grammar research interests

GE/TDA: of-phrases, mostly | Pinkie | May 21st, 2002

Hi C, J, ML and anybody else who enjoys the linguisticky bits...

SOME HARDCORE STUFF FOR COLIN (though comments from Mary Lynn, James, Simon or anybody else very welcome)...

You're interested in "possessives"? Any specific research question? My own perennial question (When can N1+of+N2 be expressed as N2+N1?) ties in very closely with this. As part of my research I'm developing a homegrown "grammatical" taxonomy of N1+of+N2 phrases (though I certainly realize that the lexical and discoursal angles are also relevant to my question). My current classification divides N1+of+N2 phrases into two basic types, which I call RELATIONAL (the phrase refers to two related entities) and SPECIFICATIONAL (the phrase refers to a single entity): compare "the molecular weight of water is..." and "the weight of water used in this experiment was...".

Of possible interest to you, Colin: compare i) "a part of the city" and ii) "the central part of the city". I'd categorize these as relational-with-specific-indefinite-reference and relational-with-specific-definite-reference; the genitive clitic can of course only be used with (ii), because the clitic requires/confers definiteness. Does this mean that (i) is not "possessive"? In fact this is really just a question of terminology: I think the important thing to note is that the genitive clitic indicates two things, "relationality" and definiteness. Compare also a) "a piece of the cake", b) ?"the piece of the cake", c) "the biggest piece of the cake", d) "a piece of cake", and e) "the biggest piece of cake": note especially (a) and (d), which I would classify as relational and specificational respectively. [I know these are constructed examples, but to illustrate what I mean with real examples would be horribly tedious!]

Anyway: have you ever come across anything along these lines? What's particularly interesting to me is the idea that the genitive clitic is perhaps explainable in basically grammatical terms, with no need to resort to semantics. For example: don't waste time wondering how a train can "possess" its arrival: rather, note that "the arrival of the train" has specific definite reference to both N1+of+N2 and to N2. [Actually, I'm simplifying massively: for example, I've not mentioned appositives e.g. the city of Rome, which I consider to be a type of specificational, and in which I think there may be various types of reference to N2; also, I haven't mentioned relationals with generic reference to N2, and various other types/features of of-phrases; etc. etc. etc.!]

As regards criteria for deciding when the clitic is permissible, you've no doubt seen Hawkins 19eightysomething in the Journal of Linguistics? This is widely cited, but I'm not very impressed with it. For example, he mentions that "the foot of the mountain" is unlikely to be expressed as "the mountain's foot". I'm afraid I must dispute this: for example, Google finds "The weight of a body would therefore be less at the top of a mountain than at the mountain's foot". In any case, I see this sort of thing as stylistic rather than "grammatical" in any strong sense: although of course there are arguments for considering grammaticality and "good style" to be extremes of the same continuum, no?

Signin' off,

Pinkie
Spain

Re: Defined text | Mary Lynn Hughes | May 20th, 2002

Was a bit amused to see you're still fretting over things like whether 'the foot of the mountain' can be expressed equally validly as 'the mountain's foot'. Glad to see that at least this time you managed to find one actual occurrence of the latter! (This reminds me of 'the valley's function'- do you remember that one?!). Sorry, couldn't resist. I'm sure this is all riveting from a GE perspective, which I know nothing about. S'pose this is where we part company, but it makes for great conversations. :-)

Mary Lynn

GE: Pinkie's hardcore | Colin Graham | May 26th, 2002

Hi, Pinkie et al,

[Skip to the end for the references!]

You're interested in "possessives"? Any specific research question?

Why/when is the 's' used in noun modification- specifically what is the prototypical form of this grammatical function? My theory is that English tends towards noun compounding, wherever possible, but that post-modified nouns tend to have greater precision. Where does the 'so-called' possessive fit in this scheme?

the genitive clitic can of course only be used with (ii), because the clitic requires/confers definiteness.

Does this mean that (i) is not "possessive"?

Not necessarily...

Is it not, rather, that English will not support two (deictics) determiners in a noun phrase? "a this part of the city" is just as unacceptable as "a the city's part". My question would be: why is it "central" and not "centre's"? And how is this different from "a part of the centre of the city", "a central part of the city", "part of the city's centre" and "the central part of a city" etc. etc.... all of which are acceptable in a given context.

In fact this is really just a question of terminology: I think the important thing to note is that the genitive clitic indicates two things, "relationality" and definiteness.

Also, as a matter of terminology, I do not like the term genitive clitic- for two reasons.

1. English does not have a case system, except perhaps as a trace in the pronouns. If it were a true case marker, then it would attach to or inflect the head noun and not the phrase as a whole.

2. To me, clitics are abbreviated verbs only. So the 's in my subject line could mean either that "Pinkie is hardcore", when the 's is a clitic, or that the hardcore is somehow related to Pinkie - the 'possessive' form. The 's used to show possession does behave like a verbal clitic, however, in that it attaches to the end of the nominal group. Morpheme is not an acceptable term to me either, since again, morphemes would attach to or inflect the head noun.

I would plump for (graphical) associative suffix for the 's ' forms used to talk about possession.

Compare also a) "a piece of the cake", b) ?"the piece of the cake", c) "the biggest piece of the cake", d) "a piece of cake", and e) "the biggest piece of cake": note especially (a) and (d), which I would classify as relational and specificational respectively. [I know these are constructed examples, but to illustrate what I mean with real examples would be horribly tedious!]

I would not question the acceptability of (b) because one of the dangers of constructed examples is that they are usually void of context, and I can think of a context where (b) could be used....

Anyway: have you ever come across anything along these lines? What's particularly interesting to me is the idea that the genitive clitic is perhaps explainable in basically grammatical terms, with no need to resort to semantics. For example: don't waste time wondering how a train can "possess" its arrival: rather, note that "the arrival of the train" has specific definite reference to both N1+of+N2 and to N2. [Actually, I'm simplifying massively: for example, I've not mentioned appositives e.g. the city of Rome, which I consider to be a type of specificational,

I would say it is also a partitive.

As regards criteria for deciding when the clitic is permissible, you've no doubt seen Hawkins 19eightysomething in the Journal of Linguistics?

This is widely cited, but I'm not very impressed with it. For example, he mentions that "the foot of the mountain" is unlikely to be expressed as "the mountain's foot". I'm afraid I must dispute this: for example, Google finds "The weight of a body would therefore be less at the top of a mountain than at the mountain's foot". In any case, I see this sort of thing as stylistic rather than "grammatical" in any strong sense: although of course there are arguments for considering grammaticality and "good style" to be extremes of the same continuum, no?

When you get into talking about grammaticality and style, then I think you're starting to talk about the defining characteristics of certain genres - e.g. the example I gave from Carter et al relating to academic papers, or using present tense to narrative storylines in movies reviews etc. Also, there is the difference between spoken and written grammar. 's is, after all, a convention of writing, and a new one at that, and is unstable and may not be necessary at all in certain written contexts e.g. mens wear, sheeps clothing, womens magazine... My money would be on mountain foot as being the "safest" to proscribe- if you're going to take that route- i.e. the compound. Compare, "the boot of the car", "the car's boot" and "the car boot" all of which are in the COBUILD online corpus.

I am primarily of the opinion that the 'possessive' form is a discourse level tool to allow the introduction of new but known participants or to create an anaphoric contrast with a previously mentioned Thing. All I have to do is show it!!!

Anyway now I know where your interests lie, I would suggest the best read is:

Taylor, John, R. (1996) Possessives in English. Oxford: OUP

He covers 'everything' including gerunds, nominalizations, specificity, blah, blah... and he's (very) readable.

If you want to bring in quantification, and you can bear a Chomskian approach, then Barker, Chris. (1995) Possessive Descriptions. Stanford, CA. CSLI Publications has some interesting ideas and approaches but is very specific and narrow in its range.

Maybe:

Heine, B. (1997) Possession. Cognitive sources, forces, and grammaticalization. Cambridge: CUP is worth a look too. I've only just got this one, so I can't make recommendations either way for your area.

Anyway,

Hope this helps.

BFN

Colin
:-)

Pinkie (associative suffix) hardcore | Pinkie | May 26th, 2002

Many thanks, Colin...

1) You don't like the term "genitive clitic"? Hmmm... I think this is rather academic: I mean, we just need a label for this, no-one's suggesting that English has Latin-like noun declension. I note that a common term is s-genitive, and that's OK by me. I like your suggested term associative suffix, or associative marker perhaps. But then why not genitive suffix?

2) I have much more serious doubts about the term "possessive", which is heavily charged with non-technical meaning: and like I say, I think some of-phrases are semantically "possessive" (e.g. "a part of the city") despite the fact that that they have no s-genitive equivalent. Perhaps we're talking at cross-purposes here? I certainly prefer a term like "associative" or "relational" to "possessive".

3) "I would not question the acceptability of [the piece of the cake] because [...] I can think of a context where [it] could be used". Yes, certainly, e.g. "Karlin pulls a chair over and sits down enjoying the piece of the cake he cut for himself". But I haven't been able to find "the piece of the cake" without subsequent postmodification, and I'd like to hear your view as to why it can't be rephrased as "the cake's piece he cut...", or simply as "the cake's piece" forgetting about the postmodification?

4) "I would say it [the city of Rome] is [...] a partitive". You've lost me there!

5) "When you get into talking about grammaticality and style, then I think you're starting to talk about the defining characteristics of certain genres..." Well yes and no. I prefer to view this as a continuum: black and white ends, a grey area in the middle. Grammaticality is about rules followed (unconsciously) by native speakers: these rules are relatively general (i.e. relatively independent of genre and register etc.), and relatively strict (so only children and non-native speakers tend to break them). Style is about rules followed more or less consciously by native speakers: these rules are genre-specific and relatively lax (it's not so obvious when someone breaks them). Still, I know you orthodox Hallidayans get your knickers in a twist when it comes to the grammaticality issue, and that you sometimes seem to want to deny its importance altogether ("...in syntax [Halliday et al. are] more interested in stating the range of options open to the speaker than in specifying how any particular set of choices from the range available is realized as a sequence of words", Sampson Schools of Linguistics 1980 p. 299; see also Butler Systemic Linguistics p. 221). Talking about grammaticality with a Hallidayan always feels rather like championing Wagner to an Israeli!

6) "'s is, after all, a convention of writing"- Disagree ENTIRELY: you can't possibly mean this. Though of course its orthography is by definition a question of written convention.

7) Thanks indeed for the refs. I've come across articles by Barker, and didn't find him very helpful: but I'm certainly grateful for the lead to Taylor.

Finally - I'm afraid that from your message I'm unable to grasp what your research interest/theory is, though this is probably my fault. I certainly agree that since reference to N2+'s+N1 is by definition specific definite (i.e. N2's is in a sense a definite determiner like "the"), then these constructions must generally be anaphoric or referring to something not yet mentioned but known.

Best,

Pinkie