The present volume contains a broad range of peer-reviewed articles dealing with syntax, phonology, morphology, semantics and acquisition of the Romance languages as well as selected papers from the special workshop dealing with linguistic change in relation to linguistic theory.
This paper studies the Romanian conjunction iar. After a general introduction to the connective and the existing approaches, we defend an approach of iar as an information structure sensitive item. We describe two constraints on the use of iar that account for a wide range of data. We then use our analysis to compare the system of Romanian conjunctions to that of other languages, especially Russian, which uses a connective that appears similar to Romanian iar.
This paper shows that French non-core datives introduced by applicative heads may attach at two different positions above the VP. If the Appl head attaches in the thematic domain, above V but below v, a new event participant is added; we refer to this configuration as the Benefactive Dative Construction. If the Appl head is attached above v, in a non-thematic environment, it does not introduce a new event participant, hence its deficient realization as se, to which configuration we refer as the Coreferential Dative Construction. We provide arguments against a low-applicative analysis (below VP), and show how to distinguish between the two types of applicative constructions (above VP) both semantically and syntactically.
In this paper, we compare two syntactic constructions involving degree adverbs in English and Québec French: the Degree Fronting (DF) construction and the Intensification at a Distance (IAD) construction. We argue that, although they display some similar properties, these similarities are superficial. We argue that, while DF can be analyzed as involving movement, IAD cannot. We propose that the quantifiers in IAD sentences are base-generated in their surface positions, and that these syntactic positions coincide with the positions that the quantifiers occupy when they are quantifying over individuals or events. Furthermore, we argue that dialectal variation in distance quantificational structures between Standard European French and Québec French is due to differences in the semantics of degree adverbs in these dialects.
This paper focuses on the sentence-internal reading of French le même (‘the same’)and addresses two main issues: a. the problem of definiteness (le même does not behave the way standard definites do); b. the problem of compositionality (the plural licenser that même needs to get interpreted is not adjacent to it). I propose that: a. le même is a complex determiner with specific properties with respect to presupposition and specificity; b. le même is an existential quantifier over a plural event that is partitioned along participants or times. This licensing condition (plural event distributed through participants or times) relates même to the notion of pluractionality.
The existing syntactic differences between European and Brazilian Portuguese have received much attention in recent literature on generative syntax. According to several authors, Brazilian Portuguese became a discourse-oriented language with topic-prominence. This article shows that many constructions used to illustrate the specificity of Brazilian Portuguese as topic-prominent can also be found in European Portuguese. Accordingly, it is argued that the main difference between the two varieties of the language is not in discourse, but in the abstract syntactic properties associated with Inflection. This argument contributes to dispensing with discourse primitives in the syntactic component.
Friedmann’s work has shown that the syntactic productions of Broca’s aphasics are much more selectively impaired than previously thought. Here we entertain the Tree Pruning Hypothesis of Friedmann and evaluate its predictions in Catalan, Galician and Spanish, under the assumption that the functional structure of the tree is that postulated by the cartographic approach. We report the results of tests on the production of sentential negation, tense and subject agreement, clitic production, question production, and relative clause production, run with fifteen agrammatic patients. Overall, the results are consistent with the Tree Pruning Hypothesis for all subjects of all these languages. Further, on the basis of various tense and aspect projections, which are all equally impaired, we conclude that there are no distinguishable levels of impairment within a field. The analysis of question production provides an argument for the merge of why questions and yes/no interrogatives in the same functional projection.
This paper analyzes gender agreement in the Determiner Phrase in Afro-Bolivian Spanish. Our data shows a case of cross-generational change, transitioning from a basilectal Afro-Bolivian variety to the more prestigious standard Bolivian Spanish. Recent minimalist models account for variability as the differential specification and computation of uninterpretable features in a derivation. We argue that dialectal variation and change in ABS, even if externally driven by social factors, follow from feature valuation and checking and from the interplay between computational and evolutionary demands.
Recent minimalist approaches have reduced case to independent primitives (agreement, Tense) – but without any connection to its morphological expression. To solve this dichotomy, we consider the Latin -s case ending. Rejecting default treatments, we conclude that -s is associated with denotational, operator properties. These can be read as the set forming operator i.e. plural; as the inclusion operator, i.e. partitive, possessor, etc. (in a word ‘oblique’); or as the quantificational closure of EPP contexts (‘nominative’). These properties are preserved in the two-case declension of medieval Gallo-Romance, and in its residues in Romansh varieties. Thus so-called case is a denotational, ‘determiner-like’ element, with consequences for the classical historical correlation between loss of Latin case and development of the Romance determiner system.
In this paper, based on Bisetto & Scalise’s (2005) classification of compounds, I discuss the variation between English and Romanian and Spanish in endocentric subordinate compounds, where the constituents are linked by complement-head relations. I argue that in this type of compounding, the different strategies employed by various languages are only Case-related, i.e. the Case of the complement can be checked by incorporation in English, de-insertion in Romance or Th(ematic) relational adjectives in Romance and English. However, de phrases and Th-adjectives differ with respect to the checking of Genitive Case. In the absence of de last resort insertion, Th-adjectives check the Genitive Case only nP internally, as a full Gen DP which is in long distance Agree with AgrP.
The aim of this paper is to provide additional arguments against the view that on the epistemic reading of modal verbs, the time of the modal is always the utterance time. The hypothesis defended, also adopted by Eide (2002, 2003) and von Fintel and Gillies (2008) is that epistemic modals can be in the scope of Tense/Aspect. Three possible translations of might have been in French (with a passé composé or an imparfait on the modal and a simple infinitival, or with a present on the modal and a perfect infinitival) are semantically differentiated. The analysis describes the distribution of past tenses on epistemic modality and explains the differences in their interpretation. Possibilities are the sort of thing that comes into and goes out of existence, that can be ‘dated’ (Mondadori 1978, p. 246) It is obvious that we don’t have a good understanding of what happens when a modal is combined with temporal operators. (Portner 2009, p. 230)
Floating parenthetical coordinate clauses exhibit a challenging behaviour: they disrupt the structure of the host sentence, do not present an overt first term, occur in different positions inside the host clause and, although notionally related to their host, they present syntactic autonomy. Taking into account data from European Portuguese, we claim that these clauses are derived from the core devices of the computational system: the coordinate structure is built up by Set Merge and takes, as first term, a null constituent denoting the host clause; then, Pair Merge operates by adjoining the parenthetical coordinated CP to a functional or verbal projection of the host sentence. Considering the autonomy of the parenthetical clause with respect to its host, we assume that this adjunction is an instance of Late Merge, a counter cyclic operation that applies at PF.
This paper examines the semantic properties of evaluative adverbs, such as unfortunately, in question environments in French. We take Bonami and Godard’s (2008) analysis for malheureusement (‘unfortunately’) in declarative sentences, where such propositional adverbs are analyzed as ancillary commitments, and extend it to a broader array of data, including polar and wh-questions. In a nutshell, we argue that malheureusement can take as input a set of propositions, which triggers an indifference interpretation of the sort that characterizes unconditional sentences. We also show that evaluatives may take a proposition as argument when they occur in polar questions, the only restriction occurring in negative polar questions, where biases are decisive in making the sentence acceptable.
By and large in the generative framework, syntactic comparisons concern isolated mechanisms within different languages in a typological perspective. However, comparison can provide more if, as dialectologists, we consider that variation between closely genealogically related languages throws light on diachronic reconstruction. Indeed, if diatopic variation represents different stages of a change, then different dialects in a homogeneous area display different stages of the evolution and the speakers’ I-language for the earlier stages becomes accessible. Considering the Null ‘Subject’ Parameter, we focus on the dialects at the boundaries of Occitania since they display subject clitics for some persons only. Our data illustrate the gradual change between a stage zero (with no subject clitics, e.g. Latin) and a stage n (with subject clitics for all persons, e.g. Standard French). Analyzing clitics as bundles of features, we propose a progression in their emergence, which ties in with a functional ‘motivational cycle’ of the morpho-syntactic elements.
In this paper we provide a syntactic analysis of free exceptive constructions headed by excepto and salvo (‘except’) in Spanish: Todos los estudiantes cantaron, {excepto/salvo} Juan (‘Every student sang, except John’). Our claim is that free exceptives are coordinated elliptical sentences attached to the CP level of a host clause that expresses a generalization statement. We frame our analysis in the Boolean Phrase Hypothesis (Munn 1993), and defend that exceptive markers head a Boolean Phrase, as other coordinating conjunctions do. We also argue that exceptive markers select for a full-fledged CP as complement, whose null C head triggers movement of an XP constituent to its Specifier position (Juan, in the example above) and also triggers a process of ellipsis in which all the syntactic material inside the TP selected by C is marked for PF-deletion, along the lines of Merchant (2003).
In Majorcan Catalan, the process of vowel reduction of the mid front vowels to schwa in unstressed position underapplies under certain circumstances: (a) in productive derived forms with an unstressed vowel located in the initial syllable of the stem which alternates with a stressed mid front vowel in the stem of the underived form (p[é]ix ‘fish’ ˜ p[e]ixet ‘fish dim.’); (b) in verbal forms with an unstressed vowel located in the initial syllable of the stem which alternates with a stressed close mid front vowel in another verbal form of the same inflectional paradigm (p[é]ga ‘(s/he) hits’ ˜ p[e]gam ‘(we) hit’); (c) in learned and loan words with an unstressed e located in the initial syllable of the stem (p[e]culiar ‘peculiar’). In this paper I propose a novel explanation of these patterns framed within a relativized version of the Transderivational Correspondence Theory (TCT) (Benua 1997/2000), the Optimal Paradigms model (OP) (McCarthy 2005), the Positional Faithfulness Theory (Beckman 1998/1999) and the prominence driven approach to vowel reduction (Crosswhite 1999/2001, 2004).
During Spain’s ‘Golden Age’ period, n-words in Spanish, such as nada ‘nothing’, changed from being negative polarity items to negative concord items. During the same period, immediately pre-verbal n-words, which previously had expressed wide-focus for VP constitutes, came to acquire a mildly emphatic interpretation, which survives into the modern language as Quantifier Fronting (Quer 2002) or ‘verum-focus’ fronting (Leonetti & Escandell Vidal 2007, 2009). This development, in which negative polarity items seem to acquire a focus feature in the context of becoming negative concord items, is of particular interest because it provides indirect support for Watanabe’s (2004) account of negative concord, in which a focus feature is crucially implicated.
This paper focuses on the morphosyntax of a kind of verbal duplication attested in River Plate Spanish (e.g. Vení acá, vení Lit: ‘come here come’). One striking property of this construction is that it requires that the duplicated verbs not be adjacent. I show that verbal duplication in River Plate Spanish instantiates a case of head copy realization and that the observed anti-adjacency effect reflects an underlying condition on non-pronunciation of heads in general. Some cases of head copy realization under apparent adjacency are independently accounted for once parts of words cannot be subject to morphological ellipsis.
In Old and Middle French it is possible to find Stylistic Fronting where both a head and a phrase have been fronted. I have examined SF constructions where the infinitive dire ‘say’ or faire ‘do’ are fronted, in particular the cases where they are preceded by their complement. In order to account for this construction, I propose that the elements are moved together by Remnant Movement. I also examine the structure of the vP, where the object or an adverb may precede the non-finite verb. This intermediary step of scrambling in the vP is necessary in order for these elements to be moved to the left periphery of the clause.
Several Romance languages show restrictions on combinations of third person direct and indirect object clitics (Bonet 1995) and combinations of such clitics involving local person direct objects (the Person Case Constraint (PCC), Bonet 1991, 1994). The former have received morphological analyses, while the later have recently been treated as syntactic. A unified, syntactic analysis of both of these restrictions is developed by extending existing analyses of the PCC. Person restrictions are derived in a syntactic configuration where DO bleeds person licensing on IO. A more complex syntactic representation of third person is proposed. The starting point for the investigation is Barceloní Catalan, which evades both restrictions by not realizing person morphology on third person indirect objects. The proposal is subsequently extended to Spanish, in particular restrictions on animate direct objects pronouns in some varieties of leísta Spanish (Ormazabal & Romero 2007). The syntactic proposal combines with the proposal for the semantics and pragmatics of person features in Sauerland (2004, 2008) to derive interpretive aspects of the leísta Spanish data.
The present volume contains a broad range of peer-reviewed articles dealing with syntax, phonology, morphology, semantics and acquisition of the Romance languages as well as selected papers from the special workshop dealing with linguistic change in relation to linguistic theory.
This paper studies the Romanian conjunction iar. After a general introduction to the connective and the existing approaches, we defend an approach of iar as an information structure sensitive item. We describe two constraints on the use of iar that account for a wide range of data. We then use our analysis to compare the system of Romanian conjunctions to that of other languages, especially Russian, which uses a connective that appears similar to Romanian iar.
This paper shows that French non-core datives introduced by applicative heads may attach at two different positions above the VP. If the Appl head attaches in the thematic domain, above V but below v, a new event participant is added; we refer to this configuration as the Benefactive Dative Construction. If the Appl head is attached above v, in a non-thematic environment, it does not introduce a new event participant, hence its deficient realization as se, to which configuration we refer as the Coreferential Dative Construction. We provide arguments against a low-applicative analysis (below VP), and show how to distinguish between the two types of applicative constructions (above VP) both semantically and syntactically.
In this paper, we compare two syntactic constructions involving degree adverbs in English and Québec French: the Degree Fronting (DF) construction and the Intensification at a Distance (IAD) construction. We argue that, although they display some similar properties, these similarities are superficial. We argue that, while DF can be analyzed as involving movement, IAD cannot. We propose that the quantifiers in IAD sentences are base-generated in their surface positions, and that these syntactic positions coincide with the positions that the quantifiers occupy when they are quantifying over individuals or events. Furthermore, we argue that dialectal variation in distance quantificational structures between Standard European French and Québec French is due to differences in the semantics of degree adverbs in these dialects.
This paper focuses on the sentence-internal reading of French le même (‘the same’)and addresses two main issues: a. the problem of definiteness (le même does not behave the way standard definites do); b. the problem of compositionality (the plural licenser that même needs to get interpreted is not adjacent to it). I propose that: a. le même is a complex determiner with specific properties with respect to presupposition and specificity; b. le même is an existential quantifier over a plural event that is partitioned along participants or times. This licensing condition (plural event distributed through participants or times) relates même to the notion of pluractionality.
The existing syntactic differences between European and Brazilian Portuguese have received much attention in recent literature on generative syntax. According to several authors, Brazilian Portuguese became a discourse-oriented language with topic-prominence. This article shows that many constructions used to illustrate the specificity of Brazilian Portuguese as topic-prominent can also be found in European Portuguese. Accordingly, it is argued that the main difference between the two varieties of the language is not in discourse, but in the abstract syntactic properties associated with Inflection. This argument contributes to dispensing with discourse primitives in the syntactic component.
Friedmann’s work has shown that the syntactic productions of Broca’s aphasics are much more selectively impaired than previously thought. Here we entertain the Tree Pruning Hypothesis of Friedmann and evaluate its predictions in Catalan, Galician and Spanish, under the assumption that the functional structure of the tree is that postulated by the cartographic approach. We report the results of tests on the production of sentential negation, tense and subject agreement, clitic production, question production, and relative clause production, run with fifteen agrammatic patients. Overall, the results are consistent with the Tree Pruning Hypothesis for all subjects of all these languages. Further, on the basis of various tense and aspect projections, which are all equally impaired, we conclude that there are no distinguishable levels of impairment within a field. The analysis of question production provides an argument for the merge of why questions and yes/no interrogatives in the same functional projection.
This paper analyzes gender agreement in the Determiner Phrase in Afro-Bolivian Spanish. Our data shows a case of cross-generational change, transitioning from a basilectal Afro-Bolivian variety to the more prestigious standard Bolivian Spanish. Recent minimalist models account for variability as the differential specification and computation of uninterpretable features in a derivation. We argue that dialectal variation and change in ABS, even if externally driven by social factors, follow from feature valuation and checking and from the interplay between computational and evolutionary demands.
Recent minimalist approaches have reduced case to independent primitives (agreement, Tense) – but without any connection to its morphological expression. To solve this dichotomy, we consider the Latin -s case ending. Rejecting default treatments, we conclude that -s is associated with denotational, operator properties. These can be read as the set forming operator i.e. plural; as the inclusion operator, i.e. partitive, possessor, etc. (in a word ‘oblique’); or as the quantificational closure of EPP contexts (‘nominative’). These properties are preserved in the two-case declension of medieval Gallo-Romance, and in its residues in Romansh varieties. Thus so-called case is a denotational, ‘determiner-like’ element, with consequences for the classical historical correlation between loss of Latin case and development of the Romance determiner system.
In this paper, based on Bisetto & Scalise’s (2005) classification of compounds, I discuss the variation between English and Romanian and Spanish in endocentric subordinate compounds, where the constituents are linked by complement-head relations. I argue that in this type of compounding, the different strategies employed by various languages are only Case-related, i.e. the Case of the complement can be checked by incorporation in English, de-insertion in Romance or Th(ematic) relational adjectives in Romance and English. However, de phrases and Th-adjectives differ with respect to the checking of Genitive Case. In the absence of de last resort insertion, Th-adjectives check the Genitive Case only nP internally, as a full Gen DP which is in long distance Agree with AgrP.
The aim of this paper is to provide additional arguments against the view that on the epistemic reading of modal verbs, the time of the modal is always the utterance time. The hypothesis defended, also adopted by Eide (2002, 2003) and von Fintel and Gillies (2008) is that epistemic modals can be in the scope of Tense/Aspect. Three possible translations of might have been in French (with a passé composé or an imparfait on the modal and a simple infinitival, or with a present on the modal and a perfect infinitival) are semantically differentiated. The analysis describes the distribution of past tenses on epistemic modality and explains the differences in their interpretation. Possibilities are the sort of thing that comes into and goes out of existence, that can be ‘dated’ (Mondadori 1978, p. 246) It is obvious that we don’t have a good understanding of what happens when a modal is combined with temporal operators. (Portner 2009, p. 230)
Floating parenthetical coordinate clauses exhibit a challenging behaviour: they disrupt the structure of the host sentence, do not present an overt first term, occur in different positions inside the host clause and, although notionally related to their host, they present syntactic autonomy. Taking into account data from European Portuguese, we claim that these clauses are derived from the core devices of the computational system: the coordinate structure is built up by Set Merge and takes, as first term, a null constituent denoting the host clause; then, Pair Merge operates by adjoining the parenthetical coordinated CP to a functional or verbal projection of the host sentence. Considering the autonomy of the parenthetical clause with respect to its host, we assume that this adjunction is an instance of Late Merge, a counter cyclic operation that applies at PF.
This paper examines the semantic properties of evaluative adverbs, such as unfortunately, in question environments in French. We take Bonami and Godard’s (2008) analysis for malheureusement (‘unfortunately’) in declarative sentences, where such propositional adverbs are analyzed as ancillary commitments, and extend it to a broader array of data, including polar and wh-questions. In a nutshell, we argue that malheureusement can take as input a set of propositions, which triggers an indifference interpretation of the sort that characterizes unconditional sentences. We also show that evaluatives may take a proposition as argument when they occur in polar questions, the only restriction occurring in negative polar questions, where biases are decisive in making the sentence acceptable.
By and large in the generative framework, syntactic comparisons concern isolated mechanisms within different languages in a typological perspective. However, comparison can provide more if, as dialectologists, we consider that variation between closely genealogically related languages throws light on diachronic reconstruction. Indeed, if diatopic variation represents different stages of a change, then different dialects in a homogeneous area display different stages of the evolution and the speakers’ I-language for the earlier stages becomes accessible. Considering the Null ‘Subject’ Parameter, we focus on the dialects at the boundaries of Occitania since they display subject clitics for some persons only. Our data illustrate the gradual change between a stage zero (with no subject clitics, e.g. Latin) and a stage n (with subject clitics for all persons, e.g. Standard French). Analyzing clitics as bundles of features, we propose a progression in their emergence, which ties in with a functional ‘motivational cycle’ of the morpho-syntactic elements.
In this paper we provide a syntactic analysis of free exceptive constructions headed by excepto and salvo (‘except’) in Spanish: Todos los estudiantes cantaron, {excepto/salvo} Juan (‘Every student sang, except John’). Our claim is that free exceptives are coordinated elliptical sentences attached to the CP level of a host clause that expresses a generalization statement. We frame our analysis in the Boolean Phrase Hypothesis (Munn 1993), and defend that exceptive markers head a Boolean Phrase, as other coordinating conjunctions do. We also argue that exceptive markers select for a full-fledged CP as complement, whose null C head triggers movement of an XP constituent to its Specifier position (Juan, in the example above) and also triggers a process of ellipsis in which all the syntactic material inside the TP selected by C is marked for PF-deletion, along the lines of Merchant (2003).
In Majorcan Catalan, the process of vowel reduction of the mid front vowels to schwa in unstressed position underapplies under certain circumstances: (a) in productive derived forms with an unstressed vowel located in the initial syllable of the stem which alternates with a stressed mid front vowel in the stem of the underived form (p[é]ix ‘fish’ ˜ p[e]ixet ‘fish dim.’); (b) in verbal forms with an unstressed vowel located in the initial syllable of the stem which alternates with a stressed close mid front vowel in another verbal form of the same inflectional paradigm (p[é]ga ‘(s/he) hits’ ˜ p[e]gam ‘(we) hit’); (c) in learned and loan words with an unstressed e located in the initial syllable of the stem (p[e]culiar ‘peculiar’). In this paper I propose a novel explanation of these patterns framed within a relativized version of the Transderivational Correspondence Theory (TCT) (Benua 1997/2000), the Optimal Paradigms model (OP) (McCarthy 2005), the Positional Faithfulness Theory (Beckman 1998/1999) and the prominence driven approach to vowel reduction (Crosswhite 1999/2001, 2004).
During Spain’s ‘Golden Age’ period, n-words in Spanish, such as nada ‘nothing’, changed from being negative polarity items to negative concord items. During the same period, immediately pre-verbal n-words, which previously had expressed wide-focus for VP constitutes, came to acquire a mildly emphatic interpretation, which survives into the modern language as Quantifier Fronting (Quer 2002) or ‘verum-focus’ fronting (Leonetti & Escandell Vidal 2007, 2009). This development, in which negative polarity items seem to acquire a focus feature in the context of becoming negative concord items, is of particular interest because it provides indirect support for Watanabe’s (2004) account of negative concord, in which a focus feature is crucially implicated.
This paper focuses on the morphosyntax of a kind of verbal duplication attested in River Plate Spanish (e.g. Vení acá, vení Lit: ‘come here come’). One striking property of this construction is that it requires that the duplicated verbs not be adjacent. I show that verbal duplication in River Plate Spanish instantiates a case of head copy realization and that the observed anti-adjacency effect reflects an underlying condition on non-pronunciation of heads in general. Some cases of head copy realization under apparent adjacency are independently accounted for once parts of words cannot be subject to morphological ellipsis.
In Old and Middle French it is possible to find Stylistic Fronting where both a head and a phrase have been fronted. I have examined SF constructions where the infinitive dire ‘say’ or faire ‘do’ are fronted, in particular the cases where they are preceded by their complement. In order to account for this construction, I propose that the elements are moved together by Remnant Movement. I also examine the structure of the vP, where the object or an adverb may precede the non-finite verb. This intermediary step of scrambling in the vP is necessary in order for these elements to be moved to the left periphery of the clause.
Several Romance languages show restrictions on combinations of third person direct and indirect object clitics (Bonet 1995) and combinations of such clitics involving local person direct objects (the Person Case Constraint (PCC), Bonet 1991, 1994). The former have received morphological analyses, while the later have recently been treated as syntactic. A unified, syntactic analysis of both of these restrictions is developed by extending existing analyses of the PCC. Person restrictions are derived in a syntactic configuration where DO bleeds person licensing on IO. A more complex syntactic representation of third person is proposed. The starting point for the investigation is Barceloní Catalan, which evades both restrictions by not realizing person morphology on third person indirect objects. The proposal is subsequently extended to Spanish, in particular restrictions on animate direct objects pronouns in some varieties of leísta Spanish (Ormazabal & Romero 2007). The syntactic proposal combines with the proposal for the semantics and pragmatics of person features in Sauerland (2004, 2008) to derive interpretive aspects of the leísta Spanish data.
The present volume contains a broad range of peer-reviewed articles dealing with syntax, phonology, morphology, semantics and acquisition of the Romance languages as well as selected papers from the special workshop dealing with linguistic change in relation to linguistic theory.