Infinitives & gerunds
| From: | R A Brown <ray@...> |
|---|
| Date: | Wednesday, October 4, 2006, 9:19 |
|---|
H. S. Teoh wrote:
> On Tue, Oct 03, 2006 at 07:49:59PM +0100, R A Brown wrote:
[snip]
>>Having looked more carefully at your examples, and looking at the TF
>>grammar
http://conlang.eusebeia.dyndns.org/fara, I think they are
>>indeed infinitives - but TF's way of handling relative clauses is
>>certainly unusual! But why not? :)
>
>
> Sounds good to me. :-) It's not like TF doesn't already have unusual
> features, such as the case system and the complements. One more wouldn't
> hurt.
>>Just to recap your examples from your previous mail:
>>{quote}
>>tara' kei uenai ibuneis ka'aman ia.
>>(she ORG) want (AUX_CVY:mushroom eat:INF_RCP) COMPL
>>She wants to eat the mushroom.
>>.....
>>
>>huu sa tapa tun na ibuneis arapan bata.
>>1sp CVY go slope RCP CVY:mushroom pick_up:RCP COMPL
>>I go to the slope (of a mountain) to pick mushrooms.
>>{/quote}
>>
>>I have no problem with these. The words concerned are clearly nouns as
>>they have the noun case postclitics;
>
>
> Um... actually, they don't. The last word in each clause is the
> complement, which is connected with the main verb rather than the
> infinitive.
D'oh!! (slaps forehead). Mea culpa! (beats breast)
I must've been having a senior moment when I wrote that. Yes, I knew the
complements were connected with the main verb and not with the
infinitive. And as for that horrid hybrid "postclitic" - I feel quite
ashamed. The word should be 'enclitic' and, in any case, it is not
relevant to what I intended (but singularly failed) to say.
What I meant to say is that the words concerned show inflexion for the
three core cases of nouns. I was referring to _ka'aman_ and _arapan_,
which are both inflected for receptive case. If these words have
receptive case inflexion then, surely, they must in some way function as
nouns.
>
>>but they also have verbal functions in that they have their own verb
>>arguments. that is, they are verbal nouns - i.e. infinitives.
>
> I'm not so sure about the noun part. The case they inflect for not
> indicating their function in the main clause, but rather the function of
> the subject NP (the first NP in the sentence) in the subclause.
That's interesting (as most things seem to be in TF :)
So in fact the subclause has no finite verb, but is consists entirely of
nominal arguments. I like it!
> Maybe they are more like participles than infinitives? But they can only
> ever take the subject NP as subject, and they don't agree in case, but
> instead mark the case of the subject in the subclause.
I think your second sentence shows why they are not participles. The
words concerned are inflected for the three core _noun_ cases, therefore
they must surely be nominal.
There are, as I have written in an earlier mail, languages that do
inflect the infinitive (e.g. Turkish) and, as far as I can see, there is
no intrinsic reason why they should not be inflected. (That neither
ancient Greek nor Latin inflected their infinitives and found other
'solutions' to the problem of non-inflexion, is *no* indication of any
universal 'rule'; it is an accident of the diachronic development of
those two languages.)
>
>>But:
>>{quote}
>>Notably, the infinitive clause is identical in form and construction
>>to a relative clause modifying a noun:
>>baan ikaren muras kuinin sei tanap buta' fei imi tsi.
>>(old_lady (CVY:shoe black own:RCP) CVY) dwell hut that in COMPL
>>The old lady who owns the black shoes lives in that hut.
>>{/quote}
>>
>>Well, yes, this is TF's infinitive. It is AFAIK a quite unique way of
>>handling relative clauses - as I said, I familiar with the idea of
>>using a participle clause to express this idea (The old lady, the
>>[one] owning the shoes, lives in that hut), but you haven't use a
>>participle. It's certainly an interesting way of dealing with things
>>:)
>
> I'm not so sure about analysing the relative verb here as an infinitive.
The infinitive still shows noun inflexion as it did in the earlier examples.
> The case clitic is modifying the head noun, not the relative clause.
> Maybe it's closer to a participle? Or maybe something else altogether.
> The way it is constructed is that the clause sits between the head noun
> and its corresponding clitic, just like any other noun modifier:
>
> baan sei
> old_woman CVY
> (The) old woman
>
> baan duru sei
> old_woman slow CVY
> The slow old woman
>
> baan tara' sei
> old_woman DEM CVY
> That old woman
>
> baan ikaren muras kuinin sei
> old_woman CVY:shoe black own:RCP CVY
> The old woman who owns the black shoes.
>
> The syntactic nesting of this last is:
>
> (baan ((ikaren muras) kuinin) sei)
>
> Since the relative clause is in adjectival position, a literal
> translation might be something like "the black-shoe-owning woman". Seems
> more like a participle than an infinitive to me, although again, not
> agreeing with the head noun in case,
Exactly! Therefore IMO not a participle.
>but instead indicating the case of the head noun inside the subclause.
What it appears to me that we have here is:
(a) A subordinative verb form which inflects for the three core _noun_
case, i.e. the infinitive.*
(b) It is used in clauses like:
tara' kei uenai ibuneis ka'aman ia.
(she ORG) want (AUX_CVY:mushroom eat:INF_RCP) COMPL
..where the whole clause functions as a noun, i.e. we have a nominal
clause. In the familiar SOV langs, it is the object of 'want' - would
that be receptive in TF?
(c) It is also used in clauses like:
baan ikaren muras kuinin sei
old_woman CVY:shoe black own:RCP CVY
The old woman who owns the black shoes.
..where the whole clause functions as an adjective, i.e. we have an
adjectival clause.
*I don't think conjugate/conjugation is the right word if we are talking
about core case forms. So in
http://conlang.eusebeia.dyndns.org/fara/verbs.html#Subordinative_Conjugations
I suggest changing:
- the heading 'Subordinative Conjugations' to 'Subordinative Inflexions'
(or 'Subordinative Inflections' if you prefer that spelling :)
- and changing the first sentence 'Verbs in a relative clause or an
infinitive clause conjugate for the 3 core noun cases' to 'Verbs in a
relative clause or an infinitive clause inflect for the 3 core noun cases.'
I think we need to distinguish between the _function_ of the subordinate
clauses and the syntax of the individual words in the clause. For
example in English we have sentences like:
The man I told you about was arrested last evening.
'I told you about' is adjectival and acts as an attribute of 'The man',
yet no word in that clause is an adjective. There is no reason, as far
as I can see, to call 'kuinin' anything else than an infinitive. Just as
there is no adjectival word or participle in English, there does not
need to be in TF either.
[snip][snip]
>>
>>I guess one could. In English dialect, forms like "I go to the slope
>>of the hill [all] for to pick mushrooms" were still around in the
>>early 20th cent - but probably died out now :=(
>
>
> Hmm, isn't that still around in English pidgins?
It may well be.
> [snip]
>
>>>Hehe... makes one wonder where the prescriptivists got that 'rule'
>>>from. :-)
>>
>>They have a habit of turning _tendencies_ into fixed 'rules' - even if
>>it means saying things like "Caesar's Latin is not as good as
>>Cicero's" - crazy ;)
>
> [...]
>
> lol...
>
> Speaking of which, I've been thinking about this idea of creating (or
> analysing) language not in terms of grammatical rules, but of
> *prototypes*. Maybe I'll write this up in a separate thread...
Please do - sounds interesting :)
--
Ray
==================================
ray@carolandray.plus.com
http://www.carolandray.plus.com
==================================
Nid rhy hen neb i ddysgu.
There's none too old to learn.
[WELSH PROVERB}
Reply