Abstract

This remark argues for the Case-theoretic concept of Case as a syntactic licensing requirement of DPs related to their case morphology, against alternatives where DPs need no syntactic licensing, or where their case morphology is unrelated to it. The argument is made from constraints on combining exceptional Case marking (ECM) and the double object construction, where there is a single object Agree/Case locus for two DPs. The mechanisms and nature of Case are briefly examined. The appendix sketches an extension to wager-class ECM.

1 Introduction

Case theory proposes that there is a syntactic property of DPs, Case, that accounts for aspects of their distribution and form that do not otherwise follow from their PF and LF content (Chomsky 1981:sec. 2.3, 1986:secs. 3.3.3.3.1, 3.5.2.5, Chomsky and Lasnik 1995:secs. 1.3, 1.4.3, Lasnik 2008). The Agree framework of the Minimalist Program implements Case through uninterpretable [uCase] on DPs that must be licensed by Agree with a [uφ] bearer, deleting [uCase] from syntax to satisfy Full Interpretation at LF and giving it a value realized as case morphology at PF (Chomsky 2000:122–123, 127–128, 2001:6–9, 2008:142, 151). However, the framework also shifts much of the work earlier done by Case to the needs of clausal functional heads, such as [uφ], opening up the possibility of reducing the role of Case in DP-licensing (as a condition on Agree or Move; Chomsky 2000:123, 2001:4, 6, 10) or even of eliminating it entirely (leaving case morphology to PF; Marantz 2000, McFadden 2004, Bobaljik 2008, Sigurðsson 2009).

This remark argues that the Case-theoretic view of Case is right: there is a DP-licensing requirement related to DP case morphology. The argument is made from constraints on combining exceptional Case marking (ECM) and the double object construction (DOC)—ECM+DOC—in (1). It builds on Case-based accounts of ECM+DOC by Pesetsky (1991) and Boeckx and Hornstein (2005), but separates Case from potentially independent factors such as adjacency.

(1)

  • a.

    the propositions which2they1 were shown t1 [t2 to be true]

  • b.

    *They1 were shown t1 [the propositions to be true]. (*Case licensing of they)

Section 2 reviews Case theory, section 3 identifies the relevant properties of ECM and the DOC, section 4 argues for Case licensing, section 5 relates it to case morphology, section 6 sets out the consequences, and the appendix extends the proposal to wager-class ECM.

2 Case Theory

Case theory has two core components, DP-licensing and DP morphology.

(2)

  • a.

    DP-licensing [The Case Filter]

    DPs must be licensed through certain syntactic dependencies. Those that are not licensed by selection (inherent Case) must be licensed by an A-dependency to the clause (structural Case). In the Agree framework, this is Agree for [uCase] valuation and deletion.

  • b.

    DP morphology

    The Case licensing of DPs influences their case morphology.

The working of structural Case is illustrated in (3). The subject of the infinitive is licensed and bears case in virtue of its clausal context. In (3a), the subject is accusative in virtue of entering an Agree relation (Agreeing) with v/VAcc of an active ECM verb; in (3b), it is nominative in virtue of Agreeing with TNom of a passive ECM verb; and in (3c), it is accusative in virtue of Agreeing with C for of the infinitive. In the absence of a suitable Agree relationship, the subject is not licensed, as when ECM verbs are replaced by control verbs in (3a–b). Case licensing/ assignment is limited by the usual conditions on syntactic dependencies: while the subject of the infinitive is licensed by the matrix TNom in (3b), it is not in (3d), because the infinitive is a sentential subject island opaque to syntactic dependencies.

(3)

  • a.

    Kate believesECM/*hopesControl [them to have been shown the proof].

  • b.

    Theyi are believedECM/*hopedControl [ti to have been shown the proof].

  • c.

    [For them to obtain funding]i is believed [ti to be difficult].

  • d.

    [PRO/*They/*Them to obtain funding]i is believed [ti to be difficult].

In the Agree framework, the distribution of the DPs in (3) is largely ensured by the properties of clausal heads independently of the Case of the DPs. The φ-probes [uφ] of T, v/V, and C find DPs to Agree with; their ‘‘edge’’ features [EF] ([EPP], [OCC]) license positions for movement; and control infinitives have PRO in virtue of properties of their C/T system, not because of the inability of control verbs to assign Case and of PRO to bear it (Sigurðsson 1991, Chomsky and Lasnik 1995:107–108, 110–124, Marantz 2000:20–21).1 Moreover, the link between licensing and case morphology has been weakened by arguments that some DPs like left conjuncts have ‘‘default case’’ unrelated to their licensing (Marantz 2000, Schütze 2001), while others have ‘‘quirky case’’ determined by selection yet are licensed in the same positions as DPs with structural Case (Schütze 1993, Marantz 2000, Sigurðsson 2009). These considerations have led to theories that eliminate Case licensing from syntax and shift case morphology to PF (Marantz 2000, McFadden 2004, Bobaljik 2008, Sigurðsson 2009), even as others retain or extend the role of syntactic Case (Bošković 1997, 2002, 2007, Lasnik 2008).

The present argument for Case follows others in the literature: take a well-formed structure, add a DP that should be licensed in it as far as PF/LF needs like thematic interpretation go, and relate the licensing failure to unavailability of case morphology. The usefulness of ECM+DOC comes from its providing controls for other factors such as selectional or movement requirements. Three examples illustrate the format of the argument and the challenges that ECM+DOC meets.

First, (4b) has been ruled out by positing that both it and they/them need Case and that unaccusatives can only Case-license one DP (Chomsky 1981:113, Burzio 1986:182). However, (4b) is also ruled out if it needs a CP associate (Bošković 1997:128–129).2

(4)

  • a.

    Theyi seem to her [ti to be tired].

  • b.

    *It seems [they/them to be tired].

Second, (5a) shows that DP arguments of nouns and adjectives need of, while DP arguments of verbs and prepositions do not. Case theory explains this by positing that DPs need Case and that V/P but not N/A assigns it. Pesetsky (1982:sec. 2.5) extends the logic to (5b), proposing that ask but not wonder assigns Case and so allows a DP complement. However, Rothstein (1992) highlights problems such as the licensing of what in (5c), returning to Grimshaw’s (1979) explanation for (5b) by c-selection.

(5)

  • a.

    proud (*of) leaving (cf. proud to leave)

  • b.

    ask/*wonder the time

  • c.

    Whati John wondered ti was how to get there.

Third, Burzio’s (1986:sec. 3.1) Generalization restricts object Case to v/VPs with an external argument. Rothstein (1992) argues that this explains the distribution of the ‘‘ECM’’ resultatives in (6) if their subject, X’s way, needs Case licensing. They are licensed in transitives, but only if another object does not use up object Case, (6a); in unergatives, which by Burzio’s Generalization can assign object Case, (6b); but not in unaccusatives, which cannot, (6c). However, it is not agreed that the subject of these resultatives is thematically independent of the matrix clause, leaving that as a potential factor in its licensing (Beavers 2012).

(6)

  • a.

    We drank (*the beer) our way through the room.

  • b.

    The riveriti thundered/cascaded (its way) down the ravine.

  • c.

    The riveri fell/flowed ti (*its way) down the ravine.

3 Selection, Case, Position, and Probes in ECM and the DOC

This section sets out the properties of ECM and the DOC that will allow controlling for selection, Case adjacency, object shift, and φ-probes, thus isolating Case.

3.1 ECM

ECM verbs like believe or prove take an infinitival complement whose subject depends solely on the infinitive for s- and c-selection, but on the ECM verb for case, agreement, and movement. When the ECM verb is active, the subject of the infinitive is accusative and adjacent to the verb, (7a), or the associate of an expletive adjacent to the verb, (7b).3 When the ECM verb is passive, the subject is an agreeing nominative in Spec,T, (7c), or in the pre-participial position as the associate of an expletive, (7d). In both active and passive, it may undergo Ā-extraction, (7e–f).4

(7)

  • a.

    We proved (*conclusively/*to the reader) the propositions/them (conclusively/to the reader) to be consistent.

  • b.

    We proved (*conclusively/*to the reader) there (*conclusively/*to the reader) to be few propositions consistent with ours.

  • c.

    The propositions/They were proven (to the reader) to be consistent.

  • d.

    There were few propositions proven (to the reader) to be consistent with ours.

  • e.

    the propositions that we proved (conclusively/to the reader) to be consistent

  • f.

    the propositions that were proven (to the reader) to be consistent

The case, agreement, and licensing of the subject of the infinitive in ECM cannot depend on the subject’s selectional relationship to the ECM verb, for it does not have one. Two other phenomena affect the subject in ECM: Case adjacency and object shift. These phenomena have been viewed as reflexes of Case licensing, but on Caseless theories they could be attributed to DP-external requirements—say, [EF] on V. To show that the impossibility of ECM+DOC is due to a Case-licensing failure, they must be controlled for, and they can indeed be separated from case and licensing.

Case adjacency refers to the impossibility of separating accusatives or associated expletives from the verb by adverbs or PPs; see (7a–b), (8a). It has been viewed as reflecting object shift, which is discussed next (Johnson 1991, Bowers 1993, Koizumi 1995), or a PF adjacency condition on Case assignment (Chomsky 1981:94, Alexiadou and Anagnostopoulou 2001:207–208, Ackema and Neeleman 2004:264–266). Nominatives, however, show no adjacency restrictions, belying any general adjacency requirement on Case licensing or case assignment. They may be separated from auxiliaries, finite verbs, and participles in both preceding and following positions (see Johnson 1991: 579–580 for (8b), Bowers 2002:203n25 for (8a) vs. (8d); contrast Culicover and Levine 2001:292 with different adverbs). Thus, nominatives have no need of their own that would impose adjacency, and they will serve here to control for it.

(8)

  • a.

    John rolled (*perfectly) the ball down the hill.

  • b.

    I knew that (probably) Gary (probably) had left.

  • c.

    There had been (at first) only a few finds (conclusively) attributed to this period.

  • d.

    Down the hill rolled (?perfectly) the ball (perfectly).

Object shift of the subject of the infinitive into the matrix clause in ECM is likewise not an intrinsic requirement of nominatives or accusatives. Lasnik’s (1999, 2001b, 2008, 2010) work on verb-particle constructions establishes that object shift of the accusative subject in active ECM exists but is optional. I illustrate by sketching two arguments. In (9), only the subject-particle order allows the subject to c-command into matrix adjuncts, indicating that in the particle-subject order the subject remains in the infinitive.5 In (10), only the particle-subject order allows the subject to take scope below the embedded negation; the subject-particle order patterns with raising in (11) in blocking it, suggesting that only in this case has the subject raised out of the infinitive (on A-movement and inverse scope, see von Fintel and Iatridou 2003:186–194 and the literature cited there).

(9)

  • a.

    The DA made no suspecti out to have been at the scene during hisi/any trial.

  • b.

    The DA made out no suspecti to have been at the scene (?*during hisi/any trial).

(10)

  • a.

    The mathematician made every even number out not to be the sum of two primes.

  • b.

    The mathematician made out every even number not to be the sum of two primes.

    ✓a, ✓b: every > not = ‘For every even number, the mathematician made out (proved) that it isn’t the sum of two primes [immediately falsified by 4 = 2 + 2].’

    *a, ✓b: not > every = ‘The mathematician made out (proved) that it isn’t the case that every even number is the sum of two primes [unlike 4 = 2 + 2].’

(11)

  • Everyonei seems not to be ti there yet. (everyone > not, *not > everyone)

  • ≠ ‘It seems that everyone is not there yet.’ (not > everyone)

Lasnik (1999, 2001b, 2008, 2010) concludes that ECM allows but does not require object shift.6 His mechanism is transposed into the Agree framework in (12). v/VAcc has the φ-probe [uφ], it Agrees with the closest DP, the DP’s [uCase] is valued to accusative, and optional [EF] on v/VAcc moves the DP.

(12)

graphic

Lasnik’s proposal makes accusative licensing and case independent of object shift. At the same time, object shift can be restricted to accusatives by putting [EF] only on active v/V as in (12)—that is, on v/V with [uφ, iAcc]—correctly barring object shift of nonaccusatives such as PPs (call (*to Kate) out (to Kate); Johnson 1991).7

Nominatives pattern with other nonaccusatives in failing to shift to the pre-particle position. This is true in expletive constructions (13a), where they prefer to precede the participle, and more strikingly so in locative inversion (13b), where they follow it (cf. Hoekstra 2004:379 (citing Den Dikken 1990), Postal 2004:49, Rezac 2006, 2010).8

(13)

  • a.

    graphic

  • b.

    graphic

Since nominatives require neither object shift nor Case adjacency for their case, agreement, and licensing, they will provide a control for both phenomena: if a nominative is barred, it is not because of these requirements. This conclusion extends to other restrictions in which nominatives do not participate (e.g., the Object Agreement Constraint barring ECM+DOC in Ormazabal and Romero 2002, 2010:228; as the constraint does not restrict passives with a nominative, it should not rule out passive ECM+DOC).

3.2 The DOC

The double object construction (DOC) combines two DPs: a higher indirect object (IO) and a lower direct object (DO). The indirect object has the same properties as the direct object of plain transitives for case, agreement, and position. In the active, it is accusative (14a). In the passive, it is nominative and agreeing, either in Spec,T as in (14b), or as the indefinite associate of the there expletive in the pre-participial position as in (14c). By contrast, the direct object is a nonagreeing accusative in both the active and the passive, with no possibility of raising and no definiteness restriction in expletive constructions. Only the indirect object shifts over particles (15a–b) and is subject to Case adjacency if accusative (15c).9

(14)

  • a.

    We showed only themIO the propositionsDO.

  • a′.

    *We showed the propositionsDO the readersIO.

  • b.

    TheyIO were shown the propositionsDO/only themDO/them allDO one by one/it and themDO at the same time.

  • b′.

    *TheyDO were shown the readersIO.

  • c.

    There were few readersIO shown the new propositionsDO.

  • c′.

    *There were new propositionsDO shown few readersIO.

(15)

  • a.

    Sam sent (%out) the stockholders (out) a schedule (*out).

    (cf. Pesetsky 1995:278)

  • b.

    The stockholders were sent (out) a/the schedule (*out).

  • c.

    Sue gave (*yesterday) Bill (?yesterday) a book.

    (Pesetsky 1995:124)

These facts indicate that the indirect object of the DOC enters into the same agreement, case, and A-movement relationships with clausal functional heads as the direct object of plain transitives, while the case and position of the direct object of the DOC have another source. This source has been argued to be the direct object’s selector (Larson 1988, Pesetsky 1995, Baker 1997, Anagnostopoulou 2001): a silent P (overt in present X with Y), or V through inherent case (as in Inuit; Bittner and Hale 1996:18), or V through incorporation (as in Mohawk or Nez Perce; Baker 1996:sec. 4.3, Deal 2010:90–91, respectively). On this analysis, the indirect object alone satisfies the needs of clausal heads pertaining to A-dependencies (e.g., [uφ] and [EF]) and any further requirements reflected in Case adjacency.10

4 ECM+DOC: DP-Licensing

The verb show participates in both ECM and the DOC, and may combine the two, but the resulting ECM+DOC construction is subject to severe limits. It is ungrammatical, as shown in (16) for actives and (17) for passives, unless the subject of the infinitive Ā-extracts, (18) (Kayne 1984: 5, Pesetsky 1991:sec. 4.1, Postal 1993).11

(16)

  • a.

    *We showed the reader the propositions to be consistent.

  • b.

    *We showed few readers there to be any propositions consistent with ours.

  • c.

    *We showed her ’em/only them/them all/it and them to be consistent.

  • d.

    *I showed him there to be mice in the basement.

    (Postal 1993:361)

  • e.

    *I guarantee you it to be possible to revise the program.

    (Postal 1993:362)

(17)

  • a.

    *The reader was shown _____ the propositions/’em/them all to be consistent.

  • b.

    *There were few readers shown any propositions to be consistent with yours.

(18)

  • a.

    (?)the propositions that we showed the reader _____ to be consistent

  • b.

    (?)the propositions that the readers were shown _____ to be consistent

From this pattern of (un)grammaticality, I will argue that the subject of the infinitive has a licensing requirement and that its satisfaction correlates with morphological case, revealing the link between licensing and case that defines Case.

The ungrammaticality of ECM+DOC (16) has already been analyzed as failure to Case-license the subject of the infinitive by Pesetsky (1991:sec. 4.1) and by Boeckx and Hornstein (2005). However, the former analysis relies on an approach where Case requires adjacency, the latter on one where it requires object shift. On Caseless theories, Case adjacency and object shift might reflect other requirements than the syntactic needs of DPs: for instance, a syntactic [EF] on a clausal head like v, or a PF adjacency requirement on the assignment of morphological case. The properties of ECM and the DOC established above permit eliminating these and other factors, to show that the culprit is a licensing requirement of the subject of the infinitive itself.

The problem in (16)–(17) is not s/c-selectional requirements, such as incompatibility of ECM and the DOC, because ECM and the DOC do combine in (18). It is also not nonselectional syntactic requirements of the matrix clause, because these are satisfied in the DOC (14)–(15) by the indirect object and they should be in ECM+DOC (16)–(17) as well: for example, [uφ] that gives rise to agreement in passives, and [EF] that requires movement to the pre-participial position in expletive passives. The problem must be some need of the subject of the infinitive that is not met in ECM+DOC, unlike in the plain DOC and ECM. In the plain DOC, the direct object is accusative through a selectional relationship to the DOC verb, but the subject of the infinitive in ECM+DOC is not selected by the DOC verb. In plain ECM, the subject of the infinitive is accusative or nominative according to the active or passive voice of the ECM verb, raising the question of what prevents it from being so in ECM+DOC, namely, accusative in active-voice (16) and nominative in passive-voice (17). The stumbling block must be the presence of the indirect object, since that is the difference between ECM and ECM+DOC. Thus, the subject of the infinitive has some requirement that the indirect object prevents from being satisfied. The requirement is unrelated to Case adjacency and object shift, if these reflect something other than matrix syntactic requirements like [EF] discussed earlier (e.g., a PF adjacency requirement on morphological case assignment), because nominatives are not subject to either phenomenon and ECM accusatives are not subject to object shift, yet both are unacceptable in (16)–(17).

The existence of a requirement needed to license the subject of the infinitive in ECM corresponds to the licensing role of Case—specifically, of Case as construed in the Agree framework, where it is independent of movements like object shift. The indirect object prevents Case licensing, as diagrammed in (19).

(19)

graphic

The next section turns to identifying the problem in (19), linking licensing to morphological case, and explaining how otherwise unavailable case and licensing are permitted by Ā-movement.

5 ECM+DOC: Case and Licensing

Case theory attributes to Case both DP-licensing and DP case morphology. To identify the problem in ECM+DOC as Case rather than some other licensing requirement of DPs, it should be relatable to case morphology. This relationship is supported by two lines of evidence: the blocking of ECM+DOC by the indirect object of the DOC rather than by any intervener, and its obviation by Ā-movement.

ECM structures, both active and passive ( = raising), are blocked by the DOC (16)–(17), but not by the prepositional object construction (POC) (20) (Bowers 1993:632; see (7) and footnote 4 above). The POC indirect object intervenes between the ECM verb and the infinitive both in linear order and in structural c-command, as shown in (21) for Condition C and variable binding.12

(20)

  • a.

    We showed few propositions (%to the reader) to be consistent (to the reader).

  • b.

    Few propositions were shown (to the reader) to be consistent (to the reader).

  • c.

    There were few propositions shown (to the reader) to be consistent (to the reader).

(21)

  • a.

    The projectk was shown to her*i [tk to be about Maryi].

  • b.

    The projectk was shown to every groupi[tk to meet itsi needs].

  • c.

    ?There is one project thatk we can show her*i [tk to be about Maryi].

  • d.

    ?There is one project thatk we can show every groupi [tk to meet itsi needs].

Therefore, the problem in ECM+DOC is not simply the intervention of a matrix argument between the ECM verb and the infinitive. Something about the indirect object or the head introducing it differs in the DOC and the POC. Case theory provides this difference by linking licensing and case. The case morphology of the ECM subject and the DOC indirect object depends on nonselectional relationships to the matrix clause: both are accusative in actives and nominative in passives. By contrast, the case of the indirect object in the POC to-PP is independent of the clause and assigned by P. Therefore, morphological case groups the ECM subject and the DOC indirect object to the exclusion of the POC indirect object. This is the same grouping as for licensing: licensing of the subject of the infinitive is blocked by the DOC but not by the POC. Case unifies these groupings by linking licensing and case, and the ungrammaticality of ECM+DOC may then be explained through competition for clausal structural Case:

  • There is only one Case licenser/assigner ĸ for both the DOC indirect object and the subject of the infinitive: v/VAcc in actives and TNom in passives.

  • ĸ can relate to only one of the two DPs. This follows if (i) φ-Agree occurs with one goal at a time and valuation by a nondefective goal deactivates the probe, or (ii) if locality limits φ-Agree to the closer DP, as under the notion of intervention-based locality where the indirect object is the closest φ-bearer (Chomsky 2000).13

The link between licensing and case is supported by the saving effect of Ā -movement in (18). Kayne (1984:5) and Pesetsky (1991:sec. 4.1.1.5) propose that Ā-movement allows Case licensing in ECM+DOC by bringing the subject of the infinitive into a local configuration with the ECM verb, a configuration it would otherwise lack. To support the idea that Ā-movement can establish new Case relations, Kayne observes that it affects case morphology in (22): Ā-movement past a transitive verb, but not an unaccusative one, assigns accusative to whom.

(22)

  • a.

    %the people whom {you say}/{they tell me}/{I believe} _____ are extremely bright

  • b.

    *the people whom it is obvious _____ like you

    (Kayne 1984:5, where the ungrammaticality of (22b) is ‘‘surmised’’)

This is a link between licensing and case, but in English it is on shaky ground. It is not clear whether the who-whom alternation ever reflects case, as (22) indicates (Lasnik and Sobin 2000). In other languages, however, Ā-movement past v/VAcc does assign otherwise unavailable accusative, confirming Kayne’s proposal. The Hungarian examples in (23) illustrate (É. Kiss 1987, Lipták 1998, Gervain 2002, Coppock 2004, Den Dikken 2006, 2009).

(23)

  • a.

    Péter-t/%Péter mond-t-am, hogy jön.

    Peter-ACC/NOM say-PAST-1SGCOMP come.3SG

    ‘It is Peter who I said is coming.’

    (Coppock 2004:8, (30))

  • b.

    Téged mondta-lak hogy szeretné-lek hogy elnök leygél.

    you.ACC said-1SG→2 that would.like-1SG→2 that president be.2SG

    ‘It is you that I said that I would like to be president.’

    (Den Dikken 2009:13, (29))

  • c.

    János két dolgot szeretne, ha sikerülne.

    Janos two thing.ACC would.like.INDEF if succeeded

    c′. János két dolog szeretné, ha sikerülne.

    Janos two thing.NOM would.like.DEF if succeeded

    ‘As for Janos, it is two things that he would like if they succeeded.’

    (É. Kiss 1987:141, Gervain 2002:31)

  • d.

    A fiúk/*fiúkat lenne jó ha úsznának.

    the boys.NOM/*ACC be.SUBJ.3SG good.SG if swim.SUBJ.3PL

    ‘It is the boys that it would be good if they swam.’

    (Anikó Lipták, pers. comm.; cf. Lipták 1998)

A finite subject focus-moved past a transitive bridge verb may surface as nominative, as when unmoved, but it may also become accusative and trigger upstairs object agreement for person (23b) and definiteness (23c). The distribution of accusative case and object agreement on focus-moved subjects has led researchers to conclude that these properties have the same source as they do on regular objects, specifically checking with AgrO (Lipták 1998) and Agree with v/ VAcc (Den Dikken 2006). Transitive objects are accusative and control obligatory object agreement on their selecting verb, as well as optional object agreement on a transitive bridge verb under focus movement past it. Subjects moved past a transitive bridge verb may be nominative without upstairs object agreement (23c′) or accusative with upstairs object agreement (23c). However, when the bridge verb is unaccusative and so cannot assign accusative or bear object agreement, subjects moved past it can only be nominative without upstairs object agreement (23d).

Similar arguments for object case/agreement through Ā-movement have been made on the basis of topicalization in Norwegian (Taraldsen 1981) and Ā-movement in Passamaquoddy (Bruening 2001:sec. 5.7), and it has been related to DP-licensing both in ECM+DOC and in ECM with wager/croire-class verbs that are discussed in the appendix (Kayne 1984:sec. 5.3, Pesetsky 1991:sec. 4.1.1.5, Ura 1993, Bošković 1997:secs. 3.3, 4.4.2). Ura, Bošković, and Bruening develop economy-based accounts. I adapt Bruening’s proposal to ECM+DOC in (24).

(24)

graphic

The ability to value [uCase] as accusative and nominative reflects interpretable properties of v/VAcc and TNom (Chomsky 2000, Pesetsky and Torrego 2004). These are shared by v/V in both active and passive structures.14 However, active v/V has [uφ] that Agrees with the closest goal, while passive v/V does not, allowing [uφ] onT to do so. In ECM+DOC, the goal of the [uφ] of v/V in actives and T in passives is the indirect object, since it is the closest [φ] bearer. Agree eliminates [uφ]onv/VAcc and TNom and [uCase] on the indirect object, but does not affect the interpretable properties of v/VAcc and TNom, including their ability to value [uCase]. For the subject of the infinitive, no [uφ] probe is left, but Ā-movement brings it to Spec,v/V (Fox 1999, Legate 2003, Sauerland 2003) and thereby allows [uCase] valuation as follows. Let us suppose that successive-cyclic Ā-movement entails Agree—for instance, for [uwh] in Chomsky 2000, since it is reflected in the morphology of v both as wh-agreement (Chung 1994, Cole and Hermon 2000) and as φ-agreement (Bruening 2001). Let us further suppose that Agree between items α, β maximizes to include all their features as free riders (see Chomsky 1995:secs. 4.4.4, 4.5.2, 2001:15–19, Bruening 2001:sec. 5.7). Then a DP that Agrees for Ā-movement through Spec,v/ VAcc can have its [uCase] valued by the interpretable properties of v/VAcc as part of that Agree.15 This Case-licenses the subject of the infinitive in ECM+DOC, in both active (18a) and passive (18b) contexts. The mechanics extend to extracted finite subjects (22)/(23), if their [uCase] need not be valued in their clause, or if it can be valued multiply. From analogues of (22a) where the Ā-moved element bears both nominative and accusative, Béjar and Massam (1999) argue for parametric availability of the latter option.16

6 Consequences

The argument from ECM+DOC has two independent steps: identifying a licensing requirement of DPs and relating it to case morphology. In ECM, the subject of the infinitive is licensed and assigned case by the matrix clause. Case unifies these relationships: in plain ECM, where both succeed; in ECM+DOC, where both are blocked by another DP with structural Case; and under Ā-movement, by which both are restored.

Case theory attributes case and licensing to a syntactic Case dependency. In this, it contrasts with proposals that DP-licensing belongs to syntax but DP case (and agreement) belong to an interpretive PF component that realizes but does not filter syntax (Marantz 2000:20). On this view, PF needs to relate DPs and case assigners (and agreement targets) across phrase-structurally unbounded domains, as in (25), where the case of ‘several boats’ within the infinitive depends on the voice of the matrix ECM verb. Such dependencies are characteristic of syntax (A/Ā-movement) and LF (quantifier-variable binding), but arguably not of PF, where independently established dependencies such as contextual allomorphy are phrase-structurally local (Rezac 2011: chap. 2). If there is syntactic case-and-licensing, as Case theory proposes, then there is no reason to attribute all case to PF and thereby require it to encompass long-distance dependencies like those in (25).

(25)

  • a.

    Hún taldi hafa verið keypta einhverja báta.

    she believed to.have been bought.MASC.PL.ACC several boats.MASC.PL.ACC

    ‘She believed several boats to have been bought.’

  • b.

    það voru taldir hafa verið keyptir einhverjir

    there were believed.MASC.PL.NOM to.have been bought several

    bátar.

    boats.MASC.PL.NOM

    ‘There were believed to have been several boats bought.’

    (Sigurðsson 1991:355–356)

The existence of the Case-theoretic relationship between case and licensing does not entail that all case morphology should straightforwardly reflect licensing dependencies. Case morphology may reflect dependencies other than [uCase] valuation by Agree, such as thematic ones (see Chomsky 2000:127 on quirky case). Systems outside core syntax may obscure [uCase] realization: it cannot surface on null operators (Chomsky and Lasnik 1995:115–116), morphology adjusts it in language-specific ways in multiple case resolution (McCreight 1988, Béjar and Massam 1999), and it may be adjusted by rules ‘‘external to the normal grammatical system’’ (Lasnik and Sobin 2000:350). Finally, there may be configurations where DPs do not need licensing. Their existence and character bear on what Case is.

Schütze (2001) argues that in certain configurations—dislocation, apposition, coordination, modification, and deletion remnants—PF assigns default case to DPs. It is not clear whether these DPs do or do not require syntactic Case licensing. If they do not, then two natural domains for configurations where DPs do require it are the following:

(26) Syntactic Case licensing is required by DPs that are

  • a.

    introduced by set but not pair Merge, or

  • b.

    goals of φ-Agree and/or (A-)movement.

(26a) attributes Case to argumental but not adjoined DPs. It covers ground similar to that covered by the view that Case is required for thematic interpretation (Chomsky 1981:176–177, 334ff., 1986:secs. 3.3.3.3.1, 3.4.3 (also see 212n71 for expletives), Chomsky and Lasnik 1995:sec. 1.4.3). (26b) views Case as a property that allows DPs to participate in certain operations, rendering them available to φ-Agree (Chomsky 2000:123) or to movement (Chomsky 2000:123, 2001:4, 6, 10, Sobin 2009). Both proposals call for a principled explanation of why these DPs need Case licensing, such as those mentioned by Chomsky (2000:148n86). Both require the DP subject of ECM infinitives to be Case-licensed syntactically, since it is argumental and relates to a T to satisfy its [EF]/[uφ]. It remains to be seen whether the two proposals can be differentiated by examining argumental DPs that do not participate in certain operations: for instance, unmoved DPs in infinitives opaque to external Agree and with no internal Agree.17

Appendix: Extension to Wager-Class ECM

The role of Ā-movement in Case licensing also appears with ECM verbs like wager or allege in (27), discussed by Postal (1974:secs. 9.3–9.4, 1993), Pesetsky (1991), and Bošković (1997). They license a weak pronoun and the trace of A-and Ā-movement, but not a full DP. Wager-class ECM thus differs both from believe-class ECM, which licenses all of the above, and from wager/believe-class ECM+DOC, which licenses only an Ā-trace.

(27)

  • a.

    We alleged them/*THEM/*the propositions to be inconsistent.

  • b.

    (?)the propositions that we alleged [ _____ to be inconsistent]

  • c.

    The propositions were alleged [ ______ to be inconsistent].

In this appendix, I shall suggest an approach to the wager limitation that assimilates it to ECM+DOC, following Pesetsky (1991:sec. 4.1.2). The guiding intuition is that both have extra structure blocking matrix v/VAcc φ-Agree, but that the extra structure of wager vanishes by the time of TNom Agree. I shall proceed on the assumption that Agree is constrained by featurerelativized minimality (Chomsky 2000:122–124), so that in ECM+DOC, the indirect object intervenes in φ-Agree with the subject of the infinitive (Boeckx and Hornstein 2005). In wager but not believe ECM, a silent N0 in [v/VAcc [N0 Inf]] will intervene in v/VAcc φ-Agree but become invisible by the time of TNom φ-Agree.

From unergatives in ergative languages, Laka (2000) argues for the existence and parameterization of a silent N0 intervener for φ-Agree. These unergatives may have the case/agreement pattern of transitives, attributed to a silent N0 visible to φ-Agree, or that of unaccusatives, attributed to a lexically incorporated N0 (eastern vs. western Basque borrokatu ‘fight’, respectively; Oyharçabal 1992).18 I propose that wager but not believe ECM verbs have N0 visible to Agree, as a lexical parameter.19 This gives the structure (28).

(28)

graphic

In (28), N0 intervenes for φ-Agree of v/VAcc of actives with the subject of Inf.20 However, during the v/V cycle, N0 incorporates into v/V (Baker 1988). This renders it invisible to higher φ-probes, since N0 in [x0 N0 v/VAcc] does not c-command Inf and perhaps loses its status as a syntactic atom (Frampton and Gutmann 1999:11) or as an intervener in XP chains (Chain Uniformity). Therefore, TNom of passives Agrees without any interference from N0 In both actives and passives, Baker-style incorporation satisfies any Case needs N0 might have.21 Thus, the N0 of wager and the indirect object of the DOC limit ECM in the same way, but the former is partly obviated by incorporation.

Ā-movement relates the infinitival subject to v/VAcc in wager ECM in the same way as in ECM+DOC, by [uCase] valuation as a free rider on Ā-Agree. Movement can also account for the licensing of weak pronouns in wager ECM (27a). Bošković (1997:58–59) proposes that weak pronouns incorporate into the verb, as the V-adjacency restrictions of footnote 8 suggest, and thereby satisfy the Case Filter. However, incorporation would be expected to license weak pronouns in active and passive ECM+DOC (16c), (17a) as well. Rather, following Johnson (1991: 621–624), I take footnote 8 to reflect a movement reserved for weak pronouns, as in Mainland Scandinavian object shift. If it targets a position between v/VAcc and N0 the raised pronouns can be reached by the φ-probe of v/VAcc for Case licensing. In ECM+DOC, this will not suffice, because there are still two DPs in need of Case, the indirect object and the infinitive subject, and but one φ-probe.

Among the weak pronouns licensed in wager ECM are expletives (29).

(29)

  • a.

    He alleged there to be stolen documents in the drawer.

  • b.

    There were alleged to be stolen documents in the drawer.

Of the Case of their associates, nothing need be said on partitive Case approaches (Belletti 1988, Lasnik 1992, 2008:28, Bošković 2007:628) or if unmoved DPs are exempt from Case licensing (Sobin 2009:sec. 5.9). Chomsky (2000, 2001) instead takes expletive associates to be nominative or accusative, as is overtly the case in Icelandic (25), with the same definiteness effect and for some speakers with overt expletives as in English (Thráinsson 2007:317–323, 482). On this view, expletive and associate in (29) should be assigned Case as a whole. This may be done without positing expletive-associate chains, by combining the notion of free riders (section 5) with the proposal that Agree unifies features (Frampton and Gutmann 2000, Pesetsky and Torrego 2007). Following Chomsky (2000:128–129, 2001:16–19), in (29) TInf Agrees for [uφ] with stolen documents and there Agrees with TInf for [uperson]. [uCase] of stolen documents Agrees as a free rider at each Agree step. This results in a single [uCase] linked to stolen documents, there, and TInf. Agree between v/VAcc and there in Spec,TInf values this single [uCase] to accusative at all three occurrences.

Stepping back from the details, the guiding intuition is that of Pesetsky 1991: wager has a structure richer than believe in such a way that a Case problem arises and is obviated by Ā-movement, as when ECM is enriched to ECM+DOC. Intervention by N0 in (28) achieves this, while cyclic incorporation limits the problem to the v/V cycle.22

Notes

I am grateful to two LI readers, for saving me from many a blunder and for seeking a clarity that though it has passed my skill to meet has yet led to much improvement, and to Anikó Lipták for generous help with Hungarian. Remaining blunders are mine.

1 The distribution of PRO may but need not be construed Case-theoretically (Chomsky and Lasnik 1995, Landau 2004).

2 See Chomsky 1995:306 for a related argument for Case from seem and an alternative explanation.

3 I follow Chomsky (2000, 2001) in describing the associate as nominative/accusative, as is overtly the case in Icelandic (25); see the appendix.

4 In active ECM, the intervention of an adverb/a PP/a particle at α in VECM-Acc- _____α-[InftAcc . . . ] is subject to variation, unlike in simple transitives where such intervention is always acceptable in V-Acc- _____α. Adverbs and PPs are accepted at α by some speakers but not others (Bowers 1993:632, Koizumi 1995:34 vs. Johnson 1991:630). Particles are acceptable, unless the ECM subject is there, in which case speakers’ judgments vary (Johnson 1991:608n22); this variation may be related to restrictions on there-adverb-T orders discussed by Cardinaletti (1997:sec. 4.2).

5 (9) and (10) illustrate distinct arguments. (9) requires post-particle subjects to be located in the infinitive while pre-particle ones need not be (but may); (10) requires pre-particle subjects to have raised out of the infinitive while post-particle ones need not have (but may). Only the argument from (9) makes use of argument-adjunct interaction, issues with which are discussed by Chomsky (1995:333, 2004:120–122) and might be relatable to a reviewer’s misgivings about Lasnik’s (1999, 2001b, 2008, 2010) judgments for (9a–b). Only (9a–b) are problematic for Johnson’s (1991) proposal that particles may remain at V or move with V to v+V; to explain (9b), the latter movement must be incompatible with object shift, perhaps because particle movement cannot cross the shifted DP.

6 Object shift is required of weak pronouns, which occur in the order V-pronoun(s)-particle, where Condition C shows them to be in the matrix clause (Lasnik 1999:200–201), but not in the order V-particle-pronoun(s), owing to a constraint of adjacency to the verb that is often viewed as prosodic (Ross 1967:sec. 3.1.1.3, Williams 1974:69–70, Zwicky 1986, Dowty 1995).

7 There is a principled reason for this restriction if [EF] projects a position but needs to rely on a probe like [uφ] to find its filler (Chomsky 2000:122, 2004:114; so also Lasnik 2001a, but not Chomsky 2008). Spec,T has been argued not to be restricted in a similar way (Collins 1997, Holmberg 2000, Bailyn 2004, Nevins 2005; but contrast Postal 2004, Williams 2006, Bruening 2010b, Slioussar 2011). If this suggestion—that [EF] projects a position but needs to rely on a probe like [uφ] to find its filler—is correct, it seems to relate to the Extended Projection Principle, which requires Spec,T to be occupied by something, including by expletives that do not appear in Spec,v (Chomsky 2000:109 (discussing (23b)), 102–105, 2004:126n45, 2008:157 and notes thereto).

8 The structure of (13b) may be sketched by combining Collins’s (1997) analyses for locative and quotative inversion with Johnson’s (1991) proposal that particles attach to V but may either raise with it or remain stranded. The order Vparticle-Nom conforms to (i), where Nom originates below the lowest position of the particle, provided it cannot shift to the pre-particle position as discussed for (12) (see Hoekstra 2004:379 for an alternative where the PP passes through the pre-particle position). It contrasts with the quotative inversion V-Nom-particle order in (iii), which conforms to (ii): in quotative inversion, the verb raises to T, but the particle may either raise with it to give V-particle-Nom, or remain stranded to give V-Nom-particle. Nom binding into the PP in (13b) does not indicate object shift (Lasnik 2008:28): see Pesetsky 1995:172–180, 228ff., Phillips 1996:44–48. (i) assumes that locative inversion uses unaccusative structure; for locative inversion with unergatives, see Culicover and Levine 2001, Doggett 2004, and Mendikoetxea 2006, where it is analyzed as topicalization + heavy NP shift, rightward Spec,v, and unaccusatives, respectively.

(i) [CP PP [TPtPP T[vP v(+particle)v [VP V(+particle) NOMtPP]]]]

(ii) [CP quoteAcc [TPtquote called(+out)T [vPMaiaNom v(+out) [VP V(+out) tquote to Tom]]]]

(iii) ‘‘Did you see him?’’ called (out) Maia (out) to Tom.

9 The symbol % indicates speaker variation.

10 I keep to grammars that allow DOC passives on the indirect but not direct object: for example, *The proposition was shown the readers (this is the grammar reflected in most of the generative literature; Postal 2004:241). In grammars with direct object passives, the direct object does participate in matrix Agree/Case relations across the goal. Various theoretical accounts of the difference between these two grammar types are available, such as inherent Case on the indirect object (Anagnostopoulou 2003:secs. 2.4, 3.2, 7.2.4). Whether such grammars should allow ECM+DOC None of the propositions were shown us to be provable depends on how the DOC blocks ECM (section 5). Grammars with indirect object–only passives show some exceptions to the particle/adverb placement in (15) (Stowell 1981:sec. 5.5, Hudson 1992:259, Farrell 2005:sec. 4.5). Some exceptions have independent explanations (Johnson 1991:625, 630); others are more difficult, notably object shift of weak pronoun themes, as in Sam handed her ?them/*the tools down (Johnson 1991:622–623). This gives rise to theories of the English DOC where both goal and theme participate in matrix Agree/ Case relations even if only the goal promotes in the passive. Some of these theories predict the impossibility of ECM+DOC along lines compatible with the present remark, including Johnson’s (1991), where the goal and theme participate in Agree/Case relations as a constituent. Others do not, including Koizumi’s (1995:sec. 2.4), where both objects obtain Case/casefrom dedicatedclausal Agrheads. To bar ECM+DOC, these theories might use the phase-theoretic identification of the ECM+DOC problem mentioned in footnote 13.

11 For show, ungrammatical ECM+DOC examples contrast with grammatical plain DOC and grammatical plain ECM examples, such as The propositions were shown to be consistent, They were shown the propositions. Similar verbs include guarantee and grant, which participate in wager-type ECM (discussed in the appendix), and assure, persuade, and convince, which take ECM infinitives only with the DOC, and conversely the DOC only with non-DP themes. DOCs may differ on properties other than the structural Case of the indirect object that matters here: for example, applicative for guarantee, causative for show (Pesetsky 1991:sec. 4.1.1.5).

12 The c-command relation IO > Inf corresponds to IO(DP) > DO in the DOC (Larson 1991:122n11), but not to DP > IO(to-PP) in the POC. DP/PP-internal arguments seem to c-command clausal ones generally: in (i), the PP c-commands into the CP for quantifier binding, also for Condition C (cf. Boeckx, Hornstein, and Nunes 2010:174n33) or negative polarity item licensing (cf. Pesetsky 1991:137, (533)). This seems independent of linear order; see, for example, (ii)—or (iii), where reconstruction gives results comparable to those with seem (on which, see Bošković 2002).

(i) The mayork {vowed/made a promise} to every soldieri PROk {to pay/that shek will pay} hisi parking.

(ii) ?The project was shown to meet itsi/k needs to every group presenti.

(iii)

  • a.

    ?[Itsi project]j was shown (seemed) to everyi group [tj to be superfluous].

  • b.

    ?[Each otheri’s aims]j were shown (seemed) to themi [tj to be compatible].

13 Agree with defective goals leaves the φ-probe unvalued, which is not relevant here. On multiple-Agree approaches, Agree is not restricted to a single/closest goal (Hiraiwa 2005). The DOC might then prevent infinitive subject licensing by trapping the subject below a phase boundary—for instance, if Appl0 of the DOC is a phase head, or if it interrupts Agree with C/TInf needed for Agree into the infinitive (cf. Pesetsky and Torrego 2001, Rackowski and Richards 2005; on Appl0 below vs. above V0, see Bruening 2010a). The relationship of this blocking effect to the nature of the indirect object as DP/PP would need further exploration. I am grateful to a reviewer for raising a phase-theoretic analysis.

14 This correctly allows accusative-assigning passives and impersonals, which have been argued to have at least quasi-expletive external arguments (Maling and Sigurjónsdóttir 2002, Szucsich 2007, Schäfer 2008; see Chomsky 2004: 126n37 on expletive external arguments). Pure unaccusatives do not assign accusative in (22) and (23), fitting a version of Burzio’s Generalization where the Case-assigning potential of v/V depends on external argument selection (Chomsky 1995:306, 316–317, 389n90).

15Chomsky (2000:110) proposes that movement through Spec,v counts as an A-position if φ-Agree occurs, and as an Ā-position otherwise. In English It was himiwhoiheitiwanted Peter to invite ti, this makes Spec,v an Ā-position for t'i, allowing it to corefer with hei, in contrast to A-movement in Heiseems to him*itito have been invited ti, where Condition B rules out coreference between himi and t'i in an A-position. In Hungarian focus movement, upstairs Spec,v should be an A-position and coreference should be disallowed if and only if the fronted argument Agrees in the upstairs clause. Lipták (1998) argues that this is so. This provides independent evidence that upstairs case/agreement in Hungarian focus movement is due to syntax rather than being established outside syntax, at PF (see section 6), since it affects interpretation (coreference).

16 Other interactions of Ā-movement and Case licensing call for comment. Ā-movement does not usually lead to undesirable DP-licensing, because of independent factors: for instance, DP complements of N/A require of by virtue of selection, and control infinitives require PRO by virtue of properties of their C/T (see Chomsky and Lasnik 1995, esp. pp. 107–108, 114–115). Whether any step of Ā-movement beyond the first should license Case (Bošković 1997:128–129) depends on whether and when spell-out of the lowest link with [uCase] causes the derivation to crash (see Chomsky 1995:300–301, Gärtner 2002). Whether TNom should license Case under Ā-movement past it depends on whether it Agrees with the Ā-moving DP: in Chomsky 2000, 2001 it does not, unlike vAcc, while in Chomsky 2008 C/TNom and v/VAcc behave symmetrically. Empirically, this may not be testable in English (see Bošković 1997:128–129, discussing Pesetsky 1991:sec. 4.1.1.5), while in Hungarian (23d) allows optional agreement with the subject, lennének ‘be.SUBJ.3PL’, that may be pertinent (Lipták 1998:154n1). Finally, where φ-and Ā-Agree are both possible as in ECM+DOC (i), φ-Agree must occur with the closest DP, giving (ia) rather than (ib), where Agree occurs with the farther DP while the higher one is Case-licensed by Ā-movement; this suggests an economy or locality condition that prefers the closer DP as the goal of the φ-probe ((ib) is not reducible to the weaker and variable degradation that the Oblique Trace Filter causes, (iib), on which see Stowell 1981:secs. 5.3.2.3, 6.4.2, Hudson 1992:258).

(i)

  • a.

    (?)What did Roger try to show his students _____ to be a liquid?

  • b.

    *Who did Roger try to show _____ glass to be a liquid?

(ii)

17 Candidates are Icelandic infinitives with quirky case subject + in-situ DP with structural Case. The DP is nominative in contexts where it cannot get infinitive-external Case. Thus, accusative under ECM in (25a) contrasts with nominative when a dative subject blocks ECM, (i), and when the infinitive is a subject, (ii) (Freidin and Sprouse 1991:sec. 6, Sigurðsson 1993:54n4, Jónsson 1996:sec. 4.7.2).

(i)

  • Ég taldi henni hafa verið {%gefnir bílarnir} / {%gefna

  • I believed her.DAT to.have been given.MASC.PL.NOM the.cars.MASC.PL.NOM given.MASC.PL.ACC bílana}.

  • the.cars.MASC.PL.ACC

  • ‘I believed her to have been given the cars.’

  • (Sigurðsson 1993:54n4)

(ii)

  • [Að PRO batna veikin] er venjulegt.

  • PRO.DAT to.recover.from the.disease.NOM is usual ‘To recover from the disease is usual.’ (Freidin and Sprouse 1991:409)

However, nominative here arguably reflects infinitive-internal φ-Agree (Sigurðsson 2000, Rezac 2004:sec. 5.4, Nomura 2005:sec. 3.6) and not default case (Frampton and Gutmann 1999:sec. 3.3.1, Hiraiwa 2005:71), since it controls φ-agreement on participles. In turn, the ungrammaticality of English *Few readers were shown there to beany propositionconsistent with ours would follow if the expletive associate requires but cannot get Case licensing, contrary to (26b).

18 Other possibilities include N0 with(out) φ and N0 (non)equidistant from v/VAcc.

19 It is tempting to link the parameter to Pesetsky’s (1991:sec. 4.1.2) identification of the distinctive characteristic of the wager-class as the selection of a human agent; this might plausibly require a transitive frame, but it is not clear how to ensure that it requires N0 visible to φ-Agree (see the discussion of Basque borrokatu ‘fight’ in the text). As reviewers emphasize, the presence of N0 with wager but not believe cannot be related to paraphrasability (wager-class make the claim, wager,*maintenance CP; believe-class make the discovery CP, have the belief CP). See Pesetsky 1991 for lists of believe-and wager-class ECM verbs, and Postal 1974:sec. 9.3–9.4, 307–309, Pesetsky 1991:sec. 2.7, Bošković 1997:sec. 3.2 for other patterns.

20 Agree with the DP sister of N0 is fine since DP and N0 are equidistant: We claimedthem.

21 If incorporation involves Agree with v/VAcc,[uCase] of N0 is deleted as a free rider.

22 For alternatives, see Pesetsky 1991:sec. 4.1.2 and Bošković 1997:secs. 3.2, 4.4.2.1. The wager-class resembles the French croire ‘believe’ class, which allows ECM only with Ā-movement. Kayne (1984:sec. 5.3), Rizzi (1990:58–59), and Bošković (1997:sec. 3.4) propose that croire complements are opaque to ECM; this is construable as intervention through D/that-like C (on which, see Rezac 2004:sec. 3.5.4, 2005:298).

References

Ackema
,
Peter
, and
Ad
Neeleman
.
2004
.
Beyond morphology: Interface conditions on word formation
.
Oxford
:
Oxford University Press
.
Alexiadou
,
Artemis
, and
Elena
Anagnostopoulou
.
2001
.
The Subject-in-Situ Generalization and the role of Case in driving computations
.
Linguistic Inquiry
32
:
193
231
.
Anagnostopoulou
,
Elena
.
2001
.
Two classes of double object verbs: The role of zero-morphology
. In
Progress in grammar
, ed. by
Marc
van Oostendorp
and
Elena
Anagnostopoulou
.
Amsterdam
:
Dutch Royal Academy Publications
. .
Anagnostopoulou
,
Elena
.
2003
.
The syntax of ditransitives
.
Berlin
:
Mouton de Gruyter
.
Bailyn
,
John Frederick
.
2004
.
Generalized inversion
.
Natural Language and Linguistic Theory
22
:
1
49
.
Baker
,
Mark
.
1988
.
Incorporation
.
Chicago
:
University of Chicago Press
.
Baker
,
Mark
.
1996
.
The polysynthesis parameter
.
Oxford
:
Oxford University Press
.
Baker
,
Mark
.
1997
.
Thematic roles and syntactic structure
. In
Elements of grammar
, ed. by
Liliane
Haegeman
,
73
137
.
Dordrecht
:
Kluwer
.
Beavers
,
John
.
2012
.
Resultative constructions
. In
The Oxford handbook of tense and aspect
, ed. by
Robert I.
Binnick
,
908
936
.
Oxford
:
Oxford University Press
.
Béjar
,
Susana
, and
Diane
Massam
.
1999
.
Multiple Case checking
.
Syntax
2
:
65
79
.
Belletti
,
Adriana
.
1988
.
The Case of unaccusatives
.
Linguistic Inquiry
19
:
1
34
.
Bittner
,
Maria
, and
Ken
Hale
.
1996
.
The structural determination of Case and agreement
.
Linguistic Inquiry
27
:
1
68
.
Bobaljik
,
Jonathan David
.
2008
.
Where’s φ?
In
Φ theory
, ed. by
Daniel
Harbour
,
David
Adger
, and
Susana
Be´jar
,
295
328
.
Oxford
:
Oxford University Press
.
Boeckx
,
Cedric
, and
Norbert
Hornstein
.
2005
.
A gap in the ECM paradigm
.
Linguistic Inquiry
36
:
437
441
.
Boeckx
,
Cedric
,
Norbert
Hornstein
, and
Jairo
Nunes
.
2010
.
Control as movement
.
Cambridge
:
Cambridge University Press
.
Bošković
,
Željko
.
1997
.
The syntax of nonfinite complementation: An economy approach
.
Cambridge, MA
:
MIT Press
.
Bošković
,
Željko
.
2002
.
A-movement and the EPP
.
Syntax
5
:
167
218
.
Bošković
,
Željko
.
2007
.
On the locality and motivation of Move and Agree
.
Linguistic Inquiry
38
:
589
644
.
Bowers
,
John
.
1993
.
The syntax of predication
.
Linguistic Inquiry
24
:
591
656
.
Bowers
,
John
.
2002
.
Transitivity
.
Linguistic Inquiry
33
:
183
224
.
Bruening
,
Benjamin
.
2001
.
Syntax at the edge: Cross-clausal phenomena and the syntax of Passamaquoddy
.
Doctoral dissertation, MIT, Cambridge, MA
.
Bruening
,
Benjamin
.
2010a
.
Ditransitive asymmetries and a theory of idiom formation
.
Linguistic Inquiry
41
:
519
562
.
Bruening
,
Benjamin
.
2010b
.
Language-particular syntactic rules and constraints: English locative inversion and do-support
.
Language
86
:
43
84
.
Burzio
,
Luigi
.
1986
.
Italian syntax: A Government-Binding account
.
Dordrecht
:
Reidel
.
Cardinaletti
,
Anna
.
1997
.
Subjects and clause structure
. In
The new comparative syntax
, ed. by
Liliane
Haegeman
,
33
63
.
London
:
Longman
.
Chomsky
,
Noam
.
1981
.
Lectures on government and binding
.
Dordrecht
:
Foris
.
Chomsky
,
Noam
.
1986
.
Knowledge of language
.
New York
:
Praeger
.
Chomsky
,
Noam
.
1995
.
The Minimalist Program
.
Cambridge, MA
:
MIT Press
.
Chomsky
,
Noam
.
2000
.
Minimalist inquiries: The framework
. In
Step by step
, ed. by
Roger
Martin
,
David
Michaels
, and
Juan
Uriagereka
,
89
156
.
Cambridge, MA
:
MIT Press
.
Chomsky
,
Noam
.
2001
.
Derivation by phase
. In
Ken Hale: A life in language
, ed. by
Michael
Kenstowicz
,
1
52
.
Cambridge, MA
:
MIT Press
.
Chomsky
,
Noam
.
2004
.
Beyond explanatory adequacy
. In
Structures and beyond
, ed. by
Adriana
Belletti
,
104
131
.
Oxford
:
Oxford University Press
.
Chomsky
,
Noam
.
2008
.
On phases
. In
Foundational issues in linguistic theory
, ed. by
Robert
Freidin
,
Carlos P.
Otero
, and
Maria Luisa
Zubizarreta
,
133
166
.
Cambridge, MA
:
MIT Press
.
Chomsky
,
Noam
, and
Howard
Lasnik
.
1995
.
The theory of principles and parameters
. In
Chomsky 1995
,
13
127
.
Chung
,
Sandra
.
1994
.
Wh-agreement and ‘‘referentiality’’ in Chamorro
.
Linguistic Inquiry
25
:
1
44
.
Cole
,
Peter
, and
Gabriella
Hermon
.
2000
.
Partial wh-movement: Evidence from Malay
. In
Wh-scope marking
, ed. by
Uli
Lutz
,
Gereon
Mu¨ller
, and
Arnim
von Stechow
,
101
130
.
Amsterdam
:
John Benjamins
.
Collins
,
Chris
.
1997
.
Local economy
.
Cambridge, MA
:
MIT Press
.
Coppock
,
Elizabeth
.
2004
.
Object agreement in Hungarian
.
Ms., Stanford University, Stanford, CA. Available at http://eecoppock.info/object-agreement-shortened.pdf
.
Culicover
,
Peter
, and
Robert D.
Levine
.
2001
.
Stylistic inversion in English: A reconsideration
.
Natural Language and Linguistic Theory
19
:
283
310
.
Deal
,
Amy Rose
.
2010
.
Ergative case and the transitive subject: A view from Nez Perce
.
Natural Language and Linguistic Theory
28
:
73
120
.
Dikken
,
Marcel den
.
1990
.
Particles and the dative alternation
. In
Leiden Conference for Junior Linguists: 11–14 December 1990, University of Leiden
, ed. by
John
van Lit
,
René
Mulder
, and
Rint
Sybesma
,
71
86
.
Leiden
:
Rijksuniversiteit Leiden
.
Dikken
,
Marcel den
.
2006
.
When Hungarians agree (to disagree): The fine art of ‘‘φ’’ and ‘‘art.’’
Ms., CUNY Graduate Center. Available at http://www.gc.cuny.edu/CUNY_GC/media/CUNY-Graduate-Center/PDF/Programs/Linguistics/Dikken/hungarian_agreement_2006.pdf
.
Dikken
,
Marcel den
.
2009
.
On the nature and distribution of successive cyclicity
. .
Doggett
,
Teal Bissell
.
2004
.
All things being unequal: Locality in A-movement
.
Doctoral dissertation, MIT, Cambridge, MA
.
Dowty
,
David
.
1995
.
Toward a Minimalist theory of syntactic structure
. In
Discontinuous constituency
, ed. by
Harry
Bunt
and
Arthur
van Horck
,
11
62
.
Berlin
:
Mouton de Gruyter
.
Farrell
,
Patrick
.
2005
.
English verb-preposition constructions: Constituency and order
.
Language
81
:
96
137
.
von Fintel
,
Kai
, and
Sabine
Iatridou
.
2003
.
Epistemic containment
.
Linguistic Inquiry
34
:
173
198
.
Fox
,
Danny
.
1999
.
Reconstruction, binding theory, and the interpretation of chains
.
Linguistic Inquiry
30
:
157
196
.
Frampton
,
John
, and
Sam
Gutmann
.
1999
.
Cyclic computation
.
Syntax
2
:
1
27
.
Frampton
,
John
, and
Sam
Gutmann
.
2000
.
Agreement is feature sharing
.
Ms., Northeastern University, Boston, MA. Available at http://www.math.neu.edu/ling/pdffiles/agrisfs.pdf
.
Freidin
,
Robert
, and
Rex A.
Sprouse
.
1991
.
Lexical case phenomena
. In
Principles and parameters in comparative grammar
, ed. by
Robert
Freidin
,
392
416
.
Cambridge, MA
:
MIT Press
.
Gärtner
,
Hans-Martin
.
2002
.
Generalized transformations and beyond
.
Berlin
:
Akademie Verlag
.
Gervain
,
Judit
.
2002
.
Linguistic methodology and microvariation in language: The case of operator-raising in Hungarian
.
Master’s thesis, University of Szeged, Hungary
.
Grimshaw
,
Jane
.
1979
.
Complement selection and the lexicon
.
Linguistic Inquiry
10
:
279
326
.
Hiraiwa
,
Ken
.
2005
.
Dimensions of symmetry in syntax
.
Doctoral dissertation, MIT, Cambridge, MA
.
Hoekstra
,
Teun
.
2004
.
Arguments and structure
.
Berlin
:
Mouton de Gruyter
.
Holmberg
,
Anders
.
2000
.
Scandinavian stylistic fronting
.
Linguistic Inquiry
31
:
445
483
.
Hudson
,
Richard
.
1992
.
So-called ‘‘double objects’’ and grammatical relations
.
Language
68
:
251
276
.
Johnson
,
Kyle
.
1991
.
Object positions
.
Natural Language and Linguistic Theory
9
:
577
636
.
Jo´nsson
,
Johannes Gı´sli
.
1996
.
Clausal architecture and case in Icelandic
.
Doctoral dissertation, University of Massachusetts, Amherst
.
Kayne
,
Richard
.
1984
.
Connectedness and binary branching
.
Dordrecht
:
Foris
.
É. Kiss
,
Katalin
.
1987
.
Configurationality in Hungarian
.
Dordrecht
:
Reidel
.
Koizumi
,
Masatoshi
.
1995
.
Phrase structure in the Minimalist Program
.
Doctoral dissertation, University of Connecticut, Storrs
.
Laka
,
Itziar
.
2000
.
Thetablind Case: Burzio’s Generalization and its image in the mirror
. In
Arguments and Case
, ed. by
Eric
Reuland
,
103
129
.
Amsterdam
:
John Benjamins
.
Landau
,
Idan
.
2004
.
The scale of finiteness and the calculus of control
.
Natural Language and Linguistic Theory
22
:
811
877
.
Larson
,
Richard
.
1988
.
On the double object construction
.
Linguistic Inquiry
19
:
335
392
.
Larson
,
Richard
.
1991
.
Promise and the theory of control
.
Linguistic Inquiry
22
:
103
139
.
Lasnik
,
Howard
.
1992
.
Case and expletives revisited
.
Linguistic Inquiry
23
:
381
405
.
Lasnik
,
Howard
.
1999
.
Chains of arguments
. In
Working Minimalism
, ed. by
Samuel David
Epstein
and
Norbert
Hornstein
,
189
215
.
Cambridge, MA
:
MIT Press
.
Reprinted in Minimalist investigations of linguistic theory, 139–157. London: Routledge (2003)
.
Lasnik
,
Howard
.
2001a
.
A note on the EPP
.
Linguistic Inquiry
32
:
356
362
.
Lasnik
,
Howard
.
2001b
.
Subjects, objects, and the EPP
. In
Subjects, objects, and the EPP: Grammatical functions, functional categories, and configurationality
, ed. by
William
Davies
and
Stanley
Dubinsky
,
103
121
.
Dordrecht
:
Kluwer
.
Lasnik
,
Howard
.
2008
.
On the development of Case theory: Triumphs and challenges
. In
Foundational issues in linguistic theory
, ed. by
Robert
Freidin
,
Carlos
Otero
, and
Maria Luisa
Zubizarreta
,
17
41
.
Cambridge, MA
:
MIT Press
.
Lasnik
,
Howard
.
2010
.
A (surprising?) consequence of single-cycle syntax
.
Linguistic Sciences
9
:
22
33
.
Lasnik
,
Howard
, and
Nicholas
Sobin
.
2000
.
The who/whom puzzle: On the preservation of an archaic feature
.
Natural Language and Linguistic Theory
18
:
343
371
.
Legate
,
Julie Anne
.
2003
.
Some interface properties of the phase
.
Linguistic Inquiry
34
:
505
516
.
Lipták
,
Anikó
.
1998
.
Focus and ECM
. In
Proceedings of CONSOLE VI
, ed. by
Tina
Cambier-Langeveld
,
Anikó
Lipták
, and
Michael
Redford
,
153
168
.
Leiden
:
HIL
.
Maling
,
Joan
, and
SigríðLur
Sigurjónsdóttir
.
2002
.
The ‘‘new impersonal’’ construction in Icelandic
.
Journal of Comparative Germanic Linguistics
5
:
97
142
.
Marantz
,
Alec
.
2000
.
Case and licensing
. In
Arguments and Case
, ed. by
Eric
Reuland
,
11
30
.
Amsterdam
:
John Benjamins
.
McCreight
,
Katherine Young
.
1988
.
Multiple case assignment
.
Doctoral dissertation, MIT, Cambridge, MA
.
McFadden
,
Thomas
.
2004
.
The position of morphological case in the derivation: A study on the syntaxmorphology interface
.
Doctoral dissertation, University of Pennsylvania, Philadelphia
.
Mendikoetxea
,
Amaya
.
2006
.
Unergatives that ‘‘become’’ unaccusatives in English locative inversion: A lexical-syntactic approach
. In
Points de vue sur l’inversion
, ed. by
Christine
Copy
and
Loucie
Gournay
,
133
155
.
Cahiers de Recherche en Grammaire Anglaise de l’Énonciation 9
.
Paris
:
Orphys
.
Nevins
,
Andrew
.
2005
.
Derivations without the Activity Condition
. In
Proceedings of the EPP/Phase Workshop
, ed. by
Martha
McGinnis
and
Norvin
Richards
,
283
306
.
MIT Working Papers in Linguistics 49
.
Cambridge, MA
:
MIT, MIT Working Papers in Linguistics
.
Nomura
,
Masashi
.
2005
.
Nominative Case and AGREE(ment)
.
Doctoral dissertation, University of Connecticut, Storrs
.
Ormazabal
,
Javier
, and
Juan
Romero
.
2002
.
Agreement restrictions
.
Ms., University of the Basque Country and University of Alcalá de Henares
.
Ormazabal
,
Javier
, and
Juan
Romero
.
2010
.
The derivation of dative alternations
. In
Argument structure and syntactic relations
, ed. by
Maia
Duguine
,
Susana
Huidobro
, and
Nerea
Madariaga
,
203
223
.
Amsterdam
:
John Benjamins
.
Oyharçabal
,
Beñat
.
1992
.
Structural Case and inherent Case marking: Ergaccusativity in Basque
. In
Syntactic theory and Basque syntax
, ed. by
Joseba A.
Lakarra
and
Jon Ortiz
de Urbina
,
309
342
.
Donostia
:
Gipuzkoako Diputazioa
.
Pesetsky
,
David
.
1982
.
Paths and categories
.
Doctoral dissertation, MIT, Cambridge, MA
.
Pesetsky
,
David
.
1991
.
Zero Syntax. Vol. 2, Infinitives
. .
Pesetsky
,
David
.
1995
.
Zero Syntax: Experiencers and cascades
.
Cambridge, MA
:
MIT Press
.
Pesetsky
,
David
, and
Esther
Torrego
.
2001
.
T-to-C movement: Causes and consequences
. In
Ken Hale: A life in language
, ed. by
Michael
Kenstowicz
,
355
426
.
Cambridge, MA
:
MIT Press
.
Pesetsky
,
David
, and
Esther
Torrego
.
2004
.
Tense, case, and the nature of syntactic categories
. In
The syntax of time
, ed. by
Jacqueline
Gue´ron
and
Jacqueline
Lecarme
,
495
537
.
Cambridge, MA
:
MIT Press
.
Pesetsky
,
David
, and
Esther
Torrego
.
2007
.
The syntax of valuation and the interpretability of features
. In
Phrasal and clausal architecture
, ed. by
Simin
Karimi
,
Vida
Samiian
, and
Wendy K.
Wilkins
,
262
294
.
Amsterdam
:
John Benjamins
.
Phillips
,
Colin
.
1996
.
Order and structure
.
Doctoral dissertation, MIT, Cambridge, MA
.
Postal
,
Paul M
.
1974
.
On raising
.
Cambridge, MA
:
MIT Press
.
Postal
,
Paul M
.
1993
.
Some defective paradigms
.
Linguistic Inquiry
24
:
347
364
.
Postal
,
Paul M
.
2004
.
Sceptical linguistic essays
.
Oxford
:
Oxford University Press
.
Rackowski
,
Andrea
, and
Norvin
Richards
.
2005
.
Phase edge and extraction: A Tagalog case study
.
Linguistic Inquiry
36
:
565
599
.
Rezac
,
Milan
.
2004
.
Elements of cyclic syntax
.
Doctoral dissertation, University of Toronto
.
Rezac
,
Milan
.
2005
.
On tough movement
. In
Minimalist essays
, ed. by
Cedric
Boeckx
,
288
325
.
Amsterdam
:
John Benjamins
.
Rezac
,
Milan
.
2006
.
The interaction of Th/Ex and locative inversion
.
Linguistic Inquiry
37
:
685
697
.
Rezac
,
Milan
.
2010
.
φ-Agree vs. φ-feature movement: Evidence from floating quantifiers
.
Linguistic Inquiry
41
:
496
508
.
Rezac
,
Milan
.
2011
.
φ-Features and the modular architecture of language
.
Dordrecht
:
Springer
.
Rizzi
,
Luigi
.
1990
.
Relativized Minimality
.
Cambridge, MA
:
MIT Press
.
Ross
,
John Robert
.
1967
.
Constraints on variables in syntax
.
Doctoral dissertation, MIT, Cambridge, MA
.
Rothstein
,
Susan
.
1992
.
Case and NP licensing
.
Natural Language and Linguistic Theory
10
:
119
139
.
Sauerland
,
Uli
.
2003
.
Intermediate adjunction with A-movement
.
Linguistic Inquiry
34
:
308
314
.
Schäfer
,
Florian
.
2008
.
The syntax of (anti-)causatives
.
Amsterdam
:
John Benjamins
.
Schütze
,
Carson T
.
1993
.
Towards a Minimalist account of quirky Case and licensing in Icelandic
. In
Papers on case and agreement II
, ed. by
Colin
Phillips
,
321
375
.
MIT Working Papers in Linguistics 19
.
Cambridge, MA
:
MIT, MIT Working Papers in Linguistics
.
Schütze
,
Carson T
.
2001
.
On the nature of default case
.
Syntax
2
:
205
238
.
Sigurðsson
,
Halldór Ármann
.
1991
.
Icelandic case-marked PRO and the licensing of lexical arguments
.
Natural Language and Linguistic Theory
9
:
327
363
.
Sigurðsson
,
Halldór Ármann
.
1993
.
Agreement as head visible feature government
.
Studia Linguistica
47
:
32
56
.
Sigurðsson
,
Halldór Ármann
.
2000
.
The locus of case and agreement
.
Working Papers in Scandinavian Syntax
65
:
65
108
.
Sigurðsson
,
Halldór Ármann
.
2009
.
The No Case Generalization
. In
Advances in comparative Germanic syntax
, ed. by
Artemis
Alexiadou
,
Jorge
Hankamer
,
Thomas
McFadden
,
Justin
Nuger
, and
Florian
Scha¨fer
,
249
278
.
Amsterdam
:
John Benjamins
.
Slioussar
,
Natalia
.
2011
.
Russian and the EPP requirement in the Tense domain
.
Lingua
121
:
2048
2068
.
Sobin
,
Nicholas
.
2009
.
Prestige case forms and the Comp-trace effect
.
Syntax
12
:
32
59
.
Stowell
,
Tim
.
1981
.
Origins of phrase structure
.
Doctoral dissertation, MIT, Cambridge, MA
.
Szucsich
,
Luka
.
2007
.
Nothing wrong with finite T: Non-agreeing accusative impersonal sentences
. In
Proceedings of Formal Approaches to Slavic Linguistics 15
, ed. by
Richard
Compton
,
Magdalena
Goledzinowska
, and
Ulyano
Savchenko
,
401
419
.
Ann Arbor
:
Michigan Slavic Publications
.
Taraldsen
,
Knut Tarald
.
1981
.
Case-conflict in Norwegian topicalization
. In
Proceedings of NELS 11
, ed. by
Victoria
Burke
and
James
Pustejovsky
,
377
398
.
Amherst
:
University of Massachusetts, Graduate Linguistic Student Association
.
Thráinsson
,
Hökuldur
.
2007
.
The syntax of Icelandic
.
Cambridge
:
Cambridge University Press
.
Ura
,
Hiroyuki
.
1993
.
On feature-checking for wh-traces
. In
Papers on case and agreement I
, ed. by
Jonathan David
Bobaljik
and
Colin
Phillips
,
215
242
.
MIT Working Papers in Linguistics 18
.
Cambridge, MA
:
MIT, MIT Working Papers in Linguistics
.
Williams
,
Edwin
.
1974
.
Rule ordering in syntax
.
Doctoral dissertation, MIT, Cambridge, MA
.
Williams
,
Edwin
.
2006
.
Subjects of different heights
. In
Proceedings of Formal Approaches to Slavic Linguistics 14
, ed. by
Hana
Filip
,
Steven
Franks
,
James E.
Lavine
, and
Mila
Tasseva-Kurktchieve
,
409
418
.
Ann Arbor
:
Michigan Slavic Publications
.
Zwicky
,
Arnold M
.
1986
.
The Unaccented Pronoun Constraint in English
. In
Ohio State University working papers in linguistics 32
, ed. by
Arnold
Zwicky
,
100
113
.
Columbus
:
Ohio State University, Department of Linguistics
.