Search

This paper aims at demonstrating the validity of a two-pronged hypothesis: (i) that the aspectual viewpoint content of the so-called ‘narrative imparfait’ (NIMPF) does not bear on the verb it marks (i.e., it does not combine with the event predicate denoted by said verb) but that, therefore, (ii) it must operate at a higher, discursive semantic level. To substantiate the above hypothesis, the paper first focuses on diachronic and synchronic evidence suggesting that the NIMPF does not contribute aspectual meaning at the sentence semantics-level – showing notably that it behaves like a ‘viewpoint neutral’ tense with respect to the verb it marks. The paper then discusses synchronic, discursive evidence supporting the view that the NIMPF actually indicates a partial, discourse-structurally incomplete, ‘ongoing’ narrative act. From these two facts, the paper concludes that NIMPF utterances refer to imperfectively viewed narrative speech act events, and constitute a separate speech act-level conventionalized reading of the imparfait, applying an imperfective viewpoint meaning to relational speech act functions, i.e., to rhetorical relations. It is argued that they should be endowed with a speech act event argument, and constitute an abstract type of event predicate which the viewpoint meaning of the NIMPF takes as its input.

This paper revisits the status of subject clitics in Spoken French by studying subject doubling as a sociolinguistic variable. In the literature, two influential analyses have been proposed to account for French subject doubling. Based on new evidence from a corpus study on the large Multicultural Paris French (MPF) corpus, we argue for an analysis reconciling these two competing views of the construction in Spoken (colloquial) French. On one hand, we provide further support to the morphological approach (Auger, 1994; Culbertson, 2010) in which subject clitics are morphological agreement markers on the verb. On the other hand, we argue based on new evidence that lexical subjects are topicalized, as in the dislocation analysis. Furthermore, we argue that Spoken French is in a diglossia situation where speakers alternate structures provided by both Standard French and Colloquial French grammars. This paper provides further evidence of how quantitative studies of language use can shed light on long-standing theoretical debates.

The paper posits the hypothesis that if a form or construction entering a process of grammaticalization has a diachronic structural pattern, the grammaticalization may become entrenched and strengthened; it can recategorize and move easily into other grammatical areas, can make more complex diachronic paths, and will progress better than a change without a pattern. The paper poses two types of structural pattern: an intraparadigmatic one and an interparadigmatic one. The empirical evidence is the grammaticalization of the predication dice que into the evidential discourse marker dizque and later into the adjective dizque. The paper examines two patterns supporting this grammaticalization: first, the various grammaticalizations undergone by the verb decir into different discourse particles, an intraparadigmatic pattern, and second, three predications that undergone the same diachronic path: que es que > quesque; puede que > pueque; vaya > ¡vaya!, an interparadigmatic pattern.

This paper investigates to what extent adverbial causal clauses and relative clauses can be reduced as one and the same phenomenon. Whereas causal clauses have always posed a challenge for a unified account of relativization and adverbial subordination in theoretical studies, typological research has long demonstrated that causal clauses are diachronically connected to relative clauses as well as to adverbial subordinates that have been theoretically analysed as relative clauses. We argue that at least some causal clauses are underlyingly relative clauses over situations (see Arsenijević 2021). Our claim is supported by the diachronic development of the Italian subordinator siccome ‘because/since’, an univerbated form morphologically composed of two items, the comparative-similative wh-pronoun come ‘how’ and the demonstrative adverbial pronoun sì ‘so’. We demonstrate that the causal subordinator is derived from the comparative-similative one via a three-stage diachronic change which is formally captured in terms of type of movement and null elements (Kayne 2005, Cinque 2020b). In so doing, our paper extends a relative clause analysis to causal clauses and adds a novel path to the diachronic development of causal clauses.

This paper discusses the correlation between clitics, DOM and ellipsis in Spanish. As observed by Cyrino & Ordóñez (2018), strict and sloppy readings can be obtained in both TP ellipsis and argument ellipsis with clitics. However, DOM animate antecedents severely limit the availability of sloppy readings in argument ellipsis with clitics. This is a new observation, and judgments about the lack of sloppy readings is often delicate. We carried out an experimental study in order to confirm the data reported in Cyrino & Ordóñez (2018). We also wanted to test whether this distinction applied to speakers of three different dialects of Spanish where distribution of clitics and doubling differs. The results show a significant statistical anti animacy effect for obtaining sloppy readings with clitics. This effect is not found with TP ellipsis. In this paper we propose a syntactic approach to account for such effect with clitics. We propose that clitics contain an empty doubled (see also Bošković 2018). We propose an analysis based on the idea that DOM objects are moved out of the vP and that the doubled DOM KP undergoes vehicle change to become a pronoun in the case of argument ellipsis.

The following paper considers the potential for optionality of anaphoric null objects in two early Romance languages: Old French and Old Tuscan. In both languages, anaphoric objects can be omitted in different syntactic contexts (e.g., in coordination with prepositional infinitives or in adjunct non-finite clauses). However, overt objects can be found in almost all contexts which also allow null objects. The current paper argues that null objects in both Old French and Old Tuscan were optional, in the sense that they were never the only option for objects, and that having a null or an overt object did not have bearing in the interpretation of the sentence.

Long-distance dependencies are a major property of human language, deriving from the successive application of movement operations across sentences. Nevertheless, this possibility seems to be restricted by the presence of barriers, which ‘trap’ constituents in more or less 'resistant' syntactic islands. Given this scenario, this paper paper investigates a specific type of island, namely factive complements, which are traditionally considered ‘weak’ islands. The main goal is thus to verify the status of factive islands in Italian, based on original experimental data. Moreover, since island effects have been generally tested against wh-extractions, a secondary goal of the present study is to verify whether other types of A’-movement, such as Focus fronting, can be restricted by island effects as well. The results obtained from two experimental tests, carried out with the collaboration of a total of 310 informants, provide evidence that Italian factive islands seem to lie somewhere between strong and weak islands, qualifying as a sort of ‘intermediate’ island. Indeed, direct objects appear to be less constrained than subjects, indirect objects and adjuncts. Furthermore, the extraction of a Focus constituent from a factive complement lowers the acceptability of the sentence, thus indicating that (factive) islands are indeed sensible to Focus fronting.

In this paper we discuss a change in the auxiliary selectional pattern of Brazilian Venetan, a heritage Italo-Romance variety spoken in southern Brazil. Venetan varieties display a default form of the past participle in constructions with postverbal subjects and a fully agreeing form in constructions with preverbal subjects: this is true both for the homeland varieties of the language, spoken in northern Italy, as well as for the heritage variety under analysis in this paper, spoken in southern Brazil. A crucial difference emerges in unaccusative constructions: while Italian Venetan uses the same form of the auxiliary BE in presence of preverbal and postverbal subjects, Brazilian Venetan uses a specialized form of the auxiliary in the constructions with default agreement on the past participle, when postverbal subjects are present. We argue that the specialized auxiliary form emerges as a necessary resumption in the case of lack of agreement. The heritage variety becomes, therefore, morphosyntactically more complex than the non-heritage counterpart.

In this paper we propose that verbal clusters with restructuring verbs contain a defective embedded clausal boundary. Additionally, this paper constitutes a methodological and empirical contribution to syntactic microvariation in Occitano-Romance varieties: we describe the synchronic variation in the expression of presuppositional postverbal negation and the position of clitics in infinitival complex structures. This empirical study contributes to the understanding of the micro-syntax of negation and clitics as independent phenomena. However, the interaction of both phenomena brings to light the structure of three types of verbal clusters: restructuring verbs, non-restructuring verbs and control predicates. Crucially, the data collected highlight that, although clitic climbing and embedded presuppositional negation are independent phenomena, they can both be explained by the transparency of the embedded clausal boundary. We claim that restructuring verbs select for a defective embedded C/T. Ultimately, the article shows how the comprehension of phenomena in closely related varieties allows us to better understand the architecture of grammar.

The central issue addressed in this paper is the formal linguistic notion of parameter as a predictor for the (non-)occurrence of multiple linguistic phenomena. We start from a parametric analysis of a microvariational data set and we introduce a way of gauging the success of that analysis. At the same time, we explore to what extent language contact—operationalized here as geographical proximity—can be seen as an explanatory factor that is complementary to the parametric analysis. Methodologically, the paper advocates for the use of k-nearest neighbors classification as an interesting new technique in the linguist’s toolkit.

This paper is about a clitic-like form lo that appears in two under-studied dialects of Mexico in the context of transitive clauses. The distribution of this clitic-like form in these dialects is at odds with Standard Mexican Spanish which does not allow it in the same context. This clitic-like form resembles the singular, masculine, accusative object clitic of Standard Spanish, but it differs in that it does not show the agreement pattern expected for object clitics. In this paper we argue that this clitic-like form is better understood as an object marker that is triggered by the lack of a positive [Participant] feature in the direct object as part of the extended projection of the Object-DP. We also propose that this marking strategy is not the result of linguistic transfer or interaction with a different language, but rather a possible development within the grammar of Spanish. This marking strategy is, in fact, an inherent strategy of Spanish, but it gets blocked by normative pressure. The fact that this strategy flourishes in dialects apart from normative/academic contexts could be an indicator that the explanation we offer is on the right track.

It is commonly assumed that the Catalan prefixes des- and es- are synonymous, since they can be found attached to the same base in change of state verbs with an apparently identical meaning; cf. desgranar and esgranar ‘to extract the grain’. The aim of this paper is to show that these two prefixes are non-trivially different, as suggested by the fact that only the latter, but not the former, is attested with an ingressive (or Goal-oriented) meaning; cf. estovar ‘to soften’. Following a nanosyntactic approach to grammar, I claim that the trees lexicalized by des- and es- are not composed of the same morphosyntactic features, which accounts for their different morphosyntactic behaviour as well as for their semantic contrasts. It is proposed that des- lexicalizes a Source Path, whereas es- lexicalizes a Goal Path in addition to a root node. The paper is also a contribution to the long-standing debate of how morphemes with similar meanings compete for insertion.

The Parallel Architecture (PA) framework (Jackendoff 2002, 2007, Culicover & Jackendoff 2005) is one of the most complete constraint-based linguistic theories that encompasses phonology, syntax and semantics. However, it lacks a fully developed model of word formation. More recently, a theory called Relational Morphology (RM) (Jackendoff & Audring 2020) has been developed, that integrates into the PA. The current study shows how the Slot Structure model (Benavides 2003, 2009, 2010), which is compatible with the PA and is based on the dual-route model and percolation of features (Pinker 1999, 2006; Huang & Pinker 2010), can provide a better account of morphology than RM, and can also be incorporated into the PA, thus contributing to make this a more explanatory framework. Spanish data are used as the basis to demonstrate the implementation of the SSM. The current paper demonstrates two key problems for RM: inconsistent and confusing coindexation, and a proliferation of schemas, and shows that these issues do not arise in the Slot Structure model. Overall, the paper points out significant drawbacks in the RM framework, while at the same time showing how the PA’s morphological component can be enriched with the Slot Structure model.

English-medium instruction (EMI) affects lecturers’ classroom practices as they face a new teaching scenario where language(s) can be used to construct lecturers’ identities. Therefore, lecturers can take up different identities because their language-choice acts depend on their communicative and identification purposes. Employing Membership-Categorisation Analysis (MCA) to examine classroom interaction, this paper examines the classroom practices of one EMI lecturer to explore how the orientation towards one language over the others implies a specific function associated with a particular identity. The alternation between languages reveals to what extent EMI lecturers accept or challenge the English-only policy and how lecturers position themselves as English-only or as translanguaging lecturers. This study documents lecturer’s teaching behaviour, particularly how L1-choice acts can be more effective for certain purposes. Studying how lecturers draw from both their EMI-lecturer identity and their L1-lecturer identity, this paper shows how multilingual practices unfold in EMI and highlights the pedagogical value of the L1, hence advocating that the use of languages other than English has, after all, a particular purpose.

This paper presents VeLeCa, a new resource on Catalan verbal inflection containing the phonological form of 174,200 word forms from 3,484 lexemes and their respective lexical and morphosyntactic values and frequencies. We describe the challenges and procedure we followed in the compilation and phonemization of this resource, and conduct a computational analysis of the Paradigm Cell Filling Problem (i.e. morphological predictive complexity) in the system to contrast it with those from related Romance languages.

This paper investigates the relationship between infinitival nominalizations and analytical passives in Brazilian Portuguese. Taking Distributed Morphology as a theoretical framework (Halle & Marantz 1993; Marantz 1997), we propose a parallel structure for these formations based on the important similarities they present. Specifically, we propose that the derivation of both these structures includes a projection of a mixed nature, a Switch head (Panagiotidis 2015), which carries features of two different categories and thus can interrupt an extended projection and begin a new one. In this case, the Switch makes the originally verbal structure to become nominal. We argue that the distinction between nominal infinitives and passives is that in the former, the Switch carries an [N] feature, while in the latter, it brings an [A] feature to the structure. Infinitival nominalizations, thus, are formally nouns, as evidenced by their compatibility with determiners and different argument positions prototypically associated with nouns. In passives, this operation results in participles, which behave like adjectives: they do not admit a determiner, they have unvalued φ-features and they need a copular verb to be projected into a verbal structure.

This paper is about the semantics of quality modal superlatives in predicative function, e.g. Italian È stata il più calma possibile (‘She was the calmest possible’), with a focus on what is the anchoring of the modal superlative, and how what is being compared is restricted. The limitations of single-anchoring in individuals when analysing modal superlatives, following the strategy generally adopted for ordinary absolute and relative superlatives, are highlighted. The alternative of using a comparison class made of world+individual pairs grouped into equivalence classes defined through the amount of the gradable property (Tovena & Fleury 2023), proves a better solution. New data on the stage-level vs. individual-level reading of the gradable adjective provide evidence supporting this form of anchoring. The stage-level reading, although predominant, is not the manifestation of the invariance of the individual being compared. Some stage-level interpretations favour selecting a plurality of individuals for comparison, with the difference that this plurality is determined across possible worlds, not in the actual world, in contrast to ordinary superlatives.

In this paper I argue that the position and realization of subjects in adverbial non-finite clauses in Italian, Spanish, Galician, and European Portuguese can provide valuable insight into the licensing conditions for pro and postverbal subjects in finite clauses in these pro-drop languages. I first provide evidence that Tense and Agreement constitute separate syntactic heads in these languages, as argued by Belletti (1990) among others, and also apply this analysis to non-finite clauses in Spanish, Galician, and Portuguese. I will also argue, after analyzing the syntactic variation exhibited by subjects in non-finite clauses (containing regular and inflected infinitives) in Spanish, Galician, and European Portuguese, that (i) postverbal overt subject DPs are licensed by a full set of φ-features in Agr (overt or covert), (ii) pro is licensed by an Agr head that is overtly realized by either overt φ-features or verb movement, and (iii) preverbal overt subject DPs are licensed by verb movement to Agr.

This paper introduces Exclamative Se Constructions (ESCs), analyzing their use in two Romance languages: Trevigiano, a northern Italian dialect, and Standard Italian. ESCs are used to express shock or surprise at someone’s statement and to challenge its accuracy. Although they resemble adverbial clauses, ESCs function autonomously as main clauses. I identify four defining properties of ESCs: adverbial clause form, form-function mismatch, main-clausality, and the anchoring of surprise to a preceding assertion rather than to the ESC’s own propositional content. I argue that ESCs function as counterarguments, specifically rebutting assumed premises using contextually relevant scales. These rebuttals consistently involve asserting a value that contrasts sharply with some initial statement, often reaching an extreme or unexpected point on the contextually relevant scale.

This paper investigates the acceptability of wh in-situ expressions in embedded contexts in French. We report on two experiments that tested the acceptability of wh in-situ and ex-situ in embedded clauses of biclausal direct wh-questions and in indirect questions, and how their acceptability is modulated by the presence of the negation in the embedded clause. In Experiment 1, the contexts favored a D-linked interpretation of the wh-element (in the sense of Pesetsky 1987), while the contexts in Experiment 2 disfavored it. Our results show that the in-situ strategy is generally more acceptable (or at least equally acceptable) than the ex-situ strategy in direct questions with long construal, regardless of D-linking and negation. By contrast, the in-situ strategy is significantly less acceptable than the ex-situ one in indirect questions, regardless of D-linking and negation. Our findings indicate that in long construal direct wh-questions, negation fails to selectively block wh in-situ. In this respect, French differs from many other languages (see e.g., Beck & Kim 1997, a.o.). We show that D-linking is not a necessary condition for in-situ in French. We discuss how our findings relate to the current debate between alternative formal analyses of wh in-situ in French.

This paper provides a fine-grained characterization of two confirmational tags in Catalan, oi? and eh?, starting with the novel claim that, despite appearances, they do not have the same distribution or meaning. While oi? is a rather unconstrained confirmational, eh? is proved to be sensitive to whether the main predicate of the sentence anchor p is objective or subjective. We ultimately propose that confirmational questions with oi? make a request for confirmation of the truth of the sentence anchor, whereas a confirmational with eh? is a request for the commitment of the addressee to the truth of the sentence anchor. Two consequences emerge from our analysis: oi? should be preferred when factual truths instead of opinions are conveyed, while eh? should be preferred in contexts where the speaker is not necessarily interested in finding out the truth of p. These predictions are tested in three experimental studies, which provide empirical support for the theoretical proposal regarding the different pragmatic contribution of both tags.

This paper examines external possession between a dative possessor and a relational noun in Spanish middle-passive configurations. The dative DP in these contexts must surface preverbally in out-of-the-blue contexts, either by itself –presumably in preverbal subject position–, or along with the theme containing the possessee. When the dative precedes the theme, it cannot be assumed that the former is left-dislocated while the latter sits in preverbal subject position, for this would imply a violation of locality under a low applicative analysis of dative possessors: Tº would skip the possessor argument in Spec,ApplP to probe the possessee in Applº’s complement position to its specifier. Rather, I provide data showing that preverbal DPs –dative or otherwise– in Spanish middle-passive contexts are clitic left-dislocations co-referring with empty pronouns in argument position. I offer a biclausal analysis of these configurations that avoids any potential intervention effects, while at the same time accounting for the extra-sentential properties of these constituents.

This paper identifies a micro-cue correlating to verb second word order (V2) in two closely related Medieval Romance languages. As V2 is asymmetrically distributed in main rather than subordinate clauses, an asymmetry would be expected in phenomena assumed to relate to V2, such as subject inversion, null subject and enclisis. The loss of that asymmetry should therefore indicate the loss of the V2 word order rule. These assumptions are tested here by a quantitative analysis of a treebank of calibrated data covering the crucial period of change (from the 14th to the 16th century) for Medieval French and Venetian. The hard quantitative evidence provided demonstrates that the main versus embedded asymmetry is indeed a micro-cue of V2 structure, and of its loss in one of the two investigated languages.

This paper shows that Spanish ‘más que’ (lit. more than) is much more than a comparative construction synchronically. Phonological, syntactic, and semantic evidence shows that various grammatically different entities hide under this single spelling. The most prominent of these is a (phonologically unstressed) negative polarity item with a meaning “only” or “just”. By means of robust synchronic and diachronic corpus evidence, this paper explores its morphosyntactic properties and geographic distribution in the modern language, as well as when and how a comparative expression with no polarity associations could come to grammaticalize into a negative polarity item.

This work is about verb nominalizations and the relation between these forms, the argument structure of their verbal bases, and verbs’ possible meanings. Since Chomsky’s seminal work (Chomsky, 1970), nominalizations have become a window to investigate fundamental aspects of argument structure and assess linguistic theory. Following such a tradition, adopting the framework of Distributed Morphology (Halle; Marantz, 1993; Marantz, 1997, 2013b) and using Brazilian Portuguese data, this paper explores typical nominal forms derived from various kinds of Brazilian Portuguese verbs to evaluate a specific proposal of constructionist argument structure theory found in Medeiros (2018). The BP data show that syntactic-semantic decompositions of verb phrases are at least partially preserved in their derived nominal forms in a very systematic manner, and that these forms maintain the complex structural meaning in all kinds of situations – even when these are interpreted as result nominals. This work also provides some innovative syntactic-semantic means to deal with result nominals interpreted as entities (as in the case of construção (construction)), showing how to obtain, compositionally, such entity readings, demonstrating that even in these cases there are clear reflexes of event meanings.

Distributed Morphology (DM) predicts that the units of syntactic derivation are smaller than words. This paper explores the implications of this prediction for language acquisition research and questions if DM is descriptively and theoretically sound when faced with acquisition phenomena. We first introduce independent evidence supporting sub-word units in acquisition: results from a computational model of lexical acquisition show that slightly more morphologically complex input data, such as Brazilian Portuguese when compared to English, cause a substantial decrease in the model’s performance; children’s early productions when acquiring polysynthetic languages reveal they are attempting to find these languages’ morphological units, instead of relying on chunks of non-analysed material; and words are shown to lack explanatory power in describing language acquisition in terms of storage, bootstrapping, or production. We then bridge the gap between DM and earlier proposals for the identification of words and formal features, briefly outlining a strategy for acquiring morphemes. Finally, we present accounts for two prevalent phenomena linked to language acquisition through the lens of DM: the overregularisation in acquiring irregular verbs, a step observed in children acquiring different languages; and the Brazilian Portuguese verbal paradigm shift, an example of morphological diachronic change. Our findings support the non-lexicalist derivation of words, highlighting that words are incompatible with acquisition from multiple perspectives. This suggests that understanding language acquisition benefits from considering the smaller, morpheme-based units predicted by Distributed Morphology.

In both the typological literature and literature on formal syntax and semantics, a division is drawn between nouns that are inalienably possessed such as body parts and kinship terms and nouns that are alienably possessed such as owned materials. In this paper I re-examine data from Spanish and Mayan languages and propose an analysis of it that emphasizes two important points regarding the roots and structures associated with inalienable and alienable possession. I first make the novel observation that various types of external possession in Spanish provide clear support for the idea that inalienable possession is structurally less complex than alienable possession: inalienable possessive relations are introduced within a complex n head that consists of a root and nominalizing head. I then explore attributive possession in Mayan languages and highlight data that leads to conflicting conclusions about where, precisely, inalienable relations such as part-of and kin-of are encoded: on n heads or on roots. I outline avenues for future research with the Mayan language family that may help elucidate which of these two analyses may ultimately be correct.

Over the last decades, the acknowledgment of the dynamic character of language has led to the focus on language evolution and change. Among the different streams of linguistics, Historical linguistics (morphosyntax), Construction Grammar and Phraseology have paid attention to these processes, studying grammaticalization, constructionalization and phraseologization, respectively. The three frameworks we are dealing with were quite different in origin, but through the adoption of cognitive, usage-based, pragmatic approaches they are tending to converge. In this paper we will outline the respective frameworks, their conception of language unit and the process of change, including its mechanisms and stages. In each section, we will include some applications in Catalan language. As we consider them to be complementary, since they account for different items of language, at the end we will suggest an integration of the three of them.

Determining the linguistic nature of light verbs – whether they are lexical or grammatical units – is still an open question. Light verbs are often characterised as delexicalized. Thus, grammaticalization would be the process by which a lexical verb (lexical unit) assumes the value of a light verb (grammatical unit). However, it has also been argued that light verbs are just regular verbs, albeit ones with little semantic specification. If they are, there can be no grammaticalization, at least in principle. Nevertheless, even if light verbs are considered lexical verbs, the way they evolve invite to relate them to grammaticalization processes. In this paper I contribute to the debate by providing evidence on the historical evolution of the Catalan verb agafar ‘to take, to catch, to pick, to get’. The conclusion is not firm, but suggests that the description of light verbs can benefit from the framework of grammaticalization.

Our aim in this paper is to explain the progressive semantic change of the Spanish verb volver from the 13th century until the 21st. We will show that at the beginning, its unique use is a transitive one, from Latin until the 14th. Then it acquired an intransitive use in the 15th. We will claim that the semantic change meaning of this verb from the transitive use to the intransitive, ending by the grammaticalization of the verb with the creation of a periphrasis (volver a + infinitive), is part of a ‘reanalysis’ process of the semantic relation between both cases the agent and the patient in the eventive structure and, consequently, of the subject and the object complement in the syntactic distribution. The semantic core which conveys a metaphorical or a spatial path toward a goal (“to make a rotary motion, to turn”) is unchanged. We claim that the telic character of the verb allowed it to impose itself and to replace tornar.

This paper analyzes the construction bien mirado as a discursive marker. In the first part, based on the data provided by the CDH and CORPES corpora, we will examine the discursive functioning of bien mirado and the delimitation problems associated with the constructions of absolute clause or free adjunct, taking especially into account the notions of subject of predication, coreferentiality, the semantic relationship between the main clause introduced by bien mirado and the previous discursive element, the position (initial, medial, or final), the parenthetical character, as well as the argumentative and informative values, along with the reformulating one. In the second part, we will instead focus on the origins of the construction, on the contrastive historical delimitation with absolute clauses and on the frequency distribution, additionally developing a hypothesis about the subjectification process involved in discourse functioning – linked to the discursive character of absolute constructions, the parenthetical nature, and the meaning of mirar as a verb of thought or judgment.

This paper investigates the structural and sociolectal limits of the use of non-standard “plain” que instead of the standard relativizer DET+que in Mexican Spanish noun clefts. Although plain que in Spanish noun clefts is considered ungrammatical by reference grammars, we find it in data from Felipe Carrillo Puerto (Quintana Roo, Yucatan) and Mexico City (CDMX). Among the speakers from CDMX, only younger speakers use plain que in our data, suggesting that it might be a more recent innovation in this variety. Among Yucatecan speakers, we find that bilingual speakers (Spanish-Maya) tend to use more plain que than monolingual Spanish speakers, and only bilingual speakers show similar age effects to those found in CDMX, whereas older monolingual Yucatecan speakers use more plain que than younger YS monolinguals. Structurally, plain que is limited to canonical clefts, which we interpret as an indication that clefts with plain que constitute functionally reanalyzed headed relatives. Moreover, we find that those speakers that use plain que do so mostly when the antecedent is specific and carries a determiner. If the antecedent is either non-specific or does not carry a determiner, DET+que is generally preferred.

In Gradient Harmonic Grammar (Smolensky & Goldrick 2016), syntactic strength is considered an inherent property of linguistic items. In this paper, I propose that syntactic strength plays a crucial role in pro-drop and in other phenomena that have often been related to a vague concept of “strength”. I suggest that pro-drop results from failed lexicalization of weak pronouns in the presence of a strong T. Subject clitics of non-pro-drop languages instead stem from the strengthening of weak pronouns in the context of a weak T. This strength-based approach to pronouns can be also extended to the independent phenomenon of mismatches between strong forms and weak semantics of pronouns, which is problematic for standard accounts à la Cardinaletti & Starke 1999 and it is not accounted for by any other theory of pro-drop. In general, such an approach is a mere extension of what has been tacitly presupposed while talking about strength for pro-drop (Rizzi 1986a; Alexiadou & Anagnostopoulou 1998), and for other phenomena (see Cardinaletti & Starke 1999 for pronouns; Chomsky 2013, 2015; Cecchetto & Donati 2022 for labeling).

This paper discusses a cluster of related constructions in Reunion Creole involving nana ‘have’. Focusing on a presentational construction that is functionally equivalent to the il y a-cleft of French, I argue that a once bi-clausal cleft has developed into a monoclausal broad focus construction in Reunion Creole. I present a Role and Reference Grammar analysis of both the bi-clausal cleft and the monoclausal construction, and in the former, I explain how the cleft relative clause differs from restrictive relative clauses.

This paper deals with a hitherto little studied type of superlative, so-called ‘superlative minimizers’, which in Spanish take the form el más mínimo N ‘the slightest N’. Several tests show that these superlatives are non-specific or non-referential, despite their necessarily definite shape. The semantic properties of these expressions cannot be accounted for using the usual tools in theories on definiteness or quantity superlatives. The aim of this work is to provide a compositional analysis that explains several properties of these superlatives, particularly their non-specific nature and restrictions on the type of noun they admit. Assuming a kind-based semantics for nouns and a state-based theory on gradability, I propose that superlative minimizers express uniqueness over state-kinds, thus giving rise to a non-specific meaning of the DP and a quantity meaning, since state-kinds are scalar-ordered in the extensional domain of nouns. As a consequence, this work straddles two lines of research: on the one hand, it broadens our knowledge on quantity superlatives (cf. Hackl 2009, Solt 2012, a.o.) with a new type subject to previously unstudied characteristics; on the other, SMs are a type of the much-studied weak definites (cf. Aguilar-Guevara & Zwarts 2010, Abbott 2014, a.o.).

The paper studies the conditions that determine the Discourse-Linked or non-Discourse-Linked status of noun-less Determiner Phrases introduced by different determiners, in Italian and in English. For instance, given the sentence Ten bombs exploded yesterday, the continuation [Three] were cluster bombs tends to have a meaning equivalent to 'three of the bombs that exploded’ (D-Linked), while [Three] will explode today is understood as 'three (different) bombs’ (non D-Linked). Beside world-knowledge, the syntax of the determiners and their position with respect to the verb affect the availability of DL/non-DL readings. This and other facts undermine an analysis cast purely in terms of semantic domain restrictions, and suggest that DL readings are due to the presence of a covert partitive structure. While perhaps intuitive, this idea faces various issues in Italian, due to its interactions with the syntax of the pro-form ne. We show that an NP-based structure for numeral and proportion-based partitives (three/half of the bombs) is actually compatible with the facts, and offers a cue on the nature of sub-DP pro-forms and their uses.

Spanish, like many other Romance and non-Romance languages, shows Differential Object Marking (DOM), i.e., some direct objects are morphologically marked by the prepositional marker a ‘to’, while others remain unmarked. The literature has proposed different sentential parameters in order to capture this variation (Fábregas 2013, among others), including topicality (see Leonetti 2004, Iemmolo 2010, among others). In addition, Laca (1995: 82f.) has argued that DOM also depends on discourse properties. She assumes that in Spanish the use of DOM with an indefinite direct object signals that more information about this object referent is to be expected in the upcoming discourse (see also Comrie 1981/1989). First empirical evidence for this hypothesis comes from DOM in Romanian (Chiriacescu & von Heusinger 2010). In this paper we explore the hypothesis that, in Spanish, human indefinite direct objects with DOM show more forward-looking potential than those without DOM. We present original results from two corpus studies and two paragraph continuation tasks. The corpus studies provide support for the discourse effect of DOM, while the paragraph continuation tasks do not, which might be due to the particular design of our experimental items. We evaluate the different parameters that contribute to the discourse prominence of a direct object with DOM and those that might mask such effects. We conclude that there is evidence that DOM contributes to discourse prominence, but that further studies are necessary.

In this paper we study a case of diachronic and diaphasic variation consisting in a subordination strategy for expressing immediate succession present in the cultivated narrative language of Classical Spanish. This strategy is based on verbal doubling (V-doubling), a pattern also found in some contemporary Atlantic creole languages. We analyze the elements that constitute this type of sequences and examine the similarities and differences between the constructions of Classical Spanish and those of the Creoles. We also compare the constructions of Classical Spanish with similar structures without V-doubling, present in all periods of Spanish, in order to detail the link of V-doubling with focalization of the end of the first event. The strategy analyzed was a transitory way to cover the absence of subordinators expressing immediate succession in an unambiguous way during the chronological period in which it existed.

The aim of this paper is to analyze a type of Spanish (and Romance) complex locative non-directional prepositional expressions taking a genitive complement, like encima/cerca de (“on top/near of”). We will center our attention on an alternation that has gone unnoticed by most theoretical and descriptive grammars: this genitive complement may appear as an argument of the main predicate and show dative case (se sentó encima de Juan, “(S)he sat on top of Juan” > se le sentó encima, “Lit. (S)he DAT sat on top”). Most of these complex PPs may correspond to what has been called ‘Axial Parts’ (Svenonius 2006, 2008), but the paradigm extends to other cases. Based on work by Larson and Samiian (2021) on the typology of nominalization in prepositions in Iranian Persian, we show that the common property of these prepositional expressions is that they contain a nominal element: a relational noun or a nominalized preposition. Apart from the dative alternation, one of the properties shown by complex locative prepositions in Spanish is that a stressed postnominal possessive is allowed, which can be masculine or feminine, giving rise to dialectal variation: detrás mío/mía, Lit. “behind mine”, cerca mío/mía, Lit. “near mine”. We will show that what determines the gender of the possessive is the nominalizer and the level of grammaticalization of the nominal head.

This paper focuses on the study of a particular phenomenon of variation related to the distribution of Spanish copulas: the (in)compatibility of event nouns with predication structures where the copula estar and an evaluative adjective are combined. Some structures, such as La boda estuvo preciosa (‘The wedding wasESTAR beautiful’) are present across all Spanish varieties, contrasting with examples such as La tormenta estuvo horrible (‘The storm wasESTAR horrible’), that are attested only in certain specific varieties, and that could be considered an instance of so-called innovative uses of estar. We give here a first picture of the extension of this phenomenon. Secondly, we develop the relevant distinction within the domain of event nouns that is needed to give an accurate explanation of the facts under examination. Finally, we show how these facts fit within the overall picture of copulative variation in Spanish, thus making a significant contribution to our broad understanding of Spanish copulas.

This paper discusses morphological alternations in the so-called strong preterits in Spanish. The starting point is the variation phenomenon known as Analogical Strong Preterits, which is characterized by the formation of 3PL on the basis of 3SG, adding the exponent /n/: dij-o-n, instead of general dije-ro-n ‘they said’. Although this is the way in which the 3PL is obtained in the rest of the paradigm, Spanish preterits regularly present the segment -ro- at the left of -n. The varieties with Analogical Strong Preterits preserve the segment -ro- in all the other verbs (canta-ro-n ‘they sang’), but shows this particular form in the case of strong preterits. Within the framework of Distributed Morphology, we explore the morphological alternations found in relation to verbal stems (dec-/dij- for √SAY) and functional morphology (-ro/-o for T/Pers). Our approach focuses on the properties of Vocabulary Items and the locality of terminal nodes for vocabulary insertion, avoiding thus any kind of post-syntactic operation.

This paper explores a newly identified contrastive topic configuration in Spanish. Coined by de Andrade (2018) for Galician and European Portuguese, the D(emonstrative)-construction features a left-dislocated topic and d(emonstrative)-pronoun resumptive. This study investigates whether the D-construction exists in Spanish, and if so, with which syntactic properties. We administered an acceptability judgment task on the D-construction, Clitic Left Dislocation (CLLD), Hanging Topic Left Dislocation (HTLD) and Focus Fronting (FF) to Spanish speakers. The task tested the role of the left-dislocate, case connectivity, subject-verb inversion, embedding, recursivity, and sensitivity to island constraints. Simple instances of the D-construction received consistently high ratings, demonstrating that it exists in Spanish. There was individual variation regarding the role of the left-dislocate and case connectivity. The D-construction did not require subject-verb inversion, was non-recursive and demonstrated selective island sensitivity. Findings for CLLD, HTLD and FF were mostly in line with previous literature. The D-construction did not exactly pattern with CLLD, HTLD, nor FF; it is characterized by a unique set of syntactic properties. We propose that both left-dislocated elements are base-generated at Spec, TopP: the fronted DP is a hanging topic, and its resumptive d-pronoun is linked to a clitic within the main clause via an A'-chain.

This paper investigates the properties of GO and TAKE used as functional verbs to express the beginning of an action or a state (inchoativity) or a sense of surprise, wonder, astonishment, or regret (mirativity) as found in a group of multiple verb constructions, in a macro-comparative perspective. Multiple Agreement Constructions (V1+mi/mu/da/če+V2[+finite]) and Pseudo-Coordination (V1+‘and’+V2[+finite]) are considered in Italo-Romance (Italian and Sicilian) and Balkan Slavic (Bulgarian), together with the Italo-Romance Infinitival Construction (V1+a+V2[-finite]), drawing from novel fieldwork data and online corpus data. GO and TAKE generally proved to be highly productive as both inchoative and mirative markers in all the languages considered. In particular, an analysis of cases in which these two functional verbs convey (to different degrees) both functions at once is provided, relying on the cognitive basis underlying the process of their grammaticalisation.

Previous literature on the typology of gerund clauses in Portuguese has overlooked a peculiar type of clauses which are always introduced by como (‘as’) and display an array of characteristics that set them apart from all other gerund clauses (and from other, somehow similar, constructions in different languages). In this paper, we provide an in-depth syntactic and semantic characterisation of these como-gerund clauses and the contexts in which they arise, highlighting their similarities and differences regarding other constructions, namely resultative and depictive secondary predicates. We put forward a proposal to deal with their syntactic configurations and the restrictions they exhibit. We also propose that como is obligatory in these clauses because it marks a type-shift operation, which gives como gerund clauses a predicative interpretation, usually found in the nominal domain.

This paper discusses the preliminary results of an exploratory study based on acceptability judgments about Pseudo-Coordination (PseCo) as found in the province of Catania (Sicily). PseCo is a monoclausal verbal periphrasis where an inflected verb (V1), usually of motion, is followed by another inflected verb (V2) with an optional connecting element between them (cf. Giusti, Di Caro and Ross 2022). In western Sicilian varieties, the V1 GO can occur as an invariable form (i.e., va-, vo-, uo-, o- as in Oppigghju u pani ‘I go and fetch the bread’) prefixed to the V2. An online anonymous questionnaire was administered to 295 participants (180 female, 115 male; Age M: 20,95 years, Age SD: 2,73 years) to assess whether (i) PseCo with V1 GO in the present indicative is productive among younger speakers; (ii) different forms of invariable V1 GO can cooccur in one and the same variety. The results show that PseCo in the province of Catania is still vital among younger generation of speakers and that invariable V1 GO generally occurs in at least two different forms but with no particular semantic specialization.

This paper proposes that Old French coda /s/ deletion (11th-13th centuries) forms part of a broader diachronic progression of ever-stricter requirements on the sonority contour of /sC/ clusters in syllable contact, reframing part of a well-known moraic analysis (Gess 1998a, 1999, and later work) of Old French coda loss phenomena. Given the multistage rollout of coda /s/ deletion as a function of the sonority of the following onset, an approach hinging on syllable contact constraints not only offers a more detailed and precise formalization of the diachrony of word-medial /sC/ in Old French, but also draws systemic connections with cognate processes affecting /sC/ clusters in early French such as prothesis and earlier Proto-French stop epenthesis. The Optimality-Theoretic analysis presented here formalizes these phenomenological links and the constraints on syllable-contact sonority using the Split Margin Approach to the Syllable (Baertsch 2002, Baertsch & Davis 2003). Rather than sonority-graded mora-licensing constraints causing Old French coda /s/ deletion, the present account argues that their ranking above Faith is instead the acquisitional outcome of the near-total absence of coda /s/ across the lexicon, as a culminative result of the progressive tightening of syllable contact requirements.

In this paper, we account for different patterns found in complement clauses of se constructions in Romanian and Spanish. In Romanian, a se construction cannot host an infinitival complement, an apparently controlled clause, whereas in Spanish a se construction can. However, when an additional se is added to the complement clause (a “double se construction”), the Romanian structure becomes grammatical, while the Spanish equivalent becomes ungrammatical. The Romanian patterns have been previously argued in Giurgea & Cotfas (2021) to be cases of control, with a failed agreement relation forcing the obligatory presence of se in the complement. We propose an alternative based on two major differences found in Romanian and Spanish. First, in se constructions, Spec, Voice is saturated by the external argument in Spanish, but it is unsaturated in Romanian. We argue that this prevents the external argument in Romanian from acting as a controller. Second, Romanian infinitival clauses appear to share properties with finite clauses, in contrast to Spanish. We argue that the grammatical Romanian double se construction is not an instance of control and suggest that it is the finite nature of the infinitival complement that allows for a double se construction.

The occurrence of the Subjunctive mood in sentences describing facts is commonly seen as problematic, given the relation between Subjunctive and non-veridicality. One line that is explored in the literature to account for the Subjunctive in complement clauses of factive verbs is to link the occurrence of this mood in such contexts to gradability of the main clause’s predicate. However, such an account faces empirical problems, and is not extendable to other contexts where the Subjunctive occurs even if the sentence describes a fact of reality. This paper proposes an account for the occurrence of Subjunctive in different kinds of factive contexts, showing that in all such cases the reason for this mood to occur follows from the general condition that leads to the use of Subjunctive, though for different reasons. Gradability of the main predicate is, in fact, one of the factors that leads to the consideration of non-p worlds, and the Subjunctive, but not the only one. For other predicates, other semantic features lead to counterfactual reasoning. Concessive clauses are another factive context where Subjunctive occurs and allow a better understanding of what triggers the Subjunctive mood and what this formal sign indicates.

Sardinian has several types of make-construction: the analytic causative constructions corresponding to French faire-inf (FI) and faire-par (FP); subject control constructions similar to the colloquial idiomatic use of the pronominal Italian verb farcela ‘be able to, to manage’ (a kind of dynamic ability); and particular impersonal constructions that express the deontic modality of possibility. Furthermore, there are several lexicalized expressions that involve make, similar to Italian far vedere ‘to show’ (lit. ‘to make see’). The impersonal construction, which I call the “impersonal causative”, is of particular interest in this paper: It always selects an infinitive (not inflected, which would be possible at least in some varieties of Sardinian), introduced by a; apparently, it shows transparency effects, but it must nevertheless be interpreted as a biclausal structure; it is, in principle, only possible with the third person of the verb make; with regard to its interpretation it encodes modal possibility and is thus more related to (less strong causative) LET than to strong causative MAKE; and it shows up in two different types, at least in some varieties, with or without number agreement (the latter also with an overt subject in a personal infinitive), which has interesting consequences for the interpretation of the argument structure of MAKE.

Campanian dialects such as Neapolitan feature a so-called ‘second form of the infinitive’ (SFI), a form consisting of the bare verbal stem, which can be used after functional verbs. This paper addresses the microvariation concerning the construction by analysing novel data from the Valle Caudina, located to the northeast of Naples. The SFI is frequently found specifically with the imperative va ‘go!’. In Neapolitan, the form has been reanalysed as an imperatival form in this context, yielding an asyndetic imperative. At a first glance, the use of the SFI in Valle Caudina looks very similar to its Neapolitan counterpart, but unlike Neapolitan, the SFI in these varieties has remained non-finite and has not been reanalysed as an imperative. These dialects can thus be considered a previous stage of the development described for Neapolitan by Ledgeway (1997, 2007, 2009). This claim finds support in the absence of metaphonetic forms - which have appeared in Neapolitan, as a consequence of the reanalysis - as well as the presence of clitic climbing. Finally, unlike Neapolitan, the SFI is becoming less productive in the varieties of the Valle Caudina.