Summary of responses as to the inquiry about causatives and the coding of participants
chao.li at AYA.YALE.EDU
Mon Aug 20 15:03:25 UTC 2012
This is a summary regarding my inquiry about the coding of participants in
morphological causatives. I apologize for not being able to get it finished
sooner. I thank Bill Croft, Matthew Dryer, Martin Haspelmath, Helle
Metslang, Geda Paulsen, Guozhen Peng, and Jae Song for providing
information with respect to the inquiry. I also thank Peter Arkadiev and
Bingfu Lu for their interest in the topic.
In the inquiry, I wondered whether there were languages that have
productive morphological causatives and a (rich) case-marking system, but
show the following pattern when the base verb of the causative is
monotransitive: the Causer of the causative sentence has the same coding as
the subject of a non-causative monotransitive sentence, the Causee has the
same coding as the object of a non-causative monotransitive sentence, and
the object of the monotransitive base now assumes the same coding as the
indirect object of a ditransitive verb.
1. Both Matthew Dryer and Martin Haspelmath suggested that my use of
“indirect object” was confusing. I agree and should have used “R” (the
Recipient argument) to avoid that confusion.
2.1 Bill Croft suggested the following works related to the inquiry:
Comrie, Bernard. 1976. The syntax of causative constructions:
cross-linguistic similarities and differences. *The Grammar of Causative
Constructions* (*Syntax and Semantics*, Vol. 6.), ed. by Masayoshi
Shibatani, 261-312. New York: Academic Press.
Cole, Peter. 1983. The grammatical role of the causee in universal grammar.
*International Journal of American Linguistics* 49.115-133.
Kemmer, Suzanne and Arie Verhagen. 1994. The grammar of causatives and the
conceptual structure of events. *Cognitive Linguistics *5.115-56.
Croft, William. 2012. *Verbs: Aspect and Causal Structure*. Oxford: Oxford
However, I did not find any good examples for the pattern in question in
the above works. The only potential examples are several examples of
Georgian causatives cited by Comrie (1976: 282-283). As far as the present
tense is concerned, the P (patient argument of a monotransitive verb), T
(theme argument of a ditransitive verb) and R are marked by the same ending
in Georgian. As a result, one might say that (1) is an example of the
pattern in question because one might be able to argue that “secretary”
actually bears the P marker and “letter” has the R marker.
(1) Adapted from Comrie’s examples of (75) and (77)
Mama-ɸ masc̩avleblis-tvis mdivan-s c̩eril-s
father-Subj teacher-for secretary-IO letter-DO
‘Father makes the secretary write the letter to the teacher.’
2.2 Both Helle Metslang and Geda Paulsen pointed me to Paulsen’s
dissertation "Causation and dominance: a study of Finnish causative verbs
expressing social dominance" (available online at
which contains causative examples like (1b) on p. 57 of the dissertation.
However, the Finnish pattern is not the same pattern as described in the
inquiry, as also pointed out by Helle Metslang.
2.3 Guozhen Peng mentioned that the following example from Jingpo had
exactly the same pattern as what was described in the inquiry. Her original
glosses and translation were in Chinese.
(2) Ma31 pheˀ55 lă31pu31 ʃă31pu31 uˀ31!
child Obj. pants cause-wear Imperative
‘Put on pants for the kids!’
However, I am not sure whether “pants” here truly have the same coding as
the R argument of a ditransitive verb.
2.4 Jae Song pointed me to Chapter 6 of his 1996 book “Causatives and
Causation: A Universal-Typological Perspective” and mentioned that Songhai
might allow the pattern in question. However, I could not find any Songhai
example in the book that shows that pattern. I also studied Tim Shopen and
Mamadou Konaré’s article “Sonrai causatives and passives” (published in
1970 in *Studies in African Linguistics*, Vol. 1, Number 2, pp.211-254) and
could not find any example of the pattern in question in it either.
2.5 Matthew Dryer mentioned that a language that has *no case marking* on
subjects or objects could be said to literally conform to what was
described in the inquiry. He also mentioned a specific language of this
type, namely Kinyarwanda. The pattern is illustrated by (3) below, which is
example (10b) of Dryer’s work “Indirect objects in Kinyarwanda revisited”
(1983) (in *Studies in Relational Grammar I*, ed. by David M. Perlmutter,
129-140. Chicago & London: University of Chicago Press).
(3) Umugabo a-r-úubak-iísh-a abaantu inzu. (Dryer
man he-PRES-build-CAUS-ASP people house
‘The man is making the people build the house.’
3. Summary of the summary: So far, I still have not found a language that
has productive morphological causatives and a (rich) *case-marking system*and
*unambiguously* shows the following pattern: in the case of a morphological
causative based on a monotransitive verb the Causer of the causative
sentence has the same coding as the Agent argument of a non-causative
monotransitive sentence, the Causee has the same coding as the Patient
argument of a non-causative monotransitive sentence, and the Patient
argument of the monotransitive base now assumes the same coding as the
Recipient argument of a ditransitive verb. The example from Georgian does
not give us unambiguous evidence for this pattern. The one from
Kinyarwanda, though literally conforming to the pattern, does not make a
distinction between P and R in terms of case marking, either. As for why
the pattern in question is so rare, Dryer speculated that this was due to
the close relation between causatives and ditransitives. According to him,
“all ditransitive verbs can be analysed as semantically causatives, where
the R is the causee and the T the P of the original verb. So *give* is
'cause to have' or 'cause to come to have', *teach* is 'cause to learn', *
show* is 'cause to see', *tell* is 'cause to hear' or 'cause to know',
of* is 'cause to believe', *persuade to* is 'cause to decide to', and so on.
I am not suggesting that all these verbs NEED to be analysed as causatives,
rather that the approximate similarity of these verbs and these causatives
means that Rs are semantically like causees of original transitive verbs.
Hence [the strong tendency of] marking such causees like Rs makes sense.”
4. Once again, I greatly appreciate all the responses with respect to my
inquiry. I would certainly continue to welcome any pointer to a language
that unambiguously shows the pattern in question. In this regard, in
addition to examples like “The man caused John to build the house”,
examples along the line of "The man caused John to push the girl" or "The
man caused John to kick the boy" would also be very relevant.
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the Lingtyp