Skip to main content
SpringerLink
Log in

A formal characterization of person-based alignment

The case of Paraguayan Guaraní

  • Published:
Natural Language & Linguistic Theory Aims and scope Submit manuscript

Abstract

We put forth a formal analysis within the Minimalist framework of argument alignment in languages with one type of direct/inverse system. Our proposal involves the cyclical application of a phase-edge Person constraint, which ensures that a [+Participant] argument (when present) is promoted from the verbal (vP) to the inflectional (IP) domain. We illustrate the proposed analysis with Paraguayan Guaraní, a language with direct/inverse alignment whose morpho-syntax has received little attention from a formal perspective. Paraguayan Guaraní does not mark tense morphologically in Infl(ection); instead, the overt realization of Infl varies depending on the person specification of the arguments. We refer to languages of this type as Generalized P(erson)-languages, in contrast to Restricted P-languages, whose direct/inverse system is limited to the vP domain and whose Infl encodes tense (e.g., Hungarian and Kashmiri). Building on insights in Ritter and Wiltschko (2014) on the anchoring function of Infl, we link the distinction between the two types of language to the presence vs. absence of an interpretable tense feature and its complementary interpretable person feature in the Infl node of the clausal structure.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Similar content being viewed by others

Notes

  1. The data reported here is based on field work conducted primarily in 2013 and 2014. Our main consultants were two women (ages 45–50), whose first language is Guaraní, and who learned Spanish for the purpose of schooling (around 7 years of age), but continued to use Guaraní in their daily lives, and who consider themselves fluent bilinguals. Both currently live in Asunción or the surrounding area (where field work was conducted). One of them (originally from Concepción) is an elementary school teacher and the other (originally from San Pedro) teaches Guaraní to foreigners. We have used elicited production as well as grammaticality judgement techniques, followed by questions to confirm subtleties of meaning and use. Our fieldwork was complemented with other sources, namely Guasch (1956), Guasch and Ortiz (2008), and Krivoshein de Canese (1998), Velázquez-Castillo (1996, 2002, 2007).

  2. Ritter and Wiltschko (2014:fn. 9) note that Aspect, like Infl, has an abstract function that can be instantiated by language-particular substantive content, and refer to Bliss et al. (2010) for an analysis of Blackfoot’s direct/inverse system as instantiating person-based Aspect. Wiltschko (2014) also analyzes the direct/inverse system of Blackfoot in terms of Aspect encoding a person-based point of view.

  3. See Macaulay (2005) on the variability and complexity of Algonquian verbal morphology, which has led to much disagreement on the primacy relations among [+Participant] arguments in this language family.

  4. This has also been suggested by Andrade Freitas (2011) for Ava Guaraní, a language closely related to Paraguayan Guaraní. See also fn. 18.

  5. In an apparent contradiction with this feature system, Bliss (2005:42, 2013:253–254) gives examples of obviative-marked 1P and 2P pronouns in Blackfoot, but she notes that Blackfoot is unique among the Algonquian languages in this respect (Bliss 2013:253) (and in fact analyzes the obviative morphemes in terms of case and number marking). Aissen (1997:706), following the Algonquianist tradition, considers the proximate/obviative distinction relevant only to third person nominals. We do not address the issue of proximate/obviative marking further given that it is orthogonal to the immediate concerns of this paper.

  6. The above approach contrasts with the cyclical application proposed by Béjar and Rezac (2009), who argue for one single domain of agreement, namely a verbal domain with multiple v-s (no Infl domain) and no syntactic object promotion. Our approach is more in line with Bruening (2001, 2005) and Bliss (2013). See Sect. 6 for further elaboration.

  7. It might be useful to draw a deeper analogy between interpretable Person and interpretable Tense at an abstract formal level. The function of interpretable Tense is to establish an ordering relation between time-arguments, namely between Speech Time and Event Time (mediated by Reichenbach’s 1947 R-time). Similarly, interpretable Person on the head of a phase has the function of ordering event-arguments, assigning structural primacy to one over the other. This is in line with Ritter and Wiltschko (2014) and Wiltschko (2014), who formalize this property in terms of the feature +/−coincidence.

  8. The Algonquian languages are known to give rise to a distinct “conjunct order” in subordinate clauses, inexistent in Paraguayan Guaraní. Whether this difference should be attributed to some other fundamental difference in the properties of subordination in the two languages remains to be investigated.

  9. Velázquez-Castillo (2007), working within a functionalist framework, proposes a different account of Paraguayan Guaraní inflection. The author argues that the language has an active/inactive inflectional system. The active reflects the presence of an “active event source” and the inactive reflects the presence of “inactive event sites” (a view first put forth for Paraguayan Guaraní in Velázquez-Castillo 2002). The former notion refers to “a participant construed as the initiator or origin of a dynamic event” and the latter to “a participant construed as containing the situation denoted by the predicate.” Velázquez-Castillo furthermore relates the active/inactive dichotomy to the notion of point of view as discussed by De Lancey (1981). We find no place for Velázquez-Castillo’s active-inactive ontology in our formal framework and this renders the comparison of the two analyses difficult (if not impossible) to undertake in a meaningful way at this point in time. It is important to note however that both analyses acknowledge the relevance of point of view.

  10. The 1PL inclusive in the direct inflectional paradigm has two allomorphs which depend on the oral vs. nasal nature of the following morpheme (throughout the paper we use the Leipzig glossing convention):

    1. (i)
      figure a
    1. (ii)
      figure b

  11. A sub-class of verbs inserts an -i between the prefix and the verbal root. This class of verbs is referred to as aireal, while the other class of verbs are known as areal. Verbal roots typically belong to one or the other of the two classes; it is a morphological property of roots not unlike Spanish verbal conjugation classes. In some cases one and the same root may have two forms, but crucially with a very different lexical meaning, e.g., -ke: ake ‘sleep’ (areal) vs. aike ‘enter’ (aireal). For the sake of simplicity and consistency, we use examples form the areal class only.

  12. Woolford (2016) also reaches the conclusion that the portmanteau morphemes of Guaraní mark only subject agreement, not agreement with both subject and object. In her terminology, Guaraní exhibits a case of morphological portmanteau agreement, which is the spell out of agreement with a single argument in the context of another feature specification. Syntactic portmanteau agreement, on the other hand, involves the spell out of multiple agreement, with features collected from all participating arguments.

  13. An overt 2P oblique pronoun can appear in multiple positions. It is interpreted as emphatic/contrastive and requires a special context. We will not discuss the syntax of emphatic pronouns, which are generally still poorly understood.

    1. (i)
      figure c

  14. We do not think that the asymmetry between 3P pronouns and [+Participant] pronouns is due to the fact that the latter are recoverable from the immediate speech act context (a suggestion made by one of the reviewers). As we know, many languages drop 3P object pronouns frequently, Brazilian Portuguese being one of them. This language is particularly relevant in that it has lost its 3P clitic (giving rise to the null 3P object drop) but has maintained its 1P and 2P clitics, which are obligatorily present in the clause (see Kato 2003).

  15. Bossong (2009) attributes the presence of -pe on direct objects in Paraguayan Guaraní to contact with Spanish. See also Shain and Tonhauser (2011), who corroborate this conclusion based on corpus studies; these authors furthermore argue that the occurrence of -pe as DOM is governed by animacy and topicality. It is possible that -pe has been adopted as a mark of [+Prox] objects, a conjecture that merits further investigation.

  16. As we said earlier, we will not address here the syntax of emphatic pronouns, which can appear in multiple positions in the clause, but we note that an oblique emphatic strong pronoun can co-occur with a clitic object pronoun in sentences with an inverse structure:

    1. (i)
      figure h

  17. The 2SG, 2PL, and 1PL object clitics in the indirect inflectional paradigm have two allomorphs, depending on whether the immediately following morpheme (root or prefix) is oral or nasal: nde (oral)/ ne (nasal), pende (oral)/ pene (nasal), ñande (oral)/ ñane (nasal). e.g., (i) vs. (ii):

    figure i

  18. As mentioned in fn. 4, an object movement analysis of the weak object pronouns in the inverse inflectional paradigm is also proposed by Andrade Freitas (2011) for Ava Guaraní. This author attributes the object preposing of [+Participant] pronouns to their specific/presuppositional nature. Such an analysis does not directly relate the particularities of the inflectional system to the reordering of arguments, and in particular it does not provide an answer for the lack of object promotion in the case of 1 > 2 (i.e., direct orders).

  19. The negative prefix nda- becomes na- in (9) because it is part of an accentual group that contains a nasal morpheme, namely mbo-; cf. the oral vs. nasal forms of the subject prefixes and object clitics mentioned in fn. 10 and 17.

  20. A reviewer asks how the overt DP subject and the pro in Spec, vP are linked, given the Phase Impenetrability Condition (PIC). If the overt DP is adjoined to IP, it is at the edge of the IP phase and the pro at the edge of the vP phase would be accessible to it. The situation will be different if the DP is in a Topic phrase, outside the IP phase. But we know that relations such as co-reference and pronominal binding do not obey the locality of phases, so an argument from PIC does not rule out an analysis of the overt DP being in Spec, TopP.

  21. Note that in direct orders, like (12) and (13), v defines two distinct agreement relations: one triggered by an interpretable and valued p-feature (with the external argument) and one triggered by an uninterpretable and unvalued p-feature (with the object). We assume that the latter relation is universal for transitive structures in all languages, including those without direct/inverse systems. On the other hand, in a Generalized P-language like Paraguayan Guaraní, the interpretable p-feature on v is inherited from Infl. The presence of two types of agreement features on the same phase head, one interpretable and one uninterpretable, is in line with similar proposals about dual agreement features on DPs, motivated by variation in semantic vs. syntactic agreement. For instance, the variation exhibited in The committee is/are meeting, can be accounted for if the collective nominal committee has dual number features, an interpretable plural and an uninterpretable singular feature, with either being able to control agreement (Wurmbrand 2012; building on Wechsler and Zlatić 2000; Wechsler 2011; a.o.).

  22. See Béjar and Rezac (2009) for a similar proposal for portmanteau morphemes but with different mechanics, which assume that a probe can value across multiple heads; see Sect. 6.

  23. On word order in Paraguayan Guaraní, see Velázquez-Castillo (1996) and Tonhauser and Colijn (2010). These works firmly establish that Paraguayan Guaraní is a VO language. On the other hand, the corpus-based study of Tonhauser and Colijn shows that familiar or discourse-old subjects can be preverbal or postverbal. In the present work, we assume that the external argument is introduced by v as a pro and that it is related to a clause peripheral position. We will not address here the non-trivial issue of how VSO orders are generated.

  24. A reviewer suggests that the complementarity that we observe in Paraguayan Guaraní may be the same phenomenon as the one observed in Celtic languages such as Irish (McCloskey and Hale 1984) and Breton (Jouitteau and Rezac 2006). We do not think the two exemplify the same phenomenon. In the Celtic language the complementarity is between a null subject that triggers agreement vs. overt DPs that do not. This is not the case for Paraguayan Guaraní: overt DP and null subjects in this language co-occur with agreement. It is only in the inverse system that the complementarity arises between a prefix and an object clitic. Note furthermore that the complementarity cannot be morphological since it involves an affix and a clitic. Only affixes can compete for the same morphological slot.

  25. As mentioned in fn. 3, Nevins and Sandalo (2011) have shown that Kadiwéu is a 2 > 1 person system, except when the object is 1PL in which case it flips to a 1 > 2 system. These authors argue that this state of affairs is due to a combination of factors: the portmanteau nature of the 1PL object (which fuses person and number), the Coherence Constraint that governs the syntax/morphology mapping (which they attribute to Trommer 2008) and the need to realize inverse marking in the language. In order to render Nevins and Sandalo’s analysis of Kadiwéu compatible with the phase-based syntactic analysis proposed here we would need to assume a phase-based syntax/morphology mapping that applies in parallel to the syntactic computation. This is a very interesting view, which we adopt for Guaraní’s compounding rule; see fn. 31.

  26. Other examples of individual-denoting triforme nouns are given below: core inalienables, such as body parts (i), kinship inalienable relations (ii), as well as what can be considered inalienables by extension as in (iii).

    1. (i)

      tesa – resa – hesa ‘eye’, topepi – ropepi – hopepi ‘eyelid’, topea – ropea – hopea ‘eyelashes’, tova – rova – hova ‘face’, tetyma – retyma – hetyma ‘leg’, tembe – rembe – hembe ‘lip’, tái – rái – hái ‘teeth’, tañyka – rañyka – hañyka ‘jaw’, tãimbira – rãimbira – hãimbira ‘gums’, tague – rague – hague ‘hair’, ta’anga – ra’anga – ha’anga ‘image’

    2. (ii)

      túa – rúa – húa ‘father’, teindy – reindy – heindy ‘sister of boy’, tovaja – rovaja – hovaja ‘brother- or sister-in-law’

    3. (iii)

      óga – róga – hóga ‘house’, okẽ – rokẽ – hokẽ ‘door’, ovetã – rovetã – hovetã ‘window’, tape – rape – hape ‘road, path’, taity – raity – haity ‘nest’, tembi’u – rembi’u – hembi’u ‘food’, tupi’a – rupi’a – hupi’a ‘egg’, tupã – rupã – hupã ‘God’, téra – réra – héra ‘name’, tuvicha – ruvicha – huvicha ‘leader’

  27. A reviewer notes that examples like those in (21) and (22) are “marked forms” and that the forms with incorporation of the inalienable noun (an alternative construction discussed further below in the text) constitute the unmarked option. Whatever the account may be for the perceived markedness of the non-incorporated form (but see fn. 29 for a suggestion), they were judged to be acceptable forms by our consultants, and in some cases, they were judged to be the preferred form, namely in the presence of a modifying adjective (see fn. 31). See also Velázquez-Castillo (1996:134–135), where both forms (incorporated and non-incorporated) are recognized.

  28. For Velázquez-Castillo, the r-forms are part of what the author calls the inactive inflectional system. See fn. 9 and 35.

  29. It is possible that an Applicative v (which introduces the affected theta role) combines more readily with the incorporated structure than with the non-incorporated counterpart, and it is the presence vs. absence of this affectedness layer of meaning that accounts for the perceived difference between the incorporated vs. non-incorporated forms; see fn. 27. We note though that one of our consultants readily accepted the presence of an optional dative argument in the non-incorporated version of the structure, see (i).

    1. (i)
      figure v

  30. Massam (2001) provides extensive arguments that pseudo-incorporation involves small nominal phrases (to the exclusion of Ds). While such nominal phrases appear in the canonical direct object position of the verb, they have predicative status. Here Paraguayan Guaraní differs from Niuean in that in the canonical direct object position the possessed nominal is a full DP; incorporated nPs have to appear pre-verbally.

  31. It is possible that there is speaker variability with regards to preference for long vs. short verb stem forms in the incorporated structures. The variability in judgements regarding forms like (i) and (ii) below possibly speak to this point. While Velázquez-Castillo (1996:144) considers incorporated forms with modification completely impossible, our consultants did not reject (i), provided that the long verbal stem form is used, although (ii) is the preferred form. This suggests that for m-compounding to apply, there must be adjacency between N and V, and m-compounding is a pre-requisite for V-stem shortening to take place. It is expected that speakers with a strong preference for the m-compounded forms will find (i) unacceptable. Dialects with optional application of m-compounding might be a reflection of language change, an issue that merits further investigation. (The morphological m-compounding rule may be assumed to apply in parallel to the syntactic computation, with the phase as its domain of application.)

    1. (i)
      figure aa
    1. (ii)

      (Ha’e) o-johéi nde rova ky’a

  32. Paraguayan Guaraní has borrowed alguno ‘someone’ and ninguno ‘no one’ from Spanish and these also trigger the r-rule (alguno roga ‘someone’s house,’ ninguno roga ‘no one’s house’).

  33. In Zubizarreta and Pancheva (2017), we extend the analysis of individual-denoting triforme nominals provided here to event-denoting triforme nominals like the one in (i). The light verb (oi)ko takes a locative complement, which contains a nominal phrase headed by a trifome noun (reka). The argument of reka is vaka: it originates within the nP. The same analysis given for inalienable Possessors extends to these cases as well.

    1. (i)
      figure ae

    The example in (ii) provided by a reviewer would be analyzed in the same way as (i). The optional -vo suffix functions here as a sufijo de finalidad (or ‘purpose suffix’).

    1. (ii)
      figure af

  34. We may think of the configuration in (31) as formally similar to the ECM construction: just like an embedded Spec of Infl is visible to the matrix v in an ECM structure, a Possessor is visible to v in case the inalienable nP is incorporated into the V-domain. The Possessor thus becomes an object of v, to the extent that the latter is defined as the DP that enters into an agreement relation with v.

  35. We have identified some triforme eventive intransitives (e.g., tasẽ ‘the cry,’ rasẽ / hasẽ ‘to cry’), which, as we argue in Zubizarreta and Pancheva (2017), are derived by incorporation of an inalienable possessed nP into v, comparable to the analysis of intransitives in Hale and Keyser (2002).

    1. (i)

      a.

      Che che.rasẽ

      ‘I cry.’

      Lit. ‘I do my cry.’

      b.

      Nde nde. rasẽ

      ‘You cry.’

      Lit. ‘You do your cry.’

      c.

      Ha’e hasẽ

      ‘(S)he cries.’

      Lit. ‘(S)he does his/her cry.’

    The process of nP incorporation into v is a productive process in Paraguayan Guaraní. In Zubizarreta and Pancheva (2017), we argue that (triforme) stative predicates, like the one in (ii) (tasy/rasy/hasy ‘sick’), which have been analyzed by Payne (1994) and others to be intransitives with Set B prefixes (see Sect. 2), are actually transitive predicates derived from property-concept denoting nominals incorporated into v, namely a v that introduces an external theta-role, i.e., the beholder of the state. Our analysis thus provides a unified formal account of individual-denoting inalienable nominals and stative predicates. Compare the form of the triforme noun in (ii d) (which has a verbal structure) with (iii) (which has a nominal structure): the former has an h-root while the latter has an r-root, as predicted.

    1. (ii)

      a.

      Che che rasy

      ‘I am sick.’

      b.

      Nde nde rasy

      ‘You are sick.’

      c.

      Ha’e hasy

      ‘(S)he is sick.’

      d.

      Maria hasy

      ‘Maria is sick.’

    1. (iii)

      Maria rasy ‘Maria’s illness’

    For an analysis of stative predicates like (ii) and (iii) within a functionalist framework that appeals to notions of “inactive event sites” (which are analyzed as taking Set B prefixes), see Velázquez-Castillo (1996, 2007) (and fn. 9). It is not clear how that framework would deal with cases like (i), given the agentive nature of the verb, not unlike that of the verb a-puka ‘laugh,’ which takes Set A prefixes, or in Velázquez-Castillo’s terminology, it belongs to the active paradigm.

  36. A reviewer suggests that the lack of lexical anaphors in Paraguayan Guaraní can be due to the presence of agreement, as argued in Woolford (1999). Woolford’s proposal is an extension of Rizzi’s (1982) ‘anaphor agreement effect,’ generalized to both subject and object agreement. These authors argue that the presence of lexical anaphors correlates with lack of agreement. Extending their ideas to the Paraguayan Guaraní and the Algonquian languages would amount to the claim that lexical reflexives are missing from both subject and object position because these languages have both subject and object agreement. But this crucially is not the case. In local (1 > 2, 2 > 1) and mixed (1/2>3, 3>1/2) configurations, the Algonquian languages have prefixes that mark 1P/2P arguments (with preference for 2P over 1P), irrespective of grammatical role. The following paradigm from Ojibwe, in (i), would leave room for a lexical anaphor to encode “You.sg saw you.pl” since only one argument is referenced by the prefix, i.e., if there were a lexical anaphor here, it would be non-agreeing.

    1. (i)
      figure al

    Even more clearly, in Paraguayan Guaraní we don’t have overt morphological object agreement, so a lexical anaphor could have been present in “You.sg saw you.pl” cases. In summary, our theory predicts the absence of lexical anaphors in Paraguayan Guaraní and Algonquian languages, a fact that otherwise would be an accident, as it would not follow from the anaphor agreement effect of Rizzi and Woolford.

  37. Note that here we have two words (one basic and one derived) to express the same reflexive meaning: (a)jahu (intransitive) and a.ñe.mbo.jahu (reflexivized transitive). A reviewer notes that there is a preference to use the simple form to express the reflexive meaning and the je-form to express the impersonal passive meaning (discussed further below in the text). We know from work in derivational morphology (dating back to Aronoff 1976) that morphologically distinct but semantically related words are in competition and that one generally wins out. We submit that this is the case here.

  38. There is of course an alternative way of expressing the same meaning, namely without incorporation of the inalienable noun phrase. These are ordinary direct order SVO transitives, with the Possessor pronoun within the post-verbal object.

    (i)

    Che a-johéi che rova

    ‘I wash my face.’

    (ii)

    Nde re-johéi nde rova

    ‘You wash your face.’

    (iii)

    Ha’e o-johéi hova

    ‘(S)he washes her/his face.’

  39. We part ways with Landau (2010) in this respect, who assumes that defective NPs introduce a person feature. We assume that the p-feature is a property of D. Our nP could be seen as equivalent to phi-P in Déchaine and Wiltschko (2002), as long as the phi-features are restricted to number and gender (they make a 3-way distinction between DP, phi-P, and NP pronouns). See also Legate (2014) for detailed discussion of implicit arguments in passives.

  40. See also Harley (2013) for a description of causatives in a Mayan language with similar properties, but with a different analysis.

  41. In Zubizarreta and Pancheva (2017), we extend the application of the r-rule to apply to small clause PP domains.

  42. For discussion of further intricacies in the object agreement system of Hungarian, see Coppock and Wechsler (2012), Coppock (2013), Bárány (2015).

  43. Mbo- combines with basic intransitives (i), as well as with nouns (ii), and with modifiers (iii), to make transitive verbs. It is a very productive process. (Examples and description are from Guasch 1956. See also Velázquez-Castillo 2002.)

    (i)

    a-sẽ ‘to go out’

    – a-mo-sẽ ‘to make go out’

     

    a-ke ‘sleep’

    – a-mo-nge ‘to put to sleep’

     

    ai-ke ‘to go in’

    – a-moi-nge ‘to make enter’

    (ii)

    y ‘water’

    – a-mbo’y ‘to make water or liquify’

     

    kuarahy ‘sun’

    – a-mbo-kuarahy ‘to put under the sun’

     

    yvyty ‘wind’

    – a-mbo-yvytu ‘to ventilate’

    (iii)

    hasy ‘sick/ difficult’

    – a-mbo-hasy ‘to make sick/ difficult’

     

    hatã ‘hard’

    – a-mbo-hatã ‘to harden’

  44. We have here two forms with the same meaning, form (50), which consists of a compound word N + V-stem (formed by m-compounding post-syntactically) and form (51), where the compound form has undergone further morphological affixation. Our consultants accepted both forms as possible, but when asked whether there is a preferred form, they both reported that the most common one to express the transparent compositional meaning is (51).

    Our point here is that both forms are generated by the computational system, and other factors interfere to block one of the forms or assign preference to one over the other (see fn. 37). Alternatively, speakers may develop specialized nuances of meaning for one of the forms.

References

  • Adger, David, and Daniel Harbour. 2007. Syntax and syncretisms of the Person Case Constraint. Syntax 10(1): 2–37.

    Article  Google Scholar 

  • Ahn, Byron. 2015. Giving reflexivity a voice: Twin reflexives in English. PhD diss., UCLA.

  • Aissen, Judith. 1997. On the syntax of obviation. Language 73: 705–750.

    Article  Google Scholar 

  • Anagnostopoulou, Elena. 2003. The syntax of ditransitive: Evidence from clitics. Berlin: de Gruyter.

    Google Scholar 

  • Anagnostopoulou, Elena. 2005. Strong and weak person restrictions. In Clitic and affix combinations: Theoretical perspectives, eds. Lorie Heggie and Francisco Ordóñez, 199–235. Amsterdam: Benjamins.

    Chapter  Google Scholar 

  • Andrade Freitas, María Luisa. 2011. Hierarquia de pessoa em Avá Guaraní: Aspectos sintáticos e morfológicos. In LIAMES: línguas indígenas americanas 11, 31–57. Campinas: UNICAMP.

    Google Scholar 

  • Aronoff, Mark. 1976. Word formation in generative grammar. Cambridge: MIT Press.

    Google Scholar 

  • Bárány, András. 2015. Differential object marking in Hungarian and the morphosyntax of case and agreement. PhD diss., University of Cambridge.

  • Barker, Chris. 2011. Possessives and relational nouns. In Semantics: An international handbook of natural language meaning, eds. Claudia Maienborn, Klaus von Heusinger, and Paul Portner, 1109–1130. Berlin: de Gruyter.

    Google Scholar 

  • Béjar, Susana, and Milan Rezac. 2003. Person licensing and the derivation of PCC effects. In Romance linguistics: Theory and acquisition, eds. Ana Teresa Pérez-Leroux and Yves Roberge, 49–62. Amsterdam: Benjamins.

    Chapter  Google Scholar 

  • Béjar, Susana, and Milan Rezac. 2009. Cyclic agree. Linguistic Inquiry 40(1): 35–73.

    Article  Google Scholar 

  • Benveniste, Emile. 1971. The nature of pronouns. In Problems in general linguistics, 217–222. Coral Gables: University of Miami Press.

    Google Scholar 

  • Bianchi, Valentina. 2006. On the syntax of personal arguments. Lingua 116(12): 2023–2067.

    Article  Google Scholar 

  • Bliss, Heather. 2005. Formalizing point-of-view: The role of sentience in Blackfoot’s direct/inverse system, MA thesis, University of Calgary.

  • Bliss, Heather. 2013. The Blackfoot configurationality conspiracy: Parallels and differences in clausal and nominal structures. PhD diss., University of British Columbia.

  • Bliss, Heather, Elizabeth Ritter, and Martina Wiltschko. 2010. A comparison of theme marking in Blackfoot and Nishnaabemwin. In Workshop on the Structure and Constituency of Languages of the Americas (WSCLA) 15, eds. Beth Rogers and Anita Szakay. Vol. 29 of Working papers in linguistics, 1–11. Vancouver: UBC.

    Google Scholar 

  • Bonet, Eulalia. 1991. Morphology after syntax: Pronominal clitics in Romance languages. PhD diss., MIT.

  • Bonet, Eulalia. 1994. The person-case constraint: A morphological approach. The morphology-syntax connection. Vol. 22 of MIT working papers in linguistics, pp. 33–52.

    Google Scholar 

  • Bošković, Željko. 2014. Now I’m a phase, now I’m not a phase: On the variability of phases with extraction and ellipsis. Linguistic Inquiry 45(1): 27–89.

    Article  Google Scholar 

  • Bossong, Georg. 2009. The typology of Tupi-Guaraní as reflected in the grammars of four Jesuit missionaries. Historiographia Linguistica 36(2/3): 225–258.

    Article  Google Scholar 

  • Bruening, Benjamin. 2001. Syntax at the edge: Cross-clausal phenomena and the syntax of Passamaquoddy. PhD diss., MIT.

  • Bruening, Benjamin. 2005. The Algonquian inverse is syntactic: Binding in Passamaquoddy. Ms., University of Delaware.

  • Chomsky, Noam. 1995. The minimalist program. Cambridge: MIT Press.

    Google Scholar 

  • Chomsky, Noam. 2000. Minimalist inquiries: The framework. In Step by step: Essays on minimalist syntax in honor of Howard Lasnik, eds. Roger Martin, David Michaels, and Juan Uriagereka, 89–155. Cambridge: MIT Press.

    Google Scholar 

  • Chomsky, Noam. 2001. Derivation by phase. In Ken Hale: A life in language, ed. Michael Kenstowicz, 1–52. Cambridge: MIT Press.

    Google Scholar 

  • Chomsky, Noam. 2008. On phases. In Foundational issue in linguistic theory: Essays in honor of Jean-Roger Vergnaud, eds. Robert Freidin, Carlos Otero, and María Luisa Zubizarreta, 133–166. Cambridge: MIT Press.

    Google Scholar 

  • Comrie, Bernard. 2013. Alignment of case marking of pronouns. In The world atlas of language structures online, eds. Matthew S. Dryer and Martin Haspelmath. Leipzig: Max Planck Institute for Evolutionary Anthropology. Available at http://wals.info/chapter/99. Accessed 12 Dec 2016.

    Google Scholar 

  • Coon, Jessica, and Omer Preminger. 2012. Towards a unified account of person splits. In 29th West Coast Conference on Formal Linguistics (WCCFL), eds. Jaehoon Choi, E. Alan Hogue, Jeffrey Punske, Deniz Tat, Jessamyn Schertz, and Alex Trueman, 310–318. Somerville: Cascadilla Press.

    Google Scholar 

  • Coppock, Elizabeth. 2013. A semantic solution to the problem of Hungarian object agreement. Natural Language Semantics 21(4): 345–371.

    Article  Google Scholar 

  • Coppock, Elizabeth, and Steven Wechsler. 2012. The objective conjugation in Hungarian: Agreement without phi-features. Natural Language and Linguistic Theory 30(3): 699–740.

    Article  Google Scholar 

  • Déchaine, Rose-Marie. 1999. What Algonquian morphology is really like: Hockett revisited. In Papers from the workshop on structure and constituency in native American languages, eds. Leora Barel, Rose-Marie Déchaine, and Charlotte Reinholtz, 25–72. Working papers in linguistics. Cambridge: MIT.

    Google Scholar 

  • Déchaine, Rose-Marie, and Martina Wiltschko. 2002. Decomposing pronouns. Linguistic Inquiry 33(3): 409–442.

    Article  Google Scholar 

  • De Lancey, Scott. 1981. An interpretation of split ergativity and related patterns. Language 57(3): 626–657.

    Article  Google Scholar 

  • De Lancey, Scott. 2002. Deixis, topicality, and the inverse. Ms., University of Oregon.

  • den Dikken, Marcel. 2007. Phase extension: Contours of a theory of the role of head movement in phrasal extraction. Theoretical Linguistics 33: 1–41.

    Article  Google Scholar 

  • Frantz, Donald G. 2009. Blackfoot grammar. Toronto: University of Toronto Press.

    Google Scholar 

  • Georgi, Doreen. 2011. Deriving the distribution of person portmanteaux by relativizing probing. In North Eastern Linguistics Society (NELS) 42. Toronto, Canada.

    Google Scholar 

  • Guasch, S. J. Antonio. 1956. El idioma guaraní: Gramática, antologia de prosa y verso. Asunción: Casa America – Moreno Hermanos.

    Google Scholar 

  • Guasch, S. J. Antonio, and S. J. Diego Ortiz. 2008. Diccionario Castellano-Guaraní, Guaraní-Castellano. Asunción: Centro de Estudios Paraguayos “Antonio Guasch”.

    Google Scholar 

  • Hale, Kenneth, and Samuel Jay Keyser. 2002. Prolegomenon to a theory of argument structure. Cambridge: MIT Press.

    Google Scholar 

  • Harley, Heidi, and Elizabeth Ritter. 2002. Person and number in pronouns: A feature-geometric analysis. Language 78(3): 482–526.

    Article  Google Scholar 

  • Harley, Heidi. 2013. External arguments and the Mirror Principle: On the distinctness of Voice and v. Lingua 125: 34–57.

    Article  Google Scholar 

  • Heim, Irene. 1982. The semantics of definite and indefinite determiners. PhD diss., University of Massachusetts at Amherst.

  • Jouitteau, Mélanie, and Milan Rezac. 2006. Deriving the complementarity effect: Relativized minimality in Breton agreement. Lingua 116(11): 1915–1945.

    Article  Google Scholar 

  • Kato, Mary. 2003. Null objects and VP ellipsis in European and Brazilian Portuguese. In Romance languages and linguistic theory 2001, eds. Josep Quer, Jan Schroten, Mauro Scorretti, Petra Sleeman, and Els Verheugd, 131–153. Amsterdam: Benjamins.

    Chapter  Google Scholar 

  • Kayne, Richard. 1975. French syntax. Cambridge: MIT Press.

    Google Scholar 

  • Kiss, Katalin É. 2013. The inverse agreement constraint in Uralic languages. Finno-Ugric Languages and Linguistics 2(1): 2–21.

    Google Scholar 

  • Klaiman, Miriam H. 1992. Inverse languages. Lingua 88(3/4): 227–261.

    Article  Google Scholar 

  • Krivoshein de Canese, Natalia. 1998. Gramática de la lengua Guaraní. Asunción: Colección Ñemitỹ.

    Google Scholar 

  • Landau, Idan. 2010. The explicit syntax of explicit arguments. Linguistic Inquiry 41(3): 357–388.

    Article  Google Scholar 

  • Legate, Julie Anne. 2014. Voice and v: Lessons from Acehnese. Cambridge: MIT Press.

    Book  Google Scholar 

  • Lockwood, Hunter T., and Monica Macaulay. 2012. Prominence hierarchies. Language and Linguistics Compass 6(7): 431–446.

    Article  Google Scholar 

  • Macaulay, Monica. 2005. On the 2 > 1 prominence hierarchy of Algonquian. LSO Workings Papers in Linguistics 5: 1–24.

    Google Scholar 

  • Massam, Diane. 2001. Pseudo noun incorporation in Niuean. Natural Language and Linguistic Theory 19(1): 153–197.

    Article  Google Scholar 

  • McCloskey, James, and Kenneth Hale. 1984. On the syntax of person-number inflection in modern Irish. Natural Language and Linguistic Theory 1(4): 487–533.

    Google Scholar 

  • Nevins, Andrew. 2007. The representation of third person and its consequences for person-case effects. Natural Language and Linguistic Theory 25(2): 273–313.

    Article  Google Scholar 

  • Nevins, Andrew, and Filomena Sandalo. 2011. Markedness and morphotactics in Kadiwéu [+ participant] agreement. Morphology 21(2): 351–378.

    Article  Google Scholar 

  • Nichols, Lynn. 2001. The syntactic basis of referential hierarchy phenomena: Clues from languages with and without morphological case. Lingua 111(4–7): 515–537.

    Article  Google Scholar 

  • Payne, Doris. 1994. The Tupí-Guaraní inverse. In Voice: Form and function, eds. Paul Hopper and Barbara Fox, 313–340. Amsterdam: Benjamins.

    Chapter  Google Scholar 

  • Perlmutter, David. 1971. Deep and surface structure constraints in syntax. New York: Rinehart and Winston Inc.

    Google Scholar 

  • Preminger, Omer. 2014. Agreement and its failures. Cambridge: MIT Press.

    Book  Google Scholar 

  • Reichenbach, Hans. 1947. Elements of symbolic logic. New York: Macmillan.

    Google Scholar 

  • Ritter, Elizabeth, and Martina Wiltschko. 2014. The composition of INFL: An exploration of tense, tenseless languages, and tenseless constructions. Natural Language and Linguistic Theory 32(4): 1331–1386.

    Article  Google Scholar 

  • Rizzi, Luigi. 1982. Issues in Italian syntax. Dordrecht: Foris.

    Book  Google Scholar 

  • Shain, Cory, and Judith Tonhauser. 2011. The synchrony and diachrony of differential object marking in Paraguayan Guaraní. Language Variation and Change 22(3): 321–346.

    Article  Google Scholar 

  • Silverstein, Michael. 1976. Hierarchy of features and ergativity. In Grammatical categories in Australian languages, ed. Robert M. W. Dixon, 112–171. Canberra: Australian Institute of Aboriginal Studies.

    Google Scholar 

  • Tonhauser, Judith, and Erika Colijn. 2010. Word order in Paraguayan Guaraní. International Journal of American Linguistics 76(2): 255–288.

    Article  Google Scholar 

  • Tonhauser, Judith. 2010. Is Paraguayan Guaraní a tenseless language? In Semantics of Under-represented Languages of the Americas (SULA) 5, 227–242. Amherst: Graduate Linguistics Student Association.

    Google Scholar 

  • Tonhauser, Judith. 2011. Temporal reference in Paraguayan Guaraní, a tenseless language. Linguistics and Philosophy 34(3): 257–303.

    Article  Google Scholar 

  • Trommer, Jochen. 2008. Coherence in affix order. Zeitschrift für Sprachwissenschaft 27(1): 99–140.

    Article  Google Scholar 

  • Ura, Hiroyuki. 1996. Multiple feature-checking: A theory of grammatical function splitting. PhD diss., MIT.

  • Velázquez-Castillo, Maura. 1996. The grammar of possession: Inalienability, incorporation, and possessor ascension in Guaraní. Amsterdam: Benjamins.

    Book  Google Scholar 

  • Velázquez-Castillo, Maura. 2002. Guaraní causative constructions. In The grammar of causation and interpersonal manipulation, ed. Masayoshi Shibatani. Vol. 48 of Typological studies in language series, 507–534. Amsterdam: Benjamins.

    Chapter  Google Scholar 

  • Velázquez-Castillo, Maura. 2007. Voice and transitivity in Guaraní. In The typology of semantic alignment systems, eds. Mark Donohue and Søren Wichmann, 380–395. Oxford: Oxford University Press.

    Google Scholar 

  • Vergnaud, Jean-Roger, and Maria Luisa Zubizarreta. 1992. The definite determiner and the inalienable construction in French and English. Linguistic Inquiry 23(4): 595–652.

    Google Scholar 

  • Wechsler, Stephen. 2011. Mixed agreement, the person feature, and the index/concord distinction. Natural Language and Linguistic Theory 29(4): 999–1031.

    Article  Google Scholar 

  • Wechsler, Stephen, and Larisa Zlatić. 2000. A theory of agreement and its application to Serbo-Croatian. Language 76: 799–832.

    Article  Google Scholar 

  • Wiltschko, Martina. 2014. The universal structure of categories: Towards a formal typology. Cambridge: Cambridge University Press.

    Book  Google Scholar 

  • Woolford, Ellen. 1999. More on the anaphor agreement effect. Linguistic Inquiry 30(2): 257–287.

    Article  Google Scholar 

  • Woolford, Ellen. 2016. Two types of portmanteau agreement: Syntactic and morphological. In Optimality-theoretic syntax, semantics and pragmatics: From uni- to bidirectional optimization, eds. Géraldine Legendre, Michael Putnam, Henriëtte de Swart, and Erin Zaroukian, 111–135. Oxford: Oxford University Press.

    Chapter  Google Scholar 

  • Wurmbrand, Susi. 2012. Agree(ment): Looking up or looking down? Lecture given at Agreement Seminar, MIT.

  • Zubizarreta, Maria Luisa, and Roumyana Pancheva. 2017, To appear. A note on the syntax of possession in Paraguayan Guaraní. In Boundaries, phases, and interfaces. Linguistics today, eds. O. Fernández Soriano, Elena Castroviejo, and Isabel Pérez Jiménez. Amsterdam: Benjamins.

    Google Scholar 

Download references

Acknowledgements

We thank José Camacho, Liliana Sanchez, and students at the USC Syntax Seminar (Spring 2014) for useful feedback at the early stages of this project, as well as Laura Kalin, Stefan Keine, and audiences at the venues where versions of this paper were presented: 4th Cambridge Comparative Syntax Conference (Cambridge University, May 2015), 25th Colloquium on Generative Grammar (Bayonne, France, May 2015), VII Encuentro de Gramática Generativa (Buenos Aires, July 2015), University of Washington at Seattle (October 2015), Workshop on Universals and Variation in Syntax (Beijing Language and Culture University, March 2016), and Universidad Catolica Pontificia del Perú (Lima, July 2016). We also thank the five anonymous NLLT reviewers, and NLLT editor Julie Legate, for valuable comments that have helped us improve this article.

Author information

Authors and Affiliations

  1. University of Southern California, Grace Ford Salvatori 301, University Park, Los Angeles, CA, 90089-1693, USA

    Maria Luisa Zubizarreta & Roumyana Pancheva

Authors
  1. Maria Luisa Zubizarreta
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Roumyana Pancheva
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Maria Luisa Zubizarreta.

Appendix: Interactions of je-Voice, transitivizer mbo- and causative -uka

Appendix: Interactions of je-Voice, transitivizer mbo- and causative -uka

We discuss first the interaction of je-Voice and the transitivizing prefix mbo-. We then turn to their interaction with the causative suffix -(u)ka. These interactions lend support to a phase-level syntactic computation of these constructions, as proposed in Sect. 5. (The function of mbo- and uka- are also analyzed in Velázquez-Castillo (2002), within a functionalist perspective.)

1.1 A.1 The transitivizer mbo- and je-Voice

Recall that Paraguayan Guaraní has a transitivizing prefix mbo- that combines with intransitives to form transitive predicates.Footnote 43 Interestingly, the reflexive forms function as intransitives for the purpose of mbo- prefixation. To exemplify, consider the case of transitives with an incorporated inalienable possessor nP, like those in (23a) and (24a), repeated in (50). As shown in (51), the same meaning can be constructed by first reflexivizing the predicate and then transitivizing it; in fact, the forms in (51) are the preferred ones, especially if the short verb stem form is used.Footnote 44 (Recall that je+hova is pronounced as jova.)

  1. (50)
    figure ax
  1. (51)
    figure ay

In line with the P-centered analysis of reflexives and causatives outlined in Sect. 5, we propose the following formal requirement for mbo-prefixation:

  1. (52)

    mbo- combines with a p-intransitive predicate, where p-(in)transitivity is understood in terms of number of DP dependents: p-intransitives have at most one DP argument.

Defective arguments that lack the D layer do not count for determining p-intransitivity. The predicate headed by je- is therefore bi-valent (it has two arguments), but it is syntactically p-intransitive (i.e., it has only one DP argument).

The derivation for an example like (51b) proceeds as follows: 1) incorporation of the nP internal argument applies; 2) je-prefixation applies. As we have seen in the previous section, je-Voice imposes that the external argument of v be an nP. Since je-Voice carries an EPP feature, it furthermore triggers object promotion to its edge, which, in the case under discussion, is the 1sg Possessor argument (che) of the incorporated inalienable nP. From that position, the Possessor argument binds the nP external argument of v. The output is as in (53), where bolds indicate the chain obtained via promotion of the Possessor argument and underlines indicate the binding of the external argument of v by the raised object.

  1. (53)

    [v che [ je-v [vP nP(x) v [VP [NP (che) rova ] héi ]… ]]]

The structure in (53) then merges with the transitivizer mbo-. Mbo- introduces a DP external argument and it enters into an agreement relation with the highest DP within its domain, namely with the 1sg Possessor DP (che) located at the edge of je-Voice. Since the external argument introduced by mbo- is a 2sg external argument, the P-constraint forces the Possessor argument (che) to move to the edge of mbo-; see (54a). The output structure combines with Infl. Due to the P-constraint, Infl agrees with 1sg Possessor DP (che) rather than with the 2sg external argument. To comply with the P-constraint, it must furthermore move to the edge of Infl, giving rise to the final structure, as shown in (54b).

  1. (54)
    figure az

The combination of mbo- with the reflexive structure illustrated above shows that object promotion can be triggered by two independent factors: by the EPP property of je-Voice (from where the object binds the external argument) and by the P-constraint, which forces the object to move further up to the edge of mbo- and eventually to the edge of Infl. (Note that this provides a further argument for object promotion triggered by two independent factors: (1) the EPP feature on je-Voice and (2) the P-constraint.)

1.2 A.2 Adding the causative suffix -(u)ka

The complex output structure in (54b) can furthermore combine with causative -(u)ka, which introduces the Causer external argument and triggers further promotion of the Possessor argument (che) to the edge of the causative -(u)ka, as illustrated below.

  1. (55)
    figure ba

1.3 A.3 Reflexivizing causative structures

Even more complex forms can be constructed by merging the reflexive je- for the second time to the output of the causative structures in (55), generating the forms in (56). (je → ñe due to the nasality of mbo-). As indicated by the gloss, reflexivization has applied twice. The Possessor object promoted to the edge of je-Voice is bound by the Causer external argument introduced by -(u)ka, while the external argument introduced by the transitivizer mbo- (the Causee) is bound by the Dat PP. As we have seen in 5.2., the Causee is a person-less argument bound by the Dative PP (when present); therefore, it does not interfere in the binding of the Possessor by the Causer.

  1. (56)
    figure bb

It is quite remarkable how complex these forms are, created via multiple cyclical syntactic processes: je-reflexivation, mbo-prefixation, causativization, and a second round of je-reflexivation. This complexity beautifully illustrates the computational nature of the mechanisms that underlie such forms and confirms the predictions of our analysis.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Zubizarreta, M.L., Pancheva, R. A formal characterization of person-based alignment. Nat Lang Linguist Theory 35, 1161–1204 (2017). https://doi.org/10.1007/s11049-016-9357-5

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s11049-016-9357-5

Keywords

Search

Navigation