edith aldridge* a minimalist approach to the emergence of...

Edith Aldridge*

A Minimalist Approach to the Emergence of Ergativity inAustronesian Languages

Abstract: This paper proposes a novel reconstruction of Proto-Austronesian as a language with accusativealignment, in contrast to previous analyses, which have reconstructed PAn with some type of ergativesystem. The methodology employed in this paper also departs from previous work in basing the proposal onsyntactic change rather than on phonological or morphological comparison. In the account of argumentlicensing in the Minimalist Program, nominative subjects are licensed by a higher functional head T thanaccusative objects, which value their case with the light verb v. This produces accusative alignment, withthe corollary that ergative systems arise only when the unmarked licensing relations are disrupted.Specifically, I propose that this disruption involves the loss of transitive v’s ability to value accusativecase on the object and the concomitant ability of this v to provide inherent licensing for the subject, therebyallowing T to exceptionally value nominative case on the object. The current paper argues that the ergativetypes of alignment observed in modern Formosan and Philippine languages are the result of consecutiveinnovations in high-order subgroups of the Austronesian family, being triggered first in irrealis clauses in adaughter of PAn, which I term “Proto-Ergative Austronesian”.

Keywords: alignment change, syntactic change, reconstruction, subgrouping, Austronesian, detransiti-vization, irrealis, nominalization, ergativity

DOI 10.1515/lingvan-2014-1011

1 Introduction

It is now virtually uncontroversial that the homeland of the Austronesian language family is the island ofTaiwan, Austronesian speakers migrating there from southeastern China roughly 6,000 years ago andsubsequently proceeding to populate the Philippines, Indonesia/Malaysia, Madagascar, and the Pacific(Adelaar 2005; citing Bellwood 1997). Less agreement, however, is found on the question of high-ordersubgrouping among the languages spoken in Taiwan. This short article summarizes three proposals, eachrefining and developing previous approaches by successively expanding the range of data under considera-tion, from phonological to morphological and then to syntactic.

2 Phonology, morphology, and the question of nuclearAustronesian

Building on earlier work by Li (1981), Tsuchida (1976, 1982), and others, Blust (1999, 2009/2013) proposesten primary subgroups of Austronesian, nine of which are comprised exclusively of languages which either

*Corresponding author: Edith Aldridge, Department of Linguistics, University of Washington, Box 352425, Seattle,WA 98195–2425, USA, E-mail: [email protected]

Linguistics Vanguard 2015; 1(1): 313–326

UnauthenticatedDownload Date | 7/19/16 2:50 AM

are or were spoken only in Taiwan. The tenth subgroup is Malayo-Polynesian, containing all Austronesianlanguages spoken outside of Taiwan. Blust’s proposal is based entirely on phonological mergers anddeletions, which are commonly acknowledged to be reliable indicators of subgrouping, since they areirreversible and consequently must be understood as innovations rather than retentions.

(1) Austronesian (Blust 1999: 45)1

PuyumaRukaiTsouic: Tsou, Kanakanavu, SaaroaNorthwest Formosan: Saisiyat, Kulon-PazihAtayalic: Atayal, SeediqWestern Plains: Thao, Taokas, Favorlang-Babuza, Papora, HoanyaBununPaiwanEast Formosan: Basay-Trobiawan, Kavalan, Amis, SirayaMalayo-Polynesian: all extra-Formosan languages

Ross (2009, 2012) takes his primary evidence from morphosyntactic paradigms to argue that there are onlyfour first-order subgroups, separating Puyuma, Rukai, and Tsou from a large subgroup called “NuclearAustronesian” (NucAn), which encompasses all other Austronesian languages. Most of the subgroups listedin (2) are equivalent to Blust’s (1999) proposal, with the exception of Tsouic, which is split apart by NucAn:Kanakanavu and Saaroa are located inside NucAn, while Tsou proper is outside.

(2) Austronesian (Ross 2009: 316)PuyumaRukaiTsouNuclear AustronesianKanakanavu, SaaroaNorthwest FormosanAtayalicWestern PlainsBununPaiwanEast FormosanMalayo-Polynesian

The primary innovation which Ross attributes to Proto-NucAn is the reanalysis of embedded nominaliza-tions as realis root clauses, revising an earlier proposal by Starosta et al. (1981, 1982), who assumed that thischange is reflected in all first-order subgroups of Austronesian. Starosta et al. proposed that the origin ofergative alignment illustrated by Tagalog in (3a, b) can be accounted for if the transitive clauses are derivedfrom copula constructions in which the nominative NP is the subject predicated of a headless relativeclause containing all other material in the clause. The aspectual and applicative affixes appearing ontransitive verbs in Philippine and most Formosan languages were reconstructed as nominalizers used inrelative clause formation. The external argument appears inside the relative clause and is marked withgenitive case, which is used for both ergative subjects (3b, c) and possessors (3d) in Philippine and mostFormosan languages today.

1 The order of subgroups listed by Blust (1999: 45) is different from that given here. I have made this change in order to makethe list more directly comparable with Ross’ (2009) subgrouping.

314 E. Aldridge: Minimalist Approach to the Emergence of Ergativity


Tagalog(3) a. D<um>ating ang babae.

<INTR.PRV>arrive NOM woman“The woman arrived.”

b. [B<in>ili ng babae] ang isda.<TR.PRV>buy GEN woman NOM fish“The woman bought the fish.”

c. [B<in>ilh-an ng babae ng isda] ang tindahan=ko.<TR.PRV>buy-APPL GEN woman GEN fish NOM store=1.SG.GEN“The woman bought a/the fish at my store.”

d. isda ng babaefish GEN woman“(the) woman’s fish”

Ross’ main revision is to show that this reanalysis is not reflected in all first-order Austronesian subgroups.For example, Puyuma verbs in relative clauses are affixed with nominalizing morphemes, while verbs inroot clauses take different affixes. The perfective aspect marker <in> in (4b) surfaces only on nominalizedverbs in Puyuma, but appears on finite verbs in Tagalog, as in (3b, c). Note further the Puyuma nominalizer–an, which has been reanalyzed as a locative applicative in Philippine languages, as in (3c).

Puyuma(4) a. tu=trakaw-aw na paisu kan isaw

3.GEN=steal-TR1 DEF.NOM money SG.OBJ Isaw“Isaw stole the money.”(Teng 2008: 147)

b. ala amuna sadru [[tu=tr<in>ekelr-an] na asi]maybe because many 3.PSR=<PRV>drink-NMLZ DEF.NOM milk“Maybe because the milk he drank is a lot.”(Teng 2008: 105)

As mentioned earlier, the main discrepancy between the Ross (2009) and Blust (1999) subgroupinghypotheses lies in accepting a Tsouic subgroup, which was first proposed by Tsuchida (1976) and hasbeen adopted by Ho (1998), Blust (1999), and Sagart (2004, 2010). But Ross (2012) argues that several of thechanges cited by Tsuchida (1976) are also observed in other subgroups and therefore could have beeninnovated independently in the languages in question, while other proposed innovations are more plau-sibly explained as instances of borrowing.

Ross (2012) further offers morphosyntactic evidence from Chang (2006) that Kanakanavu andSaaroa belong to the NucAn subgroup. For example, the latter languages reflect the reanalysis of nomina-lizations as root verbal categories. The perfective makers ni- and li- on the finite verbs in (5) are cognatewith Puyuma’s nominal perfective marker in (4b). Note also the genitive subject in the transitive clausein (5b).

(5) a. ni-pia-pacai cau tutuiPRV-CAUS-kill person pig“Someone killed the pig.”(Kanakanavu; Chang 2006: from Mei 1982: 221)

b. ƚi-ala na cucu’u ka vutukuƚuPRV-take GEN person NOM fish“The person took the fish away.”(Saaroa; Chang 2006; from Li et al. 1997: 279)

E. Aldridge: Minimalist Approach to the Emergence of Ergativity 315

By accepting with Ringe et al. (2002: 68), Thomason and Kaufman (1988: 52, 98), and Weinreich (1953: 32)that bound morphemes, in particular morphological paradigms, are neither easily borrowed nor indepen-dently innovated, Ross (2012) is able to establish with a reasonable amount of certainty that his proposedinnovation was not the result of borrowing or independent developments in the languages in question. Heis further able to determine the direction of the change, which is from nominalization to verb and not vice-versa, because the extra-NucAn languages reflect only the nominal use. However, when it comes toreconstructing the parent language of the entire family, deciding between retentions and innovations isfar less simple, as the only evidence available can be found in the languages which are in turn descendedfrom this proto-language. Ross (2012: 1255) refers to this as the “root-node limitation”.

One recourse in this case is to reconstruct the most commonly occurring forms in the daughterlanguages, the so-called “majority wins” strategy. Among the Austronesian languages of Taiwan and thePhilippines, only Rukai has canonical accusative alignment, the others being characterized by a type ofnon-accusative alignment which has been labeled a “focus system”, “voice system”, or “ergative” (seeSchachter 1976; Payne 1982; Gerdts 1988; de Guzman 1988; Shibatani 1988; Himmelmann 1999, 2005; Liao2002, 2004; Aldridge 2004, 2008; Chang 2011; and others for views and discussion). The “majority wins”premise would suggest that PAn should be reconstructed with this non-accusative type of alignment, whichis the choice opted for by Ross (2009, 2012).

On the other hand, majority rule is not an absolute deciding factor in reconstruction. Given thatdiscrepancies among sister languages must be accounted for in terms of changes from the reconstructedparent language, it is important to posit changes which are cross-linguistically plausible. In the followingsection, I propose a revision of Ross’ (2009) high-order Austronesian subgrouping and reconstruction ofPAn alignment. Specifically, I go against the grain of “majority wins” and reconstruct PAn as an accusativelanguage and show how the change to ergative alignment reflected in Tsou, Puyuma, and NucAn was anatural change conditioned by syntactic factors. I refer to the subgroup defined by this innovation as“Ergative Austronesian” (EAn).

(6) Austronesian

Rukai Ergative An

Tsou Puyuma Nuclear An

3 New proposal

Aldridge (2013, 2014) proposes that PAn was an accusative language and that this alignment is preserved inRukai. Ergative alignment was first innovated in Rukai’s sister Proto-Ergative Austronesian in irrealis clauses,which I posit was the result of a commonly observed process of detransitivization in this clause type.

3.1 Morphology of irrealis in ergative Austronesian

I first consider morphological evidence for the connection between irrealis clauses and ergative alignmentin EAn. According to Teng’s (2008) description, Puyuma is a language with ergative or split-ergativealignment. What she glosses as “intransitive” is a simple intransitive or an antipassive.2 The three different

2 Teng (2008) does not label this construction “antipassive”, but she characterizes it as having attributes commonly found inantipassive constructions.



transitive clause types differ in whether they are basic (“type 1”) transitives or applicative constructions.Applicatives are further divided into locative (“type 2 transitives”) and benefactive/instrumental (“type 3transitives”) applicatives.

Puyuma realis (Teng 2008: 147)(7) a. trakaw dra paisu i isaw

<INTR>steal INDEF.OBL money SG.NOM Isaw“Isaw stole money.”

b. tu=trakaw-aw na paisu kan isaw3.GEN=steal-TR1 DEF.NOM money SG.OBL Isaw“Isaw stole the money.”

c. tu=trakaw-ay=ku dra paisu kan isaw3.GEN=steal-TR2=1SG.NOM INDEF.OBJ money SG.OBL Isaw“Isaw stole money from me.”

d. tu=trakaw-anay i tinataw dra paisu3.GEN=steal-TR3 3.SG.NOM his.mother INDEF.OBL money“He stole money for his mother.”

Another interesting fact about Puyuma is that, like many Formosan languages, it employs different affixesin realis and irrealis mood. (8) shows an array of focus affixes employed in imperative sentences.

Puyuma imperatives (Teng 2008: 216)(8) a. trekelr

drink“Drink!”

b. pilang-u i temuu m-ukatake-TR1.IMP SG.NOM your.grandmother INTR-goi drena-drenanLOC RED-mountain

“Take your grandmother to the mountains.”

c. puka-i dra tidrul dra samayaput-TR2.IMP INDEF.OBL wasp INDEF.OBL some“Put some wasps (in).”

Examining the paradigm in (9) reveals an interesting parallel between realis and irrealis (imperative,negative, and future) transitive suffixes. The transitive realis suffixes all include a, which is followed bya glide. The glides in turn bear striking resemblance to the vowel suffixes –u and -i in the irrealis forms. Infact, all of the realis forms can be derived by adding –a to the verb root or to the TR3 base V-an andsubsequently affixing either –u (for TR1) or –i (TR2, TR3).3

(9) Puyuma verbal inflection (adapted from Ross 2009: 304)INTR TR1 TR2 TR3

Realis V V-aw V-ay V-anayImperative V V-u V-i V-anNegative V V-i V-i V-anFuture RED-V RED-V-i RED-V-i RED-V-an

3 Ross (1995, 2002) identified a similar pattern, but did not propose and explanation for it.

Regarding the origin and function of the pre-glide a, Ross (1995, 2002, 2009) reconstructs an *–a suffix asthe PAn nonfinite type 1 transitive suffix. This is understandable, given the role of the reflex in Tsou as thetype 1 transitive suffix which surfaces on nonfinite verbs. The only finite verb forms in Tsou are auxiliaryverbs. These show a transitive/intransitive4 dichotomy, but –a does not appear on auxiliaries.

Tsou (Chang 2011: 285; based on Zeitoun 2000: 93–94)(10) a. mo mo-si ta pangka to emi ‘o amo

INTR.3S INTR-put OBL table OBL wine ABS father“Father put wine on the table.”

b. i-si si-a ta pangka to amo ‘o emiTR-3.SG put-TR OBL table ERG father ABS wine“Father put the wine on the table.”

Thus, Tsou offers compelling evidence that *-a attached to nonfinite verbs. Given the potential to furtherattach irrealis affixes, as in Puyuma, I reconstruct *-a as a nonfinite irrealis suffix, which I refer to simply as“subjunctive”. Further support comes from the fact that the reflex of *-a occurs on nonfinite, hortative,imperative, or subjunctive verbs in a variety of Formosan, Philippine, and Malayic languages, as noted byRoss (2009). Below are imperative examples from Rukai.

Tanan Rukai(11) a. kanɨ-a

<M>eat-IMP

“Eat!”

b. sila’-a na avava-suseek-IMP ACC.DEF toy-2SG“Look for your toy!”

Regarding the vowels/glides, Tsou also has –i on TR2 and TR3 verbs, suggesting that this portion of theparadigm be reconstructed to a common ancestor of Puyuma and Tsou.

(12) INTR TR1 TR2 TR3Nonfinite m-V V-a V-i V-(n)eni

At this point, we have arrived at the following reconstruction of the parent of Puyuma and Tsou. The non-accusative type of alignment found in modern Puyuma is reconstructed only for irrealis contexts. Tsouretains the subjunctive forms as the basic nonfinite paradigm. The affixes have been simplified somewhat,and this may be due to a process of monophthongization, given that modern Tsou does not havediphthongs. The first column in (13) is labeled “verb” rather than “intransitive”. This is because ergativealignment is not found in realis clauses, where *M-V would have appeared on either transitive orintransitive verbs.

(13) PEAn reconstructionVerb TR1 TR2 TR3

Realis *M-V … … …Subjunctive *M-V-a *V-a-u *V-a-i *V-an-a-iImperative *V *V-u *V-i *V-an-i

4 (10a) appears to be transitive since it contains a direct object, but this is analyzed by Chang (2011) as an antipassive. Note thatthe direct object is indefinite.

I have tentatively attributed the reconstructions in (13) to Proto-Ergative Austronesian, so NuclearAustronesian languages are also predicted to reflect the irrealis transitive suffixes, and indeed they do.The realis affixes in the Atayalic language Seediq are reflexes of PAn nominalizers, as proposed by Ross(2009). The irrealis affixes are clearly related to the transitive affixes in Tsou and Puyuma.

(14) Seediq realis vs irrealis (Holmer 1996: 385)INTR TR1 TR2 TR3

Realis <m>/<m>-V V-un V-an s-VImperative V V-i V-ani ...Negative V V-i V-ani ...

On the other hand, Rukai has consistent accusative alignment and does not reflect the transitive affixes in(13), though it does retain the *-a subjunctive as an imperative, as shown above in (11).

Tanan Rukai(15) a. uduri=aku sa bɨlbɨl

plant=1SG.NOM ACC.INDEF banana“I plant bananas.”

b. labuwal=aku kɨlawalk=1SG.NOM come“I come walking.”

On this basis, it is reasonable to conclude that the transitive irrealis paradigm arose as the result of aninnovation in Proto-Ergative Austronesian, meaning that PAn can be reconstructed with accusative align-ment, which is retained in Rukai.

(16) PAn reconstructionVerb TR1 TR2 TR3

Realis *M-V … … …Subjunctive *M-V-a … … …Imperative *V … … …

Before proceeding, however, it must be acknowledged that Ross’ (2012) “root node limitation” makes itdifficult to conclude with any certainty that irrealis suffixes were innovated. Another logical possibility is toattribute (13) to PAn and posit that Rukai lost the transitive irrealis suffixes. In the next subsection, Isupport the proposed reconstruction of PAn as accusative by positing a natural syntactic change to accountfor the innovation of ergative alignment in irrealis clauses in Proto-Ergative Austronesian.

3.2 Innovation in the irrealis in proto-ergtive Austronesian

Within the Minimalist Program of Chomsky (2001) and subsequent works, the functional heads finite T andtransitive v value nominative and accusative case, respectively, with the first NP in their c-commanddomains. Consequently, nominative case generally appears on the structurally most prominent NP in theclause, specifically the subject, while accusative case is found on the highest NP within the VP, i.e. the

5 The chart in Holmer (1996: 38) does not show the forms for negation but Holmer points out on page 62 that the negator ini isfollowed by imperative verb forms.

object. In other words, this process results in accusative alignment, most notably the uniform assignment ofnominative case to the first member of the verb’s argument structure.

(17) a. She[NOM] walks.b. She[NOM] sees him[ACC].c.

TP

T[uf] vP

NP[f, NOM] v’

v[uf] VP

V NP[f,ACC]

In an ergative language, nominative case is not assigned uniformly to the subject. Rather, the subject in anintransitive clause shares the case of the object in a transitive clause.

Seediq(18) a. Wada kudurjak ka qedin=na.

PAST flee NOM wife=3S.GEN“His wife ran away.”

b. Wada bube-un na Pihu ka dangi=na.PAST hit-TR ERG Pihu NOM friend=3S.GEN“Pihu hit his friend.”

This lack of uniformity requires a modification to the analysis in (17). A straightforward approach has beenformulated by Legate (2008), which posits an exceptional type of v in transitive clauses to assign aninherent case to the subject. This v does not make accusative case available for the object, but since thesubject does not require licensing from T, T is free to look past the subject and value nominative case on theobject. See also Bok-Bennema (1991), Bittner and Hale (1996), Ura (2000), Alexiadou (2001), Aldridge(2004), Anand and Nevins (2006) for similar approaches.

(19) TP

T[uf] vP

DP[f, INH] v’

v VP

V DP[f, NOM]



Aldridge (2012) has proposed that such a parametric distinction can account for the change from accusativeto ergative alignment in nominalized clauses in Proto-Nuclear Austronesian. 6 The input to the reanalysis isa cleft construction embedding a reduced relative clause nP. Since n is a nominal category, it does not makeaccusative case available for the object. The object raises to the edge of nP in order to receive a focusinterpretation and also be case licensed by T. The subject is assigned genitive case in its base position in nP.Ergative alignment results when the cleft is reanalyzed as a monoclausal construction after loss of thecopula in T, which in turn results in relabeling (in the sense of Whitman 2000) of n as v.

(20) TP

T[uf] nP

DP[f, NOM] n’

DP[GEN] n’

n[EPP] NP

N tDP

Aldridge (2014) proposes that detransitivization of v in irrealis clauses accounts for the emergence of non-accusative alignment in Proto-Ergative Austronesian. Irrealis clauses are detransitivized in a number oflanguages,7 in the sense that structural case is not available for the object. One example is the phenomenonknown as “genitive of negation” in Slavic languages. In the following Russian examples, an object oftenreceives genitive case in the scope of sentential negation, as in (21). Since inherent case is assigned to theobject rather than the subject, alignment does not change, both transitive and intransitive subjectscontinuing to be marked nominative.

Russian (Harves 2002: 97)(21) Anna ne kupila knig.

Anna.NOM NEG bought books.GEN

“Anna did not buy any books.”

Under different conditions, the object can surface with nominative case. Objects in telic events in Finniclanguages require structural licensing,8 which is typically accusative case, as in (22a). In atelic events,objects are assigned partitive case, as in (22b). Note the object is also definite in a telic event, while it can be(and often is) indefinite in atelic events.

6 Clausal nominalizations are a common source for ergative alignment. See Johns (1992) for Inuit, Gildea (1998) for Cariban, andWhitman and Yanagida (2012) on diachronic or synchronic connections between nominalization and ergative alignment.7 Hopper and Thompson (1980) propose realis/irrealis as one of their transitivity parameters. See also Denniss (2007), Beaversand Zubair (2010), and others for analyses of irrealis clauses as intransitive.8 See Kiparsky (1998) for discussion of the correlation between case and aspect in Finnish.



Estonian (Hiietam 2004)(22) a. Poiss luges raamatu läbi.

boy.NOM read.PAST.3.SG book.ACC through

b. Poiss luges raamatut.boy.NOM read.PAST.3.SG book.PART“The boy was reading a/the book.”

Hiietam (2004) shows that accusative case is not available in imperative clauses in Estonian. The objectreceives partitive case, as expected, in the atelic clause in (23a) but appears with nominative case in thetelic clause in (23b). I interpret this as a consequence of the detransitivization of v, depriving it of the abilityto value accusative case. When the object requires structural licensing in the telic event, it must undergoAgree with T and value nominative case.

Estonian (Hiietam 2004)(23) a. Söö võileiba!

eat.2.SG.IMP sandwich.PART“Eat some sandwich!/ i.e. Do some sandwich eating!”

b. Söö võileib ära!eat.2.SG.IMP sandwich.NOM up“Eat the sandwich up!”

The emergence of non-accusative alignment in irrealis clauses in Proto-Ergative Austronesian, as proposedabove in (13), can be accounted for as a consequence of the detransitivization of irrealis v, coupled with theavailability of inherent case for the subject. Note that subjects in transitive clauses in Puyuma are genitive,while the object is nominative, as shown in (24a). The object in the transitive clause is definite, while theobject in the antipassive in (24b) is indefinite. This alternation also falls out naturally on the proposedanalysis. Just as in Finnic languages, I assume that definite objects in telic events in PEAn requiredstructural licensing, forcing the object to undergo Agree with T in irrealis clauses. But in atelic events,the indefinite object could be assigned inherent case. The transitive irrealis clauses were later reanalyzed asrealis root clauses in Puyuma. The objects in these clauses are always definite and have nominative case(Teng 2008: 147). But the inherited realis clause type, with the verb inflected with a reflex of *M-, wouldhave been retained in atelic events with inherently case marked indefinite objects. Put differently, thisanalysis accounts for why the reflex of *M- is found on intransitive or antipassive verbs in Formosan andPhilippine languages today, while transitive clauses have a different diachronic source (irrealis in Puyumaand Tsou) and nominalization in NucAn languages.

Puyuma realis (Teng 2008: 147)(24) a. tu=trakaw-aw na paisu kan isaw

3.GEN=steal-TR1 DEF.NOM money SG.OBL Isaw“Isaw stole the money.”

b. trakaw dra paisu i isaw<INTR>steal INDEF.OBL money SG.NOM Isaw“Isaw stole money.”

In this way, the emergence of ergative alignment in irrealis clauses in Proto-Ergative Austronesian can beunderstood as a natural and conditioned syntactic change, resulting from the detransitivization of v inirrealis clauses. As proposed in section 3.1, Tsou reflects the subjunctive of PEAn in embedded, nonfinitecontexts. In Puyuma, the subjunctive was reanalyzed as a finite root clause type. Aldridge (2014) proposesthat this was triggered by the loss of auxiliaries in this language, which are retained in Tsou. Evidence for

the loss of the auxiliaries in Puyuma comes in the guise of clitic placement. Subject agreement markers inTsou are all post-verbal (specifically, post-auxiliary), as can be seen in (25), as well as in (10).

Tsou(25) a. te-ta m-imo ta emi

will-3SG.BN AF-drink OBL wine"He will drink wine."(Zeitoun 1996: 517)

b. te-ta n’a ana ’e naveu?IRR-3 ASP eat.PF9 NOM rice"Is she going to eat the rice (later)?"(Huang and Huang 2003: 10)

In contrast, weak pronominal subjects in intransitive clauses in Puyuma are enclitics, while they procliticizeto the verb in transitive clauses. Note the lack of an auxiliary verb, the main verb occupying clause-initialposition.

Puyuma(26) a. bəray=ku ɖa kuraw ɖa ŋiaw

give=1.SG.NOM OBL.INDEF fish OBL.INDEF cat"I gave a fish to a cat."(Tan 1997: 11)

b. tu=trakaw-aw na paisu kan isaw3.GEN=steal-TR1 DEF.NOM money SG.OBJ Isaw“Isaw stole the money.”(Teng 2008: 147)

Starosta et al. (1982) and Ross (2002, 2006) have proposed that clitics in PAn were all enclitics and thatproclitics in the languages that have them are the result of the loss of a clause-initial auxiliary verb. I adoptthis analysis here and further suggest that the loss of the subjunctive-introducing auxiliary in Puyumaprovided the trigger for the reanalysis of subjunctive to realis root clause in this language. Specifically,without the auxiliary, the child acquiring the language did not have evidence that the verb was nonfinite(or embedded, for that matter). Consequently, they chose the default parameter (in the sense of Roberts1997, Roberts and Roussou 2003) setting and acquired these forms as finite root verbs.

4 Methodological summary

This article has summarized three approaches to high-order subgrouping in Austronesian languages. I haveshown how subgrouping hypotheses have been refined by examining different kinds of linguistic evidence:sounds, morphology, and syntax. The basic methodology is the same in all three cases. First, forms arecompared in an attempt to reconstruct the linguistic system in the language ancestral to all of the languagesunder consideration. When discrepancies are found among the related languages, reconstructions arechosen so as to account for synchronic variation by means of natural changes.

I have proposed a syntactic change that accounts for the emergence of ergative alignment inAustronesian languages through a commonly observed process of detransitivization. The naturalness ofthis change suggests that PAn should be reconstructed with accusative alignment, which is retained inRukai, while ergative alignment was innovated in Rukai’s sister, Proto-Ergative Austronesian.

9 The main verb in (25b) is glossed as “patient focus” to indicate that the semantic object is the argument with nominative casein this clause type.



References

Adelaar, Alexander. 2005. The Austronesian languages of Asia and Madagascar: a historical perspective. In AlexanderAdelaar & Nikolaus P. Himmelmann (eds.), The Austronesian languages of Asia and Madagascar, 1–42. London:Routledge.

Aldridge, Edith. 2004. Ergativity and word order in Austronesian languages. Ithaca, NY: Cornell University dissertation.Aldridge, Edith. 2008. Phase-based account of extraction in Indonesian. Lingua 118(10). 1440–1469.Aldridge, Edith. 2012. Nominalization source of ergativity in Tagalog. Paper presented at the 14th Diachronic Generative Syntax

(DIGS) conference, University of Lisbon.Aldridge, Edith. 2013. Two origins of ergativity in Austronesian languages. Paper presented at A Minimalist Workshop on

Austronesian Verbal Syntax, Institute of Linguistics, Academia Sinica, Taipei.Aldridge, Edith. 2014. Ergativity from subjunctive in Austronesian Languages. Paper presented at The 14th International

Symposium on Chinese Languages and Linguistics, Academia Sinica, TaiwanAlexiadou, Artemis. 2001. Functional structure in nominals. Amsterdam: John Benjamins.Anand, Pranav & Andrew Nevins. 2006. The locus of ergative case assignment: evidence from scope. In Alana Johns, Diane

Massam & Juvenal Ndayiragije (eds.), Ergativity: emerging issues, 3–25. Dordrecht: Springer.Beavers, John & Cala Zubair. 2010. The interaction of transitivity features in the Sinhala involitive. In Patrick Brandt & Marco

Garcia (eds.), Transitivity. Form, meaning, acquisition, and processing, 69–92. Amsterdam: John Benjamins.Bellwood, Peter. 1997. Prehistory of the Indo-Malaysian archipelago, revised edition). Honolulu, HI: University of

Hawai’i Press.Bittner, Maria & Ken Hale. 1996. The structural determination of case and agreement. Linguistic Inquiry 27. 1–68.Blust, Robert A. 1999. Subgrouping, circularity and extinction: some issues in Austronesian comparative linguistics. In Selected

Papers from the eighth International Conference on Austronesian Linguistics (Symposium Series of the Institute ofLinguistics), Elizabeth Zeitoun and Paul Jen-kuei Li (eds.), 31–94. Taipei: Institute of Linguistics (Preparatory Office),Academia Sinica.

Blust, Robert A. 2009//2013. The Austronesian languages (Pacific Linguistics 602; 2nd Edn. Asia-Pacific Linguistics 8).Canberra: Research School of Pacific and Asian Studies.(2nd edn. Collage of Asia and the Pacific), Australian NationalUniversity).

Bok-Bennema, Reineke. 1991. Case and agreement in inuit. Berlin: Foris Publications.Chang, Henry Y. 2006. Rethinking the tsouic subgroup hypothesis: a morphosyntactic perspective. In Henry Y. Chang, Lillian M.

Huang & Dah-an Ho (eds.), Streams converging into an ocean: festschrift in honor of professor Paul Jen-Kuei Li on His 70th

birthday, 565–583. Taipei: Language and Linguistics.Chang, Henry Y. 2011. Transitivity, ergativity, and the status of O in Tsou. In Jung-hsing Chang (ed.), Language and cognition:

festschrift in honor of James H-Y. Tai on His 70th birthday, 277–308. Taipei: Crane Publishing.Chomsky, Noam. 2001. Derivation by phase. In Michael Kenstowicz (eds.), Ken Hale: A life in language, 1–52. Cambridge, MA:

MIT Press.De Guzman, Videa P. 1988. Ergative analysis for Philippine languages: an analysis. In Richard McGinn (ed.), Studies in

Austronesian linguistics, 323–345. Athens, OH: Ohio University Center for International Studies.Denniss, Jessica. 2007. Antipassives in Yukulta. In Andreas Jäger & Rob Pensalfini (eds.), The University of Queensland working

papers in linguistics. http://www.library.uq.edu.au/ojs/index.php/uqwpl/indexGerdts, Donna B. 1988. Antipassives and causatives in Ilokano: evidence for an ergative analysis. In Richard McGinn (ed.),

Studies in Austronesian linguistics, 295–321. Athens, OH: Ohio University Center for International Studies.Gildea, Spike. 1998. On reconstructing grammar. New York: Oxford University Press.Harves, Stephanie. 2002. Genitive of negation and the syntax of scope. In M. van Koppen, E. Thrift, E.J. van der Torre & M.

Zimmerman (eds.), Proceedings of ConSOLE. 96–110. http://www.leidenuniv.nl/hil/sole/Hiietam, Katrin. 2004. Case marking in Estonian grammatical relations. Leeds Working Papers in Linguistics and Phonetics. Vol.

10. University of Leeds. http://www.leeds.ac.uk/linguistics/WPL/WPL.htmlHimmelmann, Nikolaus. 1999. The lack of zero anaphora and incipient person marking in Tagalog. Oceanic Linguistics 38(2).

231–269.Himmelmann, Nikolaus P. 2005. The Austronesian languages of Asia and Madagascar: typological characteristics. In Alexander

Adelaar & Nikolaus P. Himmelmann (eds.), The Austronesian languages of Asia and Madagascar, 110–181. New York:Routledge.

Ho, Dah-an. 1998. Taiwan Nandaoyu De Yuyan Guanxi [genetic relationships among the Formosan languages]. Chinese Studies16(2). 141–171.

Holmer, Arthur. 1996. A parametric grammar of Seediq. Lund: Lund University Press.Hopper, Paul & Sandra Thompson. 1980. Transitivity in grammar and discourse. Language 56. 251–299.Huang, Shuanfan & Huei-Ju Huang. 2003. On the status of reality marking in Tsou. Taiwan Journal of Linguistics 1(2). 1–34.Johns, Alana. 1992. Deriving ergativity. Linguistic Inquiry 23. 57–88.



Kiparsky, Paul. 1998. Partitive case and aspect. In M. Butt, W. Geuder (eds.), The projection of arguments, 265–307. Stanford:CSLI.

Legate, Julie. 2008. Morphological and abstract case. Linguistic Inquiry 39. 55–101.Li, Paul Jen-kuei. 1981. Reconstruction of proto-atayalic phonology. Bulletin of the Institute of History and Philology Academia

Sinica 44(2). 311–338.Li, Paul Jen-kuei, Dah-an Ho, Mei-jing Huang, Elizabeth Zeitoun & Claire Saillard (eds.). 1997. Austronesian languages in

Kaohsiung County. Kaohsiung: Kaohsiung County Government [in Chinese].Liao, Hsiu-chuan. 2002. The interpretation of Tu and Kavalan ergativity. Oceanic Linguistics 41(1). 140–158.Liao, Hsiu-chuan. 2004. Transitivity and Ergativity in Formosan and Philippine Languages. Ph.D. Dissertation, University of

Hawaii.Mei, Kuang. 1982. Pronouns and verb inflection in Kanakanavu. Tsinghua Journal of Chinese Studies, New Series 14. 207–232.Payne, Thomas. 1982. Role and reference related subject properties and ergativity in Yup’ip Eskimo and Tagalog. Studies in

Language 6(1). 75–106.Ringe, Don, Tandy Warnow & Ann Taylor. 2002. Indo-European and computational cladistics. Transactions of the Philological

Society 100(1). 59–129.Roberts, Ian. 1997. Directionality and word order change in the history of English. In Ans van Kemenade &Nigel Vincent (eds.),

Parameters of morphosyntactic change, 397–426. Cambridge: Cambridge University Press.Roberts, Ian & Anna Roussou. 2003. Syntactic change: a minimalist approach to grammaticalization. Cambridge: Cambridge

University Press.Ross, Malcolm. 1995. Reconstructing proto-Austronesian verbal morphology: evidence from Taiwan. In Paul J.-K. Li, Cheng-hwa

Tsang, Ying-kuei Huang, Dah-an Ho & Chiu-yu Tseng (eds.), Austronesian studies relating to Taiwan, 727–791. Taipei:Academia Sinica.

Ross, Malcolm. 2002. The history and transitivity of western Austronesian voice and voice-marking. In Fay Wouk & MalcolmRoss (eds.), The history and typology and western Austronesian voice systems, 17–62. Canberra: Pacific Linguistics.

Ross, Malcolm. 2006. Reconstructing the case-marking and personal pronoun systems of proto Austronesian. In Henry Y.Chang, Lillian M. Huang & Dah-an Ho (eds.), Streams converging into an ocean: festschrift in honor of professor Paul Jen-Kuei Li on His 70th birthday, 521–563. Taipei: Language and Linguistics.

Ross, Malcolm. 2009. Proto Austronesian verbal morphology: a reappraisal. In K. Alexander Adelaar & Andrew Pauley (eds.),Austronesian historical linguistics and culture history: a festschrift for Robert Blust (Pacific Linguistics 601)), 295–326.Canberra: Research School of Pacific and Asian Studies, Australian National University.

Ross, Malcolm. 2012. In defense of nuclear Austronesian (and against Tsouic). Language and Linguistics 13(6). 1253–1330.Sagart, Laurent. 2004. The higher phylogeny of Austronesian and the position of Tai-Kadai. Oceanic Linguistics 43(2). 411–444.Sagart, Laurent. 2010. Is puyuma a primary branch of Austronesian? Oceanic Linguistics 49(1). 194–204.Schachter, Paul. 1976. The subject in Philippine languages: topic, actor, actor-topic, or none of the above. In Charles Li (ed.),

Subject and topic, 491–518. New York: Academic Press.Shibatani, Masayoshi. 1988. Voice in Philippine languages. In Masayoshi Shibatani (ed.), Passive and voice, 85–142.

Amsterdam: John Benjamins.Starosta, Stanley, Andrew K. Pawley & Lawrence A. Reid. 1981. The evolution of focus in Austronesian. Paper presented at the

3rd International Conference on Austronesian Linguistics, Bali.Starosta, Stanley, Andrew K. Pawley & Lawrence A. Reid. 1982/2009. The evolution of focus in Austronesian. In Amran Halim,

Lois Carrington & S.A. Wurm (eds.), Papers from the third international conference on Austronesian linguistics. Vol. 2:tracking the travellers (Pacific Linguistics C-65), 145–170. Canberra: Research School of Pacific and Asian Studies,Australian National University.(republished in Elizabeth Zeitoun (ed.), Formosan linguistics: Stanley Starosta’s contribu-tions. Vol. 2: Publications on Formosan languages (Language and Linguistics Monograph Series C6-65). Taipei: Institute ofLinguistics, Academia Sinica, 297-328 [with an expanded version of the paper, with the same title, published for the firsttime in the same volume, 329–481]).

Tan, Cindy Ro-lan. 1997. A study of puyuma simple sentences. Taipei: National Taiwan Normal University MA Thesis.Teng, Stacy Fang-ching. 2008. A reference grammar of puyuma, an Austronesian language of Taiwan (Pacific Linguistics 595).

Canberra: Research School of Pacific an Asian Studies, Australian National University.Thomason, Sarah Grey & Terrence S. Kaufman. 1988. Language contact, creolization and genetic linguistics. Berkeley, CA:

University of California Press.Tsuchida, Shigeru. 1976. Reconstruction of Proto-Tsouic phonology. Ph. D. dissertation, Yale University. New Haven. Tokyo:

Study of Languages & Cultures of Asia & Africa, Monograph Series No.5.Tsuchida, Shigeru. 1982. A comparative vocabulary of Austronesian languages of Sinicized ethic groups in Taiwan, Part 1: West

Taiwan. Memoirs of the Faculty of Letters, No. 7. Tokyo: University of Tokyo.Ura, Hiroyuki. 2000. Checking theory and grammatical functions in universal grammar. New York: Oxford University Press.Weinreich, Uriel. 1953. Languages in contact: findings and problems. New York: Linguistic Circle of New York.Whitman, John. 2000. Relabelling. In Susan Pintzuk, George Tsoulas & Anthony Warner (eds.), Diachronic syntax: models and

mechanisms, 220–238. Oxford: Oxford University Press.



Whitman, John & Yuko Yanagida. 2012. The formal syntax of alignment change. In Charlotte Galves, Sonia Cyrino, Ruth Lopes,Filomena Sandalo & Juanito Avelar (eds.), Parameter theory and dynamics of change, 177–195. Cambridge: OxfordUniversity Press.

Zeitoun, Elizabeth. 1996. The Tsou Temporal, aspectual, and modal system revisited. Bulletin of the Institute of History andPhilology 67(3). 503–531.

Zeitoun, Elizabeth. 2000. A Tsou reference grammar. Taipei: Yuanliou Publishing Co. Series on Formosan Languages, 7.(in Chinese).



edith aldridge* a minimalist approach to the emergence of...

Documents