inchoative-causative pairs

[Martin Haspelmath]

Suzanne Kemmer wrote:
> We still  have the problem of why words meaning 'crack' in English and 
> other languages
> should be different in the intransitive vs. transitive argument frames --
> to the point where some languages resist using the verb in a 
> transitive construction at all.
This is of course an instance of the much more general issue of what are 
favorable conditions for single-word causatives.  I think it has to do 
with frequency of use: If the non-causative situation is more rarely 
expressed and the causative situation is more frequently expressed, 
we're quite likely to get a single-word causative (e.g. English break: 
the transitive 'break' is much more frequent than the intransitive 'break').

However, if the causative situation is very rarely expressed, and the 
non-causative situation is very frequent, many languages lack 
single-word causatives. For instance, causatives of agentive 
intransitives are generally quite rare ('make someone talk', 'make 
someone play', 'make someone dance'), so many languages lack simple 
causatives (cf. English *I played the child 'I made the child play'). 
The same goes for "internally caused verbs" like 'rust', 'decay', 'rot', 
and other verbal concepts that seem to be relatively rare in causative 
use, e.g. 'melt', 'freeze', and 'crack' (in the sense of 'have/give a 
fissure').

(Of course, not all languages lack simple causatives -- the European 
languages that have figured prominently in this discussion are generally 
very poor in single-word causatives. Many languages can easily 
causativize almost any verb.)
> I think it has to do with the interaction of the semantics of 
> 'cracking' (splitting along a created fissure) and the
> semantics of prototypical  transitive vs. intransitive constructions.
> The transitive frame, used protoypically with impact verbs, has a 
> strong implication
> of  total affectedness/extreme results, but the intransitive lacks 
> this property.
If this were the explanation, one would not expect 'crack' to pattern 
with 'rust', 'rot' and 'decay', because when something rots or decays, 
the affectedness is typically fairly extreme.

So I think frequency of use is the better explanation. Semantics could 
come in at the point when we ask why we rarely talk about causing a 
crack, or causing something to rot. But I'm not sure if it is relevant 
at all: Maybe causative cracking is rare because it's not so easy to 
just create a crack in a vase, as opposed to smashing it. And maybe 
causative rotting is rare because we don't generally find it desirable 
to see something rot. I don't know the explanation, but fortunately it's 
irrelevant: The explanation for the widespread absence of causative 
'crack' and 'rot' does not depend on it. If we know that they are used 
infrequently, that's sufficient to explain that languages use more 
complex (multi-word) ways of expressing causation. (In general, rarer 
meanings are expressed by longer, less compact forms.)

Martin

-- 
Martin Haspelmath (haspelmath at eva.mpg.de)
Max-Planck-Institut fuer evolutionaere Anthropologie, Deutscher Platz 6	
D-04103 Leipzig      
Tel. (MPI) +49-341-3550 307, (priv.) +49-341-980 1616

Glottopedia - the free encyclopedia of linguistics
(http://www.glottopedia.org)