Volume 36, Issue 1 - Special Issue: The interplay of theory and observation: studies in syntax and semantics; Guest Editors: Ana Maria Martins and Fernanda Pratas

Here it is proposed that Form Copy (Chomsky, Noam, T. Daniel Seely, Robert C. Berwick, Sandiway Fong, M. A. C. Huybregts, Hisatsugu Kitahara, Andrew McInnerney, Yushi Sugimoto. 2023. Merge and the strong minimalist thesis. In Cambridge elements . Cambridge University Press), applied to features, can derive a very simple form of Agree without the need for the distinction between interpretable and uninterpretable features or the Activity Condition. Furthermore, since Roberts (Roberts, Ian. 2010. Agreement and head movement: Clitics and defective goals . Cambridge, MA: MIT Press) derives head-movement from Agree, Form Copy can be similarly extended in such a way as to derive the effects of head-movement. This conclusion holds independently of whether head-movement is a purely morphophonological operation or whether there are narrow-syntactic cases. It is also shown how an account of cliticisation, both proclisis and enclisis, can be derived, updating the proposals in Roberts (Roberts, Ian. 2010. Agreement and head movement: Clitics and defective goals . Cambridge, MA: MIT Press) in terms of Form Copy.

Some grammatical phenomena are more resistant to diachronic change than others. The syntactic core is particularly resilient, raising the question why this is the case and what causes the least vulnerable properties to change. Since fundamental alterations of grammars do not occur across the lifespan of adults, first language acquisition is commonly considered to be the main locus of syntactic change. Under the assumption that language contact leads to cross-linguistic interaction, early bilinguals have been claimed to be the main agents of change. I revisit this debate, focusing on head directionality and V2. Summaries of studies of various acquisition types lead to the conclusion that reanalysis in core syntax does not happen in the course of neither monolingual nor bilingual L1 acquisition. Contrary to hypotheses entertained in diachronic linguistics, neither language contact nor structural ambiguity/complexity has this effect. For core properties to change in L1, the triggering information must be contained in the input. Insufficient exposure, as in heritage language acquisition, can cause morphosyntactic change, though not in the syntactic core. Only second language acquisition exhibits such effects. L2 learners are thus the most likely agents of fundamental syntactic change. I conclude that explanations of the resilience of syntactic phenomena cannot rely exclusively on structural aspects. It results from an interaction of syntactic and developmental factors, defined by grammatical constraint, acquisition principles, and processing demands.

Several languages allow for hyper-raising, a structure in which a DP that is interpreted as the subject of a finite complement clause is spelled-out as the subject of the matrix clause. Hyper-raising challenges certain core concepts of syntactic theory related to movement and locality. Various proposals have been made for analysing these structures, the main difference being whether the final position of the DP results from movement or not. It has often been assumed that European Portuguese, in contrast to Brazilian Portuguese, does not allow hyper-raising, or only allows it in very limited contexts. In this paper, we present empirical data extracted from written corpora and experimental results attesting to the production and acceptance of hyper-raising structures by a number of native speakers of European Portuguese. We contribute to identifying the contexts that favour hyper-raising in this variety and outline a preliminary analysis to explain what leads these speakers to produce and accept hyper-raising, while many others systematically reject it. In our proposal, this difference results from the properties of the embedded functional heads T and C, and the way in which their formal features are checked. Specifically, we propose that those speakers dissociate Case features from person-number features.

This paper presents an experimental approach to subject inversion in Brazilian Portuguese (BP). We investigated the acceptability of SV and VS sentences by two groups of speakers: monolingually-raised and bilingual heritage speakers of BP, using acceptability judgment tasks to test the effect of verb type, definiteness and pragmatic context. Results confirm that BP lost VS order with the exception of unaccusative constructions. Both speaker groups accept SV orders in all contexts, rejecting VS in sentences with transitive and unergative verbs. With respect to unaccusative verbs, pragmatic context and definiteness play a role in the acceptance of VS structures: with narrow focus on the subject, monolingual speakers accept VS order with definite and indefinite postverbal subjects. However, in all-new contexts, they tend to reject definite postverbal subjects. Given this differential behavior in the two contexts, we assume that BP exhibits two different syntactic positions for postverbal subjects in unaccusative constructions. Heritage speakers of BP are generally stricter in rejecting VS order. They do not allow for postverbal definite subjects in VS clauses independent of pragmatic context, indicating that they are progressively eliminating a residual postverbal focus position in unaccusative constructions. We take this as another indication that heritage speakers may promote and accelerate ongoing diachronic change.

In some finite clauses only one of the subjunctive and indicative moods might occur, while in others there is option between these two moods. This is the basic distinction between intensional and polarity subjunctive. This paper questions the sustainability of such division in Portuguese, providing a critical analysis of the major arguments sustaining that those two types of subjunctive need to be considered. Instead, it is shown that a semantic analysis of the indicative and subjunctive moods accounts for the distribution of these moods in European Portuguese (EP), no different explanations being needed for the cases of lexical selection and those of mood choice. In both cases Indicative signals that only p- worlds are taken in account, while Subjunctive signals the consideration of non- p worlds. This allows a comprehensive account of the EP data: Subjunctive or Indicative occurs depending on whether the meaning of the sentence leads to the consideration of non- p worlds or not. However, in other varieties of Portuguese, Indicative occurs in some sentences whose meaning involves the consideration of non -p worlds. Such is the case of Brazilian Portuguese (BP), where, in colloquial speech, the use of Indicative is common in complement clauses of querer (‘to want’), a paradigmatic domain of Subjunctive. The conjecture is made that in this variety resorting to one or another mood is used as a pragmatic, discursive, strategy, even if the meaning of the sentence leads to the other mood.

This paper discusses the contemporary use of the Portuguese simple and two compound pluperfect forms (with the auxiliary verbs ter , and haver ) of the indicative mood in written texts. Four topics are addressed. First, the claim that the simple pluperfect is currently a defective paradigm lacking the third person plural; this claim is confirmed and substantiated by looking at specific distinguishing contexts in corpora. Secondly, the competition between the three pluperfect forms, in written contemporary European and Brazilian Portuguese; it is shown that – contrary to some frequent comments in the literature – all three forms still have widespread usage, and occur in a relatively balanced way, in the type of registers considered (newspaper writing, modern literary texts, modern literary translations), in both varieties of the language. Thirdly, as regards semantic polyvalence, the possible existence of a deictic rather than anaphoric (temporal) pluperfect; it is argued that this type of pluperfect does indeed exist, and deserves separate grammatical recognition. Fourthly, the well-known cases of ambiguity in structures with pluperfect and temporal locating adjuncts, where the latter can act either as markers of the past perspective point required by the pluperfect, or as markers of the location time of the described eventualities; facts in English and Portuguese are compared, with some differences highlighted.