I'm debating whether the noun phrase "John Locke's Two Treatises of Government" is restrictive or non-restrictive in this context:

Benjamin Franklin was introduced to democratic ideals when he read one of the most influential works of his time: John Locke's Two Treatises of Government.

My instinct is that "John Locke's Two Treatises of Government" is restrictive, because the the phrase "one of the most influential works of his time" suggests that there were several influential works. Naming the specific work that Franklin had read restricts the several influential works down to just one.

However, I think it could also be argued that the prepositional phrase "of the most influential works of his time" is purely ornamental. The leading noun is "one", which is singular. The leading noun refers to just one thing, so citing "John Locke's Two Treatises of Government" isn't restricting anything.

  • 2
    "John Locke's Two Treatises of Government" is a noun phrase, not a relative clause. It's relative clauses that can be either restrictive or non-restrictive. Noun phrases can be in apposition, however, and that's what's going on here. That noun phrase is in apposition with the noun phrase one of the most influential works of his time. That's what the colon signifies here. Dec 19, 2019 at 14:38
  • 1
    I think it's probably relevant that although we could reverse the two "equated" noun phrases one of the most influential works of his time and John Locke's Two Treatises of Government in speech (also in the written form if it had separated them by a comma), we don't seem to be able to do that with the actual orthography as presented here. Dec 19, 2019 at 14:41
  • Of course, orthography is irrelevant to grammar. The intonation would be different if the phrases were swapped, so that might vary the punctuation, but punctuation is semi-random in English anyway. Dec 19, 2019 at 14:57
  • The second noun phrase in the apposition certainly specifies (singles out) the actual work from the set of alternatives identifed by the first noun phrase. This is a specifying appositive rather than a renaming appositive. Were the noun phrases swapped over (probably with the colon replaced by a comma), we'd have an appositive adding incidental detail. Dec 19, 2019 at 14:59
  • The first noun phrase one of the most influential works of his time already has narrowed the set of influential works of his time down to one work. The title simply assigns an author and name to it.
    – TimR
    Nov 11, 2023 at 17:43

1 Answer 1


As commenters have already said,

[1] John Locke's Two Treatises of Government

is not a clause, but rather a noun phrase (NP). The distinction between restrictive and non-restrictive applies to clauses, not NPs.

In your sentence, the NP [1] is not integrated into the syntactic structure of the sentence: it is a supplement.

As I said, the restrictive/non-restrictive distinction does not apply to NP supplements. However, there is another distinction that you were probably sensing: NP supplements can be either specifying or ascriptive. And in this case, [1] is specifying, which can be seen by the fact that it licenses the addition of indicator words and phrases such as namely, that is, i.e., etc.:

Benjamin Franklin was introduced to democratic ideals when he read one of the most influential works of his time, namely, John Locke's Two Treatises of Government.

Corresponding to the fact that the supplement is specifying, its anchor (the part of the main sentence to which the supplement is semantically related) is a non-referential NP. In your case, the anchor is one of the most influential works of his time. It is non-referential, and it would still be non-referential if we turned it into a definite NP by omitting one of. Such a use of an NP is called an indeterminate use, and I will say more about it below.

When we added namely, we had to change the colon into a comma, but both are acceptable as markers of supplementation. Moreover, the colon can be used as a marker of NP supplementation only in cases of specifying supplements, i.e. those which provide identifying information. For example, the comma could not be replaced by a colon in

They went to Bill Clinton, the only man who could help them.

The reason is that the supplement (in boldface) 'provides descriptive, not identifying, information—compare They went to the only man who could help them: Bill Clinton, where the supplement does identify' (CGEL, p. 1741).

If the supplement were a clause, e.g.

Benjamin Franklin was introduced to democratic ideals when he read one of the most influential works of his time, which is John Locke's Two Treatises of Government.

it would be non-restrictive, because supplements are never semantically restrictive (I will quote a relevant segment from CGEL at the very end).


In your sentence

[2] Benjamin Franklin was introduced to democratic ideals when he read one of the most influential works of his time: John Locke's Two Treatises of Government.

the NP [1] has the syntactical status of a supplement, i.e. it is not integrated into the syntactical structure of a sentence. In speech, this would be indicated by prosody and a change in intonation.

According to (CGEL (p. 1351),

Although supplements are not syntactically dependent on a head, they are semantically related to what we will call their anchor.

In [2], the anchor is the NP one of the most influential works of his time.

In the case of NP supplements with NP anchors, which is your case, there is a distinction between

Specifying and ascriptive supplements to NP anchors

In supplementations with one NP as anchor and another as supplement, the relation between the two is comparable to that between subject and predicative complement in a be clause. In particular, the distinction between specifying and ascriptive complements applies also to supplements. Compare [19i—ii], for example, where the brackets enclose the subject of the clausal construction and the anchor in supplementation, while the predicative complement and the supplement are marked in boldface:

[19]    i  a.  [The first contestant] was Lulu. [specifying]
              b.  [Kim Jones] was a quite outstanding student. [ascriptive]
           ii a.  [The first contestant], Lulu, was ushered on stage. [specifying]
              b.  [Kim Jones], a quite outstanding student, won a scholarship to MIT. [ascriptive]

In the [a] examples Lulu is interpreted as specifying who the first contestant was, while in [b] a quite outstanding student expresses a property that is ascribed to Kim Jones. One formal difference in the case of supplementation is that specifying NPs accept the indicators namely, that is, i.e., etc.: The first contestant, namely Lulu, was ushered onstage.


The construction with a specifying NP as supplement is known as apposition. More particularly, this is the supplementary type of apposition, corresponding to the integrated apposition of the opera 'Carmen' or my husband George (in [loiia/11ii] below). Thus the appositive NP can be substituted for the whole supplementation yielding an entailment of the original: [19iia] entails Lulu was ushered on stage.

                                                                                                                  (CGEL, pp. 1356–1357)

[In an earlier chapter, in the case of dependents, CGEL explained that 'appositive dependents are ones which when substituted for the matrix NP in a declarative clause systematically yield a clause which is an entailment of the original'. For example, [The verb 'use'] is transitive entails 'Use' is transitive. 'The appositive thus provides a formulation that can stand instead of the NP containing it' (p. 447).]

The anchor NP is non-referential

Note that the anchor NP in supplementary apposition is non-referential, in the following sense:

Referential expressions, we have said, are generally NPs—but NPs aren't always referential.Compare:

[2]   i  Did Mary telephone while I was out? [referential]
       ii  Did anyone telephone while I was out? [non-referential]

In [i] I refer to a certain person and ask whether this person telephoned. In [ii], by contrast, anyone is not used referentially: I am not asking whether a certain person phoned.

In your case, the anchor is one of the most influential works of his time. It is non-referential in the sense of having an indeterminate interpretation. Let me explain this more.

NPs with an indeterminate interpretation

As I said above, NP anchors of NP supplements are always non-referential, even if they are definite NPs. This can happen because such anchors can have an indeterminate interpretation. Here is an explanation from CGEL (p. 403):

[9]   i  The boy who wrote this email must be expelled.
       ii  I think Ed's CD player was stolen by a friend of his.

These examples are ambiguous, with the underlined NPs having either a referential or a non-referential interpretation. Take first the underlined NP in [i]. This could be used as a way of picking out a particular individual (Smith Junior, say); in this referential interpretation, then, I'm saying that Smith Junior must be expelled. A context for this interpretation might be one where an investigation has already taken place into who wrote the offending email, and the culprit has been apprehended.37 Alternatively, however, it could be that no investigation has yet taken place, so that the culprit is not yet known. In this case, the underlined NP can be used non-referentially to say that the individual denoted ("the x such that x is a boy who wrote this email") must be expelled, whoever this individual might turn out to be. We term this the indeterminate use of the NP. The indeterminate interpretation can be expressed unequivocally by Whichever boy wrote this email must be expelled (fused relative), or The boy who wrote this email must be expelled, whoever it is (NP + exhaustive conditional adjunct). Note, then, that the indeterminate use is akin to the descriptive use outlined in (b) above, but it is not identical: following the utterance of the sentence, the indeterminate NP does not automatically have a referential use.

37It should be noted that in the referential use, the property given in the NP does not in fact have to apply to the referent. It could be that speaker and addressee are both convinced that Smith Junior is the culprit, even though he is not; in that case the boy who wrote this email would still successfully pick out Smith Junior and I would still be saying that Smith Junior must be expelled.

Indeterminate NPs can serve as antecedent for personal pronouns (we could continue [i], for example, with He has brought the school into disrepute): as is evident from the formulation in §8.2, inability to be antecedent for an anaphoric pronoun in an independent clause is a sufficient but not a necessary condition for the non-referential use of a 3rd person NP.

The referential versus indeterminate distinction also applies in the case of indefinite NPs, as in [9ii]. In the referential use I am picking out an individual whom I could distinguish by independent properties—compare a friend of his named Jones. On the other hand, I might have no idea who the thief is, but deduce from the fact that there has been no break-in that only a friend, someone who had access to the house, could have done it. In that case, a friend of his is being used indeterminately, to indicate what property the thief must have (whoever it actually is). Of course you, the addressee, cannot judge from the utterance of [9ii] alone whether I was using the NP referentially or indeterminately; a response such as So you know who did it then? would serve to seek confirmation that a referential use had been intended.38

38The term 'attributive* is often used for what we have here called the 'indeterminate* use of NPs. We prefer the latter term in order to avoid conflict with the syntactic use of 'attributive* as applied to adjectives functioning as pre-head modifier to a noun (contrasting with their 'predicative' function). The concept of indeterminate use of an NP can be extended to pronouns, as in I wonder which of the boys told her: he must have wanted to embarrass me. The pronoun is equivalent to the indeterminate NP the boy who told her.

Supplements are semantically non-restrictive

Supplements are necessarily semantically non-restrictive (CGEL, p. 1352). In more detail (p. 1353):

[10]   i  a.   The necklace which her mother gave to her was in the safe.             [modifier]
             b.   The necklace, which her mother gave to her, was in the safe.     [supplement]
        ii   a.   They are working on a new production of the opera 'Carmen'.            [modifier]
             b.  Bizet's most popular overa, 'Carmen', was first produced in 1875.   [supplement]

In [ia] the relative clause is a modifier of the head noun necklace and serves semantically to identify which necklace is being referred to, but in [ib] it is a supplement to the anchor NP the necklace, which is assumed to be identifiable independently of the information given in the relative clause. Similarly, in [iia] the appositive Carmen is a modifier of opera, identifying which opera is being referred to, while in [iib] it is a supplement to the anchor NP Bizet's most popular opera, and since there can be only one entity satisfying that description the supplement is again non-restrictive.

However, we have noted in our description of relative clauses and appositives that the integrated construction is not necessarily semantically restrictive - see Ch. 12, §4.2, andCh. 5, §14.3, respectively. Compare, then:

[11]   i  The father who had planned my life to the point of my unsought arrival
             in Brighton
took it for granted that in the last three weeks of his legal guardianship
             I would stillact as he directed.

        ii This is my husband George.

In [i] the relative clause doesn't distinguish one father from another: the narrator has only one father, so the modifier provides non-restrictive information about him. And [ii] does not convey that the speaker has more than one husband. It is for this reason that we have departed from the traditional account of relative clauses, in which the two main constructions are distinguished as 'restrictive' and 'non-restrictive'. A distinction in terms of integrated versus supplementary reflects the semantic difference more accurately and also matches the prosodic difference that distinguishes them in speech. It enables us, moreover, to capture the similarity between the unintegrated relatives and other elements that are semantically, prosodically, and syntactically unintegrated with the rest of the sentence: these can all be subsumed under the concept of supplement.

  • In 'The Siberian ibex, Capra ibex sibirica, is a species of ibex that lives in central Asia' we have an appositive which is neither specifying (either name does that) nor ascriptive (neither noun phrase adds information (to non-Latin-scholars) other than that of the alternative name). It is a renaming appositive, a class which seems not to be mentioned here. Does CGEL refer to them separately (not under 'supplements')? Dec 19, 2019 at 17:47
  • @EdwinAshworth As best as I can tell, there is no such category in CGEL. I think CGEL would say that this is indeed a specifying supplement, because it admits the indicator phrase that is: The Siberian ibex, that is, Capra ibex sibirica, is a species of ibex that lives in central Asia. Dec 19, 2019 at 19:30
  • @EdwinAshworth I think that the possible trouble with CGEL's account is their claim that the anchor is always non-referential. I don't know how to test for it independently—my understanding of what counts as referential and what doesn't is a bit shaky—but the more I think about it, the more it seems that it is the very presence of the apposition that can turn a formerly referential NP into a non-referential (I know—circularity threatens. But hold on!). An example: Dec 19, 2019 at 19:31
  • 1
    @EdwinAshworth And so Karen implicitly gets interpreted as "x such that x is called Karen", and then Abby specifies which Karen. I don't expect anyone to be convinced by this argument—like I said, there really has to be some independent test for referentiality, or else circularity threatens. But it is at least possible that something like this is going on. Dec 19, 2019 at 19:32
  • 1
    You can also argue that either renaming is adding information, albeit only at the hyperlinguistic level (again, where names are not, or are equally, attribute-informed). But I think there needs to be a classification which includes 'mere renaming'. There will always be the possibility that someone cottons on to the second name but not the first, as you say. (Or vice versa, in which case do we regard the apposition as both defining for some and descriptive for others?) Dec 19, 2019 at 19:46

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.