scieee AI-readable full text Open interactive document viewer

Different types of nouns, different types of projections

Gondra, Ager

Abstract

This paper proposes (1) for Basque nominal domain. Expanding on the ideas of Distributed Morphology, all nouns are syntactically derived from a categorically unspecied root, which is nominalized by an n-head. With deverbal nouns, this unspecied root is rst dominated by a series of structure-creating nodes (V-v). Event nouns take an external and an internal argument due to their argument structure, and have an event reading, while result nouns have the option to take an adjunctexternal argument and an internal argument, and have a referential reading. Thus, event nouns project the v that introduces the event reading and the external argument, while result nouns project a defective v* that has a non-active reading and does not introduce an external argument (Kratzer 1996). With common nouns, however, the »RP merges directly with the little n, making impossible for them to take arguments. Finally, with result nouns and common nouns, a possessor1 is possible. Hence, two types of n are proposed: n* introduces the possessor, while n does not. The possessor raises to the pec-FP position along with the external and the internal argument.

Full text

Different types of nouns, different types of projections Ager Gondra Abstract This paper proposes (1) for Basque nominal domain. Expanding on the ideas of Distributed Morphology, all nouns are syntactically derived from a categorically unspecified root, which is nominalized by an n-head. With deverbal nouns, this unspecified root is first dominated by a series of structure-creating nodes (V-v). Event nouns take an external and an internal argument due to their argument structure, and have an event reading, while result nouns have the option to take an adjunctexternal argument and an internal argument, and have a referential reading. Thus, event nouns project the v that introduces the event reading and the external argument, while result nouns project a defective v* that has a non-active reading and does not introduce an external argument (Kratzer 1996). With common nouns, however, the √RP merges directly with the little n, making impossible for them to take arguments. Finally, with result nouns and common nouns, a possessor1is possible. Hence, two types of n are proposed: n* introduces the possessor, while n does not. The possessor raises to the Spec-FP position along with the external and the internal argument. 1The label possessor is given to any argument that has referential relation to the following nominal constituents: the kinship noun, nouns with inherent part-whole relations, and ordinary nouns. 1 Different types of nouns, different types of projections Ager Gondra (1) DP D FP F’ F’ F’ F nP nP n vP v’ v VP V’ V √RP √R Int. Arg. Ext. Arg. Poss Spec Spec Spec Keywords: Basque, syntax, nominal domain, deverbal nouns. 1 Introduction2 My hypothesis (1) is based on the proposals outlined in the Distributed Morphology (Marantz 1997), which includes the assumption that lexical categories do not have a specific categorical feature [+/-N] or [+/-V], but are originated as neutral roots and assigned a categorical feature by the syntactic structure. Thus, for the nominal structure, I propose the structure shown in (2). 2I gratefully acknowledge the help I obtained from Elena Benedicto, Karlos Arregi, Xabier Artiagoitia and Jon Ortiz de Urbina. All shortcomings are my own. Iberia: IJTL |Volume 6 (2014), 1–26 ISSN: 1989-8525 http://revistas.ojs.es/index.php/iberia/ 2 Different types of nouns, different types of projections Ager Gondra (2) nP nP n vP v’ v VP V’ V √RP √R Internal Arg. External Arg. Poss This syntactic structure is also based on the assumption that in some fundamental respects verbs and deverbal nouns seem to share argument properties. This idea was already presented by Chomsky (1970) with the following examples: (3) a. The enemy destroyed the city. b. The enemy’s destruction of the city. In (3a) we have the verb destroyed that takes two arguments: the agent the enemy and the theme the city. In (3b) there is not a verb, but a deverbal noun destruction that also takes the same arguments as the verb. These two examples may suggest that deverbal nouns show the same argument structure as verbs and, in fact, some researchers (e.g. Borer 1993, Hazout 1991) relate the verbal-like properties of deverbals to the presence of a verbal projection within the nominal structure. Under this assumption, the deverbal nouns are split into two subcategories (Grimshaw 1990); event nouns and result nouns (the eventive and referential reading). Examples of event nouns and result nouns are shown in (4a–c) and (4d–f) respectively (examples taken from Grimshaw 1990): (4) a. The felling of the trees. b. The destroying of the town. c. The development was applauded. d. The expression is desirable. e. This semester’s assignment led to disaster. f. The solution to the problem simplified the assignment. Iberia: IJTL |Volume 6 (2014), 1–26 ISSN: 1989-8525 http://revistas.ojs.es/index.php/iberia/ 3 Different types of nouns, different types of projections Ager Gondra This paper is structured as follows. Section 2 will expand on the idea of verbal projections within the nominal structure. Section 3 will propose two types of n-heads. Section 4 will show that the possessor moves to the Spec-PossP position along with the external and the internal argument. Finally, section 5 will summarize the main points of this paper. 2 Verbal projections Grimshaw (1990) shows some salient differences between the two types of deverbal nouns (event nouns and result nouns) that can be explained in terms of absence versus presence of argument structure. Event nouns contain an argument structure allowing them to take arguments, and have an eventive reading. Result nouns, on the other hand, lack an argument structure, therefore do not have to take arguments, and have a referential reading. Grimshaw provides tests to identify event nouns and result nouns. Some of these tests will be applied to identify Basque event nouns and result nouns:3 i. Result nouns can appear as predicates, whereas event nouns cannot. (5) a. *Hau This Francoren Franco.gen Gernikaren Gernika.ren suntsiketa destroying.d da be.prs.3s ‘This is Franco’s destroying of Gernika’ b. Hau This euskalkien dialect.gen sailkapena classification.d da be.prs.3s ‘This is the classification of the dialects’ The ungrammaticality of (5a) indicates that suntsiketa is an event noun, while the grammaticality of (5b) reveals that sailkapen is a result noun. ii. Implicit argument control is possible with event nouns, but result nouns do not permit it. (6) a. Gernikaren Town.gen suntsiketak destroying.d.erg euskaldunak Basque.d.pl beldurtzeko frighten.to hiru three ordu hour iraun last zuen aux.3s3s ‘The destroying of Gernika to frighten the Basques lasted three hours’ 3See Artiagoitia (2000: 137–41) for similar data. Iberia: IJTL |Volume 6 (2014), 1–26 ISSN: 1989-8525 http://revistas.ojs.es/index.php/iberia/ 4 Different types of nouns, different types of projections Ager Gondra b. *Euskalkien dialect.gen sailkapena classification.d euskaldunak Basque.d.pl goratzeko praise.to polita pretty da be.prs.3s ‘The classification of the dialects to praise the Basques is pretty’ Since suntsiketa in (6a) allows the implicit argument control euskaldunak beldurtzeko, we know that it is an event noun. Conversely, as sailkapen in (6b) does not allow the implicit argument control, we also know that it is a result noun. iii. Event nominals do not pluralize, while result nominals do. Consider the following examples: (7) a. Francoren Franco.gen Gernikaren Gernika.gen suntsiketak destroying.d.pl beldurgarriak scary.d.pl izan be ziren aux.pst.3pl ‘Franco’s destroyings of Gernika were scary’ b. Hauek These euskalkien dialect.ren sailkapenak classification.d.pl dira be.prs.1s ‘These are the classifications of the dialects’ The fact that when we pluralize suntsiketa ‘destroying’ turns into ungrammaticality indicates that suntsiketa in (7a) is an event noun. Nevertheless, sailkapen ‘classification’ can be pluralized in (7b), which shows that it is a result noun. iv. Event nouns only allow the definite articles (8a–b), while result nouns can also occur with the indefinite determiner (8c–d). (8) a. Francoren Franco.gen Gernikaren Gernika.ren suntsiketa destroying.d beldurgarria scary.d izan be zen aux.3s ‘Franco’s destroying of Gernika was scary’ b. ?Francoren Franco.gen Gernikaren Gernika.gen suntsiketa destroying bat a beldurgarria scary.d izan be zen aux.3s ‘Franco’s destroying of Gernika was scary’ c. Hau This euskalkien dialect.gen sailkapena classification.d da be.1s ‘This is the classification of the dialects’ Iberia: IJTL |Volume 6 (2014), 1–26 ISSN: 1989-8525 http://revistas.ojs.es/index.php/iberia/ 5 Different types of nouns, different types of projections Ager Gondra d. Hau This euskalkien dialect.gen sailkapen classification bat a da be.1s ‘This is a classification of the dialects’ Suntsiketa does not allow the indefinite article (8b), while sailkapena does (8d). Thus, this evidence indicates that suntsiketa in (8a–b) is an event noun, and sailkapen in (8c–d) a result noun. v. Another test that can be applied is the use of temporal constituents such as ‘three days’ with devebal nouns. An event noun may be referred by temporal constituents whereas a result noun cannot as examples (9a–b) show. (9) a. Francoren Franco.gen Gernikaren Gernika.gen suntsiketak destroying.d.erg hiru three egun day iraun last zuen aux.pst.3s3s ‘Franco’s destroying of Gernika lasted three days’ b. *Euskalkien dialect.gen sailkapenak classification.d.erg hiru three egun day iraun last zuen aux.pst.3s3s ‘The classification of the dialects lasted three days’ Event nouns, due to their eventive reading, can have a temporal constituent referring to it. However, result nouns do not allow a temporal constituent since they lack this eventive reading. Thus, we conclude that suntsiketa ‘destroying’ in (9a) is an event noun, while sailkapen ‘classification’ in (9b) is a result noun. The following tests that Grimshaw (1990) argues for the eventive/resultative distinction have been applied: (i) event nouns cannot appear as predicates, (ii) implicit argument control is not possible with result nouns, (iii) event nominals cannot be pluralized, (iv) event nouns cannot occur with the indefinite determiner and the numeral one, and (v) result nouns cannot be referred by temporal constituents such as three days. Based on their results, we can conclude that suntsiketa ‘destroying’ in the preceding examples is an event noun, while sailkapen ‘classification’ is a result noun. Additionally, like verbs, event nouns have an argument structure that must be satisfied, while result nouns lack this argument structure. According to Grimshaw (1990), event nouns always need to take an external argument and an internal argument, whereas result nouns take them optionally. Hence, we would expect for suntsiketa ‘destroying’ to have an argument structure and, therefore, to take Iberia: IJTL |Volume 6 (2014), 1–26 ISSN: 1989-8525 http://revistas.ojs.es/index.php/iberia/ 6 Different types of nouns, different types of projections Ager Gondra an external and internal argument obligatorily, while for sailkapen ‘classification’ to lack this argument structure and, therefore, to take an external argument and an internal argument optionally. Consider the following Basque examples with no little pro: (10) a. Francoren Franco.gen Gernikaren Gernika.gen suntsiketak destroying.d.erg hiru three egun day iraun last zuen. aux.pst.3s3s ‘Franco’s destroying of Gernika lasted 3 days’ b. *Suntsiketak destroying.d.erg hiru three egun day iraun last zuen. aux.pst.3s3s ‘The destroying lasted 3 days’ (11) a. Euskalkien dialect.gen sailkapena classification.d ikarragarrizkoa tremendous.d da be.3s.prs ‘The classification of the dialects is tremendous’ b. Sailkapena classification.d ikarragarrizkoa tremendous.d da be.3s.prs ‘The classification is tremendous’ The event noun suntsiketa in (10a) takes the internal argument Gernikaren in order to fulfill the argument structure requirement, because if it does not, the sentence is ruled out (10b). In (11c–d), however, we can see that it is optional for the result noun sailkapen ‘classification’ to take the internal argument Euskalkien. Assuming that arguments cannot appear as predicates with a copula since they would not be able to receive a thematic role from the head noun (Anderson 1983, Artiagoitia 2006), the following examples (12a–b) confirm that both Gernikaren in (10a) and Euskalkien ‘of the dialects’ in (11a) are the arguments of suntsiketa ‘destroying’ and sailkapen ‘classification’ respectively: (12) a. *Suntsiketa destroying.d.erg Gernikarena Gernika.gen.d da be.prs.3s ‘The destroying is of Gernika’ b. *Sailkapena classification.d euskalkiena dialect.gen.d da be.prs.3s ‘The classification is of the dialects’ Gernikaren ‘of Gernika’ in (12a) and euskalkien ‘of the dialects’ in (12b) appear as predicates with a copula and, therefore, they are not able to receive a thematic-role. The ungrammaticality of (12a–b) indicates that Gernikaren ‘of Iberia: IJTL |Volume 6 (2014), 1–26 ISSN: 1989-8525 http://revistas.ojs.es/index.php/iberia/ 7 Different types of nouns, different types of projections Ager Gondra Gernika’ and Euskalkien ‘of the dialects’ are arguments of suntsiketa ‘destroying’ and sailkapen ‘classification’ respectively. Thus, it can be concluded that Gernikaren in (10a) and Euskalkien ‘classification’ in (11a) are the internal arguments of suntsiketa ‘destroying’ and euskalkien ‘of the dialects’ respectively. In regard to external arguments, Grimshaw (1990) argues that event nouns must always take the external argument while the result nouns take it optionally. This is illustrated in the following examples4in Basque with no little pro: (13) a. *Gernikaren Gernika.gen suntsiketak destroying.d.erg hiru three egun day iraun last zuen aux.pst.3s3s ‘The destroying of Gernika lasted three days’ b. Euskalkien dialect.gen sailkapena classification.d ikarragarrizkoa tremendous.d da be.3s.prs ‘The classification of the dialects is tremendous’ Neither the event noun suntsiketa ‘destruction’ in (13a) nor the result noun sailkapen ‘classification’ in (13b) takes an external argument. However, the ungrammatically of (13a) in contrast with the grammaticality of (13b) show that event nouns have to take an external argument, while result nouns do not have to. In fact, for Grimshaw (1990) the only true external argument is the event argument that event nominals take. The external argument that result nouns take, on the contrary, is an adjunct-argument, which appears either as a by-phrase or as a possessor. Let’s contemplate the Basque sentences (14a–b): (14) a. Francoren Franco.gen Gernikaren Gernika.gen suntsiketak destroying.d.erg hiru three egun day iraun last zuen aux.pst.3s3s ‘Franco’s destroying of Gernika lasted three days’ b. Bonaparteren Bonaparte.gen euskalkien dialect.gen sailkapena classification.d ikarragarrizkoa tremendous.d da be.3s.prs ‘The classification of the dialects by Bonaparte is tremendous’ Following Grimshaw (1990), in (14a) the event noun suntsiketa takes the event argument Francoren ‘Franco’s’ whereas in (14b) the result noun sailkapen ‘classification’ takes the adjunct-argument euskalkien ‘of the dialects’. Yet how do we know that Francoren in (14a) and Bonaparteren in (14b) are different? If event nouns take a true external argument, while result nouns take an external adjunctargument, we will expect to be impossible for the external argument of event nouns 4Both subject and object genitives, as well as regular possessors, are isomorphic and equally prenominal Artiagoitia (2012b). Iberia: IJTL |Volume 6 (2014), 1–26 ISSN: 1989-8525 http://revistas.ojs.es/index.php/iberia/ 8 Different types of nouns, different types of projections Ager Gondra to appear in a predicate position, but possible for the external argument of result nouns. In fact, this is true as examples (15a–b) show: (15) a. *Gernikaren Gernika.gen suntsiketa destroying.d. Francorena Franco.gen.a da be.prs.3s3s ‘The destroying of Gernika is Franco’s’ b. Euskalkien dialect.gen sailkapena classification.d Bonaparterena Bonaparte.gen.a da be.3s.prs ‘The classification of the dialects is Bonaparte’s’ The ungrammaticality of (15a) proves that Francoren is a true argument of suntsiketa ‘destroying’. On the other hand, the ungrammaticality of (15b) indicates that Bonaparteren is not a true argument of sailkapen ‘classification’, therefore, it is an adjunct-argument. In summation, Basque event nouns have to take an internal and an event argument in order to fulfill their argument structure. Basque result nouns, on the contrary, do not have an argument structure so they are not required to take arguments. When they do, however, they take an internal argument, and an external adjunct-argument. Hence, given that deverbal nouns show these verbal-like properties, this section will argue for a VP and vP projection within the nominal constituent of such nouns. 2.1 VP projection As it has been shown in (10a–b, 11a–b) , repeated here as (16a–b, 17a–b) , event nouns must take an internal argument, while result nouns do not have to. (16) a. Francoren Franco.gen Gernikaren Gernika.gen suntsiketak destroying.d hiru three egun day iraun last zuen aux.pst.3s3s ‘Franco’s destroying of Gernika lasted 3 days’ b. Francoren Franco.gen suntsiketak Gernika.gen hiru destroying.d egun three iraun day zuen last aux.pst.3s3s ‘Franco’s destroying of Gernika lasted 3 days’ (17) a. Euskalkien dialect.gen sailkapena classification.d ikarragarrizkoa tremendous.d da be.3s.prs ‘The classification of the dialects is tremendous’ b. Sailkapena dialect.gen ikarragarrizkoa classification.d da tremendous.dbe.3s.prs ‘The classification is tremendous’ It has been suggested that the internal argument is originated inside the root phrase (√RP) and assigned the thematic role by the root-head (Alexiadou 2001, Iberia: IJTL |Volume 6 (2014), 1–26 ISSN: 1989-8525 http://revistas.ojs.es/index.php/iberia/ 9 Different types of nouns, different types of projections Ager Gondra All nominal syntactic constructions, event nouns (25a), result nouns (25b) and common nouns (29), project an nP. The following section will discuss the syntactic properties of this layer. 3 (n)oun-phrase This paper assumes the Distributed Morphology framework (Marantz 1997) in that nouns are syntactically derived and show an internal structure as the one of sentences. Under the framework of Distributed Morphology, lexical categories do not have a specific categorical feature [+/-N] or [+/-V], but the roots are interpreted as neutral. The roots are assigned a categorical feature by the syntactic structure, that is, the root is governed by a functional head that will determine the lexical category. The lexical category will manifest as a noun when the root is governed by a nominal functional head (30a) or a verb when the root is governed by a verbal functional head (30b). (30) a. nP √RP √R◦ n◦ b. vP √RP √R◦ v◦ The following subsections will address the two syntactic functions that the n-head has. The first one will deal with nominalization, while the second will discuss which are the properties of the n-head that allows or blocks it to introduce a possessor. 3.1 n-head assigns [+N] The event nouns’ and result nouns’ roots are neuter in position until they enter into relation with a higher functional head resulting in a nominalization. Marantz (1997) argues that the head responsible for assigning [+N] lexical category to the root is n. In Basque there is a long list of deverbal nouns that are derived from the root adding a nominalizer suffix. In the following examples (31) we can see some roots and their corresponding deverbal nouns after the suffix incorporation: Iberia: IJTL |Volume 6 (2014), 1–26 ISSN: 1989-8525 http://revistas.ojs.es/index.php/iberia/ 16 Different types of nouns, different types of projections Ager Gondra (31) Root Deverbal noun + D suntsi ‘destroy’ suntsiketa ‘the destruction’ azter ‘examine’ azterketa ‘the examination’ irakur ‘read’ irakurketa ‘the reading’ ebalua ‘evaluate’ ebaluaketa ‘the evaluation’ garbi ‘clean’ garbiketa ‘the cleaning’ konpon ‘fix’ konponketa ‘the fixing’ gomendio ‘advice’ gomendio ‘the advice’ sor ‘create’ sorkuntza ‘the creation’ eraiki ‘build’ eraikuntsa ‘the building’ ager ‘appear’ agerpen ‘the appearance’ aurkez ‘present’ aurkezpen ‘the presentation’ The suffix, which gets attached to the root, is situated in the n-head position and nominalizes its complement. 32 shows the syntactic structure of a deverbal noun such as suntsiketa ‘destroying’: (32) nP n suffix vP v’ v VP V’ V √RP √R Internal Arg. Spec Event Arg. Taking this into consideration and under the Distributed Morphology framework, my hypothesis suggests that also common nouns are neutral until they are assigned [+N] category by the n-head. Common nouns, as mentioned before, differ from deverbal nouns in that they lack verbal nodes (vP and VP), which explains their impossibility to take arguments. The syntactic tree (33) shows the structure of a common noun: Iberia: IJTL |Volume 6 (2014), 1–26 ISSN: 1989-8525 http://revistas.ojs.es/index.php/iberia/ 17 Different types of nouns, different types of projections Ager Gondra (33) nP n’ n◦ √RP √R◦ Poss Summarizing, we have a root that does not have a specific categorical feature [+/-N] or [+/-V] until it is assigned one by the n-head. Deverbal nouns have verbal nodes (vP and/or VP) between the √RP and nP, while common nouns do not. 3.2 Types of n-heads I follow Longobardi (2001) and Alexiadou et al. (2007) in that the possessor argument also originated inside the nominal domain and it c-commands both subject and object arguments, and also assume that the possessor is originated in the Spec-nP (Radford 2000). For the sake of this paper, the label ‘possessor’ is used to refer to the argument that has a referential relation to the following nominal constituents: nouns with inherent part-whole relations (34a–b) and ordinary nouns (34c–d): (34) a. Aulkiaren chair.d.gen lau four hanka leg ‘The chair’s four legs’ b. Aulkiaren chair.d.gen lau four hankak leg.pl ‘The chair’s four legs’ c. Neskaren girl.d.gen etxe house asko many ‘The girl’s many houses d. Neskaren girl.d.gen etxe house hau d.proximal ‘This house of the girl’ Notice in (34b, d) that possessors are compatible with determiners, which cause the examples to have a definite reading, whereas in (34a, c) there is not an overt determiner and the example does not have definite interpretation. Thus, this data provides evidence to support Artiagoitia’s (2012b) claim that as in Catalan Picallo (1991), Italian Schoorlemmer (1998), and Hungarian Szabolcsi (1994), definiteness and possessor are dislocated in Basque. Iberia: IJTL |Volume 6 (2014), 1–26 ISSN: 1989-8525 http://revistas.ojs.es/index.php/iberia/ 18 Different types of nouns, different types of projections Ager Gondra Anderson’s (1983) predicate test shows that Aulkiaren in (30a–b) and Neskaren in (34c–d) are possessive genitives because they can appear as predicates (35a–b), and therefore they receive the thematic-role neither from the V-head nor the v-head. (35) a. Lau four hankak leg.d.pl aulkiarenak chair.d.gen.ak dira be.prs.3pl ‘The four legs belong to the chair’ b. Etxe house hau d.proximal neskarena girl.d.gen.d da be.prs.3s ‘This house is the girl’s’ Additionally, possessors can co-occur with results nouns, but not with event nouns (36a–b): (36) a. *Nire poss.1s Francoren Franco.gen Gernikaren Gernika.gen suntsiketa destroying.d ‘My Franco’s destroying of Gernika’ b. Nire poss.1s Euskalkien dialect.gen sailkapena classification.d ikarragarrizkoa tremendous.d da be.3s.prs ‘My classification of the dialects is tremendous’ In order to explain why possessors can co-occur with ordinary nouns, nouns with inherent part-whole relations and result nouns, while they cannot with event nouns, this paper proposes that, parallel to the two types of v (one introducing an external argument and another not), there are also two types of n; n* does not contain an event argument in its denotation and can introduce the possessor, while n contains an event argument in its denotation and does not introduce a possessor. Expanding on Kratzer’s (1996) ‘Event Identification’, n* will be incompatible with the vP that has v as its head, whereas it will be compatible with a vP that that has a v* as its head or a common noun. n, on the other hand, with an event argument in its denotation, will match the former vP with its own e-argument. Hence, we derive the possibility of a possessor with result nouns (37a), nouns with inherent part-whole relations and ordinary nouns (37b), but not with event nouns (37c). Iberia: IJTL |Volume 6 (2014), 1–26 ISSN: 1989-8525 http://revistas.ojs.es/index.php/iberia/ 19 Different types of nouns, different types of projections Ager Gondra (37) a. nP n’ n* vP v’ v* VP V’ V √RP √R Internal Arg. Adjunct Arg Poss b. nP n’ n* √RP √R◦ Poss c. nP n vP v’ v VP V’ V √RP √R Internal Arg. Spec Event Arg. The syntactic tree (37a) predicts that result nouns can have a possessor, an adjunct argument and an internal argument co-occurring. Eguzkitza (1993) shows that this is possible with picture nouns (38a–b). In fact, not only it is possibility to Iberia: IJTL |Volume 6 (2014), 1–26 ISSN: 1989-8525 http://revistas.ojs.es/index.php/iberia/ 20 Different types of nouns, different types of projections Ager Gondra have the three genitives with picture nouns but also with result nouns (38c): (38) a. Von Von Thyssen-en Thyssen.gen Riberaren Ribera.gen eskalearen beggar.d.gen erretratua portrait.d ‘Von Thyssen’s Ribera portrait of the beggar’ (Eguzkitza 1993: 166) b. Monzonen Monzon.gen Leizarragaren Leizarraga.gen bibliaren bible.gen itzulpena translation.d ‘Monzon’s Leizega’s translation of the bible’ (Eguzkitza 1993: 170) c. Nire poss.1s Bonaparteren Bonaparte.gen Euskalkien dialect.gen hiru three sailkapen classification.d ikarragarrizkoak tremendous.d dira be.3pl ‘My three classifications of the dialects by Bonaparte are tremendous’ Given that the possessor, the external argument and the internal argument share the same case morphology (-ren), it is only the hierarchy which determines the thematic role of each of them as it will be shown in the next section. 4 The Functional Phrase After the possessor, the external argument and the internal argument DPs are originated, the unvalued Case feature [u_Case] in their respective D has to be valued. Artiagoitia (2009, 2012a) proves that both the external argument and the internal argument move to a Spec-FP7, a projection above NP and below DP, and argues that they check genitive case with the F-head under this multiple specifier configuration. Assuming that the possessor is originated in the Spec-nP, I claim that the possessor also raises to the Spec-FP position. By showing that the possessor raises to a position higher than the QP but lower than the DP, I will conclude that it moves to the Spec-FP. First, as in (1) the QP is higher in the structure than the nP, a quantifier originated in the Spec-QP (e.g. hainbat ‘many’) would precede the possessor if there were no movement involved. Nevertheless, as it can be observed in the following examples (39a–b) , a quantifier originated in the Spec-QP must appear between the possessor and the deverbal noun. (39) a. Nire poss.1s hainbat many txakurrekin dog.pl.instr gogoratu remember naiz be.1s ‘I have just remembered my many dogs’ 7Artiagoitia (2009, 2012a) calls it Spec-PossP. Iberia: IJTL |Volume 6 (2014), 1–26 ISSN: 1989-8525 http://revistas.ojs.es/index.php/iberia/ 21 Different types of nouns, different types of projections Ager Gondra b. *Hainbat many nire poss.1s txakurrekin dog.pl.instr gogoratu remember naiz be.1s ‘I have just remembered my many dogs’ In (39a) Nire ‘my’ precedes the Spec-QP hainbat ‘many’, while in (39b) it follows the quantifier. The fact that sentence (39a) is grammatical whereas (39b) is ungrammatical indicates that even though the possessor Nire is originated in the Spec-nP, this constituent must have undergone movement to a position higher than the QP. Second, assuming that the DP level is associated with referentiality (specifically, +definite) and that elements in Spec-DP will trigger a definite reading, it can be observed in (39a), an example with a –def interpretation, that the possessor does not affect the ±definiteness of the DP. Thus, it has not moved to Spec-DP position. The coordination in (40) corroborates that the possessor is not in Spec-DP as the object does not form a constituent with the determiner. (40) [Nire poss.1s txakur] dog eta and [zure poss.2s katu]-a cat-d ikusi see ditut aux.3pl3s ‘I have just seen my dog and your cat’ Hence, given that the possessor moves to a position higher than the QP but lower than the DP, we can conclude that it raises to Spec-FP along with the external and the internal argument creating a multiple specifier configuration. Richards (2001) discusses strict ordering effects among multiple specifiers of the same category and proposes the Multiple A-Spec theory, which establishes that movement to multiple specifiers of the same head has to respect superiority, therefore, creating tucking-in effect. Consider sentences (41a–c): (41) a. *Gernikaren Gernika.gen Francoren Franco.gen oin long dala time asko ago.ko suntsiketak destroying.d.erg hiru three ordu hour iraun last zuen aux.3s3s ‘Franco’s long time ago destroying of Gernika lasted three hours’ b. *Euskalkien dialect.gen Bonaparteren Bonaparte.gen oin long dala time askoko ago.ko sailkapena classification.d ikarragarrizkoa tremendous.d da be.3s.prs ‘The classification of the dialects by Bonaparte is tremendous’ Iberia: IJTL |Volume 6 (2014), 1–26 ISSN: 1989-8525 http://revistas.ojs.es/index.php/iberia/ 22 Different types of nouns, different types of projections Ager Gondra c. *Bonaparteren Bonaparte.gen Euskalkien dialect.gen nire Poss.1s hiru three sailkapen classification.d ikarragarrizkoak tremendous.d dira be.3pl ‘My three classifications of the dialects by Bonaparte are tremendous’ As the ungrammaticality of (41a–b) show, the first example (41a) with a result now and the second (41b) with an event noun, the internal argument cannot precede the external argument. Regarding the possessor, if the external and internal arguments precede it, the sentences turns into ungrammaticality as occurs in (41c). Thus, these examples show that the movement of the possessor, the external argument and the internal argument must respect the order in which they were originated; that is, the hierarchical structure <possessor> <external argument> <internal argument>. 5 Conclusion This paper has shown that in Basque event nouns project a VP and a vP (with the eventive v as its head) in which the internal and the external argument are originated respectively. Result nouns, on the other hand, project a VP and a vP (with the non-active v* as its head) and may or may not take the internal argument and the adjunct-external argument respectively. The n is responsible for assigning [+N] lexical category to its complement. We have seen that with deverbal nouns this functional head will take a verbal node (the eventive vP in the case of event nouns, and the non-action vP in the case of result nouns) as its complement, whereas with common nouns it will merge directly with the √RP. For the introduction or lack of introduction of the possessor two types of n have been suggested: one n-head introduces the possessor and does not have an event argument in its denotation making it incompatible with the eventive vP, while it will be compatible with the non-active vP and the √RP. The other n-head does not introduce a possessor and has an event argument in its denotation allowing the n to take an eventive vP as its complement. Finally, Artiagoitia (2009, 2012b) provides evidence to support the existence of a FP with multiple specifier positions, to which the external argument and the internal argument are independently raised. This study has shown that the possessor also moves to a Spec-FP position above the external argument. References Alexiadou, A. 1997. Adverb placement: A case study in anti-symmetric syntax. Amsterdam: John Benjamins. Iberia: IJTL |Volume 6 (2014), 1–26 ISSN: 1989-8525 http://revistas.ojs.es/index.php/iberia/ 23 Different types of nouns, different types of projections Ager Gondra Alexiadou, A. 2001. Functional structure in nominals. Amsterdam/Philadelphia: John Benjamins. Alexiadou, A., L. Haegeman & M. Stavrou. 2007. Noun phrase in the generative perspective. Berlin: Mouton de Gruyter. Anderson, M. 1983. Prenominal genitive NPs. The Linguistic Review 3(1), 1–24. Artiagoitia, X. 2000. Hatsarreak eta parametroak lantzen. Gasteiz: Arabako Foru Aldundia. Artiagoitia, X. 2006. Euskarazko izen-sintagma: arkitektura eta egitura funtzionala. UPV/EHU Katedra Txosten argitaragabea (Professorship Report). Artiagoitia, X. 2009. Jabetza-genitiboaren sintaxia eta erkatze anizkuna. Unpublished ms. Artiagoitia, X. 2012a. The DP hypothesis in the grammar of Basque. In U. Etxebarria, R. Etxepare & M. Uribe-Etxebarria (eds.), Noun phrases and nominalization in Basque, 21–77. Amsterdam/Philadelphia: John Benjamins. Artiagoitia, X. 2012b. Genitive case and multiple checking in Basque. In U. Etxebarria, R. Etxepare & M. Uribe-Etxebarria (eds.), Noun phrases and nominalization in Basque, 209–242. Amsterdam/Philadelphia: John Benjamins. Borer, H. 1993. Parallel morphology. Unpublished ms. Burzio, L. 1986. Italian syntax. Dordrecht: Reidel. Chomsky, N. 1970. Remarks on nominalization. In R. A. Jacobs & P. S. Rosenbaum (eds.), Readings in English transformational grammar, 181–221. Waltham, MA: Ginn & Co. Chomsky, N. 2005. On phases. Unpublished ms. Cinque, Guglielmo. 1999. Adverbs and functional heads: A cross-linguistic perspective. Oxford: Oxford University Press. Eguzkitza, A. 1993. Adnominals in the grammar of Basque. In J. I. Hualde & J. Ortiz de Urbina (eds.), Generative studies in Basque linguistics, 163–187. Amsterdam/Philadelphia: John Benjamins. Grimshaw, J. 1990. Argument structure. Cambridge, MA: MIT Press. Hazout, I. 1991. Verbal nouns: Theta theoretic studies in Hebrew and Arabic. PhD dissertation, University of Massachusetts at Amherst. Iberia: IJTL |Volume 6 (2014), 1–26 ISSN: 1989-8525 http://revistas.ojs.es/index.php/iberia/ 24 Different types of nouns, different types of projections Ager Gondra Jackendoff, R. 1997. The architecture of the language faculty. Cambridge, MA: MIT Press. Keskin, C. 2009. Subject agreement-dependency of accusative case in Turkish. Utrecht: LOT. Kratzer, A. 1996. Severing the external argument from its verb. In J. Rooryck & L. A. Zaring (eds.), Phrase structure and the lexicon, 109–137. Dordrecht: Kluwer Academic Publishers. Laka, I. 2000. Theta blind Case: Burzio’s Generalisation and its image in the mirror. In E. Reuland (ed.), Arguments and case, 103–129. Amsterdam: John Benjamins. Longobardi, G. 2001. The structure of DPs: some principles, parameters and problems. In M Baltin & C. Collins (eds.), The handbook of contemporary syntactic theory, 562–603. Oxford: Blackwell. Marantz, A. 1997. No escape from syntax: Don’t try morphological analysis in the privacy of your own lexicon. University of Pennsylvania Working Papers in Linguistics 4(2), 201–225. Picallo, C. 1991. Nominals and nominalizations in Catalan. Probus 3(3), 279–316. Radford, A. 2000. NP shells. Essex Research Reports in Linguistics 33, 2–20. Rezac, M. 2008. Phi-agree and theta-related case. In D. Harbour, D. Adger & S. Béjar (eds.), Phi theory: Phi-features across modules and interfaces Oxford Studies in Theoretical Linguistics, 83–129. Oxford: Oxford University Press. Rezac, M., P. Albizu & R. Etxepare. 2011. The structural ergative of Basque and the theory of case. Case at the interfaces, Brussels Conference on Generative Linguistics (BCGL). Richards, N. 2001. Movement in language: Interactions and architectures. Oxford: Oxford University Press. Schoorlemmer, M. 1998. Possessors, articles, and definiteness. In A. Alexiadou & C. Wilder (eds.), Possessors, predicates and movement in the determiner phrase, 55–86. Amsterdam: John Benjamins. Szabolcsi, A. 1994. The noun phrase. In K. É. Kiss & F. Kiefer (eds.), The syntactic structure of Hungarian, 55–86. San Diego: Academic Press. Iberia: IJTL |Volume 6 (2014), 1–26 ISSN: 1989-8525 http://revistas.ojs.es/index.php/iberia/ 25