Streams Converging Into an Ocean, 205-238 2006-8-005-008-000009-1
Case Marking System in Plngawan Atayal*
Lillian M. Huang Shih Chien University
Atayal is one of the Austronesian languages spoken in Taiwan. It consists of two major subgroups, namely, Squliq and C’uli’. The present paper attempts to provide a description of the case marking system (including pre-nominal case markers and the nominal case marking suffix) of Plngawan Atayal, a C’uli’ dialect spoken in Jen-ai Prefecture, Nantou County. Its population is around 1000, with less than one-third of the people being able to speak the dialect. The case markers of Plngawan Atayal will be examined in terms of their morphological forms, syntactic distribution, and syntactic/semantic functions. Furthermore, a brief comparison of the case marking systems of Wulai Atayal (Squliq), Mayrinax Atayal (C’uli’) and Plngawan Atayal will be presented.
Key words: Austronesian, Atayal, Plngawan, C’uli’, case markers
1. Introduction
Atayal is one of the Austronesian languages spoken in Taiwan. It consists of two major subgroups, namely, Squliq and C’uli’. The former is considered to be more innovative and has lost a lot of information for historical reconstruction, whereas the latter is regarded as being more conservative and useful for historical and comparative reasons. The dialectal variant investigated in this paper is that of Plngawan, a C’uli’ dialect spoken in Jen-ai Prefecture, Nantou County. Its population is around 1000, with less than one-third of the people being able to speak the dialect. The data upon which this paper is based were collected between December 2004 and December 2005, and the major informant is Mr. Watan Nawi.1
* It is my great pleasure to present this paper in honor of Professor Paul Jen-kuei Li on his 70th birthday. The paper inidcates part of the results of an on-going research project supported by the National Science Council (NSC 94-2411-H003-030). Here I would like to thank Dr. Elizabeth Zeitoun for her valuable comments on the earlier version of the paper. I am, of course, responsible for any remaining errors. 1 Mr. Watan Nawi, whose Chinese name is Nai Ming-dong (乃明東), was born in 1946. He grew up in a family where only Plngawan Atayal was spoken; consequently, he didn’t learn Mandarin until he served in the army in 1967. He is currently teaching the named dialect in Chin-ai
Lillian M. Huang
Plngawan Atayal has rarely been investigated. Not only the available works are very limited (e.g. Li 1985, Rau 2004), but none of them is wholly devoted to the study of the dialect; that is, only certain portions of the papers are related to the dialect and they mainly concern its phonological (or phonemic) aspect. The present paper attempts to provide a description of the case marking system of the named dialect, including pre-nominal case markers and a nominal case marking suffix. The pre-nominal case markers will be examined in terms of their morphological forms and syntactic as well as semantic functions. Furthermore, a brief comparison of the case marking systems of Wulai Atayal (Squliq), Mayrinax Atayal (C’uli’) and Plngawan Atayal will be presented, with regard to the number of sets of cases, their morphological and semantic/functional similarities and differences. It is hoped that this work will help Atayal speakers understand something about how their language manifests case relationship. It is also hoped that interested non-Atayal speakers and linguists will find this study useful for learning about the language.
2. Case marking system in Plngawan Atayal
Case relationship in different languages can be indicated by different ways, such as by pre-nominal markers, by nominal affixes, or even by verbal affixes.2 Like most Formosan languages, Plngawan Atayal has its nouns preceded by case markers3 though they are not always required to be present, the situation of which resembles some of the other Atayal dialects like Wulai Squliq, but differs from some like Mayrinax Atayal. These case markers function as the morphosyntactic device to designate the case roles and/or grammatical functions of noun phrases in sentences. In other words, these markers generally fulfill two functions: (i) designating semantic properties of the nouns/noun phrases they precede, and (ii) indicating their syntactic roles. In addition to such pre-nominal case markers, Plngawan Atayal, unlike all the other Atayal dialects, also utilizes a nominal affix to designate case relation, as will be discussed shortly. Below, let us first consider the Plngawan case system as presented in Table 1a, with the dot lines indicating that the restriction of using certain case markers for proper
Elementary School, Nantou County. 2 For example, in Mantauran Rukai, human objects are marked through the cliticization of -in on the noun or on the verb. For details, please refer to Zeitoun (forthcoming). 3 This is a term adopted by many Formosan linguists such as Huang (1995), Huang et al. (1998), and Zeitoun (1995, 1997a). Different terms used for such pre-nominal markers are also found, e.g. “relation markers” by Tsuchida (1976), “construction markers” by Ferrell (1979), and “determiners” by Starosta (1988).
206
Case Marking System in Plngawan Atayal nouns (including kinship terms and Neutral pronouns)4 and common nouns is not that significant any longer.
Table 1a: Case markers in Plngawan Atayal Case Nom Acc Gen Ben Ins Com Noun Proper ka ci ni ni --- ci Common Animate cika na na --- cika Inanimate naka --- na ---
Next, let us examine the following two texts with the pre-nominal case markers and the nominal case suffix underlined to examine the functions of these markers:
Text 1: saaa ci ke na itaal ‘Learning our Atayal language’5 1. ma-as=cu saaa ci ke ni yaya=mu nak. [AF-happy=1S.Nom learn.AF Acc word Gen mother=1S.Gen self] 2. k
4 Please refer to Appendix 1 for the complete personal pronominal system in Plngawan Atayal. 5 The abbreviations and symbols used in this paper are as follows. 1S: First person singular; 1PE: First person plural exclusive; 1PI: First person plural inclusive; Acc: Accusative case; Act: Active; AF: Agent focus; Aux: Auxiliary; Ben: Benefactive case; BF: Benefactive focus; Com: Comitative case; Dat: Dative case; Dist: Distal; Dyn: Dynamic; Fin: Finite; Gen: Genitive case; IF: Instrument focus; Imp: Imperative; Ins: Instrument case; Irr: Irrealis; LF: Locative focus; Loc: Locative case; NAF: Non-agent focus; Neg: Negation; Neu: Neutral; Nfin: Non-finite; Nom: Nominative case; Nrf: Nonreferential; Obl: Oblique; Part: Particle; PF: Patient focus; Prf: Perfective aspect; Prog: Progressive aspect; Prox: Proximal; Q: Question; Real: Realis; Red: Reduplicate; Rf: Referential; Top: Topic; < >: indicating the enclosed element is infix or its gloss; =: indicating the following bound pronoun is a clitic.
207
Lillian M. Huang
7. pawas-an=cu=naha ha. [sing-LF=1S.Nom=3P.Gen temporarily] 8. “ule minubah, ana=mamu m-usa inu, [child following ever=2P.Nom AF-go where 9. iya ui ci ke=ta. [Neg.Imp forget.AF Acc word=1PI.Gen] 10. ke hani a, kakinuxan=ta, iwa na hamalit ni yaya [word this Top future:life=1PI.Gen equivalent to tongue Gen mother yaa.” [father] 11. m-u=cu ci awas=naha la a, [AF-listen=1S.Nom Acc song=3P.Gen Part Top 12. lunlu-un=mu ka lak-yaki lak-yutas=mu la. [miss-PF=1S.Gen Nom deceased-grandma deceased-grandpa=1S.Gen Part] 13. m-ael=cu malahan a, sipel-un=mu ka laha. [AF-sleep=1S.Nom midnight Top dream-PF=1S.Gen Nom 3P.Neu] 14. yo na nyel pasa maha, [like Prog.Prox teach.AF say.AF 15. “iya=ta ui ka ke=ta!” [Neg.Imp=1PI.Gen forget.PF Nom word=1PI.Gen] ‘I am happy to learn my mother tongue. When I speak my language, my heart is warm. When my grandma and grandpa taught me the Atayal language, I listened very carefully and learned to speak it. Sometimes when I was lazy to learn, they would sing to me saying, “Child, wherever you go, don’t forget our language. This language is the future (hope) of our life; it is like our parents’ tongue.” When I heard their song, I would think about my deceased grandparents. When I sleep at midnight, I would dream of them. It’s like that they are teaching me by saying, “Don’t forget our language!”’
Text 2: Taken from rumui cu musa pastana yaan mu la ‘I forgot to pick up my father’ 1. m-aha=cu pastana yaa-n=mu. [AF-go=1S.Nom pick:up.AF father-Acc=1S.Gen] 2. cyel m-umarah calak. [Prog.Dist AF-work water:field] 3. uat ka aulu a-usa pastana hiyan [Neg Nom car Red-go pick:up.AF 3S.Loc
208
Case Marking System in Plngawan Atayal
4. m
In the above two texts, we may notice four case markers ka, ci, ni and na, and the nominal case marker -an. The functions of these markers and the other case markers will be discussed below.
2.1 Nominative case marking
Like the other Atayal dialects (cf. Rau 1992, Huang 1993 & 1995a), Plngawan Atayal has its verb appearing sentence initially and grammatical subject appearing sentence finally. The marker ka preceding the grammatical subject of a sentence is thus treated as the Nominative case marker in the dialect, as illustrated in Text 1.
2.1.1 Nominative case marker and semantic roles of the subject
The grammatical subject may manifest various semantic roles. For instance, in lines 12, 13 & 15 in Text 1 above, the grammatical subject of each clause (i.e. lakyaki lakyutas mu ‘my deceased grandparents’, laha ‘they’, and ke ta ‘our language’) happens to manifest the Patient participant to the event of ‘missing’, ‘dreaming’ and ‘forgetting’ respectively. However, the grammatical subject can manifest the Theme or Agent participant, as illustrated by aulu ‘car’ and mama mu ‘my uncle’ in lines 3 & 6 in Text 2, respectively. More examples with the grammatical subject manifesting the Agent participant are given below.
209
Lillian M. Huang
(1) a. m-asye ka watanagent [AF-laugh Nom Watan] ‘Watan is laughing’
b. m-asye ka yaa=muagent [AF-laugh Nom father=1S.Gen] ‘My father is laughing’
In addition, the grammatical subject may manifest Location (e.g. (2a)), Beneficiary (e.g. (2b)), and Instrument (e.g. (2c)) participant of an event.6 These different semantic roles are indicated by the verbal forms (i.e. the Patient-focus [PF] marker -un in line 12 in Text 1, the Agent-focus [AF] marker -um- in line 6 in Text 2 and m- in examples (1a-b) above, the Locative-focus [LF] marker -an in (2a), the Beneficiary/ Instrument-focus [BF/IF] marker si- in (2b-c)), which is beyond the scope of the present paper and will not be further discussed here.
(2) a. in-oh-an=mu ka yoriculocation la [Prf-go-LF=1S.Gen Nom Miaoli Part] ‘I have (already) been to Miaoli’
b. si-cau ci ucyux na nakis ka yayabenefactee=mu [BF-wrap Acc fish Gen old:man Nom mother=1S.Gen] ‘The old man wrapped fish for my mother’
c. si-pahpuy ni yaya=mu ka elohinstrument [IF-cook Gen mother=1S.Gen Nom charcoal] ‘My mother cooks with charcoal’
One more thing to point out is that the Nominative case marker ka in Plngawan Atayal can precede either proper nouns (e.g. (1a-b) and (2a-b)) or common nouns, as shown in (2c) above, and (1c) and (2d) below:
(1) c. m-asye ka nakisagent [AF-laugh Nom old:man] ‘The old man is laughing’
(2) d. si-cau ci ucyux na nakis ka kanelbenefactee hani [BF-wrap Acc fish Gen old man Nom woman this] ‘The old man wrapped fish for this woman’
6 In daily conversation, this case marker, along with others, is sometimes omitted, especially in young generation’s speech.
210
Case Marking System in Plngawan Atayal
The situation found in Plngawan Atayal above differs from that in Mayrinax Atayal (Huang 1995a:88, 91), where, for example, the Nominative case marker i can only precede personal proper nouns including kinship terms (e.g. (3a-a’)), and a/ku only common nouns (e.g. (3b-c)):
(3) Mayrinax Atayal a. m-asiaq i aay [AF-laugh Nom aay] ‘aay is laughing’ a’. m
In Wulai Atayal, on the other hand, it is noticed that the Nominative case marker used for common nouns (i.e. qu) may take the place of the one used for proper nouns (i.e. i). Compare the following sentences:
(4) Wulai Atayal a. m-syaq i/qu/ø aay [AF-laugh Nom aay] ‘aay is laughing’ b. m-ahuq lukus=myan krryax i/qu/ø yaya=mu [AF-wash clothes=1EP.Gen every:day Nom mother=1S.Gen te syaw gong. [Loc side river] ‘My mother washes our clothes at the river every day’ c. laq alay qu/ø/*i kneril qani. m-srai=sami. [good.AF very Nom woman this AF-make:friend=1PE.Nom] ‘This woman is very good; we have become friends’
As illustrated in (4a-b), proper nouns may be preceded by either i or qu or nothing, whereas common nouns can only be marked by qu or nothing, as exemplified in (4c).
211
Lillian M. Huang
To sum up the above discussion, Table 2 presents the Nominative case markers in the three named Atayal dialects (the dot lines indicating that the restriction of using certain case markers for proper nouns and common nouns is not very significant any longer):
Table 2: Nominative case markers in three Atayal dialects
Dialect Noun Plngawan Atayal Mayrinax Atayal Wulai Atayal Proper i i Common Non-referential ka a qu Referential ku
2.1.2 Nominative case marker and demonstratives
In addition to preceding proper and common nouns, the Nominative case marker ka can precede demonstrative pronouns, both proximal hani and distal haca. Examples follow:
(5) a. ima patas ka hani [who book Nom this] ‘Whose book is this?’ b. na-ni-un=mu ka haca [Red-eat-PF=1S.Gen Nom that] ‘I want to eat that’ c. in-oh-an=mu ka hani la [Prf-come-LF=1S.Gen Nom this Part] ‘I have been here (already)’ d. in-sal-an=mu ka haca la [Prf-go-LF=1S.Gen Nom that Part] ‘I have been there (already)’
Moreover, in Plngawan Atayal, the Nominative case marker and the demonstrative pronoun7 may merge into one word (i.e. ka hani > kani; ka haca > kaca), as presented below:
7 Though demonstrative pronouns and demonstrative adjectives share the same form in Plngawan Atayal, the merger of case markers with demonstrative pronouns seems to only occur in places where the demonstrative replaces the noun (i.e. demonstrative pronouns), but not in places where it modifies the noun (i.e. demonstrative adjectives). Consequently, the following phrase ka patas hani ‘this book (Nom)’ cannot be merged into *patas kani.
212
Case Marking System in Plngawan Atayal
(5) a’. ima patas kani [who book this.Nom] ‘Whose book is this?’ b’. na-ni-un=mu kaca [Red-eat-PF=1S.Gen that.Nom] ‘I want to eat that’ c’. in-oh-an=mu kani la [Prf-come-LF=1S.Gen this.Nom Part] ‘I have been here (already)’ d’. in-sal-an=mu kaca la [Prf-go-LF=1S.Gen that.Nom Part] ‘I have been there (already)’
Observing the merging of the case marker ka and the demonstrative pronouns in Plngawan Atayal, I suspect that the demonstratives qani and qasa in Wulai Atayal might have also been derived from the Nominative case marker qu and hani/hasa as well, as shown below:
(6) a. qu + hani > qani b. qu + hasa > qasa
Demonstrative pronouns being marked with case markers that are identical to those used with nouns are also found in some other Formosan languages. For example, Amis has the Nominative case marker k(u) existing in demonstrative pronouns like kuni ‘this.Nom’ and kura/kuya ‘that.Nom’, the Locative/Accusative case marker t(u)…-an in tunian ‘this.Loc’ and turaan/tuyaan ‘that.Loc’, and the Genitive case marker n(u) in nuni ‘this.Gen’ and nura/nuya ‘that.Gen’ (Huang 1995b:236-240). Similarly, Tona Rukai has the Nominative case marker ki/ko/na appearing in some demonstrative pronouns like kikay ‘this.Nom’ and kini/kioay/kia/koay ‘that.Nom’, and the Oblique case marker na in nakay ‘this.Obl’ and naoay/koay ‘that.Obl’ (Wang 2003:113).
2.2 Accusative case marking
There are two ways of marking Accusative relations in the Plngawan dialect; one is through the pre-nominal case markers, and the other is through a nominal suffix. There are two pre-nominal Accusative case markers ci and cika, and the nominal suffix is -an. Below we will examine the functions of these Accusative pre-nominal and nominal markers.
213
Lillian M. Huang
2.2.1 Pre-nominal Accusative case markers
As shown in lines 1, 2, 9 & 11 in Text 1, the marker ci precedes a noun (phrase) designating the Patient role. Like the Nominative case marker, the Accusative case marker can precede either common nouns (recall the nouns following ci in lines 1, 2, 9 & 11 in Text 1) or proper nouns (including kinship terms), as shown below:
(7) a. m-asye ci watanpatient ka temu [AF-laugh Acc Watan Nom Temu] ‘Temu is laughing at Watan’
b. cyel m-ana ci yaapatient=ni ka hiya [Prog.Dist AF-wait Acc father=3S.Gen Nom 3S.Neu] ‘He is waiting for his father’
In addition to the Accusative case marker ci, the above sentences may have the other Accusative case marker cika preceding the nouns manifesting Patient participant, as given below:
(7) a’. m-asye cika watanpatient ka temu [AF-laugh Acc Watan Nom Temu] ‘Temu is laughing at Watan (for some reason)’
b’. cyel m-ana cika yaapatient=ni ka hiya [Prog.Dist AF-wait Acc father=3S.Gen Nom 3S.Neu] ‘He is waiting for his father (for some reason)’
According to our informant Mr. Watan Nawi, the differences between (7a)-(7a’) and between (7b)-(7b’) are not apparent or significant. However, he indicates that the possible difference between such sentence pairs is that there is some additional reason implied in (7a’) and (7b’); that is, (7a’) expresses that ‘Temu is laughing at Watan for some reason’, and (7b’) ‘He is waiting for his father for some reason’; while (7a) and (7b) do not present such an implication. A similar distinction is also observed in the following sets of sentences:
(8) a. m-ahiy kariryax ci temupatient ka ule=su [AF-beat often Acc Temu Nom child=2S.Gen] ‘Your child often beats Temu’
a’. m-ahiy kariryax cika temupatient ka ule=su [AF-beat often Acc Temu Nom child=2S.Gen] ‘Your child often beats Temu (for some reason)’
214
Case Marking System in Plngawan Atayal
b. m-pa-bahiy=cu ci ulepatient [AF-Irr-beat=1S.Nom Acc child] ‘I will beat the child’
b’. m-pa-bahiy=cu cika ulepatient [AF-Irr-beat=1S.Nom Acc child] ‘I will beat the child (for some reason)’
While comparing the above examples with Mayrinax Atayal, we notice that Mayrinax Atayal also has two Accusative case markers for common nouns, i.e. cu and cku, and the distinction between them is a matter of referentiality,8 as examplified below (Huang 1995a:92):
(9) Mayrinax Atayal a. pa-aaal cu pila ku ulaqi [Irr-take.AF Acc.Nrf money Nom.Rf child] ‘The child wants money’ b. pa-aaal cku pila ku ulaqi [Irr-take.AF Acc.Rf money Nom.Rf child] ‘The child wants the money’
Since both Plngawan and Mayrinax Atayal are C’uli’ dialects, the Mayrinax examples given here somehow confirm that the difference between Plngawan sentences may also be a matter of ‘referentiality’ or ‘definiteness’. Next, consider the following set of sentences:
(10) a. cyel tehuk cika ulelocation=mu ka yumin [Prog.Dist arrive.AF Acc child=1S.Gen Nom Yumin] ‘Yumin has arrived at my child’s place’
a’.* cyel tehuk ci ulepatient=mu ka yumin [Prog.Dist arrive.AF Acc child=1S.Gen Nom Yumin] b. m
8 For a more detailed discussion of ‘referentiality’ in Mayrinax Atayal, please refer to Huang (1995a:89-109).
215
Lillian M. Huang
As the English gloss indicates, in addition to designating the Patient participant, the noun (phrase) and the preceding marker cika may designate a location, as illustrated by (10a)-(10b). However, in this case, the Accusative case marker ci is not allowed to take the place of cika, as in (10a’)-(10b’). Why ci in (10a’)-(10b’) cannot be used is perhaps because the named nouns, ule ‘child’ in (10a’) and nakis ‘old man’ in (10b’), do not refer to the human arguments but their houses, something additional other than ‘child’ and ‘old man’. Consequently, the Accusative case marker cika is required.
2.2.2 Accusative nominal suffix
In manifesting the Patient participant, Plngawan Atayal may use a noun suffixed with -an in addition to utilizing a noun preceded by the Accusative case marker, as exemplified in line 1 (i.e. yaa-n mu ‘my father’) in Text 2 above. More examples are given below:
(11) a. m-asye ci watanpatient ka temu [AF-laugh Acc Watan Nom Temu] ‘Temu is laughing at Watan’
a’. m-asye watan-anpatient ka temu [AF-laugh Watan-Acc Nom Temu] ‘Temu is laughing at Watan’
a”.* m-asye ci watan-anpatient ka temu [AF-laugh Acc Watan-Acc Nom Temu]
b. cyel m-ana ci yaapatient ka hiya [Prog.Dist AF-wait Acc father Nom 3S.Neu] ‘He is waiting for Father’
b’. cyel m-ana yaa-npatient ka hiya [Prog.Dist AF-wait father-Acc Nom 3S.Neu] ‘He is waiting for Father’
b”.* cyel m-ana ci yaa-npatient ka hiya [Prog.Dist AF-wait Acc father-Acc Nom 3S.Neu]
Note that in (11a)-(11b) the Patient participants watan and yaa are preceded by the Accusative case marker ci, while in (11a’)-(11b’), the Accusative case marker is no longer present; instead, the suffix -an is added to the nouns watan and yaa designating the Patient role of the named participant in an event. However, the nominal suffix -an cannot co-occur with the Accusative case marker ci, as illustrated by (11a”)-(11b”). The nominal suffix -an also appears in one of the four sets of the Plngawan pronouns,
216
Case Marking System in Plngawan Atayal
9 namely, Locative pronouns (e.g. kinan/kina ‘1S.Loc’, sinan/sina ‘2S.Loc’, hiyan ‘3S.Loc’). The Locative pronouns, like the nouns affixed with -an discussed above, can designate the Patient participant in an event. Examples follow:
(12) a. iya kaixu kinanpatient [Neg cheat 1S.Loc] ‘Don’t cheat me!’ b. m-pa-sarai=cu sinanpatient [AF-Irr-make:friends=1S.Nom 2S.Loc] ‘I want to make friends with you’ c. s
The reason to call such a set of pronouns ‘Locative pronouns’ is because not only the marker -an resembles the Locative Focus marker attached to verbs (e.g. inohan ‘come.LF’ in (5c), insalan ‘go.LF’ in (5d)), but they, like the nouns affixed with -an (e.g. nanean ‘table’, tataman ‘chair’, aabelan ‘bed’, patanoan ‘pillow’), can designate locations too. For instance:
(13) a. m
Note that such a Patient-case suffix -an seems only attached to personal proper nouns and kinship terms, e.g. (11a’)-(11b’), but not to common nouns, as shown in (14a’-b’):
(14) a. m-asye ci ule patient ka temu [AF-laugh Acc child Nom Temu] ‘Temu is laughing at the child’
a’.* m-asye ule-anpatient ka temu [AF-laugh child-Acc Nom Temu]
9 Our informant Mr. Temu Nawi tends to say kina/sina instead of kinan/sinan though he is perfectly aware how the two words should be pronounced.
217
Lillian M. Huang
b. m-asye ci kanelpatient ka temu [AF-laugh Acc woman Nom Temu] ‘Temu is laughing at his wife’
b’.* m-asye kanel-anpatient ka temu [AF-laugh woman-Acc Nom Temu]
The nominal suffix -an in Plngawan Atayal has various functions. In addition to marking a Patient argument, it may manifest an accompanist or a Benefactee, as shown in the following examples:
(15) a. pa-aluw=su ima-naccompanist m-usa moow liwas cuxan [Irr-accompany=2S.Nom who-Acc AF-go Puli tomorrow] ‘With whom are you going to Puli tomorrow?’
b. m-awas=minagent watan-anbenefactee [AF-sing=1PE.Nom Watan-Acc] ‘We are singing for Watan; We are giving Watan a song’
The use of an affix to mark case relationship is not found in other Atayal dialects, yet it can be observed in some other Formosan languages like Rukai (especially in the Tona, Maolin, and Mantauran dialects, see Huang et al. 1998, Wang 2003 & 2005, Zeitoun, in press) and Amis (see Huang 1995b). Below are some examples from Rukai for illustration:
(16) a. Mantauran Rukai (Huang et al. 1998: 36) maalam ipol-in taotao [love ipolo-Obl Taotao] ‘Taotao loves ipolo’ b. Mantauran Rukai (Zeitoun, in press) o-laai-a dhaan taotao-in [Dyn.Fin-buy-already house Taotao-Obl] ‘Someone bought a house for Taotao.’ c. Tona Rukai (Wang 2003:20-21) w-a-sititi ki takanaw ipol-an [Act.Dyn.Fin-Real-beat Nom Takanaw Ipolo-Obl] ‘Takanaw is beating/beat Ipolo’ d. Tona Rukai (Wang 2003:21) pa-lapo ki tatava titin-an na taokoko [Caus-Dyn.Nfin:raise Nom father mother-Obl Obl chicken] ‘Father had Mother raise chicken’
218
Case Marking System in Plngawan Atayal
As shown above, in Rukai, proper nouns or kinship terms manifesting the Patient 10 participant are marked with -in. However, unlike Plngawan Atayal, Rukai lacks pre-nominal Oblique case markers used for proper nouns, though the language does have Oblique case markers for common nouns, e.g. ko and na in Tona Rukai (Wang 2003:18). Next let us consider some Amis examples, as given below:
(17) Amis (Huang 1995b:221) a. mi-sti ci aki ci panay-an
[MI-beat Nom Aki Loc1 Panay-Loc2] ‘Aki beat Panay’ b. ma-ulah ku kaka ci panay-an
[MA-love Nom elder:sibling Loc1 Panay-Loc2] ‘Elder brother loves Panay’
Here note that Amis has its Patient participant, when manifested by a proper noun, being marked by ci…-an (tu for common nouns), and the two have to co-occur, which is different from Plngawan Atayal, where the co-occurrence of the Accusative case marker ci and the nominal suffix -an is forbidden. In Amis, without cooccurring -an, the marker ci alone is then regarded as the Nominative case marker. As for -an in Amis, just like -an in Plngawan Atayal, it also resembles the Locative marker found in the words indicating location, e.g. pananuman ‘a water container’, kafutian ‘a bed’. To sum up the present discussion on the Accusative case marking in Plngawan Atayal, we notice that the dialect has both pre-nominal Accusative case markers and a nominal Accusative suffix that cannot cooccur, the situation of which is different from Rukai and Amis.
2.3 Genitive case marking
The markers ni, na and naka are all Genitive case markers in the named dialect. The difference between them can be observed in line 1 (i.e. ke ni yaya mu ‘my mother’s language’) and line 2 (i.e. ke na itaal ‘Atayal language’) of Text 1. In other words, ni seems to be used when the named participant is manifested by a proper noun or a kinship term, while na (as well as naka to be discussed later) occurs with common
10 Zeitoun (in press) points out that such a Patient participant can be either marked on the noun or on the verb (but not on both simultaneously). Compare (16b) with the following: (16) b’ o-laai--in aan taotao [Dyn.Fin-buy-already-3S.Obl house Taotao] ‘Someone bought a house for Taotao.’
219
Lillian M. Huang nouns. However, such a constraint is not seriously obeyed. Consider:
(18) a. huil ni yaa=mu [dog Gen father=1S.Gen] ‘my father’s dog’ a’.* huil na yaa=mu [dog Gen father=1S.Gen] b. huil na ule hani [dog Gen child this] ‘this child’s dog’ b’. huil ni ule hani [dog Gen child this] ‘this child’s dog’
Note in the above phrases, while proper nouns can only be preceded by ni, common nouns can be marked by either na or ni, though na is still more often expected. In other words, the case marker used for proper nouns has a wider distribution than its counterpart used for common nouns, which is different from the situation in Wulai Atayal, as discussed in §2.1.1. The two Genitive case markers in Plngawan Atayal, i.e. ni and na, indicate the possession relationship, with the possessor following the possessed. But this is not the only function of the named markers. Consider (19a-d) which are NAF constructions and which have the Genitive case markers marking the Agent participant. Here a non-Agent participant serves as the grammatical subject (marked by the Nominative case marker ka):
(19) a. na-ni-un ni temuagent ka ucyuxpatient [Red-eat-PF Gen Temu Nom fish] ‘The fish will be eaten by Temu’
b. na-ni-un na uleagent ka ucyuxpatient [Red-eat-PF Gen child Nom fish] ‘The fish will be eaten by the child’ c. e-an ni temuagent ci pila ka ulereceipient [give-Loc Gen Temu Acc money Nom child] ‘Temu gave money to the child’ d. si-cau na nakisagent ci ucyux ka yuminbenefactee [BF-wrap Gen old:man Acc fish Nom Yumin] ‘The old man wrapped fish for Yumin’
220
Case Marking System in Plngawan Atayal
Like the Accusative case markers ci and cika, the Genitive case marker na (but not ni) in Plngawan Atayal has a counterpart naka, as given in the following examples:
(20) a. hani a, pila na laha [this Top money Gen 3P.Neu] ‘This is their money’ a’. hani a, pila naka laha [this Top money Gen 3P.Neu] ‘This is their money’ b. si-cau na nakis ci ucyux ka yumin [BF-wrap Gen old:man Acc fish Nom Yumin] ‘The old man wrapped fish for Yumin’ b’. si-cau naka nakis ci ucyux ka yumin [BF-wrap Gen old:man Acc fish Nom Yumin] ‘The old man wrapped fish for Yumin’
Whether the semantic distinction between the Genitive case markers na and naka is a matter of ‘referentiality’, like the distinction between na and nku in Mayrinax Atayal (Huang 1995a:97), is uncertain to us at this stage, which deserves future investigation.
2.4 Benefactive and instrument case markers
In addition to the above-mentioned case markers, Plngawan Atayal seems to have some more case markers, such as the Benefactive case markers ni and na which mark a Benefactee participant of an event. Examples containing these case markers follow:
(21) a. s
b. m-pa-siliy=cuagent ni/na nakisbenefactee ci huilpatient=su [AF-Irr-beat=1S.Nom Ben old:man Acc dog=2S.Gen] ‘Your dog will be beaten by me for the old man’
c. m-awas=minagent ni watanbenefactee [AF-sing=1PE.Nom Ben Watan] ‘We are singing for Watan’
d. pahpuy na ulebenefactee ka yayaagent [cook.AF Ben child Nom mother] ‘Mother cooks for the child’
221
Lillian M. Huang
Furthermore, there seems another case marker na marking an Instrument participant of an event. Since the Instrument participant is normally inanimate, ni does not serve as an Instrument case marker. Some examples are given below:
(22) a. s
b. sa-silay-un=muagent na naariinstrument ka huilpatient=su [Red-beat-PF=1S.Gen Ins stick Nom dog=2S.Gen] ‘Your dog will be beaten by me with a stick’
c. m-atas=cuagent na papatasinstrument=ni [AF-write=1S.Nom Ins pen=3S.Gen] ‘I write with his pen’
In comparison with the Nominative, Accusative and Genitive case markers, the Benefactive and Instrument case markers are seldom present in the Plngawan speakers’ daily speech. One of the reasons may be due to the fact that the participants marked by these markers tend to be more peripheral ones in a proposition (cf. ‘CIRCUMSTANTIAL VOICE’ in Huang 1993:30), unlike the Agent and Patient participant of an event. Hence we would propose that these two markers may be grouped with the Genitive case markers, not only because they share the same forms (i.e. ni and na), but also because they appear in complementary environments serving different functions; that is, the Genitive case markers mark either the Possessor or the Agent participant, while the Benefactive case marker marks the Benefactee and the Instrument case marker the Instrument. Consequently, we attempt to label them together as the Oblique case.
2.5 Comitative case markers
Furthermore, Plngawan Atayal appears to have a Comitative case marker ci which precedes a noun (phrase) designating a co-occurring participant in an event. Below we will discuss such structures with respect to the conjoined participants being Agent or Patient.
2.5.1 Comitative case markers and Agent participants
First, let us examine the following sentences containing the first person participant:
222
Case Marking System in Plngawan Atayal
(23) a. nyel=minagent m-awas ci temuaccompanist [Prog.Prox=1PE.Nom AF-sing Com Temu] ‘I am singing with Temu’
b. m
d. ma-m-aha=minagent m-ani ci temuaccompanist [Red-AF-go=1PE.Nom AF-eat Com Temu] ‘I will go to eat with Temu’
As the English gloss indicates, the Comitative case marker ci in the above examples expresses accompaniment, joining the first person pronoun and a noun which manifest two participants in an event. Also note that in (23a-d), the first person plural pronoun min is used instead of its singular counterpart cu; this is quite a unique use of ‘inclusive plural’ in all Atayal varieties (also cf. Huang 1989:128 and Huang 1995a:45). In addition, like the Accusative case markers ci and cika, the Comitative case marker ci also has a counterpart cika (e.g. (24b)), though cika does not appear as often as ci and the semantic difference between the two remains unclear to us.
(24) a. m-pa-was=minagent cuxan ci nakisaccompanist [AF-Irr-sing=1P.Nom tomorrow Com old:man] ‘I will sing with the old man tomorrow’
b. m-pa-was=minagent cuxan cika nakisaccompanist [AF-Irr-sing=1P.Nom tomorrow Com old:man] ‘I will sing with the old man tomorrow’
Similarly, when the second-person participant is involved, its plural form mamu will be used instead of su, as examples (25a-c) illustrate:
(25) a. nyel=mamuagent m-awas ci temuaccompanist a [Prog.Prox=2P.Nom AF-sing Com Temu Q] ‘Are you singing with Temu?’
11 The word mamaha ‘will go.AF’ is the only instance of repeating the first syllables of an AF verb to designate a future/irrealis event.
223
Lillian M. Huang
b. m
c. ma-m-aha=mamuagent inu cuxan ci nakisaccompanist [Red-AF-go=2P.Nom where tomorrow Com old:man] ‘Where will you go with the old man tomorrow?’
When the two involved participants are neither the speaker/the first-person participant nor the hearer/the second-person participant (i.e. they are the third-person participants), a third person plural pronoun is expected to be present. However, since the language does not have any third person bound Nominative pronouns, singular or plural (cf. Appendix 1), the third person plural Neutral pronoun laha is used. Furthermore, being a free pronoun, laha appears between the two nouns manifesting the two involved arguments, unlike the first and second person plural bound Nominative pronouns min and mamu that need to be attached to the sentence initial element (either a verb or an auxiliary). Consider the following examples:
(26) a. cyel ma-sa-siliy ka watanagent laha ci temuaccompanist [Prog.Dist Rec-Red-beat Nom Watan 3P.Neu Com Temu] ‘Watan and Temu are beating each other (with sticks)’
a’.* cyel ma-sa-siliy ka watanagent ø ci temuaccompanist [Prog.Dist Rec-Red-beat Nom Watan Com Temu]
a”.* cyel laha ma-sa-siliy ka watanagent ci temuaccompanist [Prog.Dist 3P.Neu Rec-Red-beat Nom Watan Com Temu]
b. nyel m-awas ka nakisagent laha ci uleaccompanist [Prog.Prox AF-sing Nom old:man 3P.Neu Com child] ‘The old man is singing here with the child’
b’.* nyel m-awas ka nakisagent ø ci uleaccompanist [Prog.Prox AF-sing Nom old:man 3P.Neu Com child]
b”.* nyel laha m-awas ka nakisagent ci uleaccompanist [Prog.Prox 3P.Neu AF-sing Nom old:man Com child] c. cyel m-auy ka huilagent laha ci yawaccompanist [Prog.Dist AF-dance Nom dog 3P.Neu Com cat] ‘The dog is dancing with the cat’ c’.* cyel m-auy ka huilagent ø ci yawaccompanist [Prog.Dist AF-dance Nom dog 3P.Neu Com cat] c”.* cyel laha m-auy ka huilagent ci yawaccompanist [Prog.Dist 3P.Neu AF-dance Nom dog Com cat]
224
Case Marking System in Plngawan Atayal
d. ma-m-aha tanux ka yuminagent laha ci yawaccompanist [Red-AF-go outside Nom Yumin 3P.Neu Com cat] ‘Yumin will go outside with the cat’
d’.* ma-m-aha tanux ka yuminagent ø ci yawaccompanist [Red-AF-go outside Nom Yumin 3P.Neu Com cat]
d”.* ma-m-aha laha tanux ka yuminagent ci yawaccompanist [Red-AF-go 3P.Neu outside Nom Yumin Com cat]
As illustrated above, the Comitative case marker ci in Plngawan Atayal may conjoin two proper nouns (e.g. (26b)), two common nouns (e.g. (26b-c)), or one proper and one common noun (e.g. (26d)) that manifest the participants involved in an event. However, when two common nouns are conjoined, they must manifest animate arguments, but not inanimate ones, as examplified below:
(27) a. cyel m-auy ka iluagent laha ci kukukaccompanist [Prog.Dist AF-dance Nom chicken 3P.Neu Com duck] ‘The chicken is dancing with the duck’
b. * ramas niun caru ka ilutheme laha ci kukukaccompanist [delicious.AF very Nom chicken 3P.Neu Com duck] ‘The chicken and the duck are very delicious’ c. * ramas kaal hira laha ci soni [good.AF sky yesterday 3P.Neu Com today] ‘The weather in yesterday and today is good’
Note that while ilu ‘chicken’ and kukuk ‘duck’ in (27a) refer to animate animals, the same nouns in (27b) refer to their meat and thus cannot be conjoined by the Comitative case marker. Similarly, the two inanimate common nouns hira ‘yesterday’ and soni ‘today’ in (27c) cannot be conjoined by ci either. In such a case of conjoining two inanimate common nouns, the coordinating conjunction lu needs to be used, to which we will return shortly. That two common nouns are conjoined by the Comitative case marker is not possible in other Atayal dialects. For instance, Mayrinax Atayal does not allow two common nouns, whether they manifest animate or inanimate arguments, to be conjoined by the Comitative case marker ki, as shown below:
(28) Mayrinax Atayal
a. * ta-tuti-un ni watan ku ulaqipatient ki naakisaccompanist [Red-beat-PF Gen Watan Nom child Com old:man] ‘The child will beat Watan and Yumin (with his fist)’
225
Lillian M. Huang
b. * m-ausa=cami m-aniq cu uqiluh patient ki caiaccompanist [AF-go=1PE.Nom AF-eat Acc.Nrf banana Com taro] ‘We are going to eat banana and taro’
Furthermore, while Plngawan Atayal requires the third person plural pronoun laha to appear between the two conjoined nouns, no such a requirement is observed in other Atayal dialects, as exemplified by the following Mayrinax sentences where no third person plural pronoun is allowed to appear:
(29) Mayrinax Atayal
a. * ma-quwas i limuyagent nha ki watanaccompanist [AF-sing Nom Limuy 3P.Neu Com Watan] ‘Limuy and Watan are singing’
b. * ta-tuti-un nku ulaqi i watanagent nha ki yuminaccompanist [Red-beat-PF Gen.Rf child Nom Watan 3P.Neu Com Yumin] ‘The child will beat Watan and Yumin’
One more thing worth mentioning is that in Plngawan Atayal, when the Comitative 12 case marker ci is replaced by the coordinating conjunction lu, the named pronoun laha is not allowed to appear, as shown in the following examples containing coordinating noun phrases:
(30) a. cyel ma-sa-siliy ka {watanagent ø lu temuagent} [Prog.Dist Rec-Red-beat Nom Watan Conj Temu] ‘Watan and Temu are beating each other (with sticks)’
a’.* cyel ma-sa-siliy ka {watanagent laha lu temuagent} [Prog.Dist Rec-Red-beat Nom Watan 3P.Neu Conj Temu]
12 While in Wulai and Mayrinax Atayal, the coordinating conjunction is ru, Plngawan Atayal has lu instead. It can be used to conjoin two (or more) nouns, verbs and clauses. Examples follow: (i) {yaw} lu {huil} a, ramas taxan [cat Conj dog Top good:looking] ‘The cat and the dog are beautiful’ (ii) temu a, {hapow} lu {m-aas} [Temu Top healthy.AF Conj AF-happy] ‘As for Temu, he is healthy and happy’ (iii) {m-awas ka rimuy} lu {m-auy ka iwal} [AF-sing Nom Rimuy Conj AF-dance Nom Iwal] ‘Rimuy is singing and Iwal is dancing’
226
Case Marking System in Plngawan Atayal
b. cyel m-auy ka {huilagent ø lu yawagent} [Prog.Dist AF-dance Nom dog Conj cat] ‘The dog is dancing with the cat’ b’.* cyel m-auy ka {huilagent laha lu yawagent} [Prog.Dist AF-dance Nom dog 3P.Neu Conj cat]
c. ma-m-aha tanux ka {yuminagent ø lu yawagent} [Red-AF-go outside Nom Yumin Conj cat] ‘Yumin will go outside with the cat’
c’.* ma-m-aha tanux ka {yuminagent laha lu yawagent} [Red-AF-go outside Nom Yumin 3P.Neu Conj cat]
Moreover, when the two common nouns manifest inanimate arguments, they must be conjoined by the conjunction lu, as shown below:
(31) a. ramas niun caru ka {ilu lu kukuk} [delicious.AF very Nom chicken Conj duck] ‘The chicken and the duck are very delicious’ b. ramas kaal {hira lu soni} [good.AF sky yesterday Conj today] ‘The weather in yesterday and today is good’
The last point worth mentioning here is that, when the involved participant is manifested by either the first or the second person pronoun, in addition to the plural bound Nominative pronouns (i.e. min and mamu) have to be attached to the first element of the sentence, their corresponding free Neutral pronouns may, though not necessarily, co-occur in such a construction and precede the Comitative case marker followed by the noun designating the accompanist. Compare the following sentences with (23d) and (25c):
(23) d’. ma-m-aha=minagent m-ani cami ci temuaccompanist [Red-AF-go=1PE.Nom AF-eat 1PE.Neu Com Temu] ‘I will go to eat with Temu’
(25) c’. ma-m-aha=mamuagent inu cuxan cimu ci nakisaccompanist [Red-AF-go=2P.Nom where tomorrow 2P.Neu Com old:man] ‘Where will you go with the old man tomorrow?’
Some more examples are given below for further reference:
227
Lillian M. Huang
(32) a. nyel=minagent m-awas cami ci temuaccompanist [Prog.Prox=1PE.Nom AF-sing 1PE.Neu Com Temu] ini=min pakusa. [Neg=1PE.Nom quarrel.AF] ‘I am singing with Temu; we are not quarrelling’
b. nanak=minagent m-awas cami ci temuaccompanist [only=1PE.Nom AF-sing 1PE.Neu Com Temu] ini=min pakusa. [Neg=1PE.Nom quarrel.AF] ‘I am only singing with Temu; we are not quarrelling’
2.5.2 Comitative case markers and Patient participants
Now consider the following sentences with the involved nouns manifesting the Patient participant, regardless of their being grammatical subject (e.g. (33a-b)) or object (e.g. (34a-b)):
(33) a. sa-silay-un=muagent ka watanpatient laha ci temuaccompanist [Red-beat-PF=1S.Gen Nom Watan 3P.Neu Com Temu] ‘Watan and Temu will be beaten by me (with a stick)’ b. sa-silay-un=muagent ka huilpatient laha ci yawaccompanist [Red-beat-PF=1S.Gen Nom dog 3P.Neu Com cat] ‘The dog and the cat will be beaten by me (with a stick)’
(34) a. m-pa-ahiy=cuagent ci watanpatient laha ci temuaccompanist [AF-Irr-beat=1S.Nom Acc Watan 3P.Neu Com Temu] ‘I will beat Watan and Temu’
b. m-pa-siliy=cuagent ci huilpatient laha ci yawaccompanist [AF-Irr-beat=1S.Nom Acc dog 3P.Neu Com cat] ‘I will beat the dog and the cat (with a stick)’
Again, like (30a-c), when the coordinating conjunction lu takes the place of the Comitative case marker ci, the third person plural pronoun laha is required not to appear, as shown below:
(33) a’. sa-silay-un=muagent ka watanpatient ø lu temuaccompanist [Red-beat-PF=1S.Gen Nom Watan Conj Temu] ‘Watan and Temu will be beaten by me (with a stick)’
228
Case Marking System in Plngawan Atayal
b’. sa-silay-un=muagent ka huilpatient ø lu yawaccompanist [Red-beat-PF=1S.Gen Nom dog Conj cat] ‘The dog and the cat will be beaten by me (with a stick)’
(34) a’. m-pa-ahiy=cuagent ci watanpatient ø lu temupatient [AF-Irr-beat=1S.Nom Acc Watan Conj Temu] ‘I will beat Watan and Temu’
b’. m-pa-siliy=cuagent ci huilpatient ø lu yawpatient [AF-Irr-beat=1S.Nom Acc dog Conj cat] ‘I will beat the dog and the cat (with a stick)’
Also, note that the common nouns manifesting inanimate arguments can only be conjoined by the coordinating conjunction lu, but not by the Comitative case marker ci with/without the pronoun laha. Consider:
(35) a. a-aras=misu ci {au lu box} [Red-bring.BF=1S.Gen:2S.Nom Acc wine Conj rice] ‘I will bring wine and rice for you’ a’.* a-aras=misu ci au laha ci box [Red-bring.BF=1S.Gen:2S.Nom Acc wine 3P.Neu Com rice] b. a-aras=mu ci {au lu box} ka temu [Red-bring.BF=1S.Gen Acc wine Conj rice Nom Temu] ‘I will bring wine and rice for Temu’ b’.* a-aras=mu ci au laha ci box ka temu [Red-bring.BF=1S.Gen Acc wine 3P.Neu Com rice Nom Temu] c. si-aal=cu ci {patarak lu rati} [BF-take=1S.Nom Acc chopsticks Conj bowl] ‘Please bring me the chopsticks and bowl’ c’.* si-aal=cu ci patarak laha ci rati [BF-take=1S.Nom Acc chopsticks 3P.Neu Com bowl]
2.5.3 Agent or Patient participants
There is one more thing that deserves our attention. Compare the following sets of sentences:
(36) a. m-pa-ahiy=cuagent ci watanpatient laha ci temuaccompanist [AF-Irr-beat=1S.Nom Acc Watan 3P.Neu Com Temu] ‘I will beat Watan and Temu’
229
Lillian M. Huang
a’. m-pa-ahiy=cuagent watan-anpatient laha ci temuaccompanist [AF-Irr-beat=1S.Nom Watan-Acc 3P.Neu Com Temu] ‘I will beat Watan and Temu’
b. m-pa-siliy=cuagent ci huilpatient laha ci temuaccompanist [AF-Irr-beat=1S.Nom Acc dog 3P.Neu Com Temu] ‘I will beat the dog and Temu (with a stick)’
c. m-pa-siliy=minagent ci huilpatient cami ci temuaccompanist [AF-Irr-beat=1P.Nom Acc dog 1PE.Neu Com Temu] ‘I will beat the dog with Temu (with a stick)’
d. * m-pa-bahiy=minagent ci watanaccompanist cami ci temuaccompanist [AF-Irr-beat=1P.Nom Acc Watan 1PE.Neu Com Temu]
d’. m-pa-bahiy=minagent watan-anpatient cami ci temuaccompanist [AF-Irr-beat=1P.Nom Watan-Acc 1PE.Neu Com Temu] ‘I will beat Watan with Temu’
Note that in (36a-a’) and (36b), the third person plural pronoun laha is present, and the sentences are interpreted as having a coordinated Objective phrase; that is, two Patient participants are involved in the event ‘beating’. On the other hand, in (36c) where the first person plural pronoun cami is present, the sentences are interpreted as having coordinated Agent participants in the event ‘beating’; that is, both ‘I’ and ‘Temu’ are Agents and will do the beating. However, when the involved participants are all human, as illustrated in (36d), the sentence is not accepted to our informant, though the replacement of the Accusative case marker ci with the nominal Accusative suffix -an seems to make (36d’) acceptable. The reason may be due to the same form ci shared by both the Accusative case marker and the Comitative case marker which causes confusion, unlike the nominal suffix -an which clearly indicates the Patient role of the involved participant. When the verbs are in NAF form, i.e. the non-Agent participant serves as the grammatical subject, no confusion arises, as illustrated by the following sentences:
(37) a. ba-bahiy-an=min ka watan laha ci temu [Red-beat-LF=1P.Gen Nom Watan 3P.Neu Com Temu] ‘We will beat Watan and Temu; Watan and Temu will be beaten by us’ b. ba-bahiy-an=min ka watan cami ci temu [Red-beat-LF=1P.Gen Nom Watan 1PE.Neu Com Temu] ‘I will beat Watan with Temu; Watan will be beaten by me and Temu’ b’. ba-bahiy-an=min cami ci temu ka watan [Red-beat-LF=1P.Gen 1PE.Neu Com Temu Nom Watan] ‘I will beat Watan with Temu; Watan will be beaten by me and Temu’
230
Case Marking System in Plngawan Atayal
Note that when the third person plural pronoun laha is present, e.g. (37a), the sentences are interpreted as having a coordinated subject phrase and two Patient participants are involved in the event ‘beating’. When the first person plural pronoun cami is present, e.g. (37b-b’), the sentences are interpreted as having its non-subject arguments being coordinated Agents in the event ‘beating’; that is, both ‘I’ and ‘Temu’ are Agents and will do the beating together. No confusion appears here. Further investigation concerning the interaction with the case markers and pronouns is needed. The last remark concerns the status of the Comitative case markers ci and cika. Although they share the same forms with the Accusative case markers, and can be differentiated from the latter through the co-occurring pronouns (i.e. they appear in different syntactic environments), we are not certain whether the two sets can be grouped into one, especially when taking into account the case marking systems of some other Atayal dialects like Wulai Atayal (Squliq) and Mayrinax Atayal (C’uli’), as will be discussed in §3.
2.6 Summary
To sum up the present discussion, Table 1b proposes a four-set case marking system for Plngawan Atayal, instead of a five-set one as given in Table 1a:
Table 1b: Case markers in Plngawan Atayal (revised) Case Obl Nom Acc Com Noun (Gen/Ben/Ins) Proper ni ci Common Animate ka ci na cika Inanimate cika naka ---
3. A dialectal comparison
In this section, a brief comparison of case markers among three Atayal dialects is given, including Wulai, Mayrinax and Plngawan Atayal. Below let us present the case markers in Wulai Atayal and Mayrinax Atayal first. In Huang (forthcoming), Wulai Atayal has six sets of case markers, namely, Nominative, Dative/Locative, Genitive, Benificiary, Instrument and Comitative. These markers can be further divided into two sub-sets, depending on whether they precede proper nouns or common nouns, as shown in Table 3a:
231
Lillian M. Huang
Table 3a: Case markers in Wulai Atayal (Huang forthcoming)13 Case Nom Dat/Loc Gen Ben Ins Com Noun Proper i ki ni ni --- ki Common qu i; sa; na; nqu na --- squ; te nqu
As for the case marking system in Mayrinax Atayal, Huang (1995a) proposes a nine-set system, as shown in Table 4a:
Table 4a: Case markers in Mayrinax Atayal (Huang 1995a:109)
Case Nom Acc Dat Loc Gen Ben Ins Com Neu Noun Proper i i i ki ni ni --- ki i Common Nrf a cu --- i na --- na ------Rf ku cku cku cku nku nku nku ------
The following table presents what case systems the above-named three Atayal dialects have (cf. Table 1a, Table 3a and Table 4a):
Table 5a: Case systems of Wulai, Mayrinax and Plngawan Atayal
Case Nom Acc Dat Loc Gen Ben Ins Com Neu Dialect Wulai + - + + + + + + - Mayrinax + + + + + + + + + Plngawan + + -14 - + + + + -
13 The dot lines in Table 3a indicate that the restriction of using certain case markers for proper and common nouns is not that crucial any longer. 14 Though Plngawan Atayal does not have an overt Dative or Locative case marker, the following words and sentence somehow do indicate the trace of such a case marker: sani ‘here’ and saca ‘there’ which must have undergone the derivation of *sa hani > sani, and *sa haca > saca. (i) m-umarah=cu sani kariryax lu m-umarah saca kariryax ka hiya. [AF-work=1S.Nom here every:day Conj AF-work there every:day Nom 3S.Neu] ‘I work here every day, and he works there every day’
232
Case Marking System in Plngawan Atayal
Considering the forms of the Accusative, Dative and Locative case markers in Mayrinax Atayal, and also as discussed in Huang et al. (1998:30),15 these three cases can be grouped into the label ‘Accusative case’. Similarly, as proposed in §2.4 as well as suggested in Huang et al. (1998:30), since the Genitive, Benefactive and Instrument case markers also share the same forms, and manifest either less important syntactic role (i.e. Agent without being the grammatical subject) or more peripheral semantic roles (i.e. Beneficiary and Instrument), these three cases can be labeled together as the Oblique case (or as the Genitive case if we follow the convention). Consequently, while the case system of Plngawan Atayal is reanalyzed as that given in Table 1b, the case systems of Wulai Atayal and Mayrinax Atayal can be re-presented as follows:
Table 3b: Case markers in Squliq Atayal (revised) Case Nom Acc Obl Com Noun (Dat/Loc) (Gen/Ben/Ins) Proper i ki ni ki Common qu i; sa; na; --- squ; te nqu
Table 4b: Case markers in Mayrinax Atayal (revised) (Huang et al. 1998:31)
Case Nom Acc Obl Com Neu Noun (Acc/Dat/Loc) (Gen/Ben/Ins) Proper i i; ki ni ki i Common Nrf a cu na ------Rf ku cku nku ------
Consequently, Table 5a can be re-schematized as follows:
15 Huang et al. (1998:30) point out that ‘some sets (of case) have similar forms, such as Acc & Dat, Gen & Ben & Ins. Since some of these sets may illustrate roles which resemble each other semantically (e.g. Acc in a transitive sentence vs. Dat in a ditransitive sentence), or they can be differentiated in terms of animacy/humanness (Gen vs. Ben vs. Ins), they can be analyzed as belonging to the same sets.’
233
Lillian M. Huang
Table 5b: Case systems of Wulai, Mayrinax and Plngawan Atayal (revised)
Case Nom Acc Obl Com Neu Dialect (Acc/Dat/Loc) (Gen/Ben/Ins) Wulai + + + + - Mayrinax + + + + + Plngawan + + + + -
Furthermore, we notice that the case systems in the named Atayal dialects make further distinction with regard to three properties, namely, proper/common nouns, referentiality, and animacy, as shown below.
Table 6: A comparison of the case systems in Wulai, Mayrinax and Plngawan Atayal
Criteria Proper nouns vs. Referential vs. Animate vs. Dialect Common nouns Non-referential Inanimate Wulai +/- + - Mayrinax + + - Plngawan +/- - +
4. Conclusion
To sum up the present discussion, we notice that the case systems in Wulai Atayal, Mayrinax Atayal and Plngawan Atayal all share the following characteristics:
(i) They all have Nominative, Accusative (including so-called Dative and Locative cases), Comitative, and Oblique (including Genitive, Beneficiary and Instrument cases) and these cases function similarly in the named dialects. (ii) Cases can be further differentiated depending on whether they precede proper nouns or common nouns, though such a differentiation in Wulai and Plngawan Atayal (represented by ‘+/-’) is not as significant as that in Mayrinax Atayal.
However, the three systems differ in some other aspects, as presented below:
(i) While the three dialects all have Nominative, Accusative, Comitative, and Oblique (or Genitive) cases, only Mayrinax Atayal has the Neutral case. (ii) While the case systems in the three dialects can be differentiated in terms of their preceding proper nouns or common nouns, the case markers in Mayrinax Atayal
234
Case Marking System in Plngawan Atayal
can be further categorized with respect to the notion of ‘referentiality’ and Plngawan Atayal the property of ‘animacy’. (iii) While Mayrinax Atayal does not allow the alternating use of different case markers before proper nouns and common nouns, Wulai and Plngawan Atayal enjoy more freedom, except that Wulai Atayal has the case markers intended for common nouns extending their scope to proper nouns, while Plngawan Atayal presents an opposite situation, i.e. allowing the case markers used for proper nouns to precede common nouns. (iv) While both Plngawan Atayal and Wulai Atayal seem to have their demonstratives affixed with the Nominative case marker (i.e. k- in Plngawan Atayal and so kani ‘this.Nom’ and kaca ‘that.Nom’; q- in Wulai Atayal and so qani ‘this.Nom’ and qasa ‘that.Nom’), Mayrinax Atayal does not present a similar phenomenon. (v) While Plngawan Atayal may utilize a nominal affix to indicate case relationship, Mayrinax and Wulai Atayal do not have such a nominal case affix.
235
Lillian M. Huang
Appendix 1: Pronominal systems in Plngawan Atayal
The pronominal systems in Plngawan Atayal can be sub-categorized into personal pronouns and demonstratives.
1. Personal pronouns: There are four sets of personal pronouns, two of which are bound and the other two are free, as given in Table 1.
Table 1: Personal pronouns in Plngawan Atayal
Case Bound Free Person & number Nominative Genitive Locative Neutral 1S cu mu kinan/kina kui 2S su su sinan/sina isu 3S --- ni hiyan hiya 1P-inclusive ta ta itan ita 1P-exclusive min min caminan cami 2P mamu mamu cimunan cimu 3P --- naha lahan laha 1S.Gen+2S.Nom misu 2S.Gen+1S.Nom saku 1S.Nom+3S.Gen cini 2S.Nom+3S.Gen sini 1PE.Nom+3S.Gen mini 2P.Nom+3S.Gen mani
2. Demonstratives: The dialect has the following demonstratives.
Table 2: Demonstrative pronouns in Plngawan Atayal Case Distance Nominative Locative Neutral Proximal kani sani hani Distal kaca saca haca
236
Case Marking System in Plngawan Atayal
References
Ferrell, Raleigh. 1979. Construction markers and subgrouping of Formosan languages. South-East Asian Linguistic Studies, Vol. 3, ed. by Nguyen Dang Liem, 199-211. Pacific Linguistics C-45. Canberra: Australian National University. Huang, Lillian M. 1989. The pronominal system in Atayal. Studies in English Literature and Linguistics 15:115-133. Huang, Lillian M. 1993. A Study of Atayal Syntax. Taipei: Crane. Huang, Lillian M. 1995a. A Study of Mayrinax Syntax. Taipei: Crane. Huang, Lillian M. 1995b. The case markers and pronominal system in Amis. The Journal of National Chengchi University 70:217-258. Huang, Lillian M. (forthcoming). Aboriginal languages. Monograph Series of Taipei County, Vol. 3: Peoples. Taipei: Taipei Prefecture Government. (in Chinese) Huang, Lillian M. et al. 1998. A typological overview of nominal case-marking of the Formosan languages. Proceedings of the Second International Symposium on Languages in Taiwan, ed. by Shuanfan Huang, 21-48. Taipei: Crane. Huang, Lillian M. et al. 1999. A typological overview of pronominal systems of some Formosan languages. Selected Papers from the Fifth International Conference on Chinese Linguistics, ed. by Wang, Tsao, and Lien, 165-198. Taipei: Crane. Li, Paul Jen-kuei. 1985. Linguistic criteria for classifying the Atayalic dialect groups. Bulletin of the Institute of History and Philology 56.4:699-718. Rau, Der-Hwa Victoria. 1992. A Grammar of Atayal. Taipei: Crane. Rau, Der-Hwa Victoria. 2004. Lexical similarity, sound change and intelligibility of Atayalic dialects. Papers in Austronesian Subgrouping and Dialectology, ed. by John Bowden and Nikolaus Himmelmann, 37-95. Pacific Linguistics 563. Canberra: Pacific Linguistics, Research School of Pacific and Asian Studies, Australian National University. Starosta, Stanley. 1988. A grammatical typology of Formosan languages. Bulletin of the Institute of History and Philology 59.2:541-576. Tsuchida, Shigeru. 1976. Reconstruction of Proto-Tsouic Phonology. Study of Languages and Cultures of Asia and Africa Monograph Series No. 5. Tokyo: Institute for the Study of Languages and Cultures of Asia and Africa, Tokyo University of Foreign Studies. Wang, May Hsiu-mei. 2003. Morphosyntactic Manifestations of Participants in Tona (Rukai). Taipei: National Taiwan Normal University MA thesis. Zeitoun, Elizabeth. 1995. Problèmes de linguistique dans les langues aborigènes de Taiwan [Issues on Formosan Linguistics]. Paris: Université René Diderot Paris 7 dissertation.
237
Lillian M. Huang
Zeitoun, Elizabeth. 1997. Coding of grammatical relations in Mantauran (Rukai). Bulletin of the Institute of History and Philology 68.1:149-181. Zeitoun, Elizabeth. (in press). A Grammar of Mantauran Rukai. Taipei: Institute of Linguistics, Academia Sinica.
238