Acessibilidade / Reportar erro

INTRODUCING ARGUMENTS TO THE KARITIANA VERBAL STRUCTURE

ABSTRACT:

This paper investigates how Karitiana (a Tupian language, Arikém branch, located in Rondônia, Brazil; population: 396 people - 333 speakers) introduces additional arguments into the basic verbal structures. In order to analyze and discuss the data, we adopted two theoretical frameworks, i.e., (i) the Hale and Keyser (2002)HALE, K. L.; KEYSER, S. J. Prolegomenon to a theory of argument structure. Cambridge, MA: MIT Press, 2002.'s theory of argument structure and (ii) the theory of applicative heads, elaborated by Pylkkänen (2008)PYLKKÄNEN, L. Introducing arguments. Cambridge, MA: MIT Press, 2008.. In the former proposal, Karitiana displays four verbal classes (intransitive, intransitive with oblique objects, transitive and ditransitive verbs) while, in the latter, the language has only two verbal classes (intransitive and transitive verbs). Adopting Pylkkänen's framework, the Karitiana verbal system exhibits two basic structures in which applicative or voice nuclei add additional arguments. In causatives, the voice nuclei only introduce higher arguments into the structures. However, applicative heads are divided into two types – high and low. The higher head inserts an extra argument above the VP, while the lower one inserts the argument internally into the VP.

KEYWORDS:
Argument Structure; Applicative Heads; Verbal Classes; Karitiana Language

RESUMO:

O artigo discute o processo de introdução de argumentos adicionais à estrutura básica do verbo na língua Karitiana (língua da família Tupi, subfamília Arikém, localizada em Rondônia, com uma população de 396 indivíduos e 333 falantes). Adota-se, para análise e discussão dos dados, duas propostas teóricas (i) a teoria de estrutura argumental de Hale e Keyser (2002)HALE, K. L.; KEYSER, S. J. Prolegomenon to a theory of argument structure. Cambridge, MA: MIT Press, 2002. e (ii) a proposta teórica de núcleos aplicativos de Pylkkänen (2008)PYLKKÄNEN, L. Introducing arguments. Cambridge, MA: MIT Press, 2008.. Na proposta teórica (i), Karitiana apresenta quatro tipos de verbos (1 - intransitivo, 2 - intransitivo com objeto oblíquo, 3 - transitivo e 4 - bitransitivo), já na segunda proposta, por sua vez, mais econômica, a língua é analisada como tendo apenas dois tipos verbais (1 - intransitivo e 2 - transitivo). Nesta última, o sistema verbal Karitiana apresenta duas estruturas básicas em que argumentos extras são incrementados por meio de núcleos aplicativos e Voice. Este adiciona argumentos altos na estrutura em estruturas causativas. Os aplicativos são de dois tipos, isto é, o alto que insere um argumento extra acima do VP e o baixo que adiciona argumentos extras abaixo ou internamente ao VP.

PALAVRAS-CHAVE:
Estrutura Argumental; Aplicativos; Classes Verbais; Língua Karitiana

Introduction

This article offers a theoretical and empirical analysis of the argument structure of the Karitiana verbs based on two approaches developed by (i) Hale and Keyser (2002)HALE, K. L.; KEYSER, S. J. Prolegomenon to a theory of argument structure. Cambridge, MA: MIT Press, 2002. and (ii) Pylkkänen (2008)PYLKKÄNEN, L. Introducing arguments. Cambridge, MA: MIT Press, 2008.. The former describes how verbal nuclei project their arguments, and the latter shows how the basic structure of the verb introduces their additional arguments.

The Karitiana is an endangered language, which belongs to the Arikem family, Tupian stock, located in the state of Rondônia, in Brazil. In 2016, the language comprised 333 speakers out of 396 people (ROCHA, 2018ROCHA, I. Levantamento da situação sociolinguística da língua Karitiana. In: INVENTÁRIO NACIONAL DA DIVERSIDADE LINGUÍSTICA [INDL/IPHAN/MPEG]. Levantamento regional da situação sociolinguística de 26 etnias indígenas da região de Rondônia. Belém, PA: Museu Paraense Emílio Goeldi, 2018. p. 1-67.).

Morphosyntactic tests

In this section, we show four morphosyntactic tests used to identify valence change and verbal classes in Karitiana. The mentioned tests are (i) agreement pattern; (ii) causativization; (iii) passivization, and (iv) the copular complement as a predicated syntactically intransitive.

Agreement pattern as a morphosyntactic test

Karitiana is an ergative-absolutive language in which the verbal agreement pattern1 1 The definition of agreement used is from the Case Binding theory which says that a functional nucleus agrees with different arguments in different syntactic configurations (STORTO, 1999. 154; BITTNER; HALE, 1996). refers to the absolutive argument, e.g., the verb agrees with the object in transitive sentences or with the subject in intransitive sentences (LANDIN, 1984LANDIN, D. An outline of the syntactic structure of Karitiana sentences. In: DOOLEY, R. A. Estudos sobre línguas Tupi do Brasil. Brasília: SIL Press, 1984. p. 219-254. (Série Lingüística, 11.).; STORTO, 1999STORTO, L. Aspects of a Karitiana grammar. 1999. Thesis (Ph.D. in Linguistics) - Massachusetts Institute of Technology, Cambridge, MA, 1999., 2002STORTO, L. Algumas categorias funcionais em Karitiana. In: ANPOLL. Línguas indígenas brasileiras: fonologia, gramática e história. v.1. Belém: Ed. da UFPA, 2002. p.151-164., 2005STORTO, L. Caso e concordância nas línguas tupi. Estudos Linguísticos, São Paulo, v. 34, p. 59-72, 2005., 2010STORTO, L. Copular constructions in Karitiana: a case against case movement. MIT Occasional Papers in Linguistics, Cambridge, v. 41, p. 205-226, 2010.).

Tables (1) and (2) show that the agreement pattern in Karitiana forms a method of identification of verbal classes, once the agreement marking is sensitive to the valence of the verb: transitive or ditransitive verbs agree with their objects while intransitive verbs accord with their unique arguments, that are, the subjects (ROCHA, 2011ROCHA, I. A estrutura argumental da língua Karitiana: desafios descritivos e teóricos. 220f. 2011. Dissertação (Mestrado em Semiótica e Linguística Geral) - Faculdade de Filosofia, Letras e Ciências Humanas, Universidade de São Paulo, São Paulo, 2011.; STORTO, 2001STORTO, L. Duas classes de verbos intransitivos em Karitiana (família Arikém, tronco Tupí). In: QUEIXALOS, F. (éd.). Des noms et des verbes en Tupi-Guarani: état de la question. Paris: Lincom-Europa, 2001. p. 163-180.; STORTO; ROCHA, 2015STORTO, L.; ROCHA, I. Estrutura argumental da língua Karitiana. In: STORTO, L.; FRANCHETTO, B.; LIMA, S. (ed.). Sintaxe e semântica do verbo em línguas indígenas do Brazil. Campinas: Mercado de Letras, 2015. p. 17-42.).

Table 1
The transitive verb agrees with the object (declarative mood)
Table 2
The intransitive verb agrees with the subject (assertive mood)

Causativization/transitivization as a morphosyntactic test

The causative morpheme {m-} is allowed only with intransitive roots. This morpheme works increasing the valence of the verb, and creating a space in the structure for the realization of an external argument. This fact, therefore, constitutes a morphosyntactic test capable of identifying whether a verb is intransitive or not, since if this morpheme is attached to either a transitive or ditransitive verb, the construction fails.

  • (1)

    Causativization of intransitive verbs (assertive mood)

    1. ø-pyry-som-<y>n asyryty 3-ASSERT-ripen/redden-NFUT banana “The banana ripened.”

    2. ø-pyry-m-som-<y>n asyryty taso 3-ASSERT-CAUS-ripen-NFUT banana man “The man ripened the banana.”

  • (2)

    Causativization of intransitive verbs (assertive mood)

    1. y-pyry-tarak-<a>dn yn 1-ASSERT-walk-NFUT I “I walked.”

    2. y-pyry-m-tarak-<a>n yn taso 1-ASSERT-CAUS-walk-NFUT I man “The man made me walk.”

    3. ø-pyry-m-tarak-<a>n taso yn 3-ASSERT-CAUS-walk-NFUT man I “I made the man walk.”

  • (3)

    Causativization of intransitive verbs with oblique objects

    1. y-py-so'oot-<y>n yn pikom-ty 1-ASSERT-see-NFUT I monkey-OBL “I saw the monkey.”

    2. y-py-m-so'oot-<y>n yn õwã pikom-ty 1-ASSERT-CAUS-see-NFUT I child monkey-OBL “The child made me see the monkey.”

  • (4)

    Agramaticality of a transitive verb with {m–}

    1. ø-pyr-oky-dn boroja taso 3-ASSERT-kill-NFUT snake man “The man killed the snake.”

    2. *ø-py-m-oky-dn boroja taso 3-ASSERT-CAUS-kill-NFUT snake man Intended interpretation: “The man killed the snake.”

  • (5)

    Agramaticality of a ditransitive verb with {m–}

    1. y-pyry-hit-<y>n boet-<e>ty taso 3-ASSERT-give-NFUT necklace-OBL man “The man gave me the necklace.”

    2. *y-pyry-m-hit-<y>n boet-<e>ty taso 1-ASSERT-CAUS-give-NFUT necklace-OBL man

Impersonal passives with the morpheme {a–} as a test

The impersonal passive construction forms the third morphosyntactic evidence to identify verbal classes. The impersonal passive morpheme {a-} is only used with verbs syntactically transitive. Affixing the morpheme {a-} to intransitive verbs is not, therefore, permitted in Karitiana (ROCHA, 2011ROCHA, I. A estrutura argumental da língua Karitiana: desafios descritivos e teóricos. 220f. 2011. Dissertação (Mestrado em Semiótica e Linguística Geral) - Faculdade de Filosofia, Letras e Ciências Humanas, Universidade de São Paulo, São Paulo, 2011.; STORTO, 2001STORTO, L. Duas classes de verbos intransitivos em Karitiana (família Arikém, tronco Tupí). In: QUEIXALOS, F. (éd.). Des noms et des verbes en Tupi-Guarani: état de la question. Paris: Lincom-Europa, 2001. p. 163-180.; STORTO; ROCHA, 2015STORTO, L.; ROCHA, I. Estrutura argumental da língua Karitiana. In: STORTO, L.; FRANCHETTO, B.; LIMA, S. (ed.). Sintaxe e semântica do verbo em línguas indígenas do Brazil. Campinas: Mercado de Letras, 2015. p. 17-42.).

  • (6)

    Passivization of transitive verbs is allowed

    1. ø-pyr-oky-dn boroja taso 3-ASSERT-kill-NFUT snake man “The man killed the snake.”

    2. ø-pyr-a-oky-dn boroja 3-ASSERT-PASV-kill-NFUT snake “The snake was killed.”

  • (7)

    Passivization of ditransitive verbs is allowed

    1. y-pyry-hit-<y>n boet-<e>ty taso 1-ASSERT-give-NFUT necklace-OBL man “The man gave a necklace to me.”

    2. y-pyr-a-hit-<y>n boet-<e>ty 1-ASSERT-PASV-give-NFUT necklace-OBL “The necklace was given to me.”

Examples (6a) and (7a) can have the valence decreased by using the morpheme {a-}. The external argument of the verb is deleted from the structure, and it cannot be recovered through the by-phrase expression. The result of this operation can be seen in examples (6b) and (7b), respectively, in which the verbs were passivized/intransitivized.

  • (8)

    Passivization of an intransitive verb is not allowed

    1. ø-pyry-som-<y>n asyryty 3-ASSERT-ripen/redden-NFUT banana “The banana ripened.”

    2. *ø-pyry-a-som-<y>n asyryty 3-ASSERT-PASV-ripen/redden-NFUT banana Intended: “The banana was ripened.”

  • (9)

    Passivization of an intransitive verb is not allowed

    1. y-pyry-tarak-<a>dn yn 1-ASSERT-walk-NFUT I “I walked.”

    2. *y-pyry-a-tarak-<a>dn yn 1-ASSERT-PASV-walk-NFUT I Intended: “I was walked.”

  • (10)

    Passivization of a perception intransitive verb is also not allowed

    1. y-py-so'oot-<y>n yn pikom-ty 1-ASSERT-see-NFUT I monkey-OBL “I saw the monkey.”

    2. *y-pyr-a-so'oot-<y>n yn pikom-ty 1-ASSERT-PASV-see-NFUT I monkey-OBL Intended: “The monkey was seen.”

On the other hand, intransitive verbs, as (8a), (9a), and (10a), do not allow the passivization with {a-} directly to the verbal stems. The ungrammatical examples (8b), (9b), and (10b) show the evidence that passivization is not allowed in those cases.

Passivization of intransitive verbs causativized

Once the intransitive verbs are causativized, as seen in (1b), (2b), and (3b), they will allow the use of the passive morpheme, as seen in (11), (12), and (13). Examples (8b), (9b) and (10b) are ungrammatical because the rule of passivization is that the morpheme {a-} only operates over a verb that has at least two arguments, including the external one or the causer argument.

  • (11)

    ø-pyry-a-m-som-<y>n asyryty 3-ASSERT-PASV-CAUS-ripen/redden-NFUT banana “Someone/something made the banana ripen.”

  • (12)

    y-pyry-a-m-tarak-<a>n yn 1-ASSERT-PASV-CAUS-walk-NFUT I “Someone/something made me walk.”

  • (13)

    y-pyr-a-m-so'oot-<y>n yn pikom-ty 1-ASSERT-PASV-CAUS-see-NFUT I monkey-OBL “Someone/something made me see the monkey.”

As one can see, the presence of the causative morpheme m- is evidence that the structures in (11-13) were first transitivized, in turn, to permit the passivization. Indeed, all Karitiana intransitive verbs have the same pattern shown in (11-13), as described by Rocha (2011)ROCHA, I. A estrutura argumental da língua Karitiana: desafios descritivos e teóricos. 220f. 2011. Dissertação (Mestrado em Semiótica e Linguística Geral) - Faculdade de Filosofia, Letras e Ciências Humanas, Universidade de São Paulo, São Paulo, 2011. and Storto and Rocha (2015)STORTO, L.; ROCHA, I. Estrutura argumental da língua Karitiana. In: STORTO, L.; FRANCHETTO, B.; LIMA, S. (ed.). Sintaxe e semântica do verbo em línguas indígenas do Brazil. Campinas: Mercado de Letras, 2015. p. 17-42., in which an external argument must be added to these types of verbs via the morpheme m-.

Copular constructions as a morphosyntactic test

Storto (2010)STORTO, L. Copular constructions in Karitiana: a case against case movement. MIT Occasional Papers in Linguistics, Cambridge, v. 41, p. 205-226, 2010. has shown that the nucleus of the small clause in copular constructions in Karitiana is sensitive to the verbal valence since copular sentences would be ungrammatical if we put a syntactically transitive verb in this environment. Copular constructions comprise our fourth morphosyntactic test to identify verbal classes (ROCHA, 2011ROCHA, I. A estrutura argumental da língua Karitiana: desafios descritivos e teóricos. 220f. 2011. Dissertação (Mestrado em Semiótica e Linguística Geral) - Faculdade de Filosofia, Letras e Ciências Humanas, Universidade de São Paulo, São Paulo, 2011.; STORTO, 2001STORTO, L. Duas classes de verbos intransitivos em Karitiana (família Arikém, tronco Tupí). In: QUEIXALOS, F. (éd.). Des noms et des verbes en Tupi-Guarani: état de la question. Paris: Lincom-Europa, 2001. p. 163-180.; STORTO; ROCHA, 2015STORTO, L.; ROCHA, I. Estrutura argumental da língua Karitiana. In: STORTO, L.; FRANCHETTO, B.; LIMA, S. (ed.). Sintaxe e semântica do verbo em línguas indígenas do Brazil. Campinas: Mercado de Letras, 2015. p. 17-42.).

  • (14)

    Intransitive verbs as head of the small clause in copular constructions

    1. Yn ø-na-aka-t i-tarak-<a>t I 3-DECL-COP-NFUT PART-walk-ADVZR “I walked/ I am walking.”

    2. Yn ø-na-aka-t i-otam-ø I 3-DECL-COP-NFUT PART-arrive-ADVZR “I arrived/ I am arriving.”

    3. Yn ø-na-aka-t i-so'oot-ø pikom-ty I 3-DECL-COP-NFUT PART-see-ADVZR monkey-OBL “I saw the monkey/ I am seeing the monkey.”

The group of examples in (14) shows that we can put any intransitive verbs into the small clause of the copular construction, as seen in (a) tarak ‘walk’, (b) otam ‘arrive’, and (c) so´oot ‘see.’

  • (15)

    Transitive verbs as head of the small clause in copular constructions

    1. *Yn ø-na-aka-t i-oky-t (boroja) I 3-DECL-COP-NFUT PART-kill-ADVZR snake Intended: “I killed/ am killing the monkey.”

    2. *Yn ø-na-aka-t i-'y-t (asyryty) I 3-DECL-COP-NFUT PART-eat-ADVZR banana Intended: “I ate/ am eating the banana.”

    3. *Yn ø-na-aka-t i-hit-ø(pikom) (asyryty-ty) I 3-DECL-COP-NFUT PART-give-ADVZR monkey banana-OBL Intended: I gave/ am giving the bananas to the monkey.”

Contrarily, transitive verbs are not allowed in this syntactic context as seen by the ungrammatical examples in (15) in which we elicited the copular construction with transitive verbs heading the small clauses, as seen in (a) oky ‘kill’, (b) ´y ‘eat’, and (c) hit ‘dar.’

  • (16)

    Transitive verbs passivized heading the small clause of copular construction

    1. boroja ø-na-aka-t i-a-oky-t snake 3-DECL-COP-NFUT PART-PASV-kill-ADVZR “The snake was killed.”

    2. asyryty ø-na-aka-t i-a-'y-ø banana 3-DECL-COP-NFUT PART-PASV-eat-ADVZR “The banana was eaten.”

    3. asyryty ø-na-aka-t i-a-hit-ø (pikom) banana 3-DECL-COP-NFUT PART-PASV-give-ADVZR macaco “The banana was given to the monkey.”

Examples (16), however, confirm that intransitive verbs after being passivized can occupy the position of copular complement. The verbs oky, ´y, and hit, seen in (15), were first intransitivized, in turn, to be able to head this syntactic environment. This fact justifies the use of copular sentences as a linguistic test.

Table (5) summarizes all four tests applied to identify verbal classes in the language. The first column shows the valence of the verb. The second one shows the test 1 related to the agreement. The third one tests whether a verb can or not be causativized with {m-} (test 2). The fourth one examines whether or not a verb can be passivized (test 3). Moreover, the fifth column shows which type of verb can occur in the small clause complement of the copular verb.

Table 5
Summary of the morphosyntactic tests

Hale e Keyser (2002)HALE, K. L.; KEYSER, S. J. Prolegomenon to a theory of argument structure. Cambridge, MA: MIT Press, 2002. and Pylkkänen (2008)PYLKKÄNEN, L. Introducing arguments. Cambridge, MA: MIT Press, 2008.'s theories of argument structure

Hale and Keyser (2002)HALE, K. L.; KEYSER, S. J. Prolegomenon to a theory of argument structure. Cambridge, MA: MIT Press, 2002.

According to Hale and Keyser (2002)HALE, K. L.; KEYSER, S. J. Prolegomenon to a theory of argument structure. Cambridge, MA: MIT Press, 2002., the possibility or impossibility of the transitivity alternation process is related to the properties of the verbal roots: if they are predicative or not, e.g., if the roots require an internal argument in their structure. Consequently, the syntactic behavior of a verb depends on the properties of the elements with which this verb is comprised. The predicative roots (R) project an internal argument in the structure, whose position occurs in the specifier of V2 in verbal structures of unaccusatives.

  • (17)

    Structural projections of alternating verbs:

As illustrated in (17), the complement of the head is the root (R). The head (V) functions as a verbalizer of R, that requires an internal specifier (Spec) projected via the head. The notion of the internal specifier, in this approach, is quite essential once this specifier is a pivot in the alternation causative-inchoative.

Transitive counterpart pivot Intransitive counterpart Someone/something broke the pote . broke.

The conflation operation is a process used in Hale and Keyser (2002)HALE, K. L.; KEYSER, S. J. Prolegomenon to a theory of argument structure. Cambridge, MA: MIT Press, 2002.'s theory of argument structure, and it is similar to Incorporation in Baker (1988a)BAKER, M. C. Incorporation: A theory of grammatical function changing. Chicago: University of Chicago Press. 1988a.. However, the former is more constrained than the latter. Conflation can be understood as a process of fusion of two nuclei, or a nucleus and its complement; in both cases, they should be in a relation of mutual c-command. Furthermore, the nucleus according to which the complement conflates should be empty or phonologically defective. This operation is used both in verbal derivation from a root (R) and in the process of valence increase of the verb (HALE; KEYSER, 2002HALE, K. L.; KEYSER, S. J. Prolegomenon to a theory of argument structure. Cambridge, MA: MIT Press, 2002., 12).

Conflation is restricted to the process according to which the phonological matrix of a head of a complement C is introduced into the empty phonological matrix of a head that selects (and is accordingly sister to) C (HALE; KEYSER, 2002HALE, K. L.; KEYSER, S. J. Prolegomenon to a theory of argument structure. Cambridge, MA: MIT Press, 2002., p. 12).

Conflation consists in the process of copying the p-signature of the complement into the p-signature of the head, where the latter is defective (HALE; KEYSER 2002HALE, K. L.; KEYSER, S. J. Prolegomenon to a theory of argument structure. Cambridge, MA: MIT Press, 2002., p. 63).

In Hale and Keyser's proposal, there are four types of structural projections to describe verbs in natural languages: monadic, composite dyadic, basic dyadic, and atomic structures. In this paper, we discuss only the first three, once these structures are empirically found in Karitiana, and not the fourth one.

In the monadic structure, the nucleus V projects only its complement so that there is no projection of a specifier. This structure describes both transitive and unergative verbs (HALE; KEYSER, 2002HALE, K. L.; KEYSER, S. J. Prolegomenon to a theory of argument structure. Cambridge, MA: MIT Press, 2002.).3 3 Inergative verbs do not have internal arguments. Once they do not project internal arguments, there is no nominative case assignment to the external argument (BURZIO, 1986). Regarding this fact, Hale and Keyser (2002) assume that these verbs do not project an internal specifier in their structures. Therefore, they do not participate in the causative-inchoactive alternation because of the lack of an internal specifier. The internal specifier will function as the subject in an intransitive version of an unaccusative verb, and as the object in the transitive version of those verbs.

The composite dyadic structure displays two nuclei V, one upper (V1) and another lower (V2). The latter derives the intransitive version of the alternating verb while the former generates the transitive counterpart of the verb. As mentioned, the root (R) of alternating verbs functions as a predicator, which forces the lower nucleus V2 to project the internal specifier in the structure. This position bears the element that works as the subject of the intransitive version or the object of the transitive version (HALE; KEYSER, 2002HALE, K. L.; KEYSER, S. J. Prolegomenon to a theory of argument structure. Cambridge, MA: MIT Press, 2002.).

The basic dyadic shows a P-head projecting its complement and the specifier. However, to become a verbal structure, this P-head must be taken as a complement by a monadic structure (HALE; KEYSER, 2002HALE, K. L.; KEYSER, S. J. Prolegomenon to a theory of argument structure. Cambridge, MA: MIT Press, 2002.).

The structural representations concern the relations among the heads, specifiers, and complements. According to Hale and Keyser (2002)HALE, K. L.; KEYSER, S. J. Prolegomenon to a theory of argument structure. Cambridge, MA: MIT Press, 2002., these relations allowed the composition of specific lexical structures.

Pylkkänen (2008)PYLKKÄNEN, L. Introducing arguments. Cambridge, MA: MIT Press, 2008.

According to Pylkkänen (2008)PYLKKÄNEN, L. Introducing arguments. Cambridge, MA: MIT Press, 2008., the majority of the languages uses the strategy of applicative heads to add an indirect object to the basic structure of a verb. This possibility was strongly attested in Bantu languages. In the literature of these languages, additional arguments are called applied arguments, and the resulting constructions called applied constructions.

Pylkkänen (2008)PYLKKÄNEN, L. Introducing arguments. Cambridge, MA: MIT Press, 2008. point out that English (example 19) and Chaga (example 20) show double object constructions with benefactive applied arguments. However, from those two languages, only Chaga allows to add a benefactive participant to unergative verbs, as the following examples:

  • (19)

    Applied arguments in English

    1. I baked a cake. (transitive verb)

    2. I baked him a cake.

    3. I ran (unergative verb)

    4. *I ran him

    (PYLKKÄNEN, 2008PYLKKÄNEN, L. Introducing arguments. Cambridge, MA: MIT Press, 2008., ex. 1, p. 11))

  • (20)

    Unergative verbs with applied arguments in Chaga

    1. N-ä-ï-lì-í-à m-kà k-élyá FOC-1SUB-PRES-eat-APPL-FV C1-wife C7-food “He is eating food for his wife.

    2. N-ä-i-zrìc-í-à m-bùyà FOC-1SUB-PRES-run-APPL-FV C9-friend “He is running for a friend.” (BRESNAN; MOSHI, 1990 apudPYLKKÄNEN, 2008, p. 11)

As stated by Pylkkänen, the semantic similarities between English and Chaga benefactives are not valid, once Chaga applied head relates a participant to the event described by the VP. By contrast, the English applied nucleus correlates an entity to the direct object (PYLKKÄNEN, 2008PYLKKÄNEN, L. Introducing arguments. Cambridge, MA: MIT Press, 2008.).

The author claims that, cross-linguistically, these applied constructions can be divided into two different types, as presented by the asymmetry in examples (19) and (20).

In this theoretical approach, syntactic heads combine with their complements and specifiers compositionally through the rules of Functional Application and Predicate Modification (PYLKKÄNEN, 2008PYLKKÄNEN, L. Introducing arguments. Cambridge, MA: MIT Press, 2008.):

  • (21)

    High applicative construction

The applied argument ‘wife,’ in (21), is the benefactive, which is related to the event ‘eating,’ but is not sharing any features with ‘food,’ the object of ‘eating.’ The same reasoning is true for the instrumental applicatives in Chichewa, as in (22), in which ‘knife’ has an instrumental relationship with the event of ‘molding,’ not with ‘waterpot’ (PYLKKÄNEN, 2008PYLKKÄNEN, L. Introducing arguments. Cambridge, MA: MIT Press, 2008., p. 12).

Instrumental applicative in Chichewa

  • (22)

    Mavuto a-na-umb-ir-a mpeni mtsuko Mavuto SP-PAST-mold-APPL-ASP knife waterpot “Mavuto molded the waterpot with a knife.” (BAKER, 1988b, p. 353)

In English, however, the interpretation in which the applied argument does not have a relation with the direct object in the double object constructions is not possible (cf. examples (23) and (24)).

  • (23)

    *He ate the wife food.

  • (24)

    *John held Mary the bag.

In the following section, we will present two analyses based on the two theoretical proposals seen in this section.

The Karitiana verbal classes under Hale and Keyser (2002)HALE, K. L.; KEYSER, S. J. Prolegomenon to a theory of argument structure. Cambridge, MA: MIT Press, 2002.'s approach

The Karitiana verbs can be basically described into 3 classes correlated to the 3 structures proposed by Hale and Keyser (2002)HALE, K. L.; KEYSER, S. J. Prolegomenon to a theory of argument structure. Cambridge, MA: MIT Press, 2002.: (i) the alternating intransitive verbs are described by the composite dyadic; (ii) the transitive verbs described by the monadic; and (iii) the ditransitive verbs described by the basic dyadic structures (ROCHA, 2011ROCHA, I. A estrutura argumental da língua Karitiana: desafios descritivos e teóricos. 220f. 2011. Dissertação (Mestrado em Semiótica e Linguística Geral) - Faculdade de Filosofia, Letras e Ciências Humanas, Universidade de São Paulo, São Paulo, 2011.; STORTO, 2001STORTO, L. Duas classes de verbos intransitivos em Karitiana (família Arikém, tronco Tupí). In: QUEIXALOS, F. (éd.). Des noms et des verbes en Tupi-Guarani: état de la question. Paris: Lincom-Europa, 2001. p. 163-180.; STORTO; ROCHA, 2015STORTO, L.; ROCHA, I. Estrutura argumental da língua Karitiana. In: STORTO, L.; FRANCHETTO, B.; LIMA, S. (ed.). Sintaxe e semântica do verbo em línguas indígenas do Brazil. Campinas: Mercado de Letras, 2015. p. 17-42.). We highlight the fact that Karitiana does not distinguish unergative from unaccusative verbs, once all morphosyntactic tests, which were applied to intransitive verbs, have shown that they have the same syntactic behavior in this language. Different from Karitiana, Juruna (another Tupian language) differentiates unergative from unaccusative intransitive verbs so that it displays two different causative morphemes: the causative {ma-} used with unaccusative verbs in the pars (25) and (26), and {ũ-} used with unergative verbs in (17) and (28) (FARGETTI, 2001FARGETTI, C. M. Estudo fonológico e morfossintático da língua Juruna. 317f. 2001. Tese (Doutorado em Linguística) – Instituto de Estudos da Linguagem, Universidade Estadual de Campinas, Campinas, 2001.; LIMA, 2008LIMA, S. O. de. A estrutura argumental dos verbos na língua Juruna (Yudja): da formação dos verbos para a análise das estruturas sintáticas. 297f. 2008. Tese (Doutorado em Semiótica e Linguística Geral) - Faculdade de Filosofia, Letras e Ciências Humanas, Universidade de São Paulo, São Paulo, 2008.).

Alternating unaccusative

  • (25)

    anïiyu

    3SG.sleep

    “He slept.”

  • (26)

    Sewakii-ma-iyu he anu Sewaki 3SG-CAUS-sleep he ASP “Sewaki made him sleep.” (Lit. ´Sewaki slept him.')

Unergative

  • (27)

    tahuapï

    run dog

    “Dog ran.”

  • (28)

    una apï y-ũ-tahu anu I dog 3SG-CAUS-run ASP “I made the dog run.” (FARGETTI, 2001, p. 186-190)

Examples (25-28) demonstrate that Juruna exhibits two different structures to describe the intransitive verbs: (i) monadic to describe the unergative and (ii) composite dyadic to describe alternating unaccusative verbs.

Composite dyadic verbs: unaccusatives

The composite dyadic structure describes all intransitive verbs in Karitiana. In (31), we show the argumental projection of an alternating unaccusative verb in which a predicative root (R), as otam ‘arrive’ and terekterek ‘dance,’ requires a specifier projected by the head V2. In this case, the specifier is depicted by the NP taso ‘man.’

Intransitive verbs derivation: otam and terekterek

  • (29)

    ø-na-otam-ø taso 3-DECL-arrive-NFUT man “The man arrived.”

  • (30)

    ø-na-terek.teregng-ø taso 3-DECL-dance-NFUT man “The man danced.”

Pre-conflation

In (31), we show the pre-conflation configuration, in which the root is in the position R (complement of the verbalizing head V2), where a root such as otam and terekterek projects its internal specifier (the NP taso) bridged by the head in a composite dyadic structure.

  • (31)

    Composite dyadic structure deriving unaccusative verbs from a predicative root (R)

First conflation operation

In (32), the first conflation operation occurs to form the intransitive verb version, as otam in (29) and terekterek in (30). Thus, the root R combines with the verbalizing head V2 to create the verb from the root.

  • (32)

    Unaccusative verbal structure

The following examples exhibit the transitive counterparts of the data in (29) and (30), respectively:

Deriving transitive verbs from otam and terekterek

  • (33)

    Maria ø-na-mb-otam-ø João Maria 3-DECL-CAUS-arrive-NFUT João “Maria made João arrive.” (Lit. <Maria arrived João>)

  • (34)

    Yn ø-na-m-terekteregng-ø taso I 3-DECL-CAUS-dance-NFUT man “I made the man dance.” (Lit. <I danced the man>)

Second conflation operation

In structures (35a, b), from example (33) and (43), respectively, the transitive counterpart of the verb otam and terekterek is derived using the second conflation operation. In this level of the process, the language displays a causative morpheme {m-} attached to the verb.

  • (35)

    Transitive counterpart: structure

This operation comprises the combination of the head V1 with the head V2. In this type of structure, the NP, that worked as subject of the intransitive version, functions as the object of the transitive version, as seen in examples (33) and (34) from (29) and (30), respectively.

Composite dyadic verbs with oblique object

Among the alternating unaccusative verbs in Karitiana, there is an intransitive subclass that allows a projection of an optional indirect object. Morphosyntactically, this subclass passes the same tests applied to the verbs otam and terekterek. We describe and analyze the following verbs as intransitive with optional object: a'ak ‘like sexually’; hõroj ‘lie’; kãrã ‘be jealous’; koro'oppasap ‘miss’; opihok ‘escutar’; opiso ‘ouvir’; paket ‘feel disgusted’; pasadn ‘love’; pi ‘fear, dread, be afraid of’; pypyyt ‘know, have the ability’; pyso ‘touch/feel (with hands/fings)’; pytimokokit‘get motion sickness, sicken, nauseate’; pytim ‘want’; so'oot ‘ver’; sondyp ‘know’; and tirira ‘shake. shiver’ (ROCHA, 2011ROCHA, I. A estrutura argumental da língua Karitiana: desafios descritivos e teóricos. 220f. 2011. Dissertação (Mestrado em Semiótica e Linguística Geral) - Faculdade de Filosofia, Letras e Ciências Humanas, Universidade de São Paulo, São Paulo, 2011.). For this analysis and discussion, we adopt the verb so'oot as a prototype verb of this intransitive subclass (so'oot-type).

In (36), we present the so'oot-type verb paradigm, in which the word order is always verb-initial VS (with a possibility of adding an oblique object). The order V(Oblique Object)S is also possible. In case of a transitive structure, the orders are VOS and VSO in assertive mood. The syntactic-semantic properties of this type of verb are distinguished from the otam/terekterek-type, e.g., the so'oot-type displays (i) an oblique object and (ii) a subject whose semantic role is an experiencer (ROCHA, 2011ROCHA, I. A estrutura argumental da língua Karitiana: desafios descritivos e teóricos. 220f. 2011. Dissertação (Mestrado em Semiótica e Linguística Geral) - Faculdade de Filosofia, Letras e Ciências Humanas, Universidade de São Paulo, São Paulo, 2011.).

  • (36)

    ø-py-so'oot-yn (pykom-ty) õwã 3-ASSERT-see-NFUT (monkey-OBL) child. “The child saw the monkey.”

  • (37)

    *ø-pyr-a-so'oot-yn õwã (boet-<e>ty) 3-ASSERT-PASV-see-NFUT child (necklace-OBL) (Intend: The child/necklace was seen.)

As seen in (37), the so'oot-type verb does not allow the impersonal passive formation through the morpheme {a-}, following the same pattern as otam and terekterek.

  • (38)

    y-py-m-so'oot-yn õwã yn (pikom-ty) 1-ASSERT-CAUS-see-NFUT child I (monkey-OBL) “The child made me see the monkey.”

Example (38) shows the verb so'oot being causativized by the morpheme {m-}, as predicted for an alternating unaccusative verb. Furthermore, the absolutive agreement marker, first-person {y-}, co-occurs with the absolutive argument, the pronoun yn.

  • (39)

    y-pyr-a-m-so'oot-<y>n yn (boet-<e>ty) 1-ASSERT-PASV-CAUS-see-NFUT I necklace-OBL “Lit: (someone/something made me see the necklace.”

In (37), we just saw that the morpheme {a-} could not directly passivize the verb so'oot. However, this type of verb might accept the passivization if it is causativized by {m-} beforehand, as depicted in (38). This process happens to introduce the external argument (cause or agent) to license the passive use in (39).

  • (40)

    õwã Ø-na-aka-t i-so'oot-Ø (pikom-ty) child 3-DECL-COP-NFUT PART-see-ADVZR monkey-OBL “The child saw the monkey.”

Example (40) displays a so'oot-type verb functioning as the nucleus of the complement of copular verbs. As attested in the morphosyntactic tests, only intransitive verbs can fill this syntactic position.

Adopting a structure that describes the facts observed in verbs like so'oot in (36)-(40) is relatively convoluted since the theoretical approach chosen shows the basic verbal structures: composite dyadic, basic dyadic, and monadic.

We have analyzed the so'oot-type verbs as having a composite dyadic structure, because these verbs can occur in the same environments that otam and terekterek can. However, the so'oot-type verbs semantically require an object, which syntactically appear with an oblique marker. However, the composite dyadic itself does not describe the fact that they project an oblique complement. Moreover, we could consider analyzing it using the basic dyadic structure, that projects the oblique complement, but it would not be sensible to the alternating properties of those verbs, because the basic dyadic comprises a monadic structure that takes a P-complement.

Bearing this in mind, Rocha (2011)ROCHA, I. A estrutura argumental da língua Karitiana: desafios descritivos e teóricos. 220f. 2011. Dissertação (Mestrado em Semiótica e Linguística Geral) - Faculdade de Filosofia, Letras e Ciências Humanas, Universidade de São Paulo, São Paulo, 2011. opted to use a structure derived from a composite dyadic, predicted in Hale and Keyser's (2002)HALE, K. L.; KEYSER, S. J. Prolegomenon to a theory of argument structure. Cambridge, MA: MIT Press, 2002. proposal. In this case, instead of having a root (R) as a complement of the lower verbal head (V2), it takes a P-complement. This way, we describe the process of transitivity alternation, at the same time in which the structure, adopted, allows the projection of the oblique object (cf. ex. (41)).

  • (41)

    Composite dyadic structure with a P-complement (intransitive version)

  • (42)

    Composite dyadic structure with a P-complement (transitive version)

The basic dyadic structure with a P-complement must depict the alternation that those verbs do at the same time in which the structure captures the occurrence of the indirect object.

Transitive verbs: monadic structure

Regarding the transitive verbs, the analysis adopted by Rocha (2011)ROCHA, I. A estrutura argumental da língua Karitiana: desafios descritivos e teóricos. 220f. 2011. Dissertação (Mestrado em Semiótica e Linguística Geral) - Faculdade de Filosofia, Letras e Ciências Humanas, Universidade de São Paulo, São Paulo, 2011. is that the monadic structure describes the transitive verbs. In the following examples (43)-(47) from Rocha (2011)ROCHA, I. A estrutura argumental da língua Karitiana: desafios descritivos e teóricos. 220f. 2011. Dissertação (Mestrado em Semiótica e Linguística Geral) - Faculdade de Filosofia, Letras e Ciências Humanas, Universidade de São Paulo, São Paulo, 2011., we present a paradigm of the transitive verbs.

Transitive verbs in the assertive sentence (VSO)

  • (43)

    Ø-pyry-'y-dn taso asyryty 3-ASSERT-eat-NFUT man banana “The man ate a/the banana.”

Transitive verb being passivized via {a-} in the assertive sentence (VS)

  • (44)

    Ø-pyr-a-'y-dn asyryty 3-ASSERT-PASV-eat-NFUT banana “The banana was eaten.'

Transitive verb causativized with {m-} in the assertive sentence

  • (45)

    *Ø-pyry-m-'y-dn asyryty taso 3-ASSERT-CAUS-eat-NFUT banana man

The transitive verbs in Karitiana cannot be causativized by {m-} as attested in (45). The datum in (46) is ungrammatical due to the morpheme {m-}, not by the occurrence of the passive morpheme {a-}.

Transitive verbs with the morphemes {a-} and {m-}

  • (46)

    *Ø-pyr-a-m-'y-dn asyryty 3-ASSERT-PASV-CAUS-eat-NFUT banana

Copular constructions with transitive verbs in the head of the small-clause

  • (47)

    *asyryty Ø-na-aka-t i-'y-t banana 3-DECL-COP-NFUT PART-eat-ADVZR

The datum (47) shows the constraints of a transitive verb occupying the head of the small clause that functions as a complement of the copula.

  • (48)

    Monadic structure – transitive verb

The structure in (48) describes transitive verbs as shown. These verbs, regarding their structure, project a complement, seen by the NP asyryty ‘banana’, but not a specifier, as predicted by the adopted theory (HALE; KEYSER, 2002HALE, K. L.; KEYSER, S. J. Prolegomenon to a theory of argument structure. Cambridge, MA: MIT Press, 2002.).

Basic dyadic verbs: ditransitives

The ditransitive verbs in Karitiana are: hit ‘give,’ hit ‘it ‘lend,’ and oigng ‘gift.’ All of them have the same pattern morphosyntactically, as seen by examples (49) – (53). Notice that the verb agrees with the absolutive argument in (49). As one can see in the following examples, the argument theme receives the oblique marker {-ty}. Also, the ditransitive verbs, as well as the transitive verbs, can passivize directly by {a-} (cf. ex., 50). They do not accept the causativizer morpheme {m-}, even that they have been passivized beforehand (cf. 52). Furthermore, they cannot head the small clause in a copular environment (cf. 53).

Transitive verb oigng ‘gift’ in the assertive mood

  • (49)

    y-pyr-oigng-<a>n õwã boet-<e>ty 1-ASSERT-gift-NFUT child necklace-OBL “The child gifted me with a necklace.”

Ditransitive verb with the morpheme {a-} in the assertive mood

  • (50)

    Ø-pyr-a-oigng-<a>n õwã boet-<e>ty 3-ASSERT-PASV-gift-NFUT child necklace-OBL “The child was gifted with a necklace.”

Ditransitive verb with the causative morpheme {m-} in the assertive mood

  • (51)

    *y-py-m-oigng-<a>n õwã boet-<e>ty 1-ASSERT-CAUS-gift-NFUT child necklace-OBL

Ditransitive verb with the morphemes {a-} and {m-} in the assertive mood

  • (52)

    *Ø-pyr-a-m-oigng-<a>n õwã boet-<e>ty 3-ASSERT-PASV-CAUS-gift-NFUT child necklace-OBL

Ditransitive verb heading the small clause in copular construction

  • (53)

    *taso Ø-na-aka-t i-oigng-<a>t boet-<e>ty man 3-DECL-COP-NFUT PART-gift-ADVZR necklace-OBL

In addition to presenting a morphosyntactic pattern comparable to the transitive verbs, the ditransitive ones share syntactic and semantic features with the intransitive verbs with oblique objects, e.g., the oblique arguments play a semantic role of the theme. Thus, to make an analogy to the intransitive verbs with oblique objects, which are syntactically intransitive, but semantically transitive, the ditransitive ones are syntactically transitive but semantically ditransitive (ROCHA, 2011ROCHA, I. A estrutura argumental da língua Karitiana: desafios descritivos e teóricos. 220f. 2011. Dissertação (Mestrado em Semiótica e Linguística Geral) - Faculdade de Filosofia, Letras e Ciências Humanas, Universidade de São Paulo, São Paulo, 2011.; STORTO; ROCHA, 2015STORTO, L.; ROCHA, I. Estrutura argumental da língua Karitiana. In: STORTO, L.; FRANCHETTO, B.; LIMA, S. (ed.). Sintaxe e semântica do verbo em línguas indígenas do Brazil. Campinas: Mercado de Letras, 2015. p. 17-42.).

According to the facts shown in (49)-(63), the ditransitive verbs have the GOAL argument projected higher than the THEME one in the structure. In Karitiana, the theme is marked by the oblique postposition {-ty}. Given the linguistic evidence, the structure, which best describes the syntactic and semantic behavior of these verbs, is the basic dyadic depicted in (54)

  • (54)

    Deriving basic dyadic verbs: ditransitive

The structure (54) can be decomposed as follows: the monadic structure takes as its complement a pre/postpositional head that projects a GOAL argument in its specifier, and a THEME argument in the position of the complement (54b) (HALE; KEYSER, 2002HALE, K. L.; KEYSER, S. J. Prolegomenon to a theory of argument structure. Cambridge, MA: MIT Press, 2002.). Thus, the structure (54) describes the transitivity of the verb, in addition to representing the fact that the Karitiana ditransitive verbs display the thematic argument lower in the structure, similar to the English double-object verbs. However, only in Karitiana, the thematic argument is introduced by a postposition due to the restriction of nuclear arguments.

Adding arguments

In this section, we show how additional arguments are added to the basic structures of the verbs in Karitiana according to the theoretical approach of Pylkkänen (2008)PYLKKÄNEN, L. Introducing arguments. Cambridge, MA: MIT Press, 2008.. The author assumes that verbs have a basic structure and functional nuclei that allow the introduction of non-nuclear arguments. She analyzes this functional nucleus as applicative; and the additional argument, in turn, as an applied argument.

As stated in Pylkkänen (2008)PYLKKÄNEN, L. Introducing arguments. Cambridge, MA: MIT Press, 2008., the voice and applicative heads introduce arguments4 4 Non-nuclear arguments are not required by the verb. In general, when they appear, they are introduced through a functional head in the verb, and the additional argument marked by oblique case. into the verbal structure. Thereby the voice head introduces an argument above the VP, changing the argument structure of the verb while the higher and lower applicative heads add extra arguments to the VP, not changing, in this case, the argument structure. The higher applicative relates the applied argument to the event described by the VP. On the other hand, the lower applicative adds an internal argument to the VP, relating the original argument to the applied one.

Based on the Pylkkänen (2008)PYLKKÄNEN, L. Introducing arguments. Cambridge, MA: MIT Press, 2008.'s approach, Karitiana presents only two verbal classes: transitive and intransitive ones. Thus, the additional arguments are introduced via the applicative heads. Therefore, both the verbs oky ‘kill’ and hit ‘give’ display the same basic argument structure, so that hit can have an applied argument added to its structure, predicted semantically. By contrast, both the intransitive verbs otam ‘arrive’ and so'oot ‘see’ share the same basic structure. However, as hit, so'oot can also receive an additional argument, introduced by the lower applicative head.

The constructions, which were analyzed as causatives by Rocha (2011ROCHA, I. A estrutura argumental da língua Karitiana: desafios descritivos e teóricos. 220f. 2011. Dissertação (Mestrado em Semiótica e Linguística Geral) - Faculdade de Filosofia, Letras e Ciências Humanas, Universidade de São Paulo, São Paulo, 2011., 2014ROCHA, I. Causativization in karitiana. Boletim do Museu Paraense Emílio Goeldi: Ciências Humanas, Belém, v. 9, n. 1, p. 183-197, 2014.) and Everett (2006)EVERETT, C. Gestural, perceptual and conceptual patterns in karitiana. Houston: Rice University Press, 2006., are reanalyzed herein as high applicative constructions.

Summarizing, the introducer heads of arguments:

  1. Voice: introduces the external argument to the alternating unaccusative verbs via morpheme {m-} to transitivize them;

  2. High applicative: introduces an instrumental argument to the transitive verbs via the typoong head (above the VP);

  3. Low applicative: introduces an applied argument internally to the intransitive verbs of the so'oot-type or hit via {ø}.

High applicative construction

Karitiana displays an applicative construction formed by a functional head externalized by typoong in which reading is instrumental (56). This construction has been analyzed as causative constructions by Everett (2006)EVERETT, C. Gestural, perceptual and conceptual patterns in karitiana. Houston: Rice University Press, 2006., and Rocha (2011ROCHA, I. A estrutura argumental da língua Karitiana: desafios descritivos e teóricos. 220f. 2011. Dissertação (Mestrado em Semiótica e Linguística Geral) - Faculdade de Filosofia, Letras e Ciências Humanas, Universidade de São Paulo, São Paulo, 2011., 2014ROCHA, I. Causativization in karitiana. Boletim do Museu Paraense Emílio Goeldi: Ciências Humanas, Belém, v. 9, n. 1, p. 183-197, 2014.).

In the following examples, we show how applied arguments are added to a transitive construction:

  • (55)

    õwã ø-na-okop-ø ot'ep child 3-DECL-break-NFUT bow “The child broke the bow.”

  • (56)

    taso ø-na-okop-ø typoong ot'ep õwã-ty man 3-DECL-break-NFUT APPL bow child-OBL “The man broke the bow using the child (as an instrument).”

Comparing examples (55) and (56), notice that, in the former, the subject, õwã, causes an action of 'breaking the bow', ot'ep, directly. In the latter, õwã is still the participant that executes the action of 'breaking the bow'. However, it is demoted from the position of the subject and promoted to another syntactic position, marked by an oblique suffix, and functioning as an instrument of the syntactical subject taso.

  • (57)

    *taso ø-na-okop-ø ot'ep typoong õwã-ty homem 3-DECL-break-NFUT bow APPL child-OBL Intend: “The man broke the bow using the child.”

Example (57) is ungrammatical because typoong constitutes a syntactic unit with the verb so that we cannot put any argumental element between the verb and typoong.

  • (58)

    yn a-taka-mĩ-t typoong an i-ty eu 2-DECL-hit-NFUT APPL you he/him-OBL “I hit you using him (as an instrument).” (EVERETT, 2006, p. 442)

Example (58) attests that the pronoun after typoong is the object, since the verb agrees with the absolutive argument, as we can see the agreement showing between the prefix {a-} and the pronoun an.

Observe the following examples:

  • (59)

    atykiri ø-na-otam-ø João therefore 3-DECL-arrive-NFUT João “Therefore João arrived.”

  • (60)

    *atykiri (*João) ø-na-otam-ø (*João) typoong (João) taso-ty therefore (João) 3-DECL-arrive-NFUT (João) APPL (João) man-OBL

  • (61)

    Renato ø-na-mb-otam-ø õwã Renato 3-DECL-CAUS-arrive-NFUT child “Renato made the child arrive.”

  • (62)

    Renato ø-na-mb-otam-ø typoongõwã João-ty Renato 3-DECL-CAUS-arrive-NFUT APPL child João-OBL “Renato made the child arrive using João (as an instrument).”

In example (59), we exhibited an intransitive sentence using the verb otam ‘arrive’. And so, in (60), we added the typoong head in order to form a structure with the high applicative directly to the intransitive structure, what is not allowed, as seen by the ungrammaticality of (60). Besides, to avoid the mistake that the position of the arguments has caused this ungrammaticality, we tested the word order variation with the subject in (60). However, if example (59) has been causativized before via {m-}, like (61), the applicative construction is allowed as seen in (62)

We present the syntactic structure of a high applicative head depicting the projection of typoong as seen in (56), (58), and (62). For the structure (63), we consider example (58)

  • (63)

    High applicative structure

The structure in (63) adapted from Pylkkänen (2008PYLKKÄNEN, L. Introducing arguments. Cambridge, MA: MIT Press, 2008., p. 14) captures the fact that the applied argument has a semantic of instrumental. That is, the argument, whose function is the subject, yn, does not directly affect the object an, but instead the event of “hitting you” is fulfilled by the applied argument that is introduced by typoong.

Low applicative constructions

In this section, we show an analysis of low applicative heads for the verbs of hit and so'oot-type. Even though we analyze those as involving low applicative constructions, we highlight again that hit (64) shows a basic structure transitive while so'oot (67) displays a basic structure intransitive.

  • (64)

    taso ø-naka-hit-ø õwã boet-<e>ty man 3-DECL-give-NFUT child necklace-OBL “The man gave the child a necklace.”

  • (65)

    taso ø-naka-hit-ø boet-<e>ty man 3-DECL-give-NFUT necklace-OBL “The man gave a necklace.”

  • (66)

    taso ø-naka-hit-ø õwã man 3-DECL-give-NFUT child “The man gave (something) to the child.”

  • (67)

    y-py-so'oot-<y>n yn pikom-ty 1-ASSERT-see-NFUT I monkey-OBL. “I saw the monkey.”

  • (68)

    y-py-m-so'oot-< y>n õwã pikom-ty 1-ASSERT-CAUS-see-NFUT child monkey-OBL “The child made me see the monkey.”

The structure for the verb (di)transitive hit

Before we present the structure of the verb hit in Karitiana (64), we do correlate this type of verb, in terms of structure, with the constructions of double object in English (69), since in both languages the argument with semantic role GOAL c-commands the THEME which is lower in the structure, according to the tree representation in (70). However, contrasting English to Karitiana, we can notice that the argument THEME receives an oblique marker in the latter, but not in the former. In our analysis, this argument is introduced by a low applicative head.

  • (69)

    I gave Mary a book.

  • (70)

    The GOAL higher than the THEME

The structure (71) depicts the low applicative construction proposed to describe the verbs of hit-type in Karitiana. In (71), we should consider the example in (64).

  • (71)

    Low applicative structure for verbs of hit-type in Karitiana

In (71), the applicative head selects three arguments, that are (i) the NP õwã, (ii) the NP boet marked by a postposition, and (iii) the verb hit.

Hale and Keyser (2002)HALE, K. L.; KEYSER, S. J. Prolegomenon to a theory of argument structure. Cambridge, MA: MIT Press, 2002. describe the double object construction as basic dyadic structures, which, in terms of argument position, have the configuration of a with-construction. Thus, relating to its configuration, the with-construction and double object constructions are similar to one another, except by the fact that the former shows an opened P-head while the latter does not. These structures differ from the to-dative ones in which the GOAL is lower than the THEME.

It should be noted that in an analysis based on Hale and Keyser (2002)HALE, K. L.; KEYSER, S. J. Prolegomenon to a theory of argument structure. Cambridge, MA: MIT Press, 2002.'s approach, verbs such as hit in Karitiana and to give in English are both prepositional. However, in an analysis based on the proposal of applicative heads developed by Pylkkänen (2008)PYLKKÄNEN, L. Introducing arguments. Cambridge, MA: MIT Press, 2008., there is an applicative head which adds the extra argument into the basic structure of the verb.

Structures of the verbs of so'oot-type

Karitiana displays a type of intransitive verb with an oblique object project. In previous works, Rocha (2011)ROCHA, I. A estrutura argumental da língua Karitiana: desafios descritivos e teóricos. 220f. 2011. Dissertação (Mestrado em Semiótica e Linguística Geral) - Faculdade de Filosofia, Letras e Ciências Humanas, Universidade de São Paulo, São Paulo, 2011. analyzed them as verbs composite dyadic with a P-complement in the root (R) as complement of the V2 head (HALE; KEYSER, 2002HALE, K. L.; KEYSER, S. J. Prolegomenon to a theory of argument structure. Cambridge, MA: MIT Press, 2002.). In this article, they are being analyzed as the low applicative heads according to Pylkkänen (2008)PYLKKÄNEN, L. Introducing arguments. Cambridge, MA: MIT Press, 2008.. In addition to the oblique argument, verbs of this type present a subject with experiential semantic role.

Pylkkänen's (2008)PYLKKÄNEN, L. Introducing arguments. Cambridge, MA: MIT Press, 2008. proposal for applicative heads predicts that the low applicative head will take 3 arguments: the direct, indirect objects, and the event described by the verb. It would seem counterintuitive to say that a verb like so'oot has two internal arguments; however, it is not counterintuitive due to the following fact: these verbs are of the unaccusative type in which the intransitive subject is generated as the internal argument of the verb. In addition, one can see that the subject of the intransitive version corresponds to the object of the transitive version (compare the pair of examples ((67) and (68))

  • (73)

    Structure of a low applicative head for verbs of the so'oot-type

Considering the fact that the verb so'oot can alternate between a transitive and an intransitive version, the structure that better describes this type of verb must be one that has a slot for such transitivity alternation. The adopted theory predicts that the external argument is introduced by a nucleus voice so that we project the voice head in the structure to describe the causativization via {m-} in Karitiana.

Conclusion

This article provided a study of the argument structure in Karitiana in which the tests applied were capable of identifying verbal classes in the language, showing the theoretical analysis and description given by Rocha (2011)ROCHA, I. A estrutura argumental da língua Karitiana: desafios descritivos e teóricos. 220f. 2011. Dissertação (Mestrado em Semiótica e Linguística Geral) - Faculdade de Filosofia, Letras e Ciências Humanas, Universidade de São Paulo, São Paulo, 2011. in which we found 3 verbal classes: (i) transitive (monadic), (ii) ditransitive (basic dyadic), and (iii) intransitive. The last one can be divided into two subclasses: (a) “simple” unaccusatives (composite dyadic) and (b) unaccusatives with oblique objects (composite dyadic with oblique objects) (HALE; KEYSER, 2002HALE, K. L.; KEYSER, S. J. Prolegomenon to a theory of argument structure. Cambridge, MA: MIT Press, 2002.).

We have shown a reanalysis of the Karitiana verbs, adopting the theoretical proposal of functional heads in which additional arguments are introduced via the low and high applicative heads (PYLKKÄNEN, 2008PYLKKÄNEN, L. Introducing arguments. Cambridge, MA: MIT Press, 2008.). In this proposal, Karitiana has only two verbal classes: transitive and intransitive. Therefore, the low applicative head introduces the argument oblique displayed in verbs as hit ‘give’ and so'oot ‘see’. Also, the high applicative head allows the additional arguments in typoong-structures.

  • 1
    The definition of agreement used is from the Case Binding theory which says that a functional nucleus agrees with different arguments in different syntactic configurations (STORTO, 1999STORTO, L. Aspects of a Karitiana grammar. 1999. Thesis (Ph.D. in Linguistics) - Massachusetts Institute of Technology, Cambridge, MA, 1999.. 154; BITTNER; HALE, 1996BITTNER, M.; HALE, K. L. The structural determination of case and agreement. Linguistic Inquiry, Cambridge, MA, v.27, n.1, p. 1-68, 1996.).
  • 2
    GLOSSES: 1, 2, 3: agreement of first, second, and third persons; DECL: declarative mood; NFUT: non-future tense; ASSERT: assertative mood; CAUS: causativizer; OBL: oblíque marker; PASV: passive preffix; COP: copular; PART: participle/nominalizer; ADVZR: adverbializer; PAST: past tense; FOC: focus; 1/2/3SG: first, second, and third singular persons; PRES: present tense; APPL: applicative head; FV: final vowel; C1/C7/C9: classes 1, 7, and 9; ASP: aspect; SP: preffix of subject agreement; SUB: subject.
  • 3
    Inergative verbs do not have internal arguments. Once they do not project internal arguments, there is no nominative case assignment to the external argument (BURZIO, 1986BURZIO, L. Italian syntax: A government-binding approach. Berlin: Springer Science & Business Media, 1986.). Regarding this fact, Hale and Keyser (2002)HALE, K. L.; KEYSER, S. J. Prolegomenon to a theory of argument structure. Cambridge, MA: MIT Press, 2002. assume that these verbs do not project an internal specifier in their structures. Therefore, they do not participate in the causative-inchoactive alternation because of the lack of an internal specifier. The internal specifier will function as the subject in an intransitive version of an unaccusative verb, and as the object in the transitive version of those verbs.
  • 4
    Non-nuclear arguments are not required by the verb. In general, when they appear, they are introduced through a functional head in the verb, and the additional argument marked by oblique case.

Acknowledgements

This research was carried out mostly during the doctoral study (thanks to FAPESP for funding this research through the scholarship process number (2012/06769-7), and BEPE process number (2014/14044-2)) at the Graduate in Semiótica e Linguística Geral (PPGL-DL-USP). I would like to thank MCTIC/CNPq/PCI, institutional process number (444338/2018-7), and individual process number (300667/2019-1) for the grant of the Institutional Capacity Building Program (PCI) of Museu Paraense Emílio Goeldi. I especially thank the Karitiana community for their collaboration in this work.

REFERÊNCIAS

  • BAKER, M. C. Incorporation: A theory of grammatical function changing. Chicago: University of Chicago Press. 1988a.
  • BAKER, M. C. Theta theory and the syntax of applicatives in Chichewa. Natural Language & Linguistic Theory, v. 6, n. 3, p. 353-389, 1988b.
  • BITTNER, M.; HALE, K. L. The structural determination of case and agreement. Linguistic Inquiry, Cambridge, MA, v.27, n.1, p. 1-68, 1996.
  • BURZIO, L. Italian syntax: A government-binding approach. Berlin: Springer Science & Business Media, 1986.
  • EVERETT, C. Gestural, perceptual and conceptual patterns in karitiana Houston: Rice University Press, 2006.
  • FARGETTI, C. M. Estudo fonológico e morfossintático da língua Juruna 317f. 2001. Tese (Doutorado em Linguística) – Instituto de Estudos da Linguagem, Universidade Estadual de Campinas, Campinas, 2001.
  • HALE, K. L.; KEYSER, S. J. Prolegomenon to a theory of argument structure Cambridge, MA: MIT Press, 2002.
  • LANDIN, D. An outline of the syntactic structure of Karitiana sentences. In: DOOLEY, R. A. Estudos sobre línguas Tupi do Brasil Brasília: SIL Press, 1984. p. 219-254. (Série Lingüística, 11.).
  • LIMA, S. O. de. A estrutura argumental dos verbos na língua Juruna (Yudja): da formação dos verbos para a análise das estruturas sintáticas. 297f. 2008. Tese (Doutorado em Semiótica e Linguística Geral) - Faculdade de Filosofia, Letras e Ciências Humanas, Universidade de São Paulo, São Paulo, 2008.
  • PYLKKÄNEN, L. Introducing arguments Cambridge, MA: MIT Press, 2008.
  • ROCHA, I. Levantamento da situação sociolinguística da língua Karitiana. In: INVENTÁRIO NACIONAL DA DIVERSIDADE LINGUÍSTICA [INDL/IPHAN/MPEG]. Levantamento regional da situação sociolinguística de 26 etnias indígenas da região de Rondônia Belém, PA: Museu Paraense Emílio Goeldi, 2018. p. 1-67.
  • ROCHA, I. Causativization in karitiana. Boletim do Museu Paraense Emílio Goeldi: Ciências Humanas, Belém, v. 9, n. 1, p. 183-197, 2014.
  • ROCHA, I. A estrutura argumental da língua Karitiana: desafios descritivos e teóricos. 220f. 2011. Dissertação (Mestrado em Semiótica e Linguística Geral) - Faculdade de Filosofia, Letras e Ciências Humanas, Universidade de São Paulo, São Paulo, 2011.
  • STORTO, L. Copular constructions in Karitiana: a case against case movement. MIT Occasional Papers in Linguistics, Cambridge, v. 41, p. 205-226, 2010.
  • STORTO, L. Caso e concordância nas línguas tupi. Estudos Linguísticos, São Paulo, v. 34, p. 59-72, 2005.
  • STORTO, L. Algumas categorias funcionais em Karitiana. In: ANPOLL. Línguas indígenas brasileiras: fonologia, gramática e história. v.1. Belém: Ed. da UFPA, 2002. p.151-164.
  • STORTO, L. Duas classes de verbos intransitivos em Karitiana (família Arikém, tronco Tupí). In: QUEIXALOS, F. (éd.). Des noms et des verbes en Tupi-Guarani: état de la question. Paris: Lincom-Europa, 2001. p. 163-180.
  • STORTO, L. Aspects of a Karitiana grammar 1999. Thesis (Ph.D. in Linguistics) - Massachusetts Institute of Technology, Cambridge, MA, 1999.
  • STORTO, L.; ROCHA, I. Estrutura argumental da língua Karitiana. In: STORTO, L.; FRANCHETTO, B.; LIMA, S. (ed.). Sintaxe e semântica do verbo em línguas indígenas do Brazil Campinas: Mercado de Letras, 2015. p. 17-42.

Publication Dates

  • Publication in this collection
    28 Oct 2020
  • Date of issue
    2020

History

  • Received
    12 Aug 2018
  • Accepted
    21 Feb 2019
Universidade Estadual Paulista Júlio de Mesquita Filho Rua Quirino de Andrade, 215, 01049-010 São Paulo - SP, Tel. (55 11) 5627-0233 - São Paulo - SP - Brazil
E-mail: alfa@unesp.br