is" can be justified as follows. In this case, the noun phrase (NP) is a complex phrase with a determiner, a head and a complement. The head would be "number", the determiner "a very large" and the complement "of venues and spaces that ...".

By accepting that the NP head is "number" it is possible to understand the concord between "is" and "number".

Nico
  • 2,480
  • 18
  • 47
  • Did you do that correctly? What about the other examples? How would you part it? – Maulik V Aug 20 '14 at 12:21
  • @MaulikV I've just updated my answer to address your comment. – Nico Aug 20 '14 at 12:33
  • 1
    I disagree. I would say "Despite the digital age, there are a very large number of venues and spaces that are looking for plays, and many of them are looking for new plays." Otherwise, it should read "Despite the digital age, there is a very large number of venues and spaces that is looking for plays, and many of them is looking for new plays." which is obviously wrong. In my mind, "venues and spaces" makes it plural. – CJ Dennis Sep 28 '14 at 22:31
  • @CJDennis The second example sounds off because it'd make "a large number" the subject of "is looking for plays"; "to look for" is unusual with inanimate subjects. – Nico Sep 30 '14 at 06:57
0

According to Ngram Viewer both are grammatically correct, but the use of one version or another it's really dependent on whether what is being discussed is singular or plural.

In your case these implies plural.

Lucian Sava
  • 11,458
  • 15
  • 46
  • 98
  • But these is merely reintroducing the discourse markers and even if we keep plural, it should take the singular verb is. Check my other examples; they talk about venues and spaces and sources - all are plural – Maulik V Aug 20 '14 at 12:10