Transcript Slide 1

CAUSATIVE CONSTRUCTIONS:
GENERAL INTRODUCTION AND
MAJOR ISSUES
Sajida Rehman,( [email protected])
Reading:
Paul Kroeger Analyzing Syntax: A Lexical Functional Approach,
CUP, 2004. Chapter 8.
Introduction to the presentation:
This presentation aims at:
• giving a general introduction to two different patterns of
behaviour observed in languages, with specific reference
to ‘morphological’ causatives.
•relating the behaviour of anaphoric relations within the
causative clauses to provide the evidence for the fstructure of morphological clauses
•expanding Kroeger’s study to Urdu, a South Asian
language, thus highlighting the major issues related with
the study of causation.
ANAPHORIC RELATIONS
While considering the behaviour of anaphoric pronouns(
myself, himself, themselves etc.), we should keep in mind
three major issues:
agreement
domain
prominence
It is the ‘domain’ which is of great interest for our today’s
discussion of causatives. The concept of ‘domain’ states
that for reflexive binding, a reflexive pronoun must find
an antecedent within its minimal clause.
For example:
1- Mary suspected that [ John admires himself/*herself too much.]
2- Mary waited for [John to excuse himself/*herself.]
3-I told you that [Mary would blame *myself/ /*yourself/herself.]
In English, thus, reflexive pronouns follow the rule of ‘ClauseBoundedness.’
What implications this information has for our present data
analysis?
Introduction to causation:
Causation is a ‘valence increasing’ process. In
Morphological causation, the causative form meaning
‘cause to X’ is derived from a basic predicate ‘X’ through a
regular morphological process e.g., affixation.
•Thus, in many languages, there may be just one word
conveying the meaning of two predicates, one of them being
‘cause’.
•Other causative construction types are Lexical (e.g. kill)
and Periphrastic Causatives (e.g., cause to do s.t/make do
s.t) which are both found in English. For example:
4-The player killed my cat. (Lexical Causative: just one entry
word. How many clauses the construction has?)
5- The player made my cat die. (Periphrastic Causative:
having two words to represent causation. How many clauses
the construction has?)
FURTHER ON CAUSATION:
•
In English there are NO such morphological processes for the
derivation of causatives.
•
The causer is almost always the subject of the causative
constructions.
•
Morphological causatives encode different grammatical relations
to the ‘causee’.
When the verb is intransitive from which we may derive a
causative, the causee is assigned the primary object ‘OBJ’ status.
If the verb is transitive, causee acts EITHER as:
1. The primary object (OBJ), e.g., Swahili through verb
agreement
2. The oblique (OBL using a DAT case marker), e.g.
Turkish trough case marking
•
•
FURTHER ON CAUSATION:
According to Baker(1988) and Dryer(1996), the causee status can be
predicted like this:
FIRST POSSIBILITY:-if the recipient is expressed as a primary object or
OBJ, the causee is also marked as OBJ. For example:
(6) Through verb agreement as in Swahili (Comerie,1976)
a
-Mscichana a-li-(u-)fungu-a
mlaongo.
girl
S.agr-PAST-O-agr-open-INDIC door
‘The girl opened the door.’
b -Mwalimu a-li-m-fungu-zish-a
msichana mlango.
teacher S.agr-PAST-O-agr-open-CAUS-INDIC girl
door.
‘The teacher made the girl open the door.’
(Evidence for primary status comes through ‘Passivization’ test,
The girl…./but *The door was made to be opened by the by girl by teacher.’
SECOND POSSIBILITY:- if the recipient is expressed as an oblique
(OBL)/OBJ2, the causee is also marked as OBL.
(7) Through case marking, as in Turkish (Aissen,1974; Comerie,1981)
a -Mudur mektub-u
imzala-di
director letter-ACC sign-PAST
‘The director signed the letter.’
b -Disci
mektub-u
mudur-e
imzala-t-ti
dentist letter-ACC director-DAT
sign-CAUS-PAST
‘The dentist got the director to sign the letter.’
(8) Turkish (Aissen,1974)
a -Kasap
et-I
kes-ti.
butcher meat-ACC
cut-PAST
‘The butcher cut the meat.’
b -Hasan
Kasap-a
et-i
kest-tir-di
Hasan
butcher-DAT meat-ACC cut-CAUS-PAST
‘Hasan had the butcher cut the meat.’
(Passivization: The letter/The meat was made to be cut by Hasan…/but, *Hasan
was…fails.
FURTHER ON CAUSATION
The violation of the uniqueness condition , i.e. a single clause cannot
possess more than one SUBJ, after the addition of the Causer in the
morphological causative, the rule of the ‘next available GR’ makes
causation work.
Thus, to make things simple, let’s suppose if it was possible to find
English equivalents to Turkish or Swahili causatives then our construction
should look like the following:
TURKISH:
The teacher CAUS-Open
the door-ACC
the girl-DAT.
SUBJ
OBJ
OBL
Or I hit-CAUS
Bill. ACC
John. OBL
‘I made …………… hit …………………..
SWAHILI:
The teacher
O.CAUS-open girl
Door.
SUBJ
OBJ
OBJ2
‘The teacher made the girl open the door.’
CAUSATIVE IN URDU:
If we apply the following technique, we can analyse causatives in
Urdu. Patients or the real primary objects in Urdu takes ACC-‘ko’ and
the second object like entity takes the INST(instrumental) case, which
functions as OBL. Thus,
Meine
Bill ACC. Ko
John.INST. Se (OBL) marwaya.
SUBJ.
OBJ.
OBL.
hit-CAUS-PAST
‘I made ………….. hit……………. .
• A LOGICAL QUESTION ARISES:- The English translation is
possible with two verbs, while the two languages have just one verb.
So how would it be possible to find out how many clauses each
construction possesses?
CAUSATIVE CONSTRUCTIONS WITH ANAPHORIC RELATIONS:
ENGLISH: Look more carefully at the data below:
self
self
1- I made John
[…x…… hit
Bill . ]
SUBJ
OBJ
SUBJ
OBJ
X-COMP
If replace with ‘…….self’, at John’s and Bill’s positions the results we
get are:
2- I made
John […x…… hit …himself.../*myself/. ]
3- I made
…myself….. […x…… hit
Bill . ]
Explanation of 2 above shows that clause boundedness, makes it
Possible for us to see how many clauses our constructions have. Thus,
Tests with anaphors can be a good source of evidence for determining
the number of clauses.
English causatives are bi-clausal constructions.
URDU:
The only reflexive we can replace in Urdu causative construction is
‘apne ap ko’ in position of ‘Bill.Ko’. Thus,
Meine
Bill ACC. Ko
John.INST. Se (OBL) marwaya.
SUBJ.
OBJ.
OBL.
hit-CAUS-PAST
Meine
apne aapko ACC.
John.INST. Se (OBL) marwaya.
‘I made ……john…….. hit……myself………. .
Reflexive pronoun here refers back to the SUBJ. of the whole
sentence, thus pointing out that the one verb causative constructions in
Urdu (like in Turkish) are monoclausal.
The analysis So far :
English has two verbs in causatives ,and is bi-clausal .
Turkish/Urdu have one verb and their
causatives constructions are monoclausal.
But are things so simple?
Chimwiini: (Bantu;Abasheikh,1979;cited in Marantz,1984)
Mi ni-m-big-ish-ize
John ru:hu-ya
Isg. S.agr-O-hit-CAUS-ASP John self-3sg.
‘I made John hit himself.
‘ru:hu’ is OBJ, and Chimwiini is bi-clausal.
CONCLUSION:
• Causative constructions in the world languages make
use of either one or more verbs, but the number of
clauses may not always co-ordinate with the number of
verbs present in such constructions, giving rise to either
mono-, or bi-clausal constructions.
• The bi-clausal causative constructions contain two
Subjects and two objects.
• Anaphoric relations are a good source of determining the
clause Structure of causative clauses.