LISSIM 6
June 1-15, 2012@ Kangra
Selected Essays
Reciprocal Constructions in Meeteilon and Nyishi
Atanu Saha
PhD, JNU
Research statement for LISSIM stage II selection
LISSIM has always been very inspiring and motivating platform for our growing interest In Generative syntax and Semantics. The last LISSIM (Lissim 5) was important from my academic research point of view for several reasons. These are arranged below in different sub sections:-
I am pursuing my PhD in linguistics since 2009. The topic that I am interested in is ‘Reciprocal Constructions in Meeteilon and Nyishi’. My investigation in this topic involves following thingsː
What are the reciprocal markers in Nyishi and Meeteilon? Are there more than one form to show reciprocity in these languages?
The answer to the first question is the primary reciprocal marking is formulated by verbal suffixes in these languages. Meeteilon has a verbal affix –nə which could mean several things such as cumulative, associative in different contexts. In a reciprocal situation –nə is attached to the verb root to show that the construction is reciprocal.
1. mabani wari sa-na-ri (Meeteilon)
He-two talk say-RCP-ASP
‘They (two) are talking to each other’.
In case of Nyishi the structure is complex. There are two suffixes i.e. miŋ and su which are markers for cumulativity and reflexivity in the language. When these two suffixes combine with a single root then it evokes reciprocity.
2. bulue ɟeŋ-miŋ-su-pa (Nyishi)
they Pl hit- COM-RFL-CMPL
‘They hit each other.’
Does it happen with all the verbs to show reciprocity? No, it does not. As we know in terms of argument structure of the reciprocals verbs can be categorized in at least three ways.
|
verbs that are inherently reciprocal |
verbs that can be reciprocalized[1] |
verbs that can never be reciprocalized |
|
Meet |
love |
build |
|
They meet |
they love each other |
*they build each other |
There are a couple of other issues which need to be addressed here. Nyishi and Meeteilon apart from their verbal morphology also use nominal markers in certain cases. It is worth noting because even if there is lexical reciprocal marker present in the lexicon with certain verbs they cannot be used.
3. ŋule aku akugube-ham mə-d̪uŋ-d̪en (nyishi)
we-PL one one –AFTD-ACC build-CNT-ASP
‘We are building houses for each other.’
4. ay-kʰoi amana amagi Ium səri (meeteilon)
we one-NOM ONE-POSS house build-ASP
‘We are building houses for each other.’
What can be noticed from 3 and 4 is that the predicate build is reciprocalized with the help of nominal reciprocals namely aku akugube and amana amagi. These types of constructions require special attention.
Apart from this Meeteilon also has a nominal marker məsen which comes with a set of predicates.
The second important question is these markers can also co occur on many occasions
5. makhoi amana amabu khaŋnə-i
they one one-OM know-rcp-Asp
‘They know each other.’
6. bulue aku aku-ni ceŋ-miŋ-su-d̪o
they one one-to know-COM-RFL-STAT
‘they know each other.’
It seems that stative verbs and their non-stative counterparts behave differently. The difference is hypothesized as follows ː
In Meeteilon amana amabu/gi and in Nyishi aku akuam or the nominal markers function as distributive markers just like English each other when the vP selects for them.
The formulation of English sentences were given by Heim Lasnik and May 1991(a,b)
The men saw each other.
LFː [S[NP[NP the men]1 D2]2 [vp saw [NP each2 other]3]]
Along this proposal we suggest that amana amabu aku akuam works as equivalent to the covert operator D2 as in English. It’s just a case that English lacks verbal morphology but a lot of other languages do allow such constructions.
However the interpretation also depends on the argument structure of the predicate. It is an established fact that semantics of the verb classes (action, accomplishment, state) itself play important roles in deciding the actual structure. (Levin, Rapaport, Hale and Keyser Trevis 2006 among others)
Thus when a stative verb like know for instance is reciprocalized and if the reciprocal markers co occur then the object of the verb i.e. amana amagi/bu or aku akuam/ni actually quantify over the plural subjects and allow a distributive reading[2] in case of Meeteilon and Nyishi respectively.
If the verb is non stative then there are a couple of options. The predicate can either select a mass term such as məsen ‘(among)’ or the reciprocity holds as long as there is one instance of that action happens.
We take Langendoen(1978) analysis of reciprocity
∀x Є A)(Ǝy, z Є A)(x ≠ y ^ x ≠ z ^ xRy ^ zRx)
and we know that strong meaning hypothesis SMH of Dalrymple is not going to work.
Suppose if we take the predicate as hit Meeteilon allows the following
7. makhoi meʃen phuneri
they among hit-RCP-ASP
they hit each other (among themselves or within a group)
8. makhoi amana amabu phunəri
they one-NOM one-OM hit-RCP-ASP
‘Each of them hit the other.’
Interaction with Boskovic and Starke during Lissim 5
It became evident for me to follow the verb class and their argument structure as proposed in Levin and Rapaport as I sat with Boskovic and Starke for discussion. They suggested that one of the ways this research can be accomplished is to look at sets of verbs and their configuration . If Starke’s Nano syntactic analysis is correct then we can adopt his proposal in this case saying a particular verb would yield that much part of its syntactic tree which is necessary for derivation.
Inadequacy in existing research
The question which I am trying to address is to make a distinction between distributive and collective reciprocity as natural languages allow for both. If the nominal and verbal reciprocal can co occur then the semantics is somewhat gives us either a strong or a weak effect. Distributivity strengthens the reciprocity where as collectivity weakens it further.
The linearization of the reciprocals suffixes within vP is an issue which is not discussed for these languages under current discussion. If we understand the arrangements of these affixes in a correct manner then only the syntax and semantics of the reciprocal structures in Nyishi and Meeteilon can be explained.
List of cited works
Dalrymple, M., M. Kanazawa, Y. Kim, S.A. Mchombo, and S. Peters. 1998. Reciprocal expressions and the concept of reciprocity. Linguistics and Philosophy 21(2):159-210.
Heim, Irene, Howard Lasnik, and Robert May. 1991b. On ‘reciprocal scope’. Linguistic Inquiry 22:173– 192.
Langendoen, D. Terence. 1978. The logic of reciprocity. Linguistic Inquiry 9:177–197.
Levin , Beth. 1993. English verb classes and alternations: a preliminary investigation. The chicago university Press.
Levin, Beth. Malka Rappaport Hovav. 1995. Unaccusativity: at the syntax-lexical semantics interface. MIT Press.
[1] for this I am going to Use Levin‘s 1993 idea of reciprocal alternation which says several verbs go for reciprocity when another argument is attached as a pp with the verb e.g they argued with each other.
[2] hence we get a strong reading as suggested by Nishigauchi 1992
For enquiries, write to
secretary@fosssil.in
 |
Skype Interview |
|
|
Interview List |
|
|
Latest News |
|
|
LISSIM 6 Page |
|
|
Home |