Abstract

Using data from agreement in three Algonquian languages (Ojibwe, Cheyenne, and Plains Cree), this squib shows that effects typically attributed to Chomsky’s (2000, 2001) Activity Condition (AC) can vary not only across languages, as in Baker’s (2008b) macroparametric proposal, but within a language as well. AC effects are thus another instance in which an apparent macroparameter turns out, on closer inspection, to be a microparameter instead, as in prominent cases such as the pro-drop parameter and the polysynthesis parameter (Kayne 2005, Baker 2008a).

1 Introduction

The AC restricts Agree to goals that bear an unvalued feature, typically a Case feature. Once the Case feature of a goal G has been valued by Agree, G is inactive and cannot be targeted by subsequent Agree operations. Chomsky appears to treat the AC as an invariant principle of the faculty of language, but Baker (2008b, 2013) has shown that not all languages display AC effects. Some languages, such as Hindi, Icelandic, and Turkish, are consistent with the AC in that multiple probes cannot agree with a single goal, but other languages, such as Kinande, Georgian, Burushaski, and Jarawara, do allow such multiple agreement.

In view of this variation, Baker (2008b:155–156) proposes that AC effects reflect a parameter, which he argues is a macroparameter that “governs languages as wholes, not individual heads or constructions in a language” (Baker 2013:28). Baker acknowledges that the existence of such macroparameters is controversial, as much current work attributes parametric differences to the features of particular lexical items (Borer 1984, Chomsky 1995). Nevertheless, the distribution of AC effects in the languages in Baker’s sample is consistent with the macroparametric conception of the AC.

Baker’s sample aside, there is no a priori reason why AC variation should be strictly macroparametric. If AC effects can indeed vary, as Baker establishes, what would prevent such variation from occurring within a language? In this squib, I argue that an agreement alternation in Plains Cree involves exactly such language-internal variation in the AC.

The squib proceeds as follows. Section 2 shows that agreement in the Algonquian languages involves probes on T and C, which are both able to target either the subject or the object. This configuration provides a simple test for the applicability of the AC: in a clause that contains only a single suitable goal, does T-agreement with that goal prevent subsequent C-agreement with the same goal? Section 3 shows that the answer varies across the Algonquian family: C can agree with the same goal as T in Ojibwe but not in Cheyenne. This difference is compatible with Baker’s macroparametric proposal: Ojibwe and Cheyenne would have opposite settings for the relevant macroparameter. In Plains Cree, however, the story is not as simple. Section 4 shows that Plains Cree displays both the Ojibwe pattern and the Cheyenne pattern: in present tense clauses, C can agree with the same goal as T (as in Ojibwe), but in past tense clauses, it cannot (as in Cheyenne). This variation cannot be attributed to a structural difference between present and past clauses, as the agreement patterns in the two tenses are otherwise identical. The only difference between the two tenses is that the goal of TPres is accessible to subsequent Agree operations while the goal of TPast is not. In Plains Cree, then, AC effects are microparametric: contra Baker’s proposal, the applicability of the AC is a property of individual lexical items, not of the language as a whole.

2 Background: Algonquian Agreement

The Algonquian agreement template is illustrated by the Ojibwe verb form in (1) (Nichols 1980). The verb stem is accompanied by several inflectional affixes, including two agreement markers (ni- . . . -naa ‘1pl’, -ag ‘3pl’) whose patterning and analysis will be discussed below. The order of suffixes, v–Neg–T–Mod–C, is consistent with a Mirror Principle analysis in which distance from the stem correlates with height in the syntactic structure.1

(1)

  • niwaabamigosiinaadogenag2

  • (T) V v Neg T Mod C

  • ni- waabam -igw -sii -naa -dogen -ag

  • 1- see -INV -NEG-1PL -DUB -3PL

  • ‘Maybe they don’t see us.’

The transitive forms in (2), which both involve 1pl and 3pl arguments, exemplify the patterning of the agreement markers. In both forms, the 1pl argument is indexed by a prefix-suffix combination that I will refer to as T-agreement (ni- . . . -naan ‘1pl’), while the 3pl argument is indexed by a further suffix that I will refer to as C-agreement (-ag ‘3pl’).

(2)

  • a.

    niwaabamaanaanig

    ni-waabam-aa -naan-ag

    1- see -DIR-1PL -3PL

    We see them.’ (1→3)

  • b.

    niwaabamigonaanig

    ni-waabam-igw-naan-ag

    1- see -INV-1PL -3PL

    They see us.’ (3→1)

The argument presented in this squib is valid only if C-agreement does indeed originate in a higher structural position than T-agreement. The analysis of these markers as T and C follows existing work on Algonquian languages (Bruening 2005, Hamilton 2013, and Oxford 2014 for T/Infl; Branigan and MacKenzie 1999 and Bliss 2013 for C; Halle and Marantz 1993 for both) and is consistent with the positions of the agreement suffixes relative to v, Neg, and Mod in the template in (1). As further support, consider that the T-agreement suffix is overtly marked for tense in Plains Cree (see section 4) and the C-agreement suffix can be replaced by a subjunctive or iterative complementizer suffix in embedded clauses (Ojibwe -en ‘whether’, Plains Cree -i ‘whenever’). As for the discontinuous spell-out of T-agreement as a prefix-suffix combination, this can be attributed to fission of the person features of T, as in Harbour’s (2008) approach to person prefixes across languages.

The forms in (2) demonstrate a key property of Algonquian agreement: the T and C agreement slots are not tied to particular grammatical functions. T agrees for person and number with whichever argument ranks highest on the person hierarchy SAP (speech act participant) > 3 > 3′ (Macaulay 2009), where 3 denotes a more topical proximate third person and 3′ denotes a less topical obviative third person. C agrees for gender, number, and obviation with third persons only.3 The 1pl argument in (2) is thus consistently indexed by T-agreement regardless of whether it functions as the subject, as in (2a), or the object, as in (2b). The agreement preferences of T and C can be formalized by relativizing the probes to particular feature specifications: T carries a fully articulated person probe that is best matched by an SAP but can also be satisfied by a third person when no SAP is present (Béjar and Rezac 2009, Coon and Bale 2013), while C probes for a feature that is specified only on third persons, such as the [±proximate] feature that distinguishes proximate and obviative third persons (Lochbihler 2012, Oxford 2014).

Table 1 gives additional examples of the T/C-agreement pattern in Ojibwe, along with the parallel forms in Cheyenne (Leman 1980). Although the shapes of the morphemes differ greatly in the two languages, the structure of the forms is the same: T-agreement is a prefix-suffix combination that indexes the argument that ranks highest on the person hierarchy, while C-agreement is a further suffix that indexes a third person.4

Table 1

Transitive T- and C-agreement in Ojibwe and Cheyenne

SUBJOBJOjibweCheyenne
1pl3pl ni- . . .-naan -ag ná- . . .-ne -vo -he 
 1 1PL 3PL 1 1PL 3PL INT 
2pl3pl gi- . . .-waa -ag né- . . .-vo -vo  -he 
 2 2PL 3PL 2 2PL 3PL INT 
3pl3′ o- . . .-waa  -an é- . . .-vo -vo  -he 
 3 3PL 3′ 3 3PL 3′ INT 
SUBJOBJOjibweCheyenne
1pl3pl ni- . . .-naan -ag ná- . . .-ne -vo -he 
 1 1PL 3PL 1 1PL 3PL INT 
2pl3pl gi- . . .-waa -ag né- . . .-vo -vo  -he 
 2 2PL 3PL 2 2PL 3PL INT 
3pl3′ o- . . .-waa  -an é- . . .-vo -vo  -he 
 3 3PL 3′ 3 3PL 3′ INT 

3 Macroparametric Variation: Ojibwe vs. Cheyenne

The above transitive forms tell us little about the role of the AC in Algonquian agreement because T and C are always able to target separate goals in these forms. However, in intransitive forms, where only one goal is available, the AC plays a crucial role: does T-agreement with the lone goal prevent subsequent C-agreement with that goal? This section shows that the answer varies within the Algonquian family: T-agreement prevents subsequent C-agreement in Cheyenne but not in Ojibwe. This variation is consistent with Baker’s macroparameter: the AC regulates agreement in Cheyenne but not in Ojibwe.5

Note first that not all intransitive forms give rise to an interaction between T- and C-agreement. Recall from section 2 that C-agreement targets third persons only. First/Second person intransitive forms contain no valid goal for C and thus uniformly display only T-agreement, as shown in the first two rows of table 2. In the Ojibwe 1pl form, for example, the subject is indexed by T-agreement (ni- . . . -min ‘1pl’) and C-agreement fails to appear.6

Table 2

Intransitive agreement in Ojibwe and Cheyenne

SUBJOjibweCheyenne
1pl ni- . . .-min — ná- . . .-me — -he 
 1 1PL  1 1PL INT 
2pl gi- . . .-mw — né- . . .-me — -he 
 2 2PL  2 2PL INT 
3pl  . . . -w  -ag é- . . .-vo — -he 
  3 3PL 3 3PL INT 
SUBJOjibweCheyenne
1pl ni- . . .-min — ná- . . .-me — -he 
 1 1PL  1 1PL INT 
2pl gi- . . .-mw — né- . . .-me — -he 
 2 2PL  2 2PL INT 
3pl  . . . -w  -ag é- . . .-vo — -he 
  3 3PL 3 3PL INT 

The crucial test case for AC effects involves third person intransitive forms, which do give rise to an interaction between T- and C-agreement. In such forms, T-agreement will always target the third person subject, as it is (trivially) the argument that ranks highest on the 1/2 > 3 > 3′ person hierarchy. The question is what effect this T-agreement has on subsequent C-agreement, which could also potentially be satisfied by the third person subject. Two outcomes are possible, depending on the applicability of the AC:

  1. If the AC applies: C-agreement fails. The subject is indexed by T-agreement.

  2. If the AC does not apply: C-agreement succeeds. The subject is indexed by C-agreement in addition to T-agreement.

As table 2 shows, the outcome differs in the two languages. The Cheyenne 3pl form displays the first outcome: the 3pl subject is indexed by T-agreement (é- . . . -vo ‘3pl’) and C-agreement fails to appear. This is what we expect if the AC applies: T-agreement with the subject prevents subsequent C-agreement. The Ojibwe 3pl form, on the other hand, displays the second outcome: the 3pl subject is indexed by C-agreement (-ag ‘3pl’) in addition to T-agreement (-w ‘3’). This is what we expect if the AC does not apply: T-agreement with the subject does not prevent subsequent C-agreement. We may conclude, then, that agreement is regulated by the AC in Cheyenne but not in Ojibwe. That is, T-agreement with the 3pl goal renders that goal inaccessible to C-agreement in Cheyenne but not in Ojibwe. This conclusion is consistent with Baker’s macroparametric conception of AC effects: Cheyenne and Oijbwe have opposite settings for the relevant macroparameter.7

Before we turn to a language that is not consistent with Baker’s proposal, one aspect of the Ojibwe data in table 2 warrants further comment. In the crucial 3pl form in which T and C both agree with the same argument, T-agreement is realized as the reduced form -w ‘3’ rather than the full form o- . . . -waa ‘3pl’ that we saw in table 1. The reduced realization of T-agreement in this context may reflect the dissimilatory deletion of at least some of the φ-features of T resulting from their duplication on C (see, e.g., Arregi and Nevins 2007 and Nevins 2007, 2012 on haplological dissimilation in morphosyntax). Despite its reduced form, however, it is clear that -w is still an explicit marker of third person agreement, as assumed in standard Algonquianist descriptions (Goddard 2007:209) and argued at length for Menominee by Trommer (2008). The evidence is that -w contrasts with another reduced form of T-agreement, -m, which appears when there is no argument for T to agree with in the first place. The Ojibwe intransitive forms in (3) illustrate this point. The 1pl form in (3a) displays the full-fledged T-agreement ni- . . . -min ‘1pl’. In the 3pl form in (3b), T-agreement is reduced to -w ‘3’ owing to its duplication by C-agreement (-ag ‘3pl’). In the impersonal form in (3c), where there is nothing for T to agree with, T-agreement is realized as the distinct reduced form -m.

(3)

  • a.

    ninibaamin

    ni-nibaa-min

    1- sleep -1PL

    ‘we sleep’ (1pl)

  • b.

    nibaawag

    nibaa-w-ag

    sleep -3 -3PL

    ‘they sleep’ (3pl)

  • c.

    nibaam

    nibaa-m

    sleep -M

    ‘there is sleeping’ (impers.)

Since the -m suffix in (3c) appears when T completely fails to agree, it must be analyzed as the elsewhere form of the T-agreement suffix. This means that the -w suffix in (3b) cannot be the elsewhere form, but must instead be specialized for third person. (Trommer (2008) reaches the same conclusion for the Menominee counterparts -n and -w.) The proposal that T and C both agree with the 3pl argument in (3b) is thus on firm empirical ground.

4 Microparametric Variation: Plains Cree

This section turns to Plains Cree (Wolfart 1973), another Algonquian language. Plains Cree is significant in that it displays both the Ojibwe agreement pattern and the Cheyenne agreement pattern: in present tense clauses, C-agreement can target the goal of T-agreement (as in Ojibwe), but in past tense clauses, it cannot (as in Cheyenne). The existence of both patterns within Plains Cree indicates that AC effects can be parameterized not only across languages but also within a single language, contra Baker’s macroparametric proposal.

Table 3 introduces Plains Cree agreement by comparing it with the Ojibwe transitive forms from section 2. The present tense Cree forms are identical in structure to the Ojibwe forms: T-agreement indexes the argument that ranks higher on the person hierarchy (e.g., ni- . . . -nân ‘1pl’) and C-agreement indexes a third person (-ak ‘3pl’).

Table 3

Transitive agreement in Ojibwe and Plains Cree

SUBJOBJOjibweCree presentCree preterit
1pl→3pl ni- . . .-naan -ag ni- . . .-nân -ak ni- . . .-htânân -ak 
 1 1PL 3PL 1 1PL 3PL 1 PRET.1PL 3PL 
2pl→3pl gi- . . .-waa -ag ki- . . .-wâw -ak ki- . . .-htâwâw -ak 
 2 2PL 3PL 2 2PL 3PL 2 PRET.2PL 3PL 
SUBJOBJOjibweCree presentCree preterit
1pl→3pl ni- . . .-naan -ag ni- . . .-nân -ak ni- . . .-htânân -ak 
 1 1PL 3PL 1 1PL 3PL 1 PRET.1PL 3PL 
2pl→3pl gi- . . .-waa -ag ki- . . .-wâw -ak ki- . . .-htâwâw -ak 
 2 2PL 3PL 2 2PL 3PL 2 PRET.2PL 3PL 

Plains Cree differs from Ojibwe in the existence of a set of “h-preterit” inflections in which T-agreement is combined with an overt past tense marker (Wolfart 1973, Pentland 1999, Goddard 2007). The preterit forms in table 3 display exactly the same agreement pattern as the present forms: T-agreement indexes the higher-ranked argument and C-agreement indexes a third person. The only difference from the present is the appearance of the preterit marker -htâ beside the familiar T-agreement suffixes (e.g., -htânân ‘pret.1pl’; cf. -nân ‘1pl’). I take the appearance of this additional marker to reflect fission in the spell-out of past tense T (Noyer 1997), with -htâ discharging st] and -nân discharging [1pl].

We can test the applicability of the AC in Plains Cree by considering third person intransitive forms, as we did for Ojibwe and Cheyenne in section 3. Once T-agreement has targeted the 3pl subject, can C-agreement do so as well? Unlike in Ojibwe and Cheyenne, the answer to this question in Plains Cree is neither “yes” nor “no,” but rather “it depends.” As shown in table 4, the Plains Cree intransitive agreement pattern differs between the present and the preterit. In the present, the 3pl subject is indexed by both T-agreement (-w ‘3’) and C-agreement (-ak ‘3pl’), exactly as in Ojibwe (see table 2).8 This pattern arises when the AC does not apply, thus allowing C-agreement to target the goal of T-agreement. In the preterit, on the other hand, C-agreement fails to appear and the 3pl subject is indexed only by T-agreement (o- . . . -htâwâw ‘pret.3pl’), exactly as in Cheyenne. This pattern arises when the AC does apply, thus preventing C-agreement from targeting the goal of T-agreement.9

Table 4

Intransitive agreement in Plains Cree

SUBJPresentPreterit
1pl ni- . . .-nân — ni- . . .-htânân— 
 1 1PL  1 PRET.1PL 
2pl ki- . . .-wâw — ki- . . .-htâwâw — 
 2 2PL  2 PRET.2PL 
3pl  . . .-w -ak o- . . .-htâwâw — 
  3 3PL 3 PRET.3PL 
SUBJPresentPreterit
1pl ni- . . .-nân — ni- . . .-htânân— 
 1 1PL  1 PRET.1PL 
2pl ki- . . .-wâw — ki- . . .-htâwâw — 
 2 2PL  2 PRET.2PL 
3pl  . . .-w -ak o- . . .-htâwâw — 
  3 3PL 3 PRET.3PL 

As far as the AC is concerned, then, Plains Cree behaves like two languages in one: in the present tense, it patterns exactly like Ojibwe, a language in which the AC does not apply, while in the preterit, it patterns exactly like Cheyenne, a language in which the AC does apply. In other words, Plains Cree displays a tense-based split in the applicability of the AC.

It does not seem possible to derive this split from independent factors. We cannot, for example, posit that only C-agreement is active in the present and only T-agreement is active in the preterit, because the forms in table 3 clearly demonstrate that C-agreement and T-agreement are both active in both tenses. We also cannot posit a fundamental structural difference between present and past clauses, akin to some analyses of split ergativity (e.g., Coon 2013), because the difference between the two tenses arises only in highly particular circumstances, namely, when there is just one goal whose features satisfy both probes; in all other forms, the agreement pattern is identical in both tenses (see table 3) and there is no evidence for a difference in structure. Microparameterization of the AC provides the only straightforward account: C can target the goal of T in the present but not in the preterit.

The details of the Plains Cree morphology provide some insight into the nature of this parameter. Since the structure of present and preterit clauses differs only in the choice of T (i.e., TPres vs. TPast ), it must be T that is responsible for the difference in the agreement pattern. The parameter is thus most accurately expressed in terms of T, as in (4).

(4) When T has agreed with a goal G, does G remain accessible to further agreement?

  • a.

    Yes: T in Ojibwe; TPres in Plains Cree

  • b.

    No: T in Cheyenne; TPast in Plains Cree

It remains to be determined whether the parameter settings in (4) can be derived from other factors. I note that in Plains Cree, TPast , which renders its goal inaccessible to further agreement, is more marked both featurally and morphologically than TPres , which leaves its goal accessible to further agreement. One possibility, then, is that the featural richness of an agreement head affects the accessibility of that head’s goal to subsequent agreement operations. Further investigation of this possibility must await future research.

5 Conclusion

In Ojibwe, Cheyenne, and Plains Cree, the clause contains two probes that can each target either the subject or the object. In most forms, the two probes target different arguments, but third person intransitives are a special case in which the lone argument is a valid goal for both probes. In such forms in Plains Cree, agreement by the lower probe on T prevents agreement by the higher probe on C in the past tense but not in the present. In other words, TPast renders its goal inaccessible to subsequent agreement while TPres does not. In Plains Cree, then, the applicability of the AC is a microparameter. Rather than being a property of the language as a whole (as in Baker 2008b), it is a property of particular functional heads (TPres and TPast).

This finding raises a question: if AC effects indeed reflect a microparameter, why are the languages in Baker’s sample overwhelmingly consistent with the macroparametric view? I suggest that a similar situation holds for other microparameters: languages may cluster at the two uniform extremes. Taking the traditional head parameter as an example, the existence of many uniformly head-initial and uniformly head-final languages does not preclude the possibility of a language with mixed order (although learnability may provide a bias for uniform patterning, as a reviewer suggests; cf. Lightfoot 1991). Similarly, the prevalence of languages in which all probes have unified AC behavior does not necessarily preclude the possibility of a language like Plains Cree in which the behavior of probes is mixed.

The microparametric status of the AC in Plains Cree counters Baker’s macroparametric proposal, thus weakening the case for the existence of macroparameters in general. However, the Algonquian facts support Baker’s perhaps more interesting point that the AC, at least in its applicability to agreement, is not a principle: rather than being hard-wired into the faculty of language, AC effects are a variable property of the Agree operation. The Algonquian data provide some insight into the nature of this variation.

Acknowledgments

I thank Jila Ghomeshi, Diane Massam, David Pentland, the audience at NELS 44, and two anonymous reviewers for helpful comments. All errors are my own.

Notes

1 The analysis of the inverse marker -igw as v follows Béjar and Rezac 2009 and Lochbihler 2012.

2 The following abbreviations appear in glosses: 1, 2, 3 = first, second, third person; 3ʹ = obviative; 1→3 = first person subject, third person object; DIR = direct; DUB dubitative; INT = interrogative; INV = inverse; NEG = negative; PL = plural; PRET = preterit.

3 Some Algonquian languages have overt C-agreement suffixes for both singular and plural. An example is Meskwaki (Goddard 1994): -a ‘3sg’, -aki ‘3pl’, -ani ‘3ʹsg’, -ahi ‘3ʹpl’, -i ‘0sg’, -ani ‘0pl’ (where “0” = inanimate gender). However, most of the languages, including those in this squib, have lost final vowels, which has made the 3sg and 0sg suffixes null. I use plural forms throughout the squib because the overt plural suffixes make it easiest to observe the patterning of the agreement slots, but it should be noted that singular forms display exactly the same agreement patterns as plural forms, differing only in that the overt 3pl C-agreement suffix is replaced by -

graphic
‘3sg’.

4 Cheyenne forms are shown in the interrogative mode (with -he ‘interrogative’ following the C-agreement suffix) because the indicative mode has, in some instances, been analogically reshaped to resemble noun inflection (Goddard 2000:93). Note also that Cheyenne displays two homophonous -vo suffixes owing to regular sound change from Proto-Algonquian (PA): the T-agreement suffix -vo ‘2pl/3pl’ is from PA *-wa:w ‘2pl/3pl’ and the C-agreement suffix -vo ‘3pl/3ʹ’ is from PA *-waki ‘3pl’ and *-wahi ‘3ʹpl’ (Goddard 2000:93).

5 While the AC is standardly framed in terms of Case licensing, the status of Case is unclear in a head-marking language such as Algonquian. I follow Lochbihler (2012) in taking person rather than Case to be the licensing feature in Algonquian (cf. the Person Licensing Condition in Béjar and Rezac 2003; also Kalin 2017).

6 I assume, following Schütze (1997) and Preminger (2011), that failure to agree does not crash the derivation. The fact that C-agreement completely disappears in 1pl and 2pl forms, rather than surfacing as a default form, raises the possibility of analyzing C-agreement as clitic doubling rather than agreement (after Preminger 2009). This possibility does not affect the key argument in this squib, however, because even under a clitic-doubling analysis of C-agreement, we would still need to invoke the AC in order to explain the variation in whether or not T-agreement prevents subsequent clitic doubling (i.e., C-agreement).

7 A reviewer suggests that the Ojibwe pattern in table 2 could be understood in terms of split ergativity, since the suffix -ag that indexes the 3pl intransitive subject also indexes 3pl transitive objects (table 1). This is one of several related third person agreement patterns across the Algonquian family. I have argued elsewhere (Oxford 2015) that since all of the patterns can be understood as reflecting variation in the interaction between T- and C-agreement but only the Ojibwe pattern can be understood in terms of split ergativity, it is best to regard the Ojibwe split-ergative pattern simply as an epiphenomenal result of the interaction between T and C.

8 As discussed for Ojibwe in section 3, third person T-agreement in Plains Cree surfaces as the reduced form -w when T and C both target the same goal.

9 The forms in table 4 may suggest a simpler alternative analysis in which the preterit simply lacks C-agreement altogether. Such an analysis cannot work, however, as the forms in table 3 show that C-agreement is indeed active in the preterit. C-agreement in the preterit disappears only when its putative goal has already been targeted by T-agreement, a circumstance that is best captured by the proposed AC-based analysis.

References

Arregi,
Karlos
, and
Andrew
Nevins
.
2007
. Obliteration vs. impoverishment in the Basque g-/z- constraint. In
Proceedings of the 30th Annual Penn Linguistics Colloquium
, ed. by
Tatjana
Scheffler
,
Joshua
Tauberer
,
Aviad
Eilam
, and
Laia
Mayol
,
1
14
.
Philadelphia
:
University of Pennsylvania, Penn Working Papers in Linguistics
.
Baker,
Mark C
.
2008a
. The macroparameter in a microparametric world. In
The limits of syntactic variation
, ed. by
Theresa
Biberauer
,
351
373
.
Amsterdam
:
John Benjamins
.
Baker,
Mark C
.
2008b
.
The syntax of agreement and concord
.
Cambridge
:
Cambridge University Press
.
Baker,
Mark C
.
2013
.
On agreement and its relationship to case: Some generative ideas and results
.
Lingua
130
:
14
32
.
Béjar,
Susana
, and
Milan
Rezac
.
2003
. Person licensing and the derivation of PCC effects. In
Romance linguistics: Theory and acquisition
, ed. by
Ana Teresa
Pérez-Leroux
and
Yves
Roberge
,
49
62
.
Amsterdam
:
John Benjamins
.
Béjar,
Susana
, and
Milan
Rezac
.
2009
.
Cyclic Agree
.
Linguistic Inquiry
40
:
35
73
.
Bliss,
Heather
.
2013
.
The Blackfoot configurationality conspiracy
. Doctoral dissertation, University of British Columbia, Vancouver.
Borer,
Hagit
.
1984
.
Parametric syntax: Case studies in Semitic and Romance languages
.
Dordrecht
:
Foris
.
Branigan,
Phil
, and
Marguerite
MacKenzie
.
1999
. Binding relations and the nature of pro in Innu-aimun. In
Proceedings of NELS 29
, ed. by
Pius
Tamanji
,
Masako
Hirotani
, and
Nancy
Hall
,
475
485
.
Amherst
:
University of Massachusetts, Graduate Linguistic Student Association
.
Bruening,
Benjamin
.
2005
.
The Algonquian inverse is syntactic: Binding in Passamaquoddy
. Ms., University of Delaware, Newark.
Chomsky,
Noam
.
1995
.
The Minimalist Program
.
Cambridge, MA
:
MIT Press
.
Chomsky,
Noam
.
2000
. Minimalist inquiries: The framework. In
Step by step: Essays on Minimalism in honor of Howard Lasnik
, ed. by
Roger
Martin
,
David
Michaels
, and
Juan
Uriagereka
,
89
155
.
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
.
Coon,
Jessica
.
2013
.
Aspects of split ergativity
.
New York
:
Oxford University Press
.
Coon,
Jessica
, and
Alan
Bale
.
2013
.
The inseparability of person and number in Mi’gmaq
. Paper presented at Features in Phonology, Morphology, Syntax, and Semantics, University of Tromsø, 31 October.
Goddard,
Ives
.
1994
.
Leonard Bloomfield’s Fox lexicon: Critical edition
.
Winnipeg, MB
:
Algonquian and Iroquoian Linguistics
.
Goddard,
Ives
.
2000
. The historical origins of Cheyenne inflections. In
Papers of the 31st Algonquian Conference
, ed. by
John D.
Nichols
,
77
129
.
Winnipeg
:
University of Manitoba
.
Goddard,
Ives
.
2007
. Reconstruction and history of the independent indicative. In
Papers of the 38th Algonquian Conference
, ed. by
H. C.
Wolfart
,
207
271
.
Winnipeg
:
University of Manitoba
.
Halle,
Morris
, and
Alec
Marantz
.
1993
. Distributed Morphology and the pieces of inflection. In
The view from Building 20
, ed. by
Kenneth
Hale
and
Samuel Jay
Keyser
,
111
176
.
Cambridge, MA
:
MIT Press
.
Hamilton,
Michael David
.
2013
.
An account of verbal person suffixes
. Paper presented at the 45th Algonquian Conference, Ottawa, 19 October.
Harbour,
Daniel
.
2008
. Discontinuous agreement and the syntax-morphology interface. In
Phi theory: Phi-features across modules and interfaces
, ed. by
Daniel
Harbour
,
David
Adger
, and
Susana
Béjar
,
185
220
.
Oxford
:
Oxford University Press
.
Kalin,
Laura
.
2017
.
Nominal licensing is driven by valued ( phi-)features
. Paper presented at GLOW 40, Leiden, 15 March.
Kayne,
Richard S
.
2005
. Some notes on comparative syntax, with special reference to English and French. In
The Oxford handbook of comparative syntax
, ed. by
Guglielmo
Cinque
and
Richard S.
Kayne
,
3
69
.
New York
:
Oxford University Press
.
Leman,
Wayne
.
1980
.
A reference grammar of the Cheyenne language
.
Occasional Publications in Anthropology, Linguistics series, no. 5. Greeley, CO
:
University of Northern Colorado, Museum of Anthropology
.
Lightfoot,
David
.
1991
.
How to set parameters
.
Cambridge, MA
:
MIT Press
.
Lochbihler,
Bethany
.
2012
.
Aspects of argument licensing
. Doctoral dissertation, McGill University, Montreal, QC.
Macaulay,
Monica
.
2009
.
On prominence hierarchies: Evidence from Algonquian
.
Linguistic Typology
13
:
357
389
.
Nevins,
Andrew
.
2007
.
The representation of third person and its consequences for person-case effects
.
Natural Language and Linguistic Theory
25
:
273
313
.
Nevins,
Andrew
.
2012
. Haplological dissimilation at distinct stages of exponence. In
The morphology and phonology of exponence
, ed. by
Jochen
Trommer
,
84
116
.
Oxford
:
Oxford University Press
.
Nichols,
John D
.
1980
.
Ojibwe morphology
. Doctoral dissertation, Harvard University, Cambridge, MA.
Noyer,
Rolf
.
1997
.
Features, positions, and affixes in autonomous morphological structure
.
New York
:
Garland
.
Oxford,
Will
.
2014
.
Microparameters of agreement: A diachronic perspective on Algonquian verb inflection
. Doctoral dissertation, University of Toronto, Toronto, ON.
Oxford,
Will
.
2015
. Probe competition as a source of ergative person splits. In
Proceedings of NELS 45
, ed. by
Thuy
Bui
and
Deniz
Özyıldız
,
2:213
222
.
Amherst
:
University of Massachusetts, Graduate Linguistic Student Association
.
Pentland,
David H
.
1999
. The morphology of the Algonquian independent order. In
Papers of the 30th Algonquian Conference
, ed. by
David H.
Pentland
,
222
266
.
Winnipeg
:
University of Manitoba
.
Preminger,
Omer
.
2009
.
Breaking agreements: Distinguishing agreement and clitic doubling by their failures
.
Linguistic Inquiry
40
:
619
666
.
Preminger,
Omer
.
2011
.
Agreement as a fallible operation
. Doctoral dissertation, MIT, Cambridge, MA.
Schütze,
Carson T
.
1997
.
INFL in child and adult language: Agreement, case, and licensing
. Doctoral dissertation, MIT, Cambridge, MA.
Trommer,
Jochen
.
2008
. Third-person marking in Menominee. In
Phi theory: Phi-features across modules and interfaces
, ed. by
Daniel
Harbour
,
David
Adger
, and
Susana
Béjar
,
221
250
.
Oxford
:
Oxford University Press
.
Wolfart,
H. C
.
1973
.
Plains Cree: A grammatical study
. Transactions of the American Philosophical Society, New Series, vol.
63
, part 5.
Philadelphia
:
American Philosophical Society
.