[Lexicog] Re: lexical phrase

David Frank david_frank at SIL.ORG
Wed Dec 6 16:39:01 UTC 2006


John R --

I realize that when I said that PPs "modify" nouns, that is a pretty simple-minded definition that goes back to the definitions we learned in "grammar school" (i.e., primary school) as a child. We were taught, for example, that adverbs can modify an adjective, a verb, or another adverb, and that prepositional phrases modify nouns. Of course, a lot of the defintions we learned then proved to be unsatisfactory on deeper analysis. One thing we didn't learn back then was that prepositional phrases can "modify" predicates, or even link one sentence to the previous one. If you are saying the word "modify" is too simple-minded, then I'm sure that's true. But the more technical aspects of what a PP does were out of focus of what I was trying to say.

Maybe what you are saying is that only what comes before a noun in English can be considered to modify it. You would then say that, normally, PPs and relative clauses don't modify nouns, but rather NPs. That is not how I look at it. I suppose that from your perspective, you would say that relative clauses are not part of a noun phrase, but rather modify a noun phrase, right?

The main thing I had wanted to say was that some PPs "apply" to a noun (or NP), some "apply" to a verb or VP or predicate, and some that come at the beginning of a sentence "apply" to the relation of that sentence to the previous one. You can choose whatever verb you want in place of "apply."

Part of the difference between your definitions and mine could be explained if your view of syntax is geared towards binary splits, as in immediate constitutent analysis from the American structuralist period, and in transformational-generative grammar. My view of grammar allows for multiple splits in a tree diagram. I consider a PP or a relative clause following a noun to be part of the NP.

-- David F

  ----- Original Message ----- 
  From: John Roberts 
  To: lexicographylist at yahoogroups.com 
  Sent: Wednesday, December 06, 2006 10:21 AM
  Subject: Re: [Lexicog] Re: lexical phrase

  David Frank wrote: 
    Generally, prepositional phrases modify nouns, and the set of prepositional phrases that can modify a noun is open-ended. There is a more limited set of prepositional phrases that can modify a verb. These generally have a time word or a location word as the object of the preposition, as in "in the morning" or "in the garden." And there is an even more limited set of prepositional phrases that can serve as sentence connectives, including the following:
        on the other hand
        on the contrary
        by contrast
        by this time
        at the same time
        in addition
        by the way
        for this reason

David,

What do you mean when you say, "Generally, prepositional phrases ... modify nouns"? The normal function of an adposition with regard to nominals is to indicate the semantic role an adjacent NP plays in the sentence. It doesn't modify the meaning of the NP as such. In English you can have prepositions which modify a noun, e.g.
above  items/average/normal/board/ground/weight/height

  away  match/game/team/form/fans/day/message

  below  average/par/ground/weight/height

  near  thing/miss/darkness/fall/tragedy

  opposite  side/end/way/direction/number

  past  times/issues/life/experience/estimates


through  train/road/ticket/flight/service

But only a limited number of prepositions can do this. You can also have PPs which modify nouns in English, e.g.
down-in-the-mouth expression

  in-your-face attitude

  off-the-cuff remark

  round-the-clock service

  up-to-date timetable

  up-to the-minute fashion



  The examples of PP sentence connectives you cite above are PPs of the first type, i.e. PPs with a head preposition and a complement NP. In these cases the P does not modify the head noun. Rather the P indicates the semantic role of location for the NP. But the mismatch is that the PP is not expressing a semantic role of the NP in relation to the predicate, but instead is expressing the semantic role of location in some other domain of the discourse.



  John Roberts
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.linguistlist.org/pipermail/lexicography/attachments/20061206/d4fe436c/attachment.htm>


More information about the Lexicography mailing list