• No results found

agents, or experiencers, causes and recipients.” When the subject is an

N/A
N/A
Protected

Academic year: 2022

Share "agents, or experiencers, causes and recipients.” When the subject is an "

Copied!
20
0
0

Loading.... (view fulltext now)

Full text

(1)

e-Content Submission to INFLIBNET

Subject Linguistics

Principal Investigator Prof. Pramod Pandey

Centre for Linguistics, SLL&CS,

Jawaharlal Nehru University, New Delhi-110067 pkspandey@yahoo.com

011-26741258, -9810979446

Paper Linguistic Typology and Language Universals

Paper Coordinator

Prof. Kārumūri V. Subbāro

University of Delhi (Rtd.), Delhi 110007 Postal address:

13/704 East End Apartments, MayurVihar Phase I Extension Delhi 110096

kvs2811@gmail.com

+91-11-4309-4675; 98-688-69904

Module title Non-nominative subjects in South Asian languages

Module ID Lings_P13_M12

Content Writer Prof. Kārumūri V. Subbāro

University of Delhi (Rtd.), Delhi 110007

Email id kvs2811@gmail.com

Phone +91-11-4309-4675; 98-688-69904

Reviewer

Prof. Probal Dasgupta

Linguistic Research Unit, Indian Statistical Institute 203 Barrackpore Trunk Road, Kolkata 700108

(2)

Non-nominative subjects in South Asian languages

Aim: To discuss in detail the nature of the non-nominative subject construction in South Asian languages

Objectives:

(i) Cases of non-nominative subjects in South Asian languages

(ii) Domains of occurrence of the non-nominative subject constructions (iii) The [-transitive] nature of the predicate in non-nominative subjects

(iv) Subjecthood of non-nominative subject constructions – Evidence presented (v) Conclusion

Non-nominative subjects in South Asian languages

12.1 In the module on ergativity, we’ve seen that the subject may be in a different case from the usually-found nominative case in many languages in the world. We’ve shown that the subject may be ergative case-marked depending on the nature of the predicate. In this module we shall see that the subject in South Asian languages may be dative, genitive, accusative, instrumental case-marked depending on the nature of the predicate. We shall show that the afore- mentioned non-nominative subjects exhibit syntactic properties of subject with regard to anaphora and control. When the subject is non-nominative case-marked, the verb agrees with the patient or theme. In most of the Indo-Aryan languages non-nominative subjects do not trigger agreement, though in some Eastern Indo-Aryan languages, they do. This module is organized as follows:

Section 1 is Introduction. Section 2 presents cases of non-nominative subjects from South Asian languages with suitable examples. Section 3 discusses the domains of occurrence of non- nominative subject constructions. Section 4 provides arguments in support ‘subjecthood’ of these non-nominative subject constructions. In Section 5, we show that the predicate in a non- nominative subject construction is [-transitive] and Section 6 is the conclusion.

1. Different casemarkings in non-nominative subject constructions in South Asian languages

1.1 The Subject is dative case-marked

In all Dravidian languages and in most of the Indo-Aryan languages the subject is dative case-marked when the predicate manifests possession, or if it is a psychological predicate.

(3)

Tamil (Dravidian)

1. kumāru- kku anta.p peṇṇ- ai.p piṭi- ttu Kumar- dat that girl- dat like- pst

‘Kumar liked the girl.’

(Lehmann 1989:187) Marathi (Indo-Aryan)

2. mula- la dukkha dzāla

boy- dat sadness.3sg.neut happen.pst.3sg.neut

‘The boy became sad.’ (lit: ‘Sadness happened to the boy.’

(Pandharpandey:1997:285)

The predicate piṭ ‘like’ that expresses a psychological state and emotion in (1) and the psych-predicate dukkhdzā ‘sadness happen’ in (2) require the subject of the sentence to be in the dative case in Tamil and Marathi respectively. Hence, the subject of the sentence, which is the first NP of S (Chomsky 1965) is dative case-marked.

One might wish to ask: If the subject in (1) and (2) for example is not nominative case- marked, and if is dative case-marked as in (1) and (2) above for example, why should we treat such subjects as subjects? We shall show later that such non-nominative case- marked subjects exhibit all the syntactic properties that are attributed to subjects, except for agreement in some languages.

In Telugu and Kannada (Dravidian) and in Hindi-Urdu, Punjabi, Kashmiri (Indo-Aryan) the theme NP in the Dative Subject construction is nominative case-marked. The following examples are illustrative.

Hindi-Urdu (Indo-Aryan)

(4)

3. kumār- ko [vah laṛkī]/ *[us laṛkī ko] Acchī Lagī Kumar- dat that girl f.sg/ that girl acc good.f.sg feltf.sg

‘Kumar liked the girl.’

Telugu (Dravidian)

4. kumār- ki [ā ammāyi]/ *[ā ammāyi ni] nacc.in.di Kumar- dat that girl f.sg/ that girl acc like/pst.f,sg

‘Kumar liked the girl.’

Note that in (3) and (4) the patient girl is in the nominative case, and it cannot be accusative case-marked. A dative case marker too is not permitted. As we shall demonstrate later, such predicates in non-nominative subject constructions are [- transitive], and hence, they are incapable of assigning accusative case either to a patient such as girl, people etc. or to a theme such as book, house etc.

When the subject is dative case-marked , and when the patient/theme is in the nominative case, the verb agrees in number and gender with the patient/theme and not with the dative case-marked subject, because, non-nominative subjects are incapable of triggering agreement in most of the Indo-Aryan and Dravidian languages.

The question that arises is: What happens in languages like Tamil, when the subject and patient/theme are both case-marked dative and accusative respectively ? In such cases the verb does not exhibit with any NP in the sentence, as case-marked NPs cannot trigger agreement in Dravidian and Indo-Aryan. In Tamil the verb exhibits default agreement, and the default agreement in Tamil is in the neuter.

2. In this section we shall discuss the domains of occurrence of the non-nominative subject constructions.

(5)

Domains of occurrence of the dative subject: (adapted from Subbarao and Bhaskararao 2004)

a. Psychological states and emotions b. Physiological and mental ailments c. Natural phenomena pertaining to body d. Perceiver of visual and auditory actions e. To express possession and kinship

f. Subject of predicates expressing obligation and necessity (desiderative) g. To denote a recipient

h. Acquisition of knowledge or a skill

For a detailed discussion of the nature of these predicates, see Subbarao (2012).

We shall present a few examples to explicate some of the domains.

(i) Psychological States and Emotions

In Assamese, the subject is dative case-marked, while in many Indo-Aryan and all Dravidian languages it is dative case-marked.

(ii) Assamese (Indo-Aryan)

5. tar khɔng uth.i- sil

he/she.gen anger (nom) rise.cpm- pst

‘He/she is angry.’

(Subbarao 2012)

In Malayalam (Dravidian), the subject is dative case-marked.

(6)

Malayalam (Dravidian)

6. avan. ə kōpam/dēšyam wannu he dat anger (nom) came

‘He got angry.’

(Aiswaria G. Shajan) Possession-[+concrete]

To express [+concrete] possession the subject in Hindi-Urdu (Indo-Aryan), the subject is locative case-marked, in Mizo, the subject is ergative case-marked, while in Bangla, it is genitive case-marked. Note that the theme bahutpaisā ‘a lot of money’ is [+concrete]. In Hindi-Urdu, and in most other Indo-Aryan languages, in sentences that express possession, it is the nature of the patient/theme that is instrumental in the assignment of the case marker to the possessor.

Hindi-Urdu (Indo-Aryan)

7. us amīr ādmī ke pās bahut paisā hai that rich Man gen near a lot of money is

‘That rich man has a lot of money.’

In Mizo, the subject is ergative case-marked because the verb nei ‘have’ is [+transitive].

Mizo (Kuki-Chin, Tibeto-Burman) and the other Kuki-Chin languages and most other Tibeto-Burman languages have two distinct verbs: One to denote possession and the other to denote existence.

(7)

In Mizo the verb for locational existence isɔm ‘be [+location]. In contrast, no IA or Dravidian language has a verb that specifically denotes possession. Note that the verb agrees with the ergative case-marked subject and the third person agreement marker a- is coindexed with lali-n ‘Lali-erg’.

Mizo (Tibeto-Burman)

8. [lali- nDP]i thiŋ.tlaŋ- ah dͻr kaai- nei

Lali- Erg village- Adv shop 31sg- have

Lali has a shop in the village.

(Lydia Khiangte 2014)

In sentences expressing possession, Telugu makes a distinction in case marking on the subject. If it is temporary possession, the subject is case-marked bythe locative postposition daggara/daggira ‘near’, and if it is permanent possession, it is case-marked by the dative.

Telugu (DR)

9. prastutam pratāpi Daggara ḍabbuluj lē- vu*i,j

at.present Pratap Near money.3p,nm be.not- 3p,nm

‘At present Pratap does not have any money.’

(Subbarao 2012: )

(8)

In Dravidian languages, the subject is case-marked dative in all the domains mentioned above.

Possession- Kinship

In Dravidian languages the subject is dative case-marked, when possession related to kinship is expressed. “The subject is genitive case-marked when kinship relationship is expressed in Hindi-Urdu, Punjabi, Bangla (IA) and Kokborok (TB); and dative case- marked in Dravidian languages and in Kashmiri (IA).” (Subbarao 2012:149).

Natural phenomena pertaining to body

In Hindi-Urdu, Punjabi and in all Dravidian languages , the subject is dative case-marked, while in Bangla and Assamese, it is genitive case-marked.

Hindi-Urdu (IA)

10. us ādmīi ke sāre bālj jhaṛ gayej

that man gen all Hair fall went

‘That man lost all his hair.’

Telugu (DR)

11. ā maniṣii ki Anta juttuj- u1

that man dat so much hair- quantifier mkr ūḍ- i pō- yindij

1 In Telugu (DR) -u also functions as a marker to impart quantificational interpretation. (Mayuri 2012).

(9)

fall- cpm go- pst.agr

‘That man lost all his hair.’

(Subbarao 2012)

Mizo (Tibeto-Burman)

12. lali- n thi a- Nei

Lali- Erg blood 3sg- Have

‘Lali is having her periods.’

13. lali- n nau/naute a- Nei Lali- Erg child/baby 3sg- Have

‘Lali is having a baby.’

Expression of Obligation

The subject is dative case-marked in obligative constructions in Bangla, just as in Hindi- Urdu and Punjabi.

(iii) Bangla (IA)

14. rina- ke aj pãc ṭa Dokane jete holo Rina- dat today five Cl shops.loc go.inf was

(10)

‘Rina had to go to five shops today.’

(Dasgupta 2004: 130) Hindi-Urdu (Indo-Aryan)

15. rīnā ko bagīce mẽ jā.nā Thā Rina dat garden in go.inf was

‘Rina had to go to the garden.’

In Dravidian languages the subject is nominative case-marked.

Malayalam (DR)

16. nī pōyē tīrū you go-emph must

‘You really must go.’

(Asher and Kumari 1997: 307)

In this section we’ve seen how varied the case marking on the subject is in the expression of various domains in which the non-nominative subject is. In the following sections we shall show that (i) the predicate in a non-nominative subject construction is [-transitive], and (ii) that the theme and the predicate compositionally assign the non- nominative subject case.

3. The nature of predicate in non-nominative subject constructions

In all Indo-Aryan languages except in Bangla and Assamese, and in Telugu and Kannada (Dravidian), the theme is invariably in the nominative case, and it cannot be

(11)

marked accusative. The theme is marked nominative Hindi-Urdu and Telugu, as (G ) and H in Hindi-Urdu and I- J in Telugu below illustrate.

Hindi-Urdu

17. bacco~ ko Zukām ho ga.yā

children dat cold (noun.nom,m, sg)

become go.pst,m,sg)

‘The children got cold.’

18. *bacco~ ko Zukām ko ho ga.yā children dat cold (noun) acc become go.pst

‘The children got cold.’ (intended meaning)

19. pillala- ki Jalubu cēs- in- di

children dat cold (noun.nom,3sg,nm) do- pst- 3sg,nm

‘The children got cold.’

20. *pillala- ki Jalubu ni cēs- in- di children dat cold

(noun.nom,3sg,nm)

acc do- pst- 3sg,nm

‘The children got cold.’ (intended meaning)

(12)

In Tamil and Malayalam (Dravidian) and Bangla and Assamese, the theme may be marked accusative. In Subbarao (2012) arguments were presented to show that the accusative marker is a specificity marker and hence, it is not a marker that marks an object as accusative.

Let us take the case of Malayalam.

In Malayalam in the Dative Subject construction, the theme is in the nominative case, when it is [-animate] and [-definite]. The accusative marker ye does not occur in such cases as M shows.

21. en- ik’k’ə oru Māŋŋa vēṇam I- dat one mango.nom want

‘I want a mango.’

(Jayaseelan 2004: 234)

When the theme is [+specific], then the accusative marker occurs as N shows.

Malayalam (DR)

22. kuṭṭi- k’k’ə āna- ye išṭam āyi child- dat elephant- acc liking became

‘The child liked the elephant.’

(Jayaseelan 2004: 229)

The features animacy and definiteness play a crucial role in the occurrence of the

(13)

accusative marker ye, though the predicate in non-nominative subject constrictions in [-transitive] in nature.

In Bangla (IA) the theme is which is [+animate] is accusative case-marked as in (146).

23. tomar2 ka.ke cai you.indirect who.acc want

‘Whom do you want?

(Dasgupta 2004:135)

Bangla learners/speakers of Hindi-Urdu assign accusative case marker to the theme.

while speaking Hindi-Urdu.

Hindi-Urdu spoken by a Bangla learner/speaker 24. *āp Ko kis Ko cāhiye

you Dat who Acc needed

‘Who do you want?’ (intended meaning) (Subbarao 2012:173)

In standard Hindi-Urdu (IA), in such cases the themekaun‘who’ cannot be accusative marked, and it is in the nominative case, as the predicate cāhiye ‘needed’ that takes a dative subject is [-transitive].

2Dasgupta (2004:131) points out that toma.r which is generally glossed as ‘you.gen’ should be treated as an indirect case form. For a discussion on this, see Dasgupta, Ford and Singh (2000).

(14)

Hindi-Urdu (IA)

25. āp ko kaun cāhiye you dat who. nom needed

‘Who do you want?’

In fact, as Alice Davison (pc) correctly points out: dative predicates are semantically transitive but syntactically intransitive. (See Subbarao 2012).

Another feature of the non-nominative subject that deserves mention is the semantic nature of the predicate in a non-nominative subject construction.We shall demonstrate that the predicate in all non-nominative subject constructions is [-volitional].

Davison (2015) points out: “

Subjects often have reference to agency and volition,

but not always: the subject may refer to an experiencer, a cause or even a

recipient, and in the passive, to a theme or patient. Ergative subjects may refer to

agents, or experiencers, causes and recipients.” When the subject is an

experiencer, possessor or recipient, the passive subject, or when the predicate

belongs to any of the other semantic domains mentioned above, the predicate is [-

volitional], and it lacks the feature of agentivity. Any [-volitional] predicate

cannot permit an agent that performs a deliberate action. That is why adverbs

such as

deliberately, intentionally, volitionally, willingly, willfully, purposefully, consciously, volitionally, voluntarily

cannot occur in a non-nominative subject

construction. The following examples are illustrative.

(15)

Hindi-Urdu (Indo-Aryan)

26. *pratibhā ko jānbujhkar bukhār āyā Pratibha dat deliberately fever came

‘Pratibha got the fever deliberately’

Telugu (Dravidian)

27. *saundarya ki kāvāli.ani kōpam vacc.in.di Saundarya dat deliberately anger came

‘Saundarya deliberately got angry.’

Literally: ‘Anger came to Saundarya deliberately.’

What we’ve attempted to demonstrate thus far is that a non-nominative subject does not semantically have the feature of volitionality, and hence, all predicates in a non- nominative subject constructions except for the ones in an ergative construction are [- volitional].

The crucial questions that arises now is: Are non-nominative subject really the subjects, and do they have any properties of a subject? In the following section we wish to show that though the non-nominative subject is not nominative case-marked, it exhibits all the properties of subject in the nominative case.

4. Subject properties of the non-nominative subject

The crucial property of a subject is that it is the first NP of S, it can be an antecedent to an anaphor, it can be a controller of PRO, and it can be a controller of PRO in a

(16)

conjunctive participial clause.3 In this section we show that a non-nominative subject exhibits all the properties that a nominative subject does, and hence, a non-nominative subject qualifies itself to be treated as subject.

5.1 Non-nominative subject as an antecedent to an anaphor

In the module on anaphora, we’ve provided evidence to demonstrate that subject is the antecedent to an anaphor in all South Asian languages. If a non-nominative subject too can be an antecedent to an anaphor, then, it should be treated as a subject.

Non-nominative subject as antecedent to a possessive anaphor

In T below from Hindi-Urdu the possessive anaphor apneapne ‘of selves’.m.pl’ has the dative subject bacco~ ko‘children dative’ as its antecedent, and in (P) below from Tamil (Dravidian), the nominal anaphor tan meleyetana.kku ‘self on self dat’

has the dative subject mālati-kku‘Malati-dat’ as its antecedent.

Hindi-Urdu

28 bacco~ ko apne apne ghar jā.nā paṛā

children dat selves’.m.pl selves’.m.pl houses go.inf Fell

‘The children had to go to their own homes.’

Tamil (Dravidian)

29 mālati- kku tan meleye (tana.kku) kōpam va.nt.atu

Malati- dat self on self.dat anger, 3,sg, neut came, 3,sg,neut

‘Malati got angry on herself.’

3For further details and supporting evidence, see Subbarao (2012).

(17)

(Parameswari Krishnamoorthy, p.c.)

tan meleye tanakku ‘self on self.dat’ is a complex anaphor that occurs in the direct object position, and it has the datve subject mālati-kku ‘Malati-dat’ as its antecedent. The verb exhibits 3 person singualar neuter agreement agreeing with the theme kōpam ‘anger’.

This shows that the non-nominative subject which is is dative case-marked is identical to a nominative subject in its capacity as an antecedent.

5.2 Non-nominative subject as Controller of PRO

Just as a nominative subject subject can be the contoller of PRO, so can be the dative or genitive case-marked non-nominative subject. Recall that Bangla has genitive case- marked subjects with psych-predicates. The genitive case-marked subject acts the contoller of PRO in X which is coindexed šɔb bacca ‘all children’.

Bangla (Indo-Aryan )

30. šɔb baccai- r- i [PROi sãtar kaṭ- te] bhalo lage all child- gen- emph to swim cut- imperfpple good strikes

‘All children like to swim.’

(Shukla Basu pc) (Subbarao 2012:157)

5.3 Non-nominative subject as controller of PRO in a CP-clause

In the module of ‘India as a Linguistic Area’, we’ve discussed the nature of the conjunctive participle, which is found in all South Asian languages. The subject of the conjunctive participle is PRO, which is uncase-marked ungoverned. A non-nominative subject can be the controller of PRO in a conjunctive partcipial construction.In ( ) below, Pragati is dative case marked and the subject of the conjunctive participle is PRO. PRO is

(18)

coindexed with Pragati. The dative subject pragati ko ‘Pragati dat’ is the contollr of PRO.

Hindi-Urdu (Indo-Aryan)

31. pragati Ko [PRO yah bāt sun kar] behad khušī huī

Pragati Dat this news hear cpm a lot of happiness happened

‘Having heard this news Pragati felt very happy.’

Telugu (Dravidian)

32. pragati Ki [PRO ī sangati vin- i] cālā santōšam kaligindi Pragati Dat this news hear cpm a lot of happiness happened

‘Having heard this news Pragati felt very happy.’

Just as a nominative subject, a non-nominative subject too can be the controller of PRO.

We shall demonstrate that just as a nominative subject subject can be the contoller of PRO, so can be the dative or genitive case-marked non-nominative subject in a conjunctive participial construction, in a particial adverbial clause.

A non-nominative subject can control a PRO that occurs in a dative /genitive subject clause in Indo-Aryan. (Subbarao 2012). In Korean too such a situation obtains.(Yoon, 2004: 266)

Hindi-Urdu (IA)

(19)

33. [PROi Is bāt kā patā cal- tehῑ]

dat this news gen knowledge go- right after laṛkiyõi Ko ghabṛāhaṭ huῑ

girls Dat panickiness happened

‘As soon as PROi (the girls) came to know of this, the girlsi got panicky.’

Telugu (DR)

34. [PROi ῑ sangati teliya- gānē]

dat this news know- right after

ammayili-i ki kangāru puṭṭindi girls- dat panickiness was borne

‘As soon as PROi (the girls) came to know of this, the girlsi got panicky.’

The predicates patā cal.nā ‘to come to know’ in Hindi-Urdu (Indo-Aryan) and telus

‘know’ in Telugu (Dravidian) require a dative subject. In (33) and (34) above, PRO occurs in the dative subject position.

The following Tamil examples further support our point.

(35) Pragatiintaceyti.y-ai.kket-tu cantoSappat- t- aaL Pragati this news-ACC hear-CPM {feel happy}- PST- 3.SG.F.

‘Having heard this news Pragati felt very happy.’

(36) intaceyti.y-ai.kket-tt- at- um Pragati cantooSamatai-nt-aaL

this news- ACC hear- ADJL- 3.sg.n- CONJ Pragati- DAT happy attain-PST-3.F.M ‘Having (as soon as) heard this news Pragati felt very happy.’

(20)

(37) intaceyti.y-ai.kket-tt- at- um Pragati-kku cantooSam this news- ACC hear- ADJL- 3.sg.n- CONJ Pragati- DAT happy ‘Having (as soon as) heard this news Pragati felt very happy.’

Conclusion:

In this module we have demonstrated the non-nominative subject behaves like a

nominative subject with regard to several syntactic phenomena. It can be an antecedent to ana anaphor and a controller of PRO.

References

Related documents

Failing to address climate change impacts can undermine progress towards most SDGs (Le Blanc 2015). Many activities not only declare mitigation targets but also cite the importance

The Use of Performance-Based Contracts for Nonrevenue Water Reduction (Kingdom, Lloyd-Owen, et al. 2018) Note: MFD = Maximizing Finance for Development; PIR = Policy, Institutional,

This report provides some important advances in our understanding of how the concept of planetary boundaries can be operationalised in Europe by (1) demonstrating how European

The Congo has ratified CITES and other international conventions relevant to shark conservation and management, notably the Convention on the Conservation of Migratory

INDEPENDENT MONITORING BOARD | RECOMMENDED ACTION.. Rationale: Repeatedly, in field surveys, from front-line polio workers, and in meeting after meeting, it has become clear that

➢ Application Generation: Automated tools are used for the construction of the software, to convert process and data models into prototypes. ➢ Testing and Turnover: As prototypes

While Greenpeace Southeast Asia welcomes the company’s commitment to return to 100% FAD free by the end 2020, we recommend that the company put in place a strong procurement

Women and Trade: The Role of Trade in Promoting Gender Equality is a joint report by the World Bank and the World Trade Organization (WTO). Maria Liungman and Nadia Rocha