Abstract

Applicatives of unaccusatives provide a crucial test case for the inherent-case view of ergativity. If ergative is assigned only to external arguments, in their θ-positions, there can be no “raising to ergative” in applicative unaccusatives; an internal argument subject can never receive ergative case. In this article, I present evidence from Nez Perce (Sahaptian) that this prediction is false. In Nez Perce applicative unaccusatives, the theme argument raises over the applicative argument and is accordingly marked with ergative case. Nez Perce thus demonstrates raising to ergative. Departing from Baker’s (2014) conclusions for similar phenomena in Shipibo (Panoan), I argue that apparently nonlocal movement of the theme in the raising-to-ergative pattern involves not a covert adpositional structure, but rather a response to independently motivated constraints on antilocal movement and remnant movement.

1 Introduction

The inherent-case view of ergativity holds that ergative case is assigned to an external argument in its θ-position by the v or Voice head that introduces it.1 A central prediction is therefore that ergative may only be assigned to external arguments—a prediction that Marantz (1991) dubbed the Ergative Case Generalization.

(1)

  • Ergative Case Generalization

  • Even when ergative case may go on the subject of an intransitive clause, ergative case will not appear on a derived subject. (Marantz 1991:236)

As Legate (2012) notes:

The reference [by Marantz] to the subject of an intransitive clause is to circumvent the confound of the transitivity restriction: in general, transitive verbs have a thematic subject that becomes the surface subject, making it impossible to test whether a derived subject could bear ergative case. An additional way around the confound would be a two-argument verb in which both arguments are internal, for example, the passive of a double object verb, or the applicative of an unaccusative verb. If the Ergative Case Generalization holds, the subject of such verbs would not bear ergative case, despite the presence of two DP arguments. (Legate 2012:183; emphasis added)

In this article I argue, building on work by Baker (2014, 2015), that the Ergative Case Generalization does not hold and that applicatives of unaccusatives provide crucial evidence against it. Focusing on the applicative unaccusatives of Nez Perce (Sahaptian), I demonstrate that the theme argument raises over the applicative argument and is marked with ergative case. This “raising-to-ergative” pattern shows that ergative case cannot be restricted to external arguments, and thus that the inherent-case view cannot be the (only) proper analysis of ergativity.2

The new evidence from Nez Perce replicates and expands on Baker’s argument from applicative unaccusatives in Shipibo (Panoan; Peru). Shipibo shows a canonical ergative-absolutive case alignment; all intransitive subjects bear absolutive case. Contrary to the Ergative Case Generalization, however, applicatives of unaccusatives in Shipibo feature ergative case on the theme argument—a derived (transitive) subject. Compare the applicative unaccusative in (2a), where the subject is ergative, with the basic unaccusative in (2b), where the subject is absolutive.3

(2)

  • a.

    Bimi-n-ra Rosa joshin-xon-ke.

    fruit-ERG-EV Rosa.ABS ripen-APPL-COMPL

    ‘The fruit ripened for Rosa.’

    (Baker 2014:346)

  • b.

    Kokoti-ra joshin-ke.

    fruit.ABS-EV ripen-COMPL

    ‘The fruit ripened.’

    (Baker 2014:345)

On the basis of the Shipibo facts, Baker (2014) motivates a “configurational” analysis of case assignment: ergative is a dependent case in Shipibo, rather than an inherent one.4Baker and Bobaljik (2017) go one step further, arguing that the inherent-case view should be abandoned in favor of the dependent-case view not just for Shipibo, but for ergativity altogether.

The Nez Perce facts introduced here cast light on two types of questions raised by this argument. First, what is the crosslinguistic distribution of raising to ergative in applicative unaccusatives? If Baker and Bobaljik (2017) are right that ergative is never inherent, then raising to ergative should be possible across the full range of languages with ergative case systems, notwithstanding the variety of ways in which this class is internally diverse. On the other hand, if Rezac, Albizu, and Etxepare (2014) and Coon (2016, 2017b) are right to suggest that ergative languages can be divided into an inherent-ergative class and a non-inherent-ergative class, we might expect to find correlations between raising to ergative in applicative unaccusatives and other distinctive characteristics of the two types of ergativity. Nez Perce proves useful in probing for correlations of this type, as it is unrelated to its fellow raising-to-ergative language Shipibo, and the two languages differ along several axes of variation in ergativity. Nez Perce, for instance, has a three-way ergative case system (ergative/nominative/accusative) whereas Shipibo has an ergative-absolutive case system (Rude 1985; Valenzuela 2003); Nez Perce has a syntactically based person split whereas Shipibo has no person split at all (Deal 2016b; Valenzuela 2003); Nez Perce has agreement with both subjects and objects in person and number whereas Shipibo has agreement only in number and only with subjects (Deal 2015c; Valenzuela 2003, 2010).5 However, both languages show ergative case on the theme in applicative unaccusatives. In Nez Perce, subjects of simple unaccusative clauses are nominative, (3). Applicative unaccusatives, in contrast, show ergative marking on the theme subject in Nez Perce, (4), just as in Shipibo.

(3)

  • Ha-’aayat hi-pa-pay-no’-kom.

  • PL-woman.NOM 3SUBJ-S.PL-come-FUT-CIS

  • ‘The women will come.’

(4)

  • Ha-’aayat-om nuun-e hi-pa-naas-pay-noo-yo’-kom.

  • PL-woman-ERG 1PL-ACC 3SUBJ-S.PL-O.PL-come-APPL-FUT-CIS

  • ‘The women will come to us.’

If there is a split between inherent-ergative and non-inherent-ergative languages, then, all differences between Nez Perce and Shipibo must represent diversity internal to the non-inherent-ergative class.

Second, what is the mechanism by which raising to ergative takes place in applicative unaccusatives? In particular, how does raising of the theme over the applicative argument avoid a violation of relative locality? Baker (2014) advocates a solution involving a covert adpositional structure. The Shipibo applicative argument, he proposes, is actually a PP, not a DP. The applicative argument remains in situ because the PP cannot raise to an A-position and the DP subconstituent cannot be extracted from within PP. The theme, on the other hand, is able to move to an A-position above the applicative argument because the latter, as a PP, does not constitute an intervener for A-movement. This proposal for the structure of (2a) is shown in (5).6

(5)

graphic

This structure leads us to expect that applicative arguments should exhibit behavior different from that of other objects, and similar to that of PPs, in respects independent of raising to ergative. In Shipibo, for instance, PPs show overt adpositional structure, constitute opaque domains for case assignment, and behave unlike objects for switch-reference. Applicative arguments, however, show none of these PP behaviors (Baker 2014:n23). They lack any visible adpositional marking and behave like ordinary DP objects for case assignment and switch-reference. Such facts raise the suspicion that applicative arguments are really just DPs after all—in which case, a different explanation will have to be found for the locality behavior of raising to ergative. The alternative I propose is based on a principle of antilocality that regulates movement between specifier positions (Bošković 2015, 2016, Brillman and Hirsch 2016, Erlewine 2016): the theme raises because the applicative argument is too close to the immediate landing site (Spec,v). In addition to the core locality facts of raising to ergative, this proposal combines with Müller’s (1996) constraint on remnant movement to provide a natural account of possessor raising from applicative arguments in Nez Perce—a pattern that proves challenging for the covert-PP proposal.

The article is structured as follows. In section 2, I lay out the basics of case and agreement, unaccusativity, and applicative constructions in Nez Perce. In section 3, I argue that the theme argument indeed raises above the applicative argument in applicative unaccusatives and that Nez Perce exemplifies raising to ergative (contra the Ergative Case Generalization). Section 4 is devoted to the theoretical questions raised by this movement: why does the theme argument move, and why can’t the applicative argument move instead? I advance an antilocality-based solution, drawing on evidence from possessor raising (Deal 2013). In section 5, I conclude by considering the consequences of raising to ergative for the analysis of ergative case.

2 Ergativity, Unaccusativity, and Applicatives in Nez Perce

Nez Perce is a Sahaptian language spoken natively by no more than 30 individuals in Idaho, Washington, and Oregon, USA. Except where otherwise indicated, the data here come from fieldwork on the Nez Perce Reservation in Lapwai, Idaho. Data are presented in the practical orthography used by the language program of the Nez Perce Tribe of Idaho. A table of correspondences to IPA is given in the appendix to Deal 2016b.

As demonstrated in Rude 1985, 1992, Crook 1999, and Deal 2010b, the language allows considerable word order freedom at the clausal level, and pronominal subjects, objects, and possessors of all persons are often omitted. Omitted arguments are indicated here by pro in Nez Perce examples, with the gloss line reflecting the person and number features conveyed by the speaker’s translation. For ease of reading, I follow a convention of placing pros in SVO order.

2.1 Case and Agreement

Nez Perce is well-known for its tripartite ergative case alignment: intransitive subjects, transitive subjects, and objects are all marked distinctly in the 3rd person. Nominative is unmarked; accusative is marked by ne and allomorphs; and ergative is marked by nim and allomorphs. The case-marking system is described and analyzed in Rude 1985, 1986, Woolford 1997, Crook 1999, Carnie and Cash Cash 2006, and Deal 2010a,b, 2016b.

(6)

  • Angel hi-pnim-se.

  • Angel.NOM 3SUBJ-sleep-IMPERF

  • ‘Angel is sleeping.’

(7)

  • Angel-nim hi-naas-wapayata-ca ma-may’as-na.

  • Angel-ERG 3SUBJ-O.PL-help-IMPERFPL-child-ACC

  • ‘Angel is helping the children.’

In contrast to 3rd person pronouns and DPs, local persons show a nominative-accusative case alignment—a fact that I show in Deal 2016b to be syntactic, rather than morphological, in nature. Accordingly, the clearest examples of raising to ergative feature 3rd person raised arguments.

The case system coexists with a nominative-accusative system of verb agreement. Verbal affixes directly distinguish 3rd from non-3rd person and plural from nonplural number. Nonplural number and local person are not marked on the verb overtly; however, restrictions on the use of plural agreement partially distinguish 1st from 2nd person. The overt markers consist primarily of the five prefixes listed in (8).7 Restrictions on the use and cooccurrence of agreement affixes are described in Deal 2015c.

(8)

  • Agreement prefixes

  • hi- 3rd person subject

  • ’e- 3rd person object

  • pee- 3rd person subject and 3rd person object

  • pe- plural subject

  • nees- plural object

Subject agreement is present in all clauses in Nez Perce, regardless of the case marking of the subject (nominative or ergative). In both intransitive (9) and transitive (10), the subject controls the 3rd person subject agreement prefix hi and the plural subject agreement prefix pe.8 (The object, being local and singular, controls no overt agreement in (10).)

(9)

  • [Háama kaa ’áayat ] hi-pa-’ác-

    graphic
    -a.

  • [man.NOM and woman.NOM] 3SUBJ-S.PL-enter-P-REM.PAST

  • ‘A man and a woman came in.’

(10)

  • [Matt kaa George-nim] hi-pa-’yáa

    graphic
    -n-a ’iin-e cepéeletp’et-pe.

  • [Matt.NOM and George-ERG] 3SUBJ-S.PL-find-P-REM.PAST 1SG-ACC picture-LOC

  • ‘Matt and George found me in the picture.’

The generalization is that subject agreement is controlled by the highest argument in the c- command domain of T, regardless of its case value (Deal 2010b).9

Object agreement and accusative case are tightly correlated in Nez Perce. As discussed in Deal 2013, a single vP may contain only one accusative-marked DP, and it is always the second-highest DP that is marked in this way. It is this DP, furthermore, that controls object agreement. In a simplex monotransitive, the theme is marked accusative and the agent is marked ergative. The theme controls object agreement (in (11), the plural object agreement prefix nees).

(11)

  • Angel-nimagent hi-nees-cewcew-téetu nuun-etheme.

  • Angel-ERG 3SUBJ-O.PL-call-HAB.PRES 1PL-ACC

  • ‘Angel calls us.’

In a simplex ditransitive, the goal c-commands the theme, and the goal is marked accusative. The agent is ergative, and the theme is unmarked (nominative).10

(12)

  • ’Aayat-onmagent pe-’eny-

    graphic
    -e haacwal-agoal tam’aamiintheme.

  • woman-ERG 3/3-give-P-REM.PAST boy-ACC cake.NOM

  • ‘The lady gave the boy cake.’

Just as the goal receives the only accusative case, so it alone controls object agreement on the verb. In (13), the plural goal controls the plural object agreement prefix neec (the preglottal allomorph of nees). In (14), the goal is local and singular and triggers no overt agreement; there is visibly no agreement with the plural theme.

(13)

  • Beth-nimagent hi-neec-’ni-

    graphic
    -ye lepit picaloo-nagoal hipttheme.

  • Beth-ERG 3SUBJ-O.PL-give-P-REM.PAST two kitten-ACC food.NOM

  • ‘Beth gave the two kittens food.’

(14)

  • ’Iin-egoal Beth-nimagent hi-’ni-

    graphic
    -ye lepit picalootheme.

  • 1SG-ACC Beth-ERG 3SUBJ-give-P-REM.PAST two kitten.NOM

  • ‘Beth gave me the two kittens.’

The generalization is that the DP bearing accusative case and controlling object agreement is the highest DP in the c-command domain of v (Deal 2013).

2.2 Diagnosing Unaccusativity

Case and agreement in Nez Perce behave identically for all intransitive predicates: the subject is nominative (unmarked) and controls subject agreement for person and number. Unergatives may nevertheless be distinguished from unaccusatives via a pattern of participle formation. The participle in question, formed by suffixation of -in’ to a verb stem, has been described as a passive (Rude 1985). Semantically, the -in’ participle forms a one-place predicate that holds of the verb’s internal argument. If the verb is transitive, the participle describes the object.

(15)

  • a.

    pro ’a-lawlimq-

    graphic
    -a piskis-ne.

    1SG 3OBJ-fix-P-REM.PAST door-ACC

    ‘I fixed the door.’

  • b.

    Piskis hii-we-s lawlimq-in’.

    door.NOM 3SUBJ-be-PRES fix-PART

    ‘The door is fixed.’

(16)

  • Situation under discussion: a boy has climbed a ladder.

  • {Hiicanwaas / #haacwal} hii-we-s hiicay-iin’.

  • {ladder.NOM / #boy.NOM} 3SUBJ-be-PRES climb-PART

  • ‘The ladder / #The boy is climbed.’

Some intransitive verbs form -in’ participles as well. In this case, the sole argument position of the verb is abstracted over; thus, the subjects of these verbs are treated like the objects of transitives. This commonality between objects and a subclass of intransitive subjects allows us to diagnose unaccusativity. Examples (17)–(18) contrast unaccusatives, which form -in’ participles, with unergatives, which do not.

(17)

  • pro lilooy-nin’ / *tiy’-iin’ wee-s.

  • 2SG be.happy-PART / *laugh-PART be-PRES

  • ‘You are happy / *laughed.’

(18)

  • pro hii-we-s paay-nin’ / *kuu-yiin’.

  • 3 SG 3SUBJ-be-PRES come-PART / *go-PART

  • ‘He is come / *gone.’

A partial list of Nez Perce intransitives that can be categorized as unergative or unaccusative using this diagnostic is given in (19) and (20).11

(19)

  • Unaccusatives (form -in’ participles)

  • k’oomay ‘be sick’, lilooy ‘be happy’, moolat ‘boil over’, paay ‘come/arrive’, peeleey ‘be lost’, niktiik ‘lose weight’, tin’uki ‘die’, tiye

    graphic
    ti ‘belch’, wiyoos ‘stretch out (e.g., of clothing)’, wuy ‘run away/escape’, ’eys ‘be glad’, ’ileese ‘make noise’

(20)

  • Unergatives (cannot form -in’ participles)

  • kuu ‘go/do’, misemi ‘tell lies’, tiy’e ‘laugh’, weec’ey ‘jump’, wehi ‘bark’, wewiiti ‘travel downstream’, wii ‘cry/meow’, ’imisq’uleey ‘make a verbal mistake’, ’ipsqikey’k ‘walk around’

This unaccusativity test is an instance of the broadly attested attributive participle diagnostic, discussed by Rosen (1984) for Italian and Hoekstra (1984) for Dutch and subsequently replicated in a range of languages (see, e.g., Haspelmath 1994).12 The distribution of -in’ can be straightforwardly captured on Kratzer’s (1996) proposal for vP structure, according to which theme arguments alone are arguments of the verb root. An unaccusative or transitive verb root is of type

graphic
e,st
graphic
(where s is the type of events); -in’ combines with the root and returns a property of individuals (type
graphic
e,t
graphic
). An unergative root, on the other hand, is of type
graphic
s,t
graphic
and therefore barred by a type mismatch from combining with -in’.

Syntactically, the -in’ participle is an adjective, like its crosslinguistic correlates. There are several indications that -in’ participles are not (passive) verbs. First, they may appear in prenominal position, between a noun and a numeral or quantifier, (21). This position is not available to relative clauses (which are never prenominal; Deal 2016a) or to verbs in Nez Perce.

(21)

  • a.

    lepit lawlimq’-in ’aatamoc

    two fix-PART car

    ‘two repaired cars’

  • b.

    naaqc lilooy-nin’ haacwal

    one be.happy-PART boy

    ‘one happy boy’

Similarly, like adjectives, they require copulas when used predicatively. Nez Perce does not use auxiliary verbs. Contrast (22a), featuring an -in’ participle and an obligatory copula, with (22b), where the same root is used verbally and no copula is present.

(22)

  • a.

    Haacwal *(hii-we-s) lilooy-nin’.

    boy.NOM 3SUBJ-be-PRES be.happy-PART

    ‘The boy is happy.’

  • b.

    Haacwal hi-llooy-ca.

    boy.NOM 3SUBJ-be.happy-IMPERF

    ‘The boy is happy.’

2.3 The Structure of Applicatives

Nez Perce has several productive applicative suffixes (Rude 1985, Deal 2010b:sec. 1.7.4.2): -aapiik ‘away from [DP]’, -aatk ‘as [DP] passes’, and -uu ‘toward [DP]’.13 Here and throughout, I exemplify with -uu. In the typology of Pylkkänen (2008), this and other Nez Perce applicatives are high applicatives. Crucially, they may attach to unergative verbs, such as wii ‘cry’ and kuu ‘do/go’. Compare the simplex predicates in the (a) examples with their applicative counterparts in the (b) examples.14

(23)

  • a.

    Kit’ic hi-wii-qa-na.

    Kit’ic.NOM 3SUBJ-cry-HAB.PAST-REM.PAST

    ‘Kit’ic used to meow.’

  • b.

    Kit’ic-nim pee-wii-nuu-qa-na Besi-ne.

    Kit’ic-ERG 3/3-cry-APPL-HAB.PAST-REM.PAST Bessie-ACC

    ‘Kit’ic used to meow at Bessie.’

(24)

  • a.

    Haacwal hi-kuu-

    graphic
    -ye Harold-
    graphic
    -px.

    boy.NOM 3SUBJ-go-P-REM.PAST Harold-OBL-to

    ‘The boy went over to Harold.’

  • b.

    Haacwal-nim pee-ky-uu-

    graphic
    -ye Harold-ne.

    boy-ERG 3/3-go-APPL-P-REM.PAST Harold-ACC

    ‘The boy went over to Harold.’

Added to a transitive verb, the argument introduced by the applicative occupies a structural position between the agent and the theme. This position may be diagnosed by accusative case and object agreement. Recall that both agreement and accusative case are restricted to the highest object within vP. When an applicative is added to a transitive verb, the applicative argument bears accusative case and controls object agreement. The theme argument is nominative (unmarked) and controls no agreement. Contrast the nominative case of the theme in the applicative transitive (25) with the accusative case of the theme in the basic transitive (26).

(25)

  • Pit’in-im ha-’ayato-na hi-naac-’nahpayk-oo-

    graphic
    -ya Fido.

  • girl-ERGPL-woman-ACC 3SUBJ-O.PL-bring-APPL-P-REM.PAST Fido.NOM

  • ‘The girl brought Fido to the women.’

(26)

  • pro paa-’nahpayk-

    graphic
    -a Fido-ne.

  • 3SG 3/3-bring-P-REM.PAST Fido-ACC

  • ‘She brought Fido.’

In the basic transitive, the theme is local to v and shows object case and agreement. The addition of an applicative disrupts this relationship, placing the applicative argument closest to v. This indicates that the applicative projection sits above VP, but below vP, as shown in (27). Agreement relations holding in this structure are indicated with dashed lines.15

(27)

graphic

This structure extends straightforwardly to applicatives of unergatives, where VP simply lacks an internal argument.16

(28)

  • Haama-nm pee-msem-uu-

    graphic
    -ye ’ip-ne.

  • man-ERG 3/3-tell.lies-APPL-P-REM.PAST 3SG-ACC

  • ‘The man lied to her/him.’

(29)

graphic

3 Raising to Ergative

From the perspective of the transitive applicative structure (27), the behavior of applicative unaccusatives is perhaps surprising. In the applicative of a transitive, as we just saw, the theme is strictly nominative; in the applicative of an unaccusative, by contrast, the theme is ergative. The applicative argument remains accusative. This pattern is exemplified in (4) and in (30)–(31).17 Throughout this section, theme arguments are boldfaced, and applicative arguments are underlined. (Examples of this type show full flexibility in surface word order, like other Nez Perce clauses.)

(30)

  • Taamsas-nim pee-’leese-nuu-

    graphic
    -ye Harold-ne.

  • Taamsas-ERG 3/3-make.noise-APPL-P-REM.PAST Harold-ACC

  • ‘Taamsas made noise at Harold.’

(31)

  • Angel-na pa-pay-noo-

    graphic
    -ya sik’eem-nim.

  • Angel-ACC 3/3-come-APPL-P-REM.PAST horse-ERG

  • ‘The horse came to Angel.’

Beyond case, there are also differences between applicative unaccusatives and applicative transitives in terms of the theme’s agreement. In the applicative of a transitive, the theme controls no agreement. In the applicative of an unaccusative, on the other hand, the theme controls subject agreement, whereas the applicative argument controls object agreement. In (32), the theme is 3rd person and the applicative argument is 1st person. Accordingly, the verb bears overt agreement only with the 3rd person theme subject; recall that there is no direct marking of local person features on the verb.18

(32)

  • a.

    K’olalk’olal-nim hi-’leese-nuu-

    graphic
    -ye pro.

    bell-ERG 3SUBJ-make.noise-APPL-P-REM.PAST 1SG

    ‘The bell rang at me.’

  • b.

    ’Eetee-x pee

    graphic
    wiy’ew’eet-unm hi-pay-noo-sa pro!

    INFER-1 thief-ERG 3SUBJ-come-APPL-IMPERF 1SG

    ‘Surely a thief is coming in on me!’

In (33), the theme is local person and the applicative argument is 3rd person. Accordingly, the verb bears overt agreement only with the 3rd person applicative object.19

(33)

  • a.

    c’alawi ’ee ’a-pay-noo-yo’qa Meeli-ne cik’eet-

    graphic
    -pe

    if 2SG.CLITIC 3OBJ-come-APPL-MODAL Mary-ACC night-OBL-at

    ‘if you came up to Mary at night’

  • b.

    pro ’a-pay-noo-toq-

    graphic
    -a pit’ini-ne.

    1SG 3OBJ-come-APPL-REST-P-REM.PAST girl-ACC

    ‘I came back to the girl.’

Finally, in (34) the theme is 3rd person singular and the applicative argument is 1st person plural. The verb shows plural object agreement with the applicative argument and 3rd person subject agreement with the theme.

(34)

  • ’Inpew’eet-umkiye hi-nees-pay-noo-yo’!

  • policeman-ERG 1PL.INCL.CLITIC 3SUBJ-O.PL-come-APPL-FUT

  • ‘A cop will come to us!’

Across these examples, theme arguments participate in subject agreement in exactly the same way as external arguments do in unergatives and transitives. Supposing, as above, that subject agreement is controlled by the highest DP in the c-command domain of T, this suggests that the theme argument raises over the applicative argument.

What is the landing site of this movement? Legate (2003) and I (Deal 2009) argue that all verbal projections contain a phasal vP layer, even in passives and unaccusatives.20 I indicate the nonthematic v head characteristic of unaccusatives as v~, following the convention in Deal 2009; I assume that this head obligatorily bears an EPP feature in Nez Perce. The structure of example (30) is thus as shown in (35). (The motivation for movement of the theme, rather than the applicative argument, is taken up in section 4.)

(35)

graphic

Movement of the theme to the nonthematic specifier of v~ produces the standard configuration for subject and object agreement: both T and v agree with the highest DPs in their domains.

The structure in (35) receives further support from two types of binding phenomena. The first involves Condition C. In (36) and (37), the theme contains a possessor R-expression coindexed with a pronominal applicative argument. These examples are well-formed.

(36)

  • Harold-nimik’olalk’olal-nim pee-’leese-nuu-

    graphic
    -ye proi.

  • Harold-GEN bell-ERG 3/3-make.noise-APPL-P-REM.PAST 3SG

  • ‘Haroldi’s bell made noise at himi.’

(37)

  • ’ip-nei pa-pay-noo-

    graphic
    -ya Angel-nimisik’em-nim.

  • 3SG-ACC 3/3-come-APPL-P-REM.PAST Angel-GEN horse-ERG

  • ‘Angeli’s horse came to heri.’

If the theme argument remained in situ, this configuration would be expected to produce a Condition C violation, given that applicatives attach above VP. Indeed, in an applicative transitive, coindexation of a theme possessor R-expression with an applicative argument produces ungrammaticality.

(38)

  • pro[Angeli-nim hipt ] ’aw-’nahpayk-oo-yo’ proj/*i.

  • 1SG [Angel-GEN food.NOM] 3OBJ-bring-APPL-FUT 3SG

  • ‘I will bring Angeli’s food to him/her/itj/*i.’

(39)

  • pro’ip-nei/*j ’ew-’nik-uu-se [Keelpin-imjkuus ].

  • 1SG 3SG-ACC 3OBJ-place-APPL-IMPERF [Calvin-GEN water.NOM]

  • ‘I am putting Calvinj ’s water out for him/her/iti/*j.’

The absence of a parallel Condition C effect in (36) and (37) therefore supports the claim that the theme subject moves out of the c-command domain of the applicative argument.

The second argument draws on the interaction of binding and case. As Rude (1985) and I (Deal 2010a,b) discuss, ergative and accusative case marking are obligatorily absent in Nez Perce when the subject binds the (highest) object’s possessor. Both arguments become nominative and only the subject agrees with the verb. This pattern is dubbed the extended reflexive in Deal 2010a (following Aissen 1999); see the analysis there and in Deal 2010b:187–338. This pattern extends to instances of subjects binding possessors of applicative arguments, as shown in (40b).

(40)

  • a.

    Angeli [proipike ] hi-muu-n-e.

    Angel.NOM [3SG(GEN) mother.NOM] 3SUBJ-call-P-REM.PAST

    ‘Angeli called heri mother.’

  • b.

    Angeli hi-ky-uu-

    graphic
    -ye [proisik’em ].

    Angel.NOM 3SUBJ-go-APPL-P-REM.PAST [3SG(GEN) horse.NOM]

    ‘Angeli went over to heri horse.’

Coindexation between the subject possessor and the object (or applicative argument), on the other hand, does not affect case or agreement. Observe the ergative suffix and object agreement (in the form of 3-on-3 portmanteau pee) reappearing in (41).

(41)

  • a.

    [proi pike-pim ] pee-muu-n-e proi.

    [3SG (GEN) mother-ERG] 3/3-call-P-REM.PAST 3SG

    ‘Heri mother called heri.’

  • b.

    [proi sik’eem-nim] pee-ky-uu-

    graphic
    -ye proi.

    [3SG(GEN) horse-ERG ] 3/3-go-APPL-P-REM.PAST 3SG

    ‘Heri horse went over to heri.’

The contrast between (40) and (41) shows that the case and agreement patterns are affected only when a possessor is c-commanded by a coindexed argument. Coindexation itself is insufficient—it is binding that crucially matters.

Against this backdrop, observe that applicatives of unaccusatives give rise to the extended-reflexive pattern: when the theme argument binds the possessor of the applicative argument, ergative and accusative case and object agreement are lost. In parallel to the contrast between (40) (with binding) and (41) (with coreference), binding examples (42) and (43) contrast with examples (36) and (37), where the theme possessor is merely coreferential with the applicative argument and we find the full ergative/accusative case pattern.

(42)

  • Waaqo’ ’im-’tooti hi-pay-noo-

    graphic
    -ki-ka [proi
    graphic

  • now 2SG-father.NOM 3SUBJ-come-APPL-P-TRANS-REM.PAST [3SG(GEN) DEM

  • graphic
    pineexsin ].

  • white daughter.in.law.NOM]

  • ‘Now [your father]i went off to hisi white (duck) daughter-in-law.’

  • (Aoki and Walker 1989:14)

(43)

  • proi [’ip-nimisik’em ] hi-pay-noo-

    graphic
    -ya.

  • 3SG [3SG-GEN horse.NOM] 3SUBJ-come-APPL-P-REM.PAST

  • ‘Shei came to heri horse.’

The extended-reflexive pattern would not be expected if the theme remained in situ in these examples. There is no c-command relation between the possessor of the applicative argument and the base position of the theme; therefore, there can be no binding. When the theme moves past the applicative argument to Spec,v~, however, it comes to c-command the possessor of the applicative argument. This c-command relation establishes the binding configuration that proves crucial to determining the extended-reflexive pattern.

Overall, both for binding and for agreement, the theme of an applicative unaccusative behaves exactly like any other subject, and the applicative argument itself behaves exactly like any other (highest) object. These facts receive a straightforward analysis in view of the structure in (35), in which the theme raises past the applicative argument.

Returning now to the Ergative Case Generalization, it can hardly be a coincidence that theme movement occurs in exactly those structures in which the theme also receives ergative case. Ergative case in transitives is always assigned to the highest DP in the clause; this generalization, we now see, applies to applicative unaccusatives as well. The generalization would not be captured if (for instance) the theme argument were assigned case in its base position and, subsequently, independently moved to Spec,v~.21 The facts suggest instead that ergative case in Nez Perce is assigned to the highest DP by a mechanism independent of θ-assignment. The theme receives ergative case no lower than its derived position on the vP edge.

In conclusion: Nez Perce demonstrates raising to ergative.

4 Inversion, Locality, and Antilocality

I turn now to the question of locality in the raising-to-ergative structure (35). Why does the theme move over the applicative argument? Why isn’t it the applicative argument that raises to vP to satisfy the EPP feature of v~?

As discussed above, for Baker (2014) the solution to the parallel puzzle in Shipibo goes by way of a covert adpositional structure present in the specifier of the applicative. His proposal for the structure of (44) is repeated in (45). The PP structure, Baker proposes, prevents the applicative argument from undergoing movement and from interfering in movement of the theme.

(44)

  • Bimi-n-ra Rosa joshin-xon-ke.

  • fruit-ERG-EV Rosa.ABS ripen-APPL-COMPL

  • ‘The fruit ripened for Rosa.’

  • (Baker 2014:346)

(45)

graphic

Yet there is a curious shortage of independent evidence for the proposed PP structure in Shipibo, and the same can be said about a potential counterpart PP structure in Nez Perce. In Nez Perce, as in Shipibo, the applicative argument has the surface morphosyntax of a DP, not a PP. The best candidates for PPs in Nez Perce are oblique phrases formed with a series of bound morphemes, for example, -ki ‘with (an instrument)’, -(p)kin’ix ‘from’, -laykin ‘near’, -pe ‘on/at’, -wecet ‘because of’, -(p)x ‘to/than’, -’ayn ‘for’. These elements assign oblique case to their complements; oblique case is overtly marked only for pronouns.

(46)

  • a.

    ’ip-ním-x

    3SG-OBL-to

    ‘to him/her/it’

  • b.

    ’ip-ním-wecet

    3SG-OBL-because

    ‘because of him/her/it’

  • c.

    ’ip-nim-pé

    3SG-OBL-on

    ‘on him/her/it’

Oblique phrases do not participate in verbal agreement and do not count as arguments for the determination of transitivity; for example, their presence does not render the subject ergative.

(47)

  • Haacwal hi-kuu-

    graphic
    -ye Harold-
    graphic
    -px.

  • boy.NOM 3SUBJ-go-P-REM.PAST Harold-OBL-to

  • ‘The boy went over to Harold.’

Oblique phrases are also opaque for complementizer agreement. In Nez Perce, complementizer agreement occurs on a variety of C-domain elements (e.g., relativizer ke, inferential evidential ’eete) and indexes features of either the subject or the object; this is an omnivorous (person) agreement pattern (cf. Nevins 2011). The system is described and analyzed in Deal 2015b. We see omnivorous agreement in 1st person with subjects and objects in (48); compare the ungrammaticality of agreement with the oblique in (49).

(48)

  • a.

    ke-x kaa pro ’e-cewcew-tée-’nix Angel-ne

    C-1 then 1PL 3OBJ-call-HAB.PRES-S.PL Angel-ACC

    ‘when we call Angel’

  • b.

    ke-x kaa Angel-nim hi-nees-cewcew-téetu nuun-e

    C-1 then Angel-ERG 3SUBJ-O.PL-call-HAB.PRES 1PL-ACC

    ‘when Angel calls us’

(49)

  • ke(*-x) kaa qetu haamti’c pro hi-wleeke’yk-sa-qa ’iin-im-x

  • C(*-1) then more fast 3SG 3SUBJ-run-IMPERF-REC.PAST 1SG-OBL-than

  • ‘when she was running faster than me’

In all of these respects, applicative arguments behave unlike obliques. If the applicative argument were a PP, we would expect its DP subconstituent to be marked with oblique case (even if the adposition itself were covert); however, the applicative argument is marked accusative. Compare the form of the pronoun in applicative unaccusative (50) with its PP counterparts in (46).

(50)

  • Wa’yaat-

    graphic
    -kin’ix wi-weepcux-nim ’ip-ne pa-pay-noo-
    graphic
    -ya.

  • far-OBL-from PL-wise-ERG 3SG-ACC 3/3-come-APPL-P-REM.PAST

  • ‘From afar, the wise ones came to him.’

  • (Axtell, Parsons, and Watters 1967)

We would also not expect the applicative argument to participate in object agreement or to render the clause transitive for the purposes of ergative case assignment. Finally, we would not expect the applicative argument to be visible for complementizer agreement. Yet applicative arguments fully participate in the omnivorous agreement system just as ordinary objects do.

(51)

  • ke-x kaa sik’eem-nim hi-pay-noo-

    graphic
    -ya pro

  • C-1 then horse-ERG 3SUBJ-come-APPL-P-REM.PAST 1SG

  • ‘when the horse came to me’

(52)

  • ’Eetee-x pee

    graphic
    wiy’ew’eet-unm hi-pay-noo-sa pro!

  • INFER-1 thief-ERG 3SUBJ-come-APPL-IMPERF 1SG

  • ‘Surely a thief is coming in on me!’

These facts show that the initial challenges for the PP analysis in Shipibo are also challenges in Nez Perce.22

Nez Perce also allows us to mount an argument against the PP analysis from the interaction of applicative unaccusatives with possessor raising. This argument has two prongs, the first concerning the general question of why PPs are opaque for movement. While Baker (2014) does not state why his PP structures prevent material within PP from moving, one reasonable hypothesis is simply that they are phases without escape hatches.23 If PP is a phase without an escape hatch, we expect both that the DP complement of P will not be able to move out of PP and that material internal to this DP will not be able to move out of PP. Yet Nez Perce not only permits but indeed requires subextraction from applicative arguments under certain circumstances. In particular, when the applicative argument contains a (free) possessor, the possessor must undergo possessor raising.

Possessor raising in Nez Perce is described in Deal 2013. When the highest DP in the c-command domain of v contains an unbound possessor in Spec,D, the possessor obligatorily moves to the specifier of a functional head, µ, which attaches directly below vP. The µ head is realized morphologically as a suffix on the verb, -ey’/-en’i.24 In its raised position, the possessor is the highest DP in the c-command domain of v, and so receives accusative case and controls object agreement. The structure of possessor raising in simplex transitive (53) is shown in (54).

(53)

  • Háama-pim hi-nées-wewkuny-en’y-

    graphic
    -e ha-háacwal-na láwtiwaa.

  • man-ERG 3SUBJ-O.PL-meet-µ-P-REM.PASTPL-boy-ACC friend.NOM

  • ‘The man met the boys’ friend.’

(54)

graphic

In addition to possessor raising from a theme, as in (53)–(54), it is perfectly possible to have possessor raising out of an applicative argument, even in the applicative of an unaccusative. In (55) and (56), formed from unaccusative roots, the suffix -ey’ realizes µ. Example (55) shows that the raised possessor may be discontiguous with the possessum, as is standardly the case in possessor raising (but is not otherwise permitted in Nez Perce; Deal 2013:399–400). Note that themes continue to control subject agreement in these examples, as in other applicative-of-unaccusative examples. The structure of (55) is shown in (57) (assuming no PP is present in Spec,ApplP).

(55)

  • Ko-nim ha-’ayato-na hi-nees-’ileese-nuu-ey’-se pi’amkin.

  • DEM-ERGPL-woman-ACC 3SUBJ-O.PL-make.noise-APPL-µ-IMPERF meeting.NOM

  • ‘That person is making noise at the ladies’ meeting.’

(56)

  • pro pe-’eys-nuu-ey’-se B.-ne miy’ac.

  • 3SG 3/3-be.glad-APPL-µ-IMPERF B-ACC child.NOM

  • ‘She is being nice to B’s child.’

(57)

graphic

These examples show that the motivation for theme movement cannot simply be a fully syntactically opaque PP in the applicative specifier.25

This brings us to the second prong of the argument from possessor raising, which concerns the potential for further raising of the raised possessor. The crux of the matter here is intervention: even if possessor raising were somehow able to escape a Baker-style PP structure in Spec, ApplP, the raised possessor itself should count as a DP intervener for movement of the theme. (This possessor is, after all, able to undergo A-movement; it A-moves to Spec,µ.) Nevertheless, there is no bleeding of theme movement in possessor-raising constructions like (55) and (56), and no raising to subject of the possessor DP. This reveals that factors other than PP status are needed in order to prevent nontheme elements from undergoing A-movement to subject in at least some theme-raising structures. A full explanation for theme raising therefore cannot be given simply on the basis of applicative specifiers being PPs.

If the applicative argument is indeed a DP, then, rather than a PP, what prohibits it from moving to satisfy the EPP feature on v~? I would like to propose that it is the height of this argument, rather than its categorial status, that imposes the crucial constraint. Beginning with simplex structure (35) (without possessor raising), the applicative argument attaches immediately subjacent to v, and thus movement from Spec,Appl to Spec,v is too short. It violates an antilocality constraint independently motivated on the basis of Ā-extraction data by Bošković (2015, 2016), Brillman and Hirsch (2016), and Erlewine (2016). I present a generalized version of Erlewine’s formulation in (58).26

(58)

  • Generalized Spec-to-Spec Antilocality

  • Movement of a phrase from the Spec,XP must cross a maximal projection other than XP.

(59)

  • Definition of crossing

  • Movement from position A to position B crosses C if and only if C dominates A but does not dominate B.

Movement from Spec,Appl to Spec,v~ in (35) crosses ApplP, but no other maximal projection.27 This violates Generalized Spec-to-Spec Antilocality. I suggest that the inability of the applicative argument to undergo movement frees the theme argument to move in its stead: because the applicative argument, in virtue of its position, cannot move, it also does not serve as an intervener for movement.

For possessor-raised applicative unaccusatives, Generalized Spec-to-Spec Antilocality plays a role in blocking movement not of the applicative argument, but of its raised possessor. In structure (57), the raised possessor in Spec,µ is too close to Spec,v~ to undergo movement. The remnant applicative argument itself is far enough from v~ to move there, but may only move in keeping with general constraints on remnant movement. One such constraint is proposed by Müller (1996) under the heading of Unambiguous Domination (presented here in its derivational version; p. 376).

(60)

  • Unambiguous Domination

  • In a structure . . . [A . . . [B . . .] . . .] . . . , A and B may not undergo the same kind of movement.

Movement to Spec,µ and to Spec,v~ are both instances of A-movement: each plays a decisive role in determining the case, agreement, and binding behavior of the moving DP. Given that a subconstituent of the applicative argument undergoes A-movement, Unambiguous Domination implies that the applicative argument will not itself be able to undergo A-movement. This means that both the highest and second-highest DPs in the c-command domain of v in structure (57) are unable to satisfy the EPP feature on v~. Again we see that DPs incapable of undergoing movement do not serve as interveners. Only the theme argument is capable of undergoing movement, and therefore there is no obstacle to theme movement into Spec,vP.

The resulting analysis, like Baker’s PP analysis before it, comes with consequences for the theory of (defective) intervention. I have argued that applicative arguments are DPs, not PPs, in Nez Perce; one DP nevertheless fails to intervene on the movement of another DP if it is itself blocked from movement by antilocality or by the Unambiguous Domination requirement on remnant movement. These results suggest either (less radically) that defective intervention is possible only on the basis of other properties of the intervener, such as PP status (e.g., Preminger 2014), or (more radically) that defective intervention does not exist in grammar (Bruening 2014)—a choice point whose resolution must await fuller attention in subsequent work.28

5 Ergativity: Larger Consequences

The elimination of the inherent case analysis for Nez Perce, a raising-to-ergative language, raises two types of further questions, with which I conclude this article. One concerns the underlying typology of ergativity: how many mechanisms give rise to ergative systems? The second concerns the nature of ergativity in languages to which the inherent-case analysis cannot be applied.

5.1 Are There Inherent-Ergative Languages?

There is substantial agreement in the ergativity literature that ergativity is not one but many phenomena.29 This raises a basic question: is there more than one way to assign ergative case? Applicative unaccusatives have the potential to show that indeed there is—that some languages call for an inherent-case analysis, whereas others cannot be so analyzed.

Some potential evidence that there are indeed inherent-ergative languages comes from Massam’s (2006) work on applicatives in Niuean. In contrast to the Shipibo/Nez Perce pattern, in this language the applicative of an unergative features ergative on the agent subject, (61), but the applicative of a (putative) unaccusative does not allow ergative on the theme subject, (62).

(61)

  • Ne tohitohi aki [e Sione] [e pene].

  • PAST writing with.APPL [ERG.PROPER Sione] [ABS.COMMON pen ]

  • ‘Sione was writing with a pen.’

  • (Massam 2006:33)

(62)

  • Fakamafana aki [e poko] [e hita ].

  • CAUSE.warm with.APPL [ABS.COMMON room] [ABS.COMMON heater]

  • ‘The room is warm with the heater.’

  • (Massam 2006:34)

Does this constitute the required evidence that ergativity arises by an inherent-case-based mechanism in some languages, as Legate (2012) suggests? While I can contribute nothing decisive about Niuean, I would like to suggest (in keeping with remarks by Massam) that the presence of the causative in the purportedly unaccusative example poses a potential confound. As Massam (1998) discusses, the aki applicative is not possible for an unaccusative verb in the absence of the causative marker: “aki cannot attach to a semantically nonagentive verb” (p. 12). If examples like (62) were not unaccusative, but in fact transitive, with a null ergative causer argument, then there would be no strong reason to think Niuean is different from Nez Perce after all. I take this conclusion to largely dovetail with Massam’s (2015) recent proposal that the causative head in examples like (62) functions much like an external argument.

I suggest, then, that as far as applicative unaccusatives are concerned, the case remains open on Baker and Bobaljik’s (2017) proposal that ergative is never in fact an inherent case. While it is certainly true that the inherent-ergative view is broadly compatible with evidence from various languages (see, e.g., Legate 2012 on Warlpiri, Coon 2017a on Ch’ol), the ideal argument for crosslinguistic variation in the nature of ergative case will come from apples-to-apples comparison of applicative unaccusatives in languages where the raised theme subject does and does not receive ergative. At present, pending further investigation of (62), there is no clear case of a language that shows the second type of pattern.

5.2 Is Ergative Dependent or (Merely) Structural?

For Baker (2014), the inapplicability of the inherent-case analysis to Shipibo constitutes core evidence for an alternative, dependent-case analysis, based on rules of case assignment like (63).

(63)

  • Rule for assignment of dependent ergative

  • If there are two distinct argumental NPs in the same phase such that NP1 c-commands NP2, then value the case feature of NP1 as ergative unless NP2 has already been marked for case. (Baker 2014:343)

The dependent-case proposal for ergative (and accusative) has garnered considerable attention in recent years, in a development that should be of interest to all who are concerned with the nature of crosslinguistic variation.30 Given that case systems vary, rules like (63) must hold in some languages and not in others, but notably, these rules are not themselves properties of any lexical item. Thus, the endorsement of a dependent-case theory of this type brings as a consequence a retreat from the Chomsky-Borer conjecture about linguistic variation—a retreat that, of course, Baker himself has advocated for many years (e.g., Baker 1996, 2008).31

Yet the dependent-case analysis is not the only noninherent approach to ergative case remaining on the table. Indeed, a variety of proposals for a merely structural ergative are also compatible with the raising-to-ergative pattern.32 In Deal 2010a,b, for instance, I proposed a reductionist view according to which the Nez Perce ergative suffix is a portmanteau of subject and object agreement features transferred onto a DP: it is inserted on DPs that agree with T and occupy the specifier of a v head that agrees with an object. This configuration holds for themes raised to vP just as it does for agents that originate there. In Deal 2010a,b, I show how this view explains the connection between case and binding (“extended reflexive”) discussed in section 3; in Deal 2016b, I show how it can be extended to capture the syntactic nature of person-based split ergativity. On this view, the properties of Nez Perce that differentiate it from other types of case systems may be stated in relatively quotidian terms: they concern properties of agreement probes and of vocabulary items. Empirically, the choice between this or another “merely structural” approach to ergativity and the dependent-case approach should be made on the basis of properties beyond raising to ergative, such as the interaction of case with binding and with person. To my knowledge, the dependent-case view does not at present provide a natural way for these phenomena to be accommodated.

Notes

1 See Woolford 1997, 2006, Aldridge 2008, Legate 2008, and references in Woolford 2006 and Deal 2015a, 2016c. Hereafter, I refer to the head that introduces external arguments as v.

2 Applicative unaccusatives therefore complement the evidence for raising to ergative in Basque infinitivals (Rezac, Albizu, and Etxepare 2014) and small clauses (Artiagoitia 2001).

3 The case difference here is not due to the lexical choice of word for ‘fruit’; see Valenzuela 2003:322 on the absence of noun-based case splits in Shipibo, along with Loriot, Lauriault, and Day 1993:118.

4 On dependent-case approaches to ergative, see Yip, Maling, and Jackendoff 1987, Marantz 1991, Baker 2014, 2015, Baker and Bobaljik 2017.

5 An additional likely instance of the raising-to-ergative pattern occurs in applicative unaccusatives in Ixil Mayan (Imanishi 2017), which presents further points of variation from both Nez Perce and Shipibo: it is a purely head-marking ergative-absolutive language with split ergativity based only on aspect. No unaccusativity diagnostic has been explicitly proposed for Ixil, however. (Unaccusativity tests elsewhere in Mayan are discussed in Imanishi 2014. For an unaccusativity test in Shipibo, see Baker 2014:370; for Nez Perce, see section 2.2 below. Tests are needed due to variability in the unaccusativity status of translation equivalents across languages, on which see Rosen 1984 and Levin and Rappaport Hovav 1995. I take no stand here on the nature of this variability.)

6 For readability, I have modified Baker’s original proposal to show head-initiality.

7 Subject number may alternatively be marked as a suffix, depending on the aspect; see Deal 2015c.

8Pe harmonizes to pa in these examples. Nez Perce has a dominant-recessive harmony system whereby recessive vowels /æ/ (orthographic e) and /u/ harmonize to /a/ and /o/ in the presence of a strong vowel within the word. The fifth vowel, /i/, is transparent. See Nelson 2013.

Example (10) demonstrates “unbalanced coordination” (Johannessen 1998), where the case marker appears only once, at the end of the coordinate structure; see the discussion of Nez Perce coordinations in Deal 2015a, 2016b. This pattern is possible both for ergative and for accusative.

9 This generalization sets aside potential A-scrambling of the object over the subject, which does not affect the agreement system. See Deal 2017 for discussion and analysis.

10 Nez Perce has no structural dative case. See Deal 2013 for Condition C evidence that the goal c-commands the theme in Nez Perce ditransitives.

11 Nez Perce verbs come in two morphological classes, “S-class” and “C-class” (Aoki 1994; see discussion in Deal and Wolf 2016). In the only previous proposal for an unaccusativity diagnostic in Nez Perce, Cash Cash (2004) proposes, on the basis of semantic properties, that C-class intransitives are unaccusative. This proposal largely lines up with (19) and (20), but there are exceptions: for example, tiye

graphic
ti ‘belch’ is an unaccusative S-class verb, whereas wii ‘cry/meow’ is an unergative C-class verb.

12 Some crosslinguistic work on this topic has concluded that attributive participle formation from intransitives requires not only unaccusativity, but also telicity (e.g., Levin and Rappaport Hovav 1995:151). (Other work has maintained that all unaccusatives are telic in some languages, such as Dutch (van Hout 2004).) For Nez Perce, formal telicity tests have yet to be systematically explored, but initial impressions do not support a recharacterization of the -in’ participle data in terms of telicity: for example, among the verbs listed in (19), paay ‘come/arrive’ and tin’uki ‘die’ are likely telic, whereas k’oomay ‘be sick’ and lilooy ‘be happy’ are likely atelic.

13Rude (1985) and I (Deal 2010b) have also analyzed an additional affix, -ey’, as an affectee/benefactive applicative; however, see Deal 2013 for evidence that this element is not an applicative but a case-assigning head µ, (see Johnson 1991).

14 The -uu suffix appears in allomorphs -uu and -nuu depending on the stem class of the preceding element. See Deal and Wolf 2016 for discussion of this pattern as well as the allomorphy of kuu ‘do/go’.

15 The nominative form of the theme here might be analyzed in either of two ways: either as reflecting a [nom] feature assigned in syntax (e.g., by Appl), or as a morphological default, reflecting the absence of any case assignment (see Schütze 2001, Preminger and Kornfilt 2015). Previous research has indicated that both analyses are required for certain portions of Nez Perce grammar: nominative on intransitive subjects involves a [nom] feature, whereas nominative on left-peripheral DPs in the hanging-topic left-dislocation (HTLD) construction is a morphological default (Deal 2016a). The difference in the status of nominative is revealed by case attraction of relative pronouns. In Deal 2016a, I show that relative clauses modifying intransitive subjects allow their relative pronouns to undergo case attraction to nominative, suggesting that a [nom] feature associated with the subject participates in (case-overwriting) Agree. By contrast, no overwriting to nominative is possible in relative clauses adjoined to base-generated left-peripheral nominatives in HTLD; this follows if no [nom] feature is present in HTLD nominatives. For themes of ditransitives (e.g., (25)), preliminary investigations suggest that case overwriting to nominative is indeed possible, revealing a [nom] feature in this context as well, though I remain officially agnostic on this point pending further confirmation of these findings.

16 This structure reveals that Appl cannot be taken to obligatorily assign nominative to a theme. A reviewer observes a connection between this finding and Bošković’s (2011) argument that the Inverse Case Filter does not hold (i.e., there is no requirement that every case-assigning head must assign its case). Suppose, then, that Appl always has a case to assign, and fails to assign it only when there is no DP within its c-command domain (as in structure (29)). It follows that themes that raise to ergative (as discussed in section 3) will initially be assigned nominative and subsequently have their case overwritten. This connects with the proposal for case overwriting discussed in Deal 2016a.

Within the family of proposals that take Appl to be a case assigner in structures like (27), as discussed in footnote 15, the major alternative is to say that Appl comes in two varieties. One assigns [nom] and is present in (27). The other does not, and is present in (29) as well as (potentially) cases of raising to ergative. On this view, Appl assigns [nom] only on an as-needed basis.

17 For Shipibo, Baker (2014:366) notes that applicatives of unaccusatives are accepted by only two of his three consultants; I have found no similar variation in Nez Perce. Examples of this type are readily obtained in elicitation and are attested in corpora of different types (see, e.g., (42), (50), and dictionary entries in Aoki 1994:515).

18 The presence of a 1st person argument is marked in (32b) on the inferential evidential ’eete. As discussed in Deal 2015b and in section 4 below, this item participates in a system of omnivorous person agreement that, in Nez Perce, is characteristic of the C system.

19 On the clitic pronouns seen in (33a) and (34), see Deal 2016b.

20 This point will prove crucial for the antilocality-based account of raising to ergative given in section 4. That is of course not to say that all controversies related to the issue are settled; extensive further discussion of vP phases (and how we might diagnose them) may be found in Fox 1999, Nissenbaum 2000, Sauerland 2003, Rackowski and Richards 2005, Richards and Biberauer 2005, Den Dikken 2006, Johnson 2007, Newell 2008, Baker 2014, Van Urk and Richards 2015, and Keine 2016, among others.

21 This type of proposal would presumably require the theme to move if and only if it previously obtained ergative case. (If themes could generally move above other objects, we would expect this movement to bleed Condition C in (38)–(39), for instance.) Note that even on theories that allow movement to be sensitive to case assignment (e.g., Otsuka 2006, Preminger 2014, Deal 2016c), it is not possible for a movement rule to apply to an ergative-marked theme but not to a nominative-marked theme. Any movement that applies to ergatives must also apply to nominatives (though not vice versa).

22 The Nez Perce facts might productively be contrasted with the situation in certain varieties of British English, where Myler (2013) finds both syntactic and semantic evidence for covert adpositions.

23 An alternative approach might appeal to a morphological ban on P-stranding. Such a ban could prevent overt movement of the applicative DP, but presumably not covert movement. See Deal 2013, 2017 for evidence of covert A- movement in Nez Perce.

24 On the allomorphy of µ, see Deal and Wolf 2016. In Deal 2013, I argue that µ is a case assigner and is structurally present only when needed for case assignment purposes.

25 This argument is weakened if some alternative conception of PP opacity could be given such that complements of P cannot extract, but subconstituents of P’s complement can. Abels’s (2003) conception of antilocality could be called on, for instance, to exactly this effect, provided PPs in the relevant structure are phases.

26 This constraint is “generalized” in that it applies both to A- and to Ā-movement; the application to A-movement is in keeping with early work by Bošković (1997) (though the version of antilocality used in that work is slightly different). Recent critical discussion of antilocality in Ā-extraction can be found in Baier 2017 and Henderson and Coon 2018. See Grohmann 2011 for discussion of various formulations of antilocality constraints, along with historical antecedents.

27 As Baier (2017:368) notes, antilocality accounts are inherently “fragile”; they are easily disturbed by the discovery of new projections in the region of interest, rendering two positions farther apart than initially thought. As a theoretical matter, I suggest we should welcome this aspect of the theory: fragility means straightforward falsifiability, which is a virtue. As an empirical matter, I am not aware of any projections between Appl and v in Nez Perce other than µ, discussed just below.

28 The antilocality approach has a variety of additional consequences worthy of further attention; for instance, as a reviewer points out, it is incompatible with the derivation of Shipibo experiencer constructions posited by Baker (2014), which features Spec,V-to-Spec,v movement. If (58) holds as a general principle of grammar, then (as the reviewer suggests) a reasonable first hypothesis is that an additional projection is present in the Shipibo vP, rendering the movement in question sufficiently long. Among the most plausible candidates for a vP-internal projection of this type is inner aspect (Travis 2010). If Shipibo experiencers originate in Spec,V, then the presence of AspP between VP and vP will allow experiencers to move to Spec,v in a way that respects antilocality. ApplP must be located above AspP, however, blocking similar movement from Spec,Appl. I am not aware of any Shipibo-internal evidence that speaks either for or against this hypothesis. Further alternatives are of course possible as well, such as a treatment of experiencers as base-generated in Spec,v (Kratzer 1996). This possibility requires more deviation from Baker 2014 than does the first option sketched above: it requires some rethinking of Baker’s tests for internal vs. external arguments, for instance, as well as an alternative analysis of case in Shipibo experiencer constructions. Overall, there is clearly more to be done empirically to assess the prospects for Shipibo of an antilocality principle of grammar.

31 The term Borer-Chomsky conjecture is Baker’s (2008), with reference to Borer 1984 and Chomsky 1995. Baker’s formulation is given in (i).

  • (i)

    Borer-Chomsky conjecture

    All parameters of variation are attributable to differences in the features of particular items (e.g., the functional heads) in the lexicon. (Baker 2008:353)

Versions of dependent-case theory more compatible with the Borer-Chomsky conjecture might of course in principle be developed, perhaps with Bittner and Hale 1996b as a central antecedent.

32 On structural approaches to ergative, see Bobaljik and Branigan 2006, Deal 2010a,b, Rezac, Albizu, and Etxepare 2014, Erlewine 2016, and Clem to appear, among others.

Acknowledgments

Thanks first and foremost to my Nez Perce teachers Florene Davis and Bessie Scott for sharing their language with me. Thanks as well to the anonymous LI reviewers, and to Nico Baier, Jessica Coon, and Tammy Stark for comments on the manuscript. This research was assisted by an ACLS Fellowship from the American Council of Learned Societies.

Abbreviations used in the glosses are as follows: 3/3 = 3rd person subject and 3rd person object (portmanteau); 3OBJ = 3rd person object; 3SUBJ = 3rd person subject; ABS = absolutive; ACC = accusative; APPL = applicative; C = complementizer; CAUSE = causative; CIS = cislocative; COMPL = completive; DEM = demonstrative; ERG = ergative; EV = evidential; FUT = future; GEN = genitive; HAB = habitual; IMPERF = imperfective; INCL = inclusive; INFER = inferential marker; LOC = locative; µ = functional head present in possessor raising; NOM = nominative; OBL = oblique; O.PL = plural object; P = perfect/perfective aspect; PART = participle; PL = plural; PRES = present; REC.PAST = recent past; REM.PAST = remote past; REST = restitutive; SG = singular; S.PL = plural subject; TRANS = translocative

References

Abels,
Klaus
.
2003
.
Successive cyclicity, anti-locality, and adposition stranding
.
Doctoral dissertation, University of Connecticut, Storrs
.
Aissen,
Judith
.
1999
.
Markedness and subject choice in Optimality Theory
.
Natural Language and Linguistic Theory
17
:
673
711
.
Aldridge,
Edith
.
2008
.
Generative approaches to ergativity
.
Language and Linguistics Compass
2
:
966
995
.
Aoki,
Haruo
.
1994
.
Nez Perce dictionary
.
Berkeley
:
University of California Press
.
Aoki,
Haruo
, and
Deward
Walker
.
1989
.
Nez Perce oral narratives
.
Berkeley
:
University of California Press
.
Artiagoitia,
Xabier
.
2001
.
Seemingly ergative and ergatively seeming
. In
Features and interfaces in Romance
, ed. by
Julia
Herschensohn
,
Enrique
Mallén
, and
Karen
Zagona
,
1
22
.
Amsterdam
:
John Benjamins
.
Axtell,
William
,
William
Parsons
, and
Sam M.
Watters
.
1967
.
Nee mi putimt ki wanipt (Nez Perce Presbyterian hymnbook)
.
Caldwell, ID
:
Caxton Printers
.
Baier,
Nico
.
2017
.
Antilocality and antiagreement
.
Linguistic Inquiry
48
:
367
377
.
Baker,
Mark
.
1996
.
The polysynthesis parameter
.
Oxford
:
Oxford University Press
.
Baker,
Mark
.
2008
.
The macroparameter in a microparametric world
. In
The limits of syntactic variation
, ed. by
Theresa
Biberauer
,
351
373
.
Amsterdam
:
John Benjamins
.
Baker,
Mark
.
2012
.
On the relationship of object agreement and accusative case: Evidence from Amharic
.
Linguistic Inquiry
43
:
255
274
.
Baker,
Mark
.
2014
.
On dependent ergative case (in Shipibo) and its derivation by phase
.
Linguistic Inquiry
45
:
341
379
.
Baker,
Mark
.
2015
.
Case: Its principles and its parameters
.
Cambridge
:
Cambridge University Press
.
Baker,
Mark
, and
Jonathan David
Bobaljik
.
2017
.
On inherent and dependent theories of ergative case
. In
The Oxford handbook of ergativity
, ed. by
Jessica
Coon
,
Diane
Massam
, and
Lisa
Travis
,
111
134
.
Oxford
:
Oxford University Press
.
Baker,
Mark
, and
Nadya
Vinokurova
.
2010
.
Two modalities of case assignment in Sakha
.
Natural Language and Linguistic Theory
28
:
593
642
.
Bittner,
Maria
, and
Ken
Hale
.
1996a
.
Ergativity: Toward a theory of a heterogeneous class
.
Linguistic Inquiry
27
:
531
604
.
Bittner,
Maria
, and
Ken
Hale
.
1996b
.
The structural determination of case and agreement
.
Linguistic Inquiry
27
:
1
68
.
Bobaljik,
Jonathan David
, and
Philip
Branigan
.
2006
.
Eccentric agreement and multiple case-checking
. In
Ergativity: Emerging issues
, ed. by
Alana
Johns
,
Diane
Massam
, and
Juvénal
Ndayiragije
,
47
78
.
Dordrecht
:
Springer
.
Borer,
Hagit
.
1984
.
Parametric syntax
.
Dordrecht
:
Foris
.
Bošković,
Željko
.
1997
.
The syntax of nonfinite complementation: An economy approach
.
Cambridge, MA
:
MIT Press
.
Bošković,
Željko
.
2011
.
On unvalued uninterpretable features
. In
NELS 39: Proceedings of the 39th Annual Meeting of the North East Linguistic Society
, ed. by
Suzi
Lima
,
Kevin
Mullin
, and
Brian
Smith
,
1
:
109
120
.
Amherst
:
University of Massachusetts, Graduate Linguistic Student Association
.
Bošković,
Željko
.
2015
.
From the Complex NP Constraint to everything: On deep extractions across categories
.
The Linguistic Review
32
:
603
669
.
Bošković,
Željko
.
2016
.
On the timing of labeling: Deducing comp-trace effects, the Subject Condition, the Adjunct Condition and tucking in from labeling
.
The Linguistic Review
33
:
17
66
.
Brillman,
Ruth
, and
Aron
Hirsch
.
2016
.
An anti-locality account of English subject/non-subject asymmetries
. In
CLS 50: Proceedings of the Fiftieth Annual Meeting of the Chicago Linguistic Society
, ed. by
Ross
Burkholder
et al
,
73
87
.
Chicago
:
University of Chicago, Chicago Linguistic Society
.
Bruening,
Benjamin
.
2014
.
Defects of defective intervention
.
Linguistic Inquiry
45
:
707
719
.
Carnie,
Andrew
, and
Phillip
Cash Cash
.
2006
.
Tree-geometric relational hierarchies and Nuumiipuutímt (Nez Perce) case
. In
Ergativity: Emerging issues
, ed. by
Alana
Johns
,
Diane
Massam
, and
Juvénal
Ndayiragije
,
229
244
.
Dordrecht
:
Springer
.
Cash Cash,
Phillip
.
2004
.
Nez Perce verb morphology
.
Ms., University of Arizona, Tucson
.
Chomsky,
Noam
.
1995
.
The Minimalist Program
.
Cambridge, MA
:
MIT Press
.
Clem,
Emily
.
To appear
.
Amahuaca ergative as agreement with multiple heads
.
Natural Language and Linguistic Theory
.
Coon,
Jessica
.
2016
.
Case discrimination in caseless languages
.
Colloquium handout, Stanford University, May 2016
.
Coon,
Jessica
.
2017a
.
Little-v agreement and templatic morphology in Ch’ol
.
Syntax
20
:
101
137
.
Coon,
Jessica
.
2017b
.
Two types of ergative agreement: Implications for case
. In
A Pesky set: Papers for David Pesetsky
, ed. by
Claire
Halpert
,
Hadas
Kotek
, and
Coppe
van Urk
,
361
370
.
MIT Working Papers in Linguistics 80
.
Cambridge, MA
:
MIT, MIT Working Papers in Linguistics
.
Crook,
Harold David
.
1999
.
The phonology and morphology of Nez Perce stress
.
Doctoral dissertation, UCLA, Los Angeles, CA
.
Deal,
Amy Rose
.
2009
.
The origin and content of expletives: Evidence from “selection.”
Syntax
12
:
285
323
.
Deal,
Amy Rose
.
2010a
.
Ergative case and the transitive subject: A view from Nez Perce
.
Natural Language and Linguistic Theory
28
:
73
120
.
Deal,
Amy Rose
.
2010b
.
Topics in the Nez Perce verb
.
Doctoral dissertation, University of Massachusetts, Amherst
.
Deal,
Amy Rose
.
2013
.
Possessor raising
.
Linguistic Inquiry
44
:
391
432
.
Deal,
Amy Rose
.
2015a
.
Ergativity
. In
Syntax–theory and analysis. An international handbook
, ed. by
Artemis
Alexiadou
and
Tibor
Kiss
,
1
:
654
707
.
Berlin
:
Mouton de Gruyter
.
Deal,
Amy Rose
.
2015b
.
Interaction and satisfaction in ϕ-agreement
. In
NELS 45
, ed. by
Thuy
Bui
and
Deniz
Özyıldız
,
1
:
179
192
.
Amherst
:
University of Massachusetts, Graduate Linguistic Student Association
.
Deal,
Amy Rose
.
2015c
. A note on Nez Perce verb agreement, with sample paradigms. In
Proceedings of the International Conference on Salish and Neighbouring Languages 50
, ed. by
Natalie
Weber
,
Erin
Guntly
,
Zoe
Lam
, and
Sihwei
Chen
,
389
413
.
Vancouver
:
University of British Columbia Working Papers in Linguistics
. https://lingpapers.sites.olt.ubc.ca/icsnl-volumes/.
Deal,
Amy Rose
.
2016a
.
Cyclicity and connectivity in Nez Perce relative clauses
.
Linguistic Inquiry
47
:
427
470
.
Deal,
Amy Rose
.
2016b
.
Person-based split ergativity in Nez Perce is syntactic
.
Journal of Linguistics
52
:
533
564
.
Deal,
Amy Rose
.
2016c
.
Syntactic ergativity: Analysis and identification
.
Annual Review of Linguistics
2
:
165
185
.
Deal,
Amy Rose
.
2017
. Covert hyperraising to object. In
NELS 47
, ed. by
Andrew
Lamont
and
Katerina
Tetzloff
,
1
:
257
270
.
Amherst
:
University of Massachusetts, Graduate Linguistic Student Association
.
Deal,
Amy Rose
, and
Matthew
Wolf
.
2016
. Outwards-sensitive phonologically-conditioned allomorphy in Nez Perce. In
The morphosyntax-phonology connection
, ed. by
Vera
Gribanova
and
Stephanie
Shih
,
29
60
.
Oxford
:
Oxford University Press
.
Dikken,
Marcel den
.
2006
.
A reappraisal of vP being phasal: A reply to Legate
.
Ms., City University of New York
.
Erlewine,
Michael Yoshitaka
.
2016
.
Anti-locality and optimality in Kaqchikel agent focus
.
Natural Language and Linguistic Theory
34
:
429
479
.
Fox,
Danny
.
1999
.
Reconstruction, binding theory, and the interpretation of chains
.
Linguistic Inquiry
30
:
157
196
.
Grohmann,
Kleanthes K
.
2011
. Anti-locality: Too-close relations in grammar. In
The Oxford handbook of linguistic Minimalism
, ed. by
Cedric
Boeckx
,
260
290
.
Oxford
:
Oxford University Press
.
Haspelmath,
Martin
.
1994
. Passive participles across languages. In
Voice: Form and function
, ed. by
Barbara
Fox
and
Paul J.
Hopper
,
151
177
.
Amsterdam
:
John Benjamins
.
Henderson,
Robert
, and
Jessica
Coon
.
2018
.
Adverbs and variability in Kaqchikel agent focus: A reply to Erlewine (2016)
.
Natural Language and Linguistic Theory
36
:
149
173
.
Hoekstra,
Teun
.
1984
.
Transitivity: Grammatical relations in Government-Binding Theory
.
Dordrecht
:
Foris
.
van Hout,
Angeliek
.
2004
. Unaccusativity as telicity checking. In
The unaccusativity puzzle: Explorations of the syntax-lexicon interface
, ed. by
Artemis
Alexiadou
,
Elena
Anagnostopoulou
, and
Martin
Everaert
,
60
83
.
Oxford
:
Oxford University Press
.
Imanishi,
Yusuke
.
2014
.
Default ergative
.
Doctoral dissertation, MIT, Cambridge, MA
.
Imanishi,
Yusuke
.
2017
. Derived ergative in Ixil. In
WCCFL 34: Proceedings of the 34th West Coast Conference on Formal Linguistics
, ed. by
Aaron
Kaplan
,
Abby
Kaplan
,
Miranda K.
McCarvel
, and
Edward J.
Rubin
,
259
266
.
Somerville, MA
:
Cascadilla Proceedings Project
.
Jenks,
Peter
, and
Hannah
Sande
.
2017
. Dependent accusative case and caselessness in Moro. In
NELS 47
, ed. by
Andrew
Lamont
and
Katerina
Tetzloff
,
2
:
109
118
.
Amherst
,
University of Massachusetts, Graduate Linguistic Student Association
.
Johannessen,
Janne Bondi
.
1998
.
Coordination
.
Oxford
:
Oxford University Press
.
Johns,
Alana
.
1992
.
Deriving ergativity
.
Linguistic Inquiry
23
:
57
88
.
Johns,
Alana
.
2000
. Ergativity: A perspective on recent work. In
The first Glot International state-of-the-article book
, ed. by
Lisa L.-S.
Cheng
and
Rint
Sybesma
,
47
73
.
New York
:
Mouton de Gruyter
.
Johnson,
Kyle
.
1991
.
Object positions
.
Natural Language and Linguistic Theory
9
:
577
636
.
Johnson,
Kyle
.
2007
. In search of phases. In
Phrasal and clausal architecture
, ed. by
Simin
Karimi
,
Vida
Samiian
, and
Wendy
Wilkins
,
146
166
.
Amsterdam
:
John Benjamins
.
Keine,
Stefan
.
2016
.
Probes and their horizons
.
Doctoral dissertation, University of Massachusetts, Amherst
.
Kratzer,
Angelika
.
1996
. Severing the external argument from its verb. In
Phrase structure and the lexicon
, ed. by
Johan
Rooryck
and
Laurie
Zaring
,
109
137
.
Dordrecht
:
Kluwer
.
Legate,
Julie Anne
.
2003
.
Some interface properties of the phase
.
Linguistic Inquiry
34
:
506
516
.
Legate,
Julie Anne
.
2008
.
Morphological and abstract case
.
Linguistic Inquiry
39
:
55
101
.
Legate,
Julie Anne
.
2012
.
Types of ergativity
.
Lingua
122
:
181
191
.
Levin,
Beth
.
1983
.
On the nature of ergativity
.
Doctoral dissertation, MIT, Cambridge, MA
.
Levin,
Beth
, and
Malka
Rappaport Hovav
.
1995
.
Unaccusativity
.
Cambridge, MA
:
MIT Press
.
Levin,
Theodore
.
2017
.
Successive-cyclic case assignment: Korean nominative-nominative case-stacking
.
Natural Language and Linguistic Theory
35
:
447
498
.
Levin,
Theodore
, and
Omer
Preminger
.
2015
.
Case in Sakha: Are two modalities really necessary?
Natural Language and Linguistic Theory
33
:
231
250
.
Loriot,
James,
Erwin
Lauriault,
and
Dwight
Day
.
1993
.
Diccionario Shipibo-Castellano
.
Yarinacocha, Peru
:
Instituto Lingüístico de Verano
.
Marantz,
Alec
.
1991
. Case and licensing. In
Eastern States Conference on Linguistics
, ed. by
Germán
Westphal
,
Benjamin
Ao
, and
Hee-Rahk
Chae
,
234
253
.
Ithaca, NY
:
Cornell University, Cornell Linguistics Club
.
Massam,
Diane
.
1998
.
Instrumental aki and the nature of Niuean transitivity
.
Oceanic Linguistics
37
:
12
28
.
Massam,
Diane
.
2006
. Neither absolutive nor ergative is nominative or accusative. In
Ergativity: Emerging issues
, ed. by
Alana
Johns
,
Diane
Massam
, and
Juvénal
Ndayiragije
,
27
46
.
Dordrecht
:
Springer
.
Massam,
Diane
.
2015
. Applicatives as secondary predicates. In
AFLA 21
, ed. by
Amber
Camp
,
Yuko
Otsuka
,
Claire
Stabile
, and
Nozomi
Tanaka
,
185
199
.
Asia-Pacific Linguistics
. http://users.clas.ufl.edu/potsdam/papers/AFLA21.pdf.
Müller,
Gereon
.
1996
.
A constraint on remnant movement
.
Natural Language and Linguistic Theory
14
:
355
407
.
Myler,
Neil
.
2013
.
On coming the pub in the North West of England: Accusative unaccusatives, dependent case, and preposition incorporation
.
Journal of Comparative Germanic Linguistics
16
:
189
207
.
Nelson,
Katherine
.
2013
.
An acoustic analysis and cross-linguistic study of the phonemic inventory of Nez Perce
.
Doctoral dissertation, Rice University, Houston, TX
.
Nevins,
Andrew
.
2011
.
Multiple Agree with clitics: Person complementarity vs. omnivorous number
.
Natural Language and Linguistic Theory
29
:
939
971
.
Newell,
Heather
.
2008
.
Aspects of the morphology and phonology of phases
.
Doctoral dissertation, McGill University, Montreal
.
Nissenbaum,
Jon
.
2000
.
Investigations of covert phrase movement
.
Doctoral dissertation, MIT, Cambridge, MA
.
Otsuka,
Yuko
.
2006
. Syntactic ergativity in Tongan: Resumptive pronouns revisited. In
Ergativity: Emerging issues
, ed. by
Alana
Johns
,
Diane
Massam
, and
Juvénal
Ndayiragije
,
79
109
.
Dordrecht
:
Springer
.
Paul,
Ileana
, and
Lisa
Travis
.
2006
. Ergativity in Austronesian languages: What it can do, what it can’t, but not why. In
Ergativity: Emerging issues
, ed. by
Alana
Johns
,
Diane
Massam
, and
Juvénal
Ndayiragije
,
315
336
.
Dordrecht
:
Springer
.
Polinsky,
Maria
.
2016
.
Deconstructing ergativity: Two types of ergative languages and their features
.
Oxford
:
Oxford University Press
.
Poole,
Ethan
.
2015
. A configurational account of Finnish case. In
University of Pennsylvania Working Papers in Linguistics 21.1
,
article 26
. https://repository.upenn.edu/pwpl/vol21/iss1/26.
Preminger,
Omer
.
2014
.
Agreement and its failures
.
Cambridge, MA
:
MIT Press
.
Preminger,
Omer,
and
Jaklin
Kornfilt
.
2015
. Nominative as no case at all: An argument from raising-to-accusative in Sakha. In
Proceedings of the 9th Workshop on Altaic Formal Linguistics (WAFL 9)
, ed. by
Andrew
Joseph
and
Esra
Predolac
,
109
120
.
MIT Working Papers in Linguistics 76
.
Cambridge, MA
:
MIT, MIT Working Papers in Linguistics
.
Pylkkänen,
Liina
.
2008
.
Introducing arguments
.
Cambridge, MA
:
MIT Press
.
Rackowski,
Andrea
, and
Norvin
Richards
.
2005
.
Phase edge and extraction: A Tagalog case study
.
Linguistic Inquiry
36
:
565
599
.
Rezac,
Milan
,
Pablo
Albizu
, and
Ricardo
Etxepare
.
2014
.
The structural ergative of Basque and the theory of Case
.
Natural Language and Linguistic Theory
32
:
1273
1330
.
Richards,
Marc
, and
Theresa
Biberauer
.
2005
. Explaining Expl. In
The function of function words and functional categories
, ed. by
Marcel
den Dikken
,
115
153
.
Amsterdam
:
John Benjamins
.
Rosen,
Carol G
.
1984
. The interface between semantic roles and initial grammatical relations. In
Studies in Relational Grammar 2
, ed. by
David M.
Perlmutter
and
Carol G.
Rosen
,
38
77
.
Chicago
:
University of Chicago Press
.
Rude,
Noel
.
1985
.
Studies in Nez Perce grammar and discourse
.
Doctoral dissertation, University of Oregon, Eugene
.
Rude,
Noel
.
1986
.
Topicality, transitivity, and the direct object in Nez Perce
.
International Journal of American Linguistics
52
:
124
153
.
Rude,
Noel
.
1992
. Word order and topicality in Nez Perce. In
Pragmatics of word order flexibility
, ed. by
Doris L.
Payne
,
193
208
.
Amsterdam
:
John Benjamins
.
Sauerland,
Uli
.
2003
.
Intermediate adjunction with A-movement
.
Linguistic Inquiry
34
:
308
314
.
Schütze,
Carson T.
2001
.
On the nature of default case
.
Syntax
4
:
205
238
.
Travis,
Lisa deMena
.
2010
.
Inner aspect
.
Dordrecht
:
Springer
.
van Urk,
Coppe
, and
Norvin
Richards
.
2015
.
Two components of long-distance extraction: Successive cyclicity in Dinka
.
Linguistic Inquiry
46
:
113
155
.
Valenzuela,
Pilar M.
2003
.
Transitivity in Shipibo-Konibo grammar
.
Doctoral dissertation, University of Oregon, Eugene
.
Valenzuela,
Pilar M.
2010
. Ergativity in Shipibo-Konibo, a Panoan language of the Ucayali. In
Ergativity in Amazonia
, ed. by
Spike
Gildea
and
Francesc
Queixalos
,
65
96
.
Amsterdam
:
John Benjamins
.
Wiltschko,
Martina
.
2006
. On “ergativity” in Halkomelem Salish (and how to split and derive it). In
Ergativity: Emerging issues
, ed. by
Alana
Johns
,
Diane
Massam
, and
Juvénal
Ndayiragije
,
197
228
.
Dordrecht
:
Springer
.
Wood,
Jim
.
2017
.
The accusative subject generalization
.
Syntax
20
:
249
291
.
Woolford,
Ellen
.
1997
.
Four-way case systems: Ergative, nominative, objective and accusative
.
Natural Language and Linguistic Theory
15
:
181
227
.
Woolford,
Ellen
.
2006
.
Lexical case, inherent case, and argument structure
.
Linguistic Inquiry
37
:
111
130
.
Yip,
Moira
,
Joan
Maling
, and
Ray
Jackendoff
.
1987
.
Case in tiers
.
Language
63
:
217
250
.