The discussions of transitive, intransitive, definite, indefinite and
stative conjugations have confused me. The only guide I have to all
this is an acquaintance with the dying Hungarian contrasts
(transitive v. intransitive) and Hebrew contrasts (non-stative v.
stative).
There is some connection between this and voice. I think the obvious
voices are active, passive and anti-passive. Keeping things simple,
e.g. ignoring instruments and indirect objects, a transitive verb has
two arguments - actor and patient, and an intransitive verb has one
arguments, the 'actor', which is usable if infelicitous for stative
verbs. (In some languages, it is very difficult to distinguish
adjectives from verbs: in others, e.g. Proto-Indo-European, it can be
difficult to distinguish adjectives from nouns. A few languages,
e.g. Japanese, have both types, one class closed - i.e. not admitting
new members, and the class open - i.e. admitting new members.) Then
there are four basic cases:
A. transitive + actor + patient =>
Use active.
B. transitive + actor =>
Use anti-passive (typically the same as active in 'active'
languages). Some languages, e.g. Chinese, require a dummy object,
e.g. 'I am reading.' is expressed as 'I am reading characters'.
C. transitive + patient =>
Use passive (typically the same as the active in 'ergative'
languages). I do not regard the expression of the actor as basic to
this construct.
D. intransitive + actor =>
Use active.
In many (all?) languages, some verbs (e.g. 'cook', 'widen' in
English) are used both transitively and intransitively, though often
(in every language?) an explicit causative construction may be used,
and not infrequently unrelated verbs are used, e.g. 'kill' and 'die'.
The pattern of case markings may vary (typically between active and
ergative languages), sometimes depending on extraneous features -
pronoun v. noun, present v. past test. (The latter distinction in
Hindi arose because the transitive+actor+patient semantics are
expressed using what was the synthetic passive, so as to dispense
with the old Indic finite verb.) Several languages have an explicit
marker, on the object, for a definite object (e.g. Turkic, Farsi,
Hebrew, Bulgarian(?)).
How does all this relate to 'transitive', 'intransitive' etc
conjugations in Nostratic languages?
There may be other voices, e.g. reflexive in Semitic, middle in Indo-
European, and possibly even some strange voice for the English 'get'
passive, e.g. 'I got splashed on the way to work', as opposed to 'I
was splashed on the way to work'. Reflexive forms frequently become
passive in meaning.