Re: Ergative?

At 12:51 pm +0100 30/9/01, Michael Poxon wrote:
>I think part of the problem with this is terminology; "passive" is a term
>coined from its relevance to the classical languages, especially Latin, in
>the days before Ergative languages were encountered, and the lack of any
>indication of ergativity in English. When an ergative notion is encountered,
>the passive voice, often employing the word "by" - used also for
>instrumentality, tends to be employed. Thus we get sentences like "The house
>was built by my father". Now, is this passive, ergative, or instrumental?
Eh? <scratches head in puzzlement>

The house   was built     by my father
 Domus    aedificata est  a meo patre

Apart from the fact that the Latin would normally (tho not necessarily)
take the word order "domus a meo patre aedificata est", how are two
constructions different?

I don't understand how a sentence can by instrumental.  A noun phrase may
be, but not a sentence.  In fact the sentence contains no instrumental noun
phrase in either language.

Some languages have a separate instrumental case for nouns.  It seems that
ProtoIndoEuropean had such a case.  In Latin it is expressed simply by the
ablative case.  In English is most commonly used "with" is used (e.g. "He
was killed _with a blunt instrument_").

"by my father"/"a meo patre" is the agent.


>In
>terms of English grammar, it is a passive form;
Exactly!  the active being: My father built the house.

but one could also say that
>in purely semantic terms it is ergative:
>My father-erg (performer of transitive verb)
>Built (transitive verb)
>The house
No - that is the active sentense.  If the verb is transitive, the performer
of the verb is traditionally said to be in the nominative case; simply by
labelling it 'ergative' makes no difference.  It reminds me of a student, a
few years back, that thought by changing the extension of a Pascal source
code text file from .pas to .exe it would automatically become compiled
object code (I kid you not)!   Simply by changing a label does not affect
the construction one way or t'other.

>Or in a language which marks for accusatives:
>My father
>Built
>The house-acc (receiver of performed action)
Irrelevant whether the accusative/object is morphologically marked or not.

>So the way you see ergativity is largely dependent on the way you want to
>view it (sounds a bit like the Heisenberg thing!)
Not it ain't.

The English verb, like Latin, quite clearly has two distinct voices: active
& passive -
  ACTIVE                   PASSIVE
(he) built        >>      (it) was built
aedificauit       >>      aedificatum est

The fact that the Romans had apparently not met ergative languages is quite
irrelevent in the context of English.

In an ergative language the verb in "The house was built by my father"
would have the *same* morphological form as an intransitive verb in a
sentence like: "The house tumbled down".  In English it does not.

In an eragative language:
- the subject of an intransitive & the object of a transitive verb (from
the point of view of an active language like Latin & English) is treated
the same way; it is commonly referred to as being in the 'absolute' form or
'absolute' case.
- all transitive verbs, then, correspond semantically to passives in a
language like English but are morphologically _active_.
- the 'subject' of the transitive verb is expressed by a form called the
'agent'; but this would not AFAIK ever be morphologically expressed by a
prepositional phrase (as it is in English & Latin).

Some linguists consider sentences like: "The window broke" (cf. active: The
tree broke the window) and "This advertisement reads well" (cf. active: She
reads the advertisement with interest) to be ergative in English; but I
fail to see how otherwise English is ergative.

Ray.

=========================================
A mind which thinks at its own expense
will always interfere with language.
                     [J.G. Hamann 1760]
=========================================

Replies