Compound tenses may display double agreement in non-standard varieties of Spanish. Harris & Halle (2005) present a body of new data for affirmative imperatives, where third person plural -n is reduplicated (once or twice) or switches places with a clitic (metathesis). Kayne (2008) proposes a syntactic reinterpretation of the data, analyzing imperatives as compound tenses with silent auxiliaries (Kayne 1992). The contending assumptions in these works concern a long standing debate on whether agreement morphology is a product of syntactic operations or the syntax-phonology interface. This paper defends the former view building on an independent proposal by Alcázar and Saltarelli (2008a,b), who identify a prescriptive light verb in imperative clauses. We extend the analysis to imperative expressions with first and third person subjects, proposing that these imperative clauses feature an additional causative head.
In this paper we provide a principled account of the syntactic properties of Spanish Comparative Correlatives (CCs) within Principles and Parameters theory. CCs have lately been at the centre of the debate between Construction Theory proponents, who claim construction status for them because of their idiosyncratic syntax and semantics, and supporters of UG-based syntax.1We contribute to this debate by showing that, despite appearances, Spanish CCs have a regular internal and external syntax. Assuming a cartographic approach to the syntax of Topic and Focus, we argue that their informational properties are the clue to their macrostructure. Specifically, we propose that C1, the first clause of CCs, is a subordinate clause that sits in the specifier position of the topic Phrase of the main clause and is followed by a focus-fronted constituent which occupies the specifier position of the focus Phrase of the main clause. We also show that our analysis can be extended to other sentence-initial adverbial adjunct clauses.
The goal of this paper is to describe the distribution and to explain the nature of the vowels that appear in preverbal position in main questions in many Northern Italian dialects. We use data primarily from the town of Donceto in the province of Piacenza, as well as data from many other Northern Italian dialects, including other Emilian dialects, Piedmontese dialects, many Veneto dialects, Friulian dialects, and standard Italian. We suggest that the preverbal vowels are the spell-out of functional heads of the CP and IP layers, and that they should be distinguished from true subject clitic pronouns. Furthermore, the functional vowels can realize different functional heads in one and the same dialect, and they can have a different distribution in different dialects.
This paper discusses the peculiar ability of certain deictic locatives (like lá ‘there’) to appear left-adjacent to the verb in European Portuguese. We propose that leftward movement of lá (and similar deictic locatives) is middle scrambling, understood as movement to Spec,TP. In order to explain why lá-preposing to Spec,TP is not always permitted, we elaborate on the hypothesis of Costa & Martins (2003, 2004) that in EP the strong nature of the polarity-encoding head Σ requires it to be ‘lexicalized’ either by syntactic merger or by morphological merger under adjacency. Middle scrambling is barred whenever Σ and V must be adjacent. The analysis derives the particular syntax of the deictic locatives (in different clausal structures, including restructuring infinitives) and its puzzling parallelism with clitic placement. Finally, we suggest that speaker/utterance-anchorage is what links together deictic locatives and tense, enabling the former to enter the syntactic domain of the latter.
We provide a principled account of the morphosyntax-semantics interface of non-root modals in two Romance languages (Spanish/French) vs. English. While English modals are morphologically impoverished, Romance modals are fully inflected for tense and aspect and the possible combinations of tense and aspect constrain the range of construals available: epistemic vs. metaphysical. We uniformly derive the range of possible construals from a restricted set of assumptions: (i) Tense/t°, Modal/m°, Aspect/asp° and v° each contribute to the temporal calculus of the clause in which they occur a time argument projected in the syntax as a Zeit-P; (ii) zeit-ps can enter into anaphoric and scopal dependencies. This proposal derives the temporal construals of non-root modals from a single phrase-structure (Tense-P > Modal-P > Aspect-P) without appealing to dedicated hierarchies of functional projections. Syntactic movement of time arguments (Zeit-Ps) and/or temporal heads (t°/asp°) ultimately accounts for cross-linguistic variation in the morphosyntax of these construals.
This paper analyzes degree constructions in Romanian and describes a visibility requirement based on representations at the level of Logical Form. The proposal thus follows the mechanism of LF-binding (Hulk & Verheugd 1994). We transfer this insight to the extraction and binding of degrees, by observing certain correlations between constructions for which a parameter of degree binding has been suggested (Beck, Oda & Sugisaki 2004), and the overt realization of the same key constructions in Romanian. The analysis is developed in terms of last-resort insertion of functional material within the adjectival shell which is bound over at LF.
The Romanian presumptive verbal paradigm (aux + be + present/past participle) is puzzling in several respects: (i) it is the only modal/temporal/aspectual construct which allows the present participle; (ii) it can use any of the modal auxiliaries in the language, in order to assemble verbal forms which convey indirect evidentiality; (iii) with the past participle, indirect evidential meanings and other modal meanings create syncretism. A problem these characteristics pose is to understand the nature of indirect evidentiality, and its mapping to the morphology. This paper proposes a morpho-semantics analysis of the presumptive; the essential part of the account is that aspectual heads can be interpreted modally, in the domain of worlds (Iatridou 2000, Izvorski 1997). The specific semantics of the participles, as well as the contribution of be derive indirect evidentiality, defined as speaker’s non-awareness of the eventuality itself.
Harris & Halle (2005) present a carefully worked out analysis of certain nonstandard Spanish phenomena involving pronominal clitics and the verbal plural morpheme -n. At issue are plural imperatives in combination with one or more object clitics. In this paper, I suggest that Harris & Halle’s primarily morphological approach to these phenomena should be replaced by a more syntactic approach. The latter seems more revealing and more likely to tie in to other aspects of Spanish grammar (and to aspects of the grammar of other languages/dialects).
Building on previous findings on tense construals and on constraints on modal-temporal configurations, this paper presents a semantic analysis of subjunctive tenses in Spanish which departs from the defective-tense hypothesis. The subjunctive tense system is analysed in a parallel way to the indicative system, and the peculiarities of the imperfect subjunctive are shown to mirror those of the imperfect indicative. Key to the analysis is the notion of ‘fake past’, a tense anchored to an interval Tx distinct from the time of utterance, which can be either temporally or modally bound. The semantic contribution of subjunctive tenses is examined first in root contexts and subsequently in argument clauses. The possibility of temporal disharmony and its limits are interpreted on the basis of Sequence of Tense principles and of temporal restrictions on modal bases.
This work proposes an analysis of Italian nouns. It explores the concept of the “final vowel” and claims that it is an analyzable object which is active in the formation of nouns in the language. The paper suggests that each “final vowel” is a complex morphophonological object (in the spirit of Kaye, Lowenstamm & Vergnaud 1985, 1990) and that only a syntactic approach to noun formation (Halle & Marantz 1993) can fully account for the distribution of such morphophonological complexes. On a more general level, the analysis depicted explains the behavior and the formation of non-derived simple nouns in Italian.
One known puzzle in Creole systems is that temporal interpretation seems to be constrained by stativity (Bickerton 1974). For decades, the relevant division has been, roughly: bare stative verbs mean present, bare nonstatives mean past. In Capeverdean, a Portuguese-based Creole, we do indeed have: N sabe risposta “I know the answer”, N kume pexe “I ate fish”. The above generalization, however, is inaccurate: most Capeverdean statives pattern with nonstatives in this respect. Crucially, also sabe “know” may pattern with nonstatives, challenging further this traditional view. In this paper I argue that the distinct temporal readings above can only be explained via the internal structure of events. A Become subevent (Dowty 1979) accounts for N sabe risposta – “I got to know the answer”, with its consequent state (Moens & Steedman 1988) being “[now] I know.” In contrast, there is no consequent state as “I eat fish” for “I ate fish” (cf. “I’ve eaten.”).
This paper pursues an analysis of verbs like Italian mordicchiare (nibble) as event-internal pluractional verbs that denote composite single events where the predicate is distributed on the fragments of one entity, and grammaticise a local form of number through the part-of relation. This opens the possibility of reading number marking in aspectual terms, whereby fragmenting is a form of modification that perturbs the mapping between event and object.
Compound tenses may display double agreement in non-standard varieties of Spanish. Harris & Halle (2005) present a body of new data for affirmative imperatives, where third person plural -n is reduplicated (once or twice) or switches places with a clitic (metathesis). Kayne (2008) proposes a syntactic reinterpretation of the data, analyzing imperatives as compound tenses with silent auxiliaries (Kayne 1992). The contending assumptions in these works concern a long standing debate on whether agreement morphology is a product of syntactic operations or the syntax-phonology interface. This paper defends the former view building on an independent proposal by Alcázar and Saltarelli (2008a,b), who identify a prescriptive light verb in imperative clauses. We extend the analysis to imperative expressions with first and third person subjects, proposing that these imperative clauses feature an additional causative head.
In this paper we provide a principled account of the syntactic properties of Spanish Comparative Correlatives (CCs) within Principles and Parameters theory. CCs have lately been at the centre of the debate between Construction Theory proponents, who claim construction status for them because of their idiosyncratic syntax and semantics, and supporters of UG-based syntax.1We contribute to this debate by showing that, despite appearances, Spanish CCs have a regular internal and external syntax. Assuming a cartographic approach to the syntax of Topic and Focus, we argue that their informational properties are the clue to their macrostructure. Specifically, we propose that C1, the first clause of CCs, is a subordinate clause that sits in the specifier position of the topic Phrase of the main clause and is followed by a focus-fronted constituent which occupies the specifier position of the focus Phrase of the main clause. We also show that our analysis can be extended to other sentence-initial adverbial adjunct clauses.
The goal of this paper is to describe the distribution and to explain the nature of the vowels that appear in preverbal position in main questions in many Northern Italian dialects. We use data primarily from the town of Donceto in the province of Piacenza, as well as data from many other Northern Italian dialects, including other Emilian dialects, Piedmontese dialects, many Veneto dialects, Friulian dialects, and standard Italian. We suggest that the preverbal vowels are the spell-out of functional heads of the CP and IP layers, and that they should be distinguished from true subject clitic pronouns. Furthermore, the functional vowels can realize different functional heads in one and the same dialect, and they can have a different distribution in different dialects.
This paper discusses the peculiar ability of certain deictic locatives (like lá ‘there’) to appear left-adjacent to the verb in European Portuguese. We propose that leftward movement of lá (and similar deictic locatives) is middle scrambling, understood as movement to Spec,TP. In order to explain why lá-preposing to Spec,TP is not always permitted, we elaborate on the hypothesis of Costa & Martins (2003, 2004) that in EP the strong nature of the polarity-encoding head Σ requires it to be ‘lexicalized’ either by syntactic merger or by morphological merger under adjacency. Middle scrambling is barred whenever Σ and V must be adjacent. The analysis derives the particular syntax of the deictic locatives (in different clausal structures, including restructuring infinitives) and its puzzling parallelism with clitic placement. Finally, we suggest that speaker/utterance-anchorage is what links together deictic locatives and tense, enabling the former to enter the syntactic domain of the latter.
We provide a principled account of the morphosyntax-semantics interface of non-root modals in two Romance languages (Spanish/French) vs. English. While English modals are morphologically impoverished, Romance modals are fully inflected for tense and aspect and the possible combinations of tense and aspect constrain the range of construals available: epistemic vs. metaphysical. We uniformly derive the range of possible construals from a restricted set of assumptions: (i) Tense/t°, Modal/m°, Aspect/asp° and v° each contribute to the temporal calculus of the clause in which they occur a time argument projected in the syntax as a Zeit-P; (ii) zeit-ps can enter into anaphoric and scopal dependencies. This proposal derives the temporal construals of non-root modals from a single phrase-structure (Tense-P > Modal-P > Aspect-P) without appealing to dedicated hierarchies of functional projections. Syntactic movement of time arguments (Zeit-Ps) and/or temporal heads (t°/asp°) ultimately accounts for cross-linguistic variation in the morphosyntax of these construals.
This paper analyzes degree constructions in Romanian and describes a visibility requirement based on representations at the level of Logical Form. The proposal thus follows the mechanism of LF-binding (Hulk & Verheugd 1994). We transfer this insight to the extraction and binding of degrees, by observing certain correlations between constructions for which a parameter of degree binding has been suggested (Beck, Oda & Sugisaki 2004), and the overt realization of the same key constructions in Romanian. The analysis is developed in terms of last-resort insertion of functional material within the adjectival shell which is bound over at LF.
The Romanian presumptive verbal paradigm (aux + be + present/past participle) is puzzling in several respects: (i) it is the only modal/temporal/aspectual construct which allows the present participle; (ii) it can use any of the modal auxiliaries in the language, in order to assemble verbal forms which convey indirect evidentiality; (iii) with the past participle, indirect evidential meanings and other modal meanings create syncretism. A problem these characteristics pose is to understand the nature of indirect evidentiality, and its mapping to the morphology. This paper proposes a morpho-semantics analysis of the presumptive; the essential part of the account is that aspectual heads can be interpreted modally, in the domain of worlds (Iatridou 2000, Izvorski 1997). The specific semantics of the participles, as well as the contribution of be derive indirect evidentiality, defined as speaker’s non-awareness of the eventuality itself.
Harris & Halle (2005) present a carefully worked out analysis of certain nonstandard Spanish phenomena involving pronominal clitics and the verbal plural morpheme -n. At issue are plural imperatives in combination with one or more object clitics. In this paper, I suggest that Harris & Halle’s primarily morphological approach to these phenomena should be replaced by a more syntactic approach. The latter seems more revealing and more likely to tie in to other aspects of Spanish grammar (and to aspects of the grammar of other languages/dialects).
Building on previous findings on tense construals and on constraints on modal-temporal configurations, this paper presents a semantic analysis of subjunctive tenses in Spanish which departs from the defective-tense hypothesis. The subjunctive tense system is analysed in a parallel way to the indicative system, and the peculiarities of the imperfect subjunctive are shown to mirror those of the imperfect indicative. Key to the analysis is the notion of ‘fake past’, a tense anchored to an interval Tx distinct from the time of utterance, which can be either temporally or modally bound. The semantic contribution of subjunctive tenses is examined first in root contexts and subsequently in argument clauses. The possibility of temporal disharmony and its limits are interpreted on the basis of Sequence of Tense principles and of temporal restrictions on modal bases.
This work proposes an analysis of Italian nouns. It explores the concept of the “final vowel” and claims that it is an analyzable object which is active in the formation of nouns in the language. The paper suggests that each “final vowel” is a complex morphophonological object (in the spirit of Kaye, Lowenstamm & Vergnaud 1985, 1990) and that only a syntactic approach to noun formation (Halle & Marantz 1993) can fully account for the distribution of such morphophonological complexes. On a more general level, the analysis depicted explains the behavior and the formation of non-derived simple nouns in Italian.
One known puzzle in Creole systems is that temporal interpretation seems to be constrained by stativity (Bickerton 1974). For decades, the relevant division has been, roughly: bare stative verbs mean present, bare nonstatives mean past. In Capeverdean, a Portuguese-based Creole, we do indeed have: N sabe risposta “I know the answer”, N kume pexe “I ate fish”. The above generalization, however, is inaccurate: most Capeverdean statives pattern with nonstatives in this respect. Crucially, also sabe “know” may pattern with nonstatives, challenging further this traditional view. In this paper I argue that the distinct temporal readings above can only be explained via the internal structure of events. A Become subevent (Dowty 1979) accounts for N sabe risposta – “I got to know the answer”, with its consequent state (Moens & Steedman 1988) being “[now] I know.” In contrast, there is no consequent state as “I eat fish” for “I ate fish” (cf. “I’ve eaten.”).
This paper pursues an analysis of verbs like Italian mordicchiare (nibble) as event-internal pluractional verbs that denote composite single events where the predicate is distributed on the fragments of one entity, and grammaticise a local form of number through the part-of relation. This opens the possibility of reading number marking in aspectual terms, whereby fragmenting is a form of modification that perturbs the mapping between event and object.