Re: Existential voice

On Monday, May 2, 2005, at 02:13 , Gregory Gadow wrote:

[snip]
> What has my attention at the moment is the "existential voice." The
> construct in English of "it is raining" has always bothered me: what is
> "it", and why should "it" be raining
"it" isn't!

The _it_ in such sentences is known as a 'non-referential dummy subject'.
That is:
- non-referential: it does not refer to anything.
- dummy subject: modern (but not Old) English requires a grammatical
subject even if there is no actual subject. Such a subject is known, for
obvious reasons, as a 'dummy' subject. In English this is usually _it_ as
in "It is raining", "It is obvious they have gone" etc, but in existential
sentences English uses _there_ as in "There is a fly in my soup."

Many natlangs and conlangs do not require such dummies and do not use them.
  In fact, in such languages the use of any pronoun or other dummy element
would be wrong; verbs in such languages are said to be _impersonal_
because they simply have no subject.

> and not me or the dog or Aunt Matilda's Studebaker?
Eh??? None of them would not make sense! The ancient Greeks occasionally
said 'Zeus is raining' (and there's a joke in one of Aristophanes' plays
that he does so by pissing into a chamber pot with a perforated bottom to
it), but normally the verb was impersonal even then.

> My solution was to create a distinct voice for when
> a verb has no specific do-er or done-to, the existential voice.
Why? Why not just use _impersonal_ verbs, i.e. verbs with no grammatical
subject as very many natlangs and quite a few conlangs do, i.e. just say
"is-raining"? Cf.
Latin - pluit (*_id pluit_ just would not make sense! - the verb cannot
have a subject)
Italian - piove
Spanish - llueve
Malay - hujan
Mandarin - xia4yu3

Esperanto - pluvas
Novial - pluva
etc.
etc.

The grammatical term an 'existential sentence' is one that asserts the
existence or non-existence of an entity. In English it normally begins
with 'there is/ there are' (or other appropriate tense), in French with
'il y a' (note the non-referential dummy 'il'), in Latin with just plain
'est/ sunt' (usually first in the clause), in Welsh with special form of
"to be", namely 'oes' (and no subject) etc.

IMO to create a special voice to do what many languages do quite well with
impersonal verbs seems a bit like using a sledge hammer to crack a nut.

[snip]
> Rather than creating something just for weather, I want to extend this
> voice to include political and social climate. That would give expressions
> that are succinct in Glörsa but a bit awkward to translate in to English:
>
> dhëghösan [DeGo'sAn] - to make war on
bellum gerere  (Latin)
> dhëghösawne [DeGo'sAh\nE] - "it is warring" (the countryside is at war)
bellum geritur

> yövïnan [jovi'nAn] - to harvest
metere
> yövïnawne [jovi'nAh\nE] - "it is harvesting" (the crops are ripe and
> everyone is busy harvesting them)
metitur - impersonal passive, quite common in Latin.

Indeed, the impersonal passive is extremely useful for when for when "a
verb has no specific do-er or done-to". It occurs also in German, for
example:
Gestern wurde getanzt - yesterday [it] was danced = 'Yesterday there was
dancing'
..and in Turkish:
Otobüsë binildi - on-the-bus (dative) [it] was boarded = 'People got on
the bus'

> I know what I'm trying to say, but the use looks very strange. Even
> stranger was when I put the verb "to hope" in this voice
Works fine with an impersonal passive 'speratur' - there's hope, we have
hope, people are hoping etc.  (no specific 'hoper' expressed)

[snip]
> What I would like to ask is,
>
> * Am I using the term "voice" correctly?
Trask's definition of voice is probably as good as any:
"The grammatical category expressing the relationship between, on the one
hand, the participant roles of the NP arguments of the verb, and on the
other hand, the grammatical relations born by those same NPs."

I suppose it can be argued that you are proposing to grammaticalize a form
for your verbs that expresses a lack of participant roles (or at least
specific participant roles) so it may be termed 'voice'. But it does seem
to me, for reasons I have explained, unnecessary.

Most, possibly all, languages have a set of verbs which have no direct
object and a set of verbs which may have direct objects. Thus a set of
verbs with no "done-to".

Very many languages allow verbs to be used _impersonally_ , that is with
no subject or "doer". Therefore there already exists in natlangs (and some
conlangs) _impersonal intransitive verbs_ (including things like
"is-raining", "is-hailing", "is-snowing" etc) which have "no specific
do-er or done-to".  Quite a few languages allow transitive verbs to be
used impersonally in the passive voice for when there is "no specific
do-er or done-to", e.g.
itur (from "to go") we're going, they're going, people are going, etc "on
va"

> * Given my description, does it look like I'm using this voice correctly?
Yes, IMO.

> * Are there any natlangs with something similar that I could use as a
> model?
See all the comments above   :)

Ray
===============================================
http://home.freeuk.com/ray.brown
ray.brown@freeuk.com
===============================================
Anything is possible in the fabulous Celtic twilight,
which is not so much a twilight of the gods
as of the reason."      [JRRT, "English and Welsh" ]

Replies