Alan wrote:Yes, I do believe I see what you're getting at: if we can somehow show that the sentence
[1] I heard students talking about how fit he seems.
has differing semantic entailments from simple
[2] I heard students.
then we will reckon the phrase 'talking...seems' a complement rather than a postmodifier.
The problem is that it really doesn't differ in any significant way: when we say that we hear someone, we may be referring to any sound that that person, directly or indirectly, produces, from spoken words to rustling leaves caused by his/her footfall. In [1] we know that what we hear is words, but that is because of the use of the verb 'talk' and has nothing inherently to do with the meaning of the word 'hear'.
There are, therefore, no real grounds here for reckoning the participial phrase to be other than a structurally optional postmodifier..
That's great explaination! I see what you mean. As I read this part, I realize that a complement is something more than a postmodifier, which means a complement must have a connection with a main verb. And meanwhile, a modifier is just connected with a post/former referent specified by it.
ex) He made me (to be) a doctor.
The structure is "S+V+O+C". The main difference between a simple modifier and a complement is a main verb "made" influences the connection between "me" and "(to be) a doctor". That is, "a doctor" has relations not only with "me" but also with "made". The "made" match "me" with "(to be) doctor".
I believe that is major difference. The main condition making difference between a simple modifier and a complement is a relationship with a main verb.
Alan wrote:As far as the question 'what kind of modifier?' is concerned, note, firstly, that to remove a restrictive clause or participial phrase from a sentence, even though it most often will leave an incomplete-sounding sentence, does not necessarily do so. In, for instance,
I have a brother who once served in India.
we could happily remove the relative clause and still leave the meaningful and sensible sentence
I have a brother.
Yet we still consider the relative clause restrictive, and therefore do not divide it off with commas, since it is still needed, effectively, to justify my mentioning of the brother: that is to say, the purpose of the sentence is presumed to be to communicate the fact that a brother of mine served in India, rather than to emphasize his existence. The addressee will infer from this that I probably have several brothers (or, at least, more than one), of whom however only one - the one here identified by the restrictive clause - served in India.
Compare this with
I have a brother, who once served in India.
The use of a nonrestrictive clause immediately changes the addressee's perception of the sentence to one in which the existence of my brother is deemed the main point, and the fact of his army service is of no more than passing interest. The addressee is likely to conclude from this that I have only one brother..
I think "comma" is also an element of a structure. In this respect, a slight difference in a nuance is enough to change the structure because the main intention a writer want to expression will be different , depending on whether he uses a comma or not.
Alan wrote:Note also that, in constructions involving verbs of perception (such as [1]), commas are not used (making the phrase technically 'restrictive'), even though the participial phrase is clearly functionally closer to a nonrestrictive.
It is interesting. I think the sentence
[1] I heard students talking about how fit he seems.
is in the case of a "strictive clause", although it has to depend on a context. It is because the main context in the reading stuff is about his healthy condition. If the writer want to show us just him, the "deep structure" is a nonrestrictive, because the main point is just "him".