John Benjamins Publishing Company
This is a contribution from Romance Linguistics 2010. Selected papers from the 40th Linguistic Symposium on Romance Languages (LSRL), Seattle, Washington, March 2010. Edited by Julia Herschensohn. © 2011. John Benjamins Publishing Company This electronic file may not be altered in any way. The author(s) of this article is/are permitted to use this PDF file to generate printed copies to be used by way of offprints, for their personal use only. Permission is granted by the publishers to post this file on a closed server which is accessible to members (students and staff) only of the author’s/s’ institute, it is not permitted to post this PDF on the open internet. For any other use of this material prior written permission should be obtained from the publishers or through the Copyright Clearance Center (for USA: www.copyright.com). Please contact
[email protected] or consult our website: www.benjamins.com Tables of Contents, abstracts and guidelines are available at www.benjamins.com
Negative imperatives in Portuguese and other Romance languages* Rerisson Cavalcante
Universidade de São Paulo
1. Introduction This paper discusses the interaction between the expression of imperative mood and sentential negation in some Romance languages: on the one hand, European Portuguese (EP), Spanish, and Italian, in which true imperative forms are incompatible with (preverbal) negation; and, on the other hand, Brazilian Portuguese (BP), in which both true and suppletive imperative forms co-occur with negation. The main goal is to present data that can help evaluate previous analyses on the lack of negated imperative verbs in some languages and explain the unexpected behavior of BP. The paper is organized in the following way: in Section 2, I present an overview of the types of imperative forms and their behavior in negative commands across languages; in Sections 3 and 4, I discuss three previous analyses of the phenomena; in Section 5, I introduce Bošković’s (2004) PF merger analysis of the ban of negative imperatives in Greek and I show how it can account for the problematic data presented here and the differences between BP and the other languages under discussion. In Section 6, I summarize the basic points of the paper. 2. Imperative forms and negation 2.1 Types of imperative forms and their distribution in negative commands To express imperative mood, languages use verbal forms that either are specific of imperative contexts or are borrowed from other moods (indicative and
* The present paper is part of my Ph.D. research project about sentential negation and constituent negation in Brazilian Portuguese, supported by FAPESP (2006/00965–2 and 2008/00073–0) and CAPES (0306/10–0). Special thanks to Jairo Nunes, Željko Bošković, Marcelo Ferreira, Esmeralda Negrão, Marcus Lunguinho and Bradley Larson for comments and suggestions; to Isabella Fortunato, Leonor Simioni and Roberta D’Alessandro for helping with Italian examples.
© 2011. John Benjamins Publishing Company All rights reserved
Rerisson Cavalcante
subjunctive, for example). The specific forms have been referred to as true imperatives and the borrowed ones, suppletive imperatives (cf. Rivero 1994; Han 1999 and others). Spanish, for example, has three distinct verbal forms to express second person singular in imperative, indicative and subjunctive. English, on the other hand, makes use of the same verbal form in imperative, indicative and (semantic) subjunctive, as shown in Table 1. Table 1. True and suppletive imperatives in Spanish and English Imperative
Indicative
Subjunctive
Spanish
da
das
des
English
give
give
give
Studies on the imperative mood have shown that these forms are incompatible with sentential negation, specifically with preverbal negative markers (cf. Rivero 1994; Han 1999, 2001; Zeijlstra 2006; and others). In other words, true imperatives occur exclusively in affirmative commands. In negative commands, on the other hand, the imperative form is replaced with suppletive forms. Examples (1), (2) and (3) show that, in the presence of the negative marker, the true imperative verb is replaced with subjunctive forms in both Spanish (cf. (1)) and EP (cf. (2)) and with infinitive forms in Italian (cf. (3)).1 (1) Spanish a. ¡(*No) lee lo! neg read-imp it-acc b. ¡No lo leas! neg it-acc read-subj “Don’t read it!” (2) EP a. (*Não) conta (tu) a verdade! neg tell-imp thou the truth
b. Não contes (tu) a verdade! neg tell-subj thou the truth “Don’t tell the truth!”
. Zanuttini (1997) and Kayne (1992) propose that Italian suppletive imperatives make use of a null modal verb, which takes the infinitive as complement.
© 2011. John Benjamins Publishing Company All rights reserved
Negative imperatives in Portuguese and other Romance languages
(3) Italian a. (*Non) telefona le! neg call-imp her b. Non telefonar le! neg call-inf her “Don’t call her!”
However, BP behaves differently from these languages. Three facts are of interest for this work. First of all, in BP the imperative verb does not present any incompatibility with negation. Example (4) shows that the (so-called) true imperative verb can appear with either a preverbal or a post-sentential negative marker.
(4) Negative and “true” imperative (BP)
Não conta (tu/você) a verdade pra ele (não)! neg tell-imp thou you the truth to him neg “Don’t tell him the truth!”
Second, despite the compatibility between true forms and negation, the suppletive/subjunctive forms are still available in negative imperatives, as shown in (5). This is unexpected since the use of suppletive forms is analyzed as a consequence of the impossibility of the true forms in this context.
(5) Negation and suppletive imperative (BP)
Não conte (tu/você) a verdade pra ele (não)! neg tell-subj thou you the truth to him neg “Don’t tell him the truth!”
Third, more surprisingly, suppletive forms alternate with true forms even in affirmative imperatives, as shown in (6), which is impossible in EP, Spanish and Italian. As shown in (7), the suppletive forms cannot appear in affirmative command when a true form is available in these languages.
(6) Suppletive forms in affirmative imperatives (BP)
Conta / conte a verdade pra ele! tell-imp tell-subj the truth to him “Tell him the truth!”
(7) Lack of suppletive forms in affirmative imperatives.
a. *¡(lo) leas (lo)! (Spanish) it-acc read-subj it-acc “Read it!”
© 2011. John Benjamins Publishing Company All rights reserved
Rerisson Cavalcante
b. *Contes (tu) a verdade! (EP) tell-subj thou the truth
“Tell the truth!”
c. *Telefonar le! (Italian) call-inf her “Call her!”
What could explain the optionality between suppletive and true imperatives in both negative and affirmative commands in BP? 2.2 The nature of imperative forms in BP A possible clue to explain these differences between EP (and the other languages) and BP is the reduction of verbal morphology that has occurred in the latter. The loss of the second person morphology in BP rendered the indicative form identical to imperative form, as shown in Table 2. Table 2. Verbal forms for second person in EP and BP Indicative
Subjunctive
Imperative
EP
(tu) cantas
(tu) cantes
canta (tu)
BP
(tu/você) canta
(tu/você) cante
canta [=IND]/cante [=SUBJ] (tu/você)
Table 2 shows that there is currently no specific verbal form for the imperative mood in BP (cf. Cavalcante 2008a, 2010; see also Postma & Wurff 2007: 242). Thus, children acquiring BP arguably identify both forms used in imperative contexts as suppletive forms, borrowed either from the indicative or from the subjunctive. Therefore, we can say that BP imperatives do not show incompatibility with negation because there are no true imperative forms at all in this language, despite the fact that there are forms that phonologically correspond to EP true imperative forms.2
. There are BP dialects (specially in the state of São Paulo) in which the subjunctive form can be replaced with the indicative form in (non-imperative) subjunctive contexts. (i) Você quer que eu faça? you want that I do.subj (ii) % Você quer que eu faço? you want that I do.ind “Do you want me to do it?”
© 2011. John Benjamins Publishing Company All rights reserved
Negative imperatives in Portuguese and other Romance languages
However this is just a descriptive observation. It is still necessary to explain what the formal property is that differentiates BP from languages like EP and Spanish and how this property is related to the transparency in verbal morphology. In the following sections I will discuss four kinds of analyses for the lack of negated true imperatives: (i) a movement analysis; (ii) a semantic analysis; (iii) a selectional analysis; (iv) and a PF merger analysis. I will present data which bring problems for the three first analyses and argue that the PF merger analysis can adequately account for the behavior of negative imperative in BP and other Romance Languages.
3. Movement based analyses The two first analyses to consider here are based on constraints on movement. The difference between them is the nature of the postulated constraint: syntactic in the first case and semantic in the second one. 3.1 Movement is blocked by syntactic constraints Rivero (1994) proposes that the imperative feature is codified in the head of CP and that a (true) imperative verb has to move to this position to check its own imperative feature (cf. (8a)). According to this view, the lack of negative imperatives would be derived from the Head Movement Constraint (HMC): the Neg head, between CP and TP, blocks the verb movement to C0 (cf. 8b) and leads the derivation to crash.3 (8) a.
CP C verb + T [IMP]
TP C [IMP]
verb [IMP]
T
VP T
…….
. Some arguments in favor of her analysis are the VS inversion and the V-clitic order in imperatives in languages with SV and clitic-V order in declarative sentences.
© 2011. John Benjamins Publishing Company All rights reserved
Rerisson Cavalcante
b.
CP C verb + T [IMP]
NegP C [IMP]
Neg
TP
não/no(n) T verb T [IMP]
VP …..
Han (1999, 2001), however, argues against this analysis based on the fact that the negative marker is a clitic in languages like Spanish, Portuguese, and Italian. As a clitic, it should not block verb movement, but move with the verb as a complex, as we can see in cases of Aux-to-Comp movement in Italian interrogatives, as shown in (9a) from Rizzi (1982). The clitic nature of the negative marker is also true for Portuguese and Spanish as shown in (9b) and (9c), in which the adjacency between the marker and the verb can only be disrupted by (complement) clitics. (9) Italian a. (Non) avendo Gianni fatto questo, … neg having G. done this “(Not) having done this, John…” Portuguese b. Eu (ontem) não (*ontem) te saí I-non yesterday neg yesterday thou.acc left “I did not see you yesterday” Spanish c. (Ayer) no (*ayer) lo vi yesterday neg yesterday 3-sg.acc saw-1-sg “I did not see him yesterday”
3.2 Movement is blocked by semantic constraints Han (1999, 2001) agrees with Rivero (1994) with respect to the requirement of verb movement to C0 in order for the imperative feature to be checked. However, based on the clitic nature of the negative markers in the relevant languages, she proposes that when the complex Neg+V moves to C0 to check the imperative feature of the verb, the negative marker ends up taking scope over and canceling the illocutionary (imperative) force of the sentence, yielding an illegitimate semantic interpretation at LF and causing the derivation to crash.
© 2011. John Benjamins Publishing Company All rights reserved
Negative imperatives in Portuguese and other Romance languages
(10) a. *No cuenta lo neg tell.imp 3-sg-acc “I don’t order you to tell it” b. *NEG > IMP
CP
C no cuenta Neg verb
NegP C [IMP]
Neg no(n)
TP T
verb T [imp]
… …..
3.3 Problems with V-to-C analyses At first sight, an analysis based on V-to-C movement could explain the differences between BP and the other languages, as BP does not have V-to-C movement (cf. Torres Morais 1993; Lopes Rossi 1993). However, this proposal faces a conceptual problem and an empirical one. Conceptually, the idea that the negation cancels the illocutionary force of the sentence by taking scope over it is at odds with standard assumptions on the left periphery. For instance, it is a standard assumption that a negative marker can move (with V) to C in other types of sentences, like declarative and interrogative ones, without canceling declarative or interrogative force, as in the example (11) (see also (9a–b). However, if negation in C0 could cancel the directive force, it should also cancel declarative and interrogative force. There is no a priori reason to assume that the imperative force is specially sensitive to negation in some way that declarative and imperative force is not. (11) a. What don’t you understand? b. [CP what [C′ don’t C0 [TP you (…) ]]]
Empirically, the analysis predicts that any negative elements generated above C0 (or dislocated to a position higher than C0) should cancel imperative force. However, examples like (12) (from Zanuttini 1995), show that dislocated negative elements do not block true imperative forms in Italian. (12) A nessuno di-llo! To nobody say.imp-it.acc “Don’t say it to anybody!”
© 2011. John Benjamins Publishing Company All rights reserved
Rerisson Cavalcante
If the imperative verb has to be in C0 in Italian, then a dislocated negative expression, as a nessuno in (12), would be at least as high as Spec,CP and c-command the imperative force encoded in C0. 3.4 Additional data and an additional problem An even more problematic fact is that the movement approaches do not explain why phrasal negative adverbs such as Portuguese and Spanish nunca or Italian mai (“never”) also block true negative imperatives (cf. (13–15)), as I pointed out in Cavalcante (2008a). (13) Spanish a. ¡(*Nunca) cuenta lo! never tell.imp 3-sg.acc “(Never) tell it!” b. ¡Nunca lo cuentes! never 3-sg.acc tell.imp “Never tell it!” (14) Italian a. (*Mai) critica un componente della sua famiglia! never criticize.imp a member of.the your family “(Never) criticize a member of your family!” b. Mai criticare un componente della sua famiglia! never criticize.inf a member of-the your family “Never criticize a member of your family!” (15) European Portuguese a. (*Nunca) bebe (tu) antes de dirigir! never drink.imp thou before of drive “(Never) drink before driving!” b. Nunca bebas (tu) antes de dirigir! never drink.subj thou before of drive “Never drink before driving!”
In Cavalcante (2008a), I observed that, if both nunca and mai (“never”) are not heads, they should not block verb movement to C0. In Han’s (1999, 2001) analysis, the derivation in (16a) should be licit and the sentence in (16b) should be acceptable in EP, but it is not.4 . The same problem may be found in BP. I have been arguing (Cavalcante 2007, 2008b) that the post-sentential negative marker não in BP should be analyzed as generated in a category of the CP-system rather than as a (simple) VP adjunct or a IP-related category. Arguments for this analysis came from the facts that the BP post-sentential negative marker is sensitive
© 2011. John Benjamins Publishing Company All rights reserved
Negative imperatives in Portuguese and other Romance languages
(16) European Portuguese CP a.
C bebe drink.imp
TP C [IMP]
TP
AdvP nunca (DP) (tu)
TP T
VP
bebe drink.imp
……
b. *Bebe nunca (tu) antes de dirigir! drink.imp never thou before of drive “Don’t ever drink before driving”
In the next section, I will review and discuss a third kind of analysis for the lack of negated true imperative verbs across languages.5
to CP specification. It is incompatible (or marginal) with embedded sentences, with WHquestions and topicalization/focalization. If this analysis is correct, the post-sentential negative in BP should also take scope over C0 canceling the imperative force. The semantic analysis incorrectly predicts that sentences like (i) should be unacceptable independently of the post-verbal position of the negative marker. (i) (Não) sai / saia não! neg leave.ind leave.sub neg . In BP there is a dialectal variation in the acceptability of the suppletive imperative taken from indicative with preverbal nunca (“never”): (i) *Nunca conta isso! never tell.ind this “Never tell this!” (ii) ? Nunca faz isso never do.ind this (iii) ? Nunca (mais) faz isso never (more) do.ind this “Never do this (anymore)!” (iv) Nunca esquece de mim, tá? never forget.ind of me is “Never forget me, ok?” The existence of this variation show that the difference between EP and BP regarding to nunca with imperatives is more complex, but it is out of the limits of this paper and it will be addressed in a future work.
© 2011. John Benjamins Publishing Company All rights reserved
Rerisson Cavalcante
4. A selectional analysis To support her analysis of negative imperatives in Italian dialects, Zanuttini (1995) mentions two facts regarding crosslinguistic distribution of negative markers with respect to mood: (i) in Latin there were two different negative markers, one of which, ne, was used only in prohibitive sentences, i.e. in negative imperative and subjunctive sentences. (ii) Sadock & Zwicky (1985) show that, crosslinguistically, when a language have two morphologically distinct negative markers, they are sensitive to mood specifications and have a specialized distribution. Zanuttini (1995) then proposes that Italian has maintained the Latin distinction and has two (homophonous) negative markers: one that is used in nonimperative contexts and compatible with all kinds of verbal forms; and another one used in imperative contexts and is sensitive to mood/modality specifications. More specifically, she proposes that the negative marker used in imperatives requires a complement with a mood feature and that true imperative forms lack (tense or) mood specification.6 Once they cannot be taken as complement by these type of negative markers because of the absence of mood specification, the lack of negated imperative verbs would follow. This proposal could in principle explain the case of BP once both indicative and subjunctive forms, used as imperatives, would have a mood feature, which can satisfy the selectional requirement of the negative marker. However, if we were to assume Zanuttini’s (1995) analysis, it would be necessary to stipulate that EP and Spanish also have two different (preverbal) negative markers, as in Italian. A problem for this assumption is that it would also be necessary to stipulate that EP and Spanish have two kinds of nunca (“never”), and Italian, two kinds of mai (“never”) in order to explain why (preverbal) negative adverbials also block the true (mood-less) imperative forms.7
. In her first analysis (Zanuttini 1991), she assumes that imperatives lack Tense (i.e. they do not have TP). In this latter work, she assumes that the missing feature is Mood and not Tense. . One alternative would be to assume that even in sentences like (13)–(15), with a preverbal negative adverb but no negative marker in Neg0, the empty negative head in imperative may also have selectional properties which require a mood specialized complement. Note, however, that there would be a difference between an overt and non-overt negative marker, but between two non-overt negative markers, since non-imperative sentence with negative adverbs do not require subjunctive verbs, as shown in (i). (i) Ela nunca dorme / *durma tarde (Portuguese) she never sleep.ind sleep.subj late
© 2011. John Benjamins Publishing Company All rights reserved
Negative imperatives in Portuguese and other Romance languages
5. A PF merger analysis To solve the problems presented here, I assume Bošković’s (2004) analysis of negative imperatives in Greek. He proposes that the incompatibility between the negative marker and a true imperative form in Greek is due to m orphological considerations. More specifically, the imperative morpheme (in C0) behaves as an affix in PF and must merge with the (imperative) verb under adjacency. When the negative marker is present between the imperative morpheme and the verb, the adjacency is broken resulting in an illformed sentence, as shown in (17). (17) Greek a. Diavase! ler.imp “Read!” b. *Den/mi diavase! neg ler.imp “Don’t read!” c. Na mi diavazis! subj neg ler.subj “Don’t read!” d. [XP MIMP [NegP mi [TP diavase ]]]
Following Bošković’s (2004) analysis, I would like to propose that in EP, Spanish, and Italian, the imperative morpheme (in C)8 should be analyzed as a PF affix, which must merge with the (imperative) verb under adjacency. Assuming that nunca and mai appear in a specifier position, i.e. Spec,NegP (see Belletti 1990; Mioto 1992), they should also block the adjacency between the imperative morpheme and the verb. According to this proposal, it is not necessary to postulate V-to-C movement in imperatives. Even the postverbal position of the subject could be derived from this adjacency requirement, as I pointed in Cavalcante (2008a), as we will see below. At first sight, this proposal faces the same problem of how to deal with languages that lack true imperative forms, like BP. If we assume that the difference between those languages is that BP does not have the imperative morpheme, we
. Bošković proposes that the imperative morpheme is generated in a higher position and admits the possibility that it could be C0, but his analysis is not committed to this.
© 2011. John Benjamins Publishing Company All rights reserved
Rerisson Cavalcante
have a new9 problem: we would have to postulate that this type of languages simply does not have formal imperative sentences. However, Bošković (personal communication) suggests that it is possible to assume a parametric difference based not on the presence/absence of the imperative morpheme, but on its nature. The imperative morpheme would be present in all languages, but it might work either as a (PF) affix or as an independent morpheme. According to this idea, differently from the case of the other languages under discussion, BP does not have an adjacency requirement. The imperative morpheme in C0 is an independent morpheme that encodes the imperative force, but it is not an affix in PF and does not require adjacency with a verbal form. According to this view, the change in the nature of the imperative morpheme in C0 is probably a consequence of the loss of verbal morphology in BP,10 mentioned in Section 2.2. Independent evidence for this proposal is provided by the fact that BP allows preverbal subjects in imperatives (Cavalcante & Simioni 2009) (cf. (18)), which are banned in EP, in which the imperative subjects are either null or postverbal (cf. Han 1998), as shown in (19). (18) Preverbal (contrastive) subjects in BP imperatives A: — Abra / abre a porta! open.subj open.ind the door “Open the door!” B: — Eu não. VOCÊ abra / abre a porta! I neg you open.subj open.ind the door “Not me! YOU open the door!” (19) Subjects in EP imperatives Lava (tu) os pratos! wash.imp thou the dishes “You wash the dishes!”
(Han 1999: 141)
. Actually, this problem is also present in the version of movement analysis proposed by Rivero (1994) and Han (1999). According to both analyses, to explain the lack of incompatibility between imperatives and negation in languages that do not have true imperatives, it is necessary to postulate that these languages do not have sentences with imperative features at all. . This relation between the loss of verbal morphology and the change of the nature of the imperative affix (in C0) becomes more likely if we assume Chomsky’s (2001) proposal that C0 is the real place where the phi features (of T0) are generated.
© 2011. John Benjamins Publishing Company All rights reserved
Negative imperatives in Portuguese and other Romance languages
It follows directly from the analysis above that preverbal subjects block the adjacency between the IMP morpheme and the imperative verb. Because of the adjacency requirement in EP, preverbal subjects is ruled out by PF. Once this requirement does not exist in BP, preverbal subjects are available. As for the postverbal position in EP, we can account for it by assuming the recourse of PLC (Pronounce Lower Copy), proposed by Franks (1998), which Bošković (2004) uses to explain the change in the Clitic-Verb order to Verb-Clitic order in Greek imperatives. According to the proposal, PF can pronounce a lower copy of a syntatic expression in order to avoid a phonological violation. In Greek non-imperative sentences display only Clitic-Verb order, as shown in (20). However, in imperative sentences we observe the mirror image. Only Verb-Clitic order is available, as shown in (21). (20) Greek: V-Clitic order in indicatives a. To diavasa 3-sg-acc read.ind b. *Diavasa to read.ind 3-sg.acc “I read it” (21) Greek: Clitic-V order in imperatives a. *To diavase! 3-sg.acc read.imp b. Diavase to! read.imp 3-sg.acc “Read it!”
Assuming PLC, Bošković (2004) proposes that the clitic blocks the adjacency between the affix and the verb, leading to a violation in PF. To solve this problem, PF deletes the higher copy of the clitic and pronounces a lower one, as shown in the derivation in (22). (22) [XP MIMP [TP to diavase to ]]]
Following the same idea, I propose that the postverbal position of subjects in EP can be accounted for by the pronunciation of a lower copy of the subject, as shown in (23), in order for avoid a violation in PF. (23) [XP MIMP [TP subject verb [VP subject (…) ]]]
© 2011. John Benjamins Publishing Company All rights reserved
Rerisson Cavalcante
6. Summarizing In this section, I summarize the main conclusions on the interaction between the imperative verbs and negative markers: a. BP lost second person morphology and the distinction between indicative and imperative verbs. As a consequence, the distinction between true and suppletive imperatives was also lost. b. The imperative morpheme in C0 is a PF affix in EP, Italian, and Spanish and requires adjacency with the imperative verb. But in BP the imperative morpheme is an independent morpheme. c. Negative adverbs such as nunca and mai (“never”) block the adjacency between the verb and the imperative affix (in EP, Spanish and Italian). d. Preverbal subjects are not available in imperatives in EP, Spanish, and Italian, because of the adjacency requirement on the imperative affix, but are available in BP, because its imperative C0 is not an affix.
References Belletti, Adriana. 1990. Generalized Verb Movement. Torino, Rosenberg & Sellier. Bošković, Željko. 2004. “On the Clitic Switch in Greek Imperatives”. Balkan Syntax and Semantics. ed. by Olga Mišeska Tomić, 269–291. Amsterdam: John Benjamins. Cavalcante, Rerisson. 2007. A Negação Pós-Verbal no Português Brasileiro. Master’s Thesis. Universidade Federal da Bahia. Cavalcante, Rerisson. 2008a. “Imperativo e Negação em Português”. Paper presented at the II Workshop do Projeto Temático Sintaxe Gerativa do Português Brasileiro na Entrada do Século XXI: Minimalismo e Interfaces. University of São Paulo, Brazil, November 2008. Cavalcante, Rerisson. 2008b. “Post-Sentential Negation in Brazilian Portuguese”. Paper presented at the VII Workshop on Formal Linguistics, Universidade Federal do Paraná, Curitiba, August 2008. Cavalcante, Rerisson. 2010. “Interação entre Imperativo e Negação”. Revista de Estudos da Linguagem 8:2. 9–36. Cavalcante, Rerisson & Leonor Simioni. 2009. “Sobre a Ordem VS em Sentenças Imperativas no Português Brasileiro”. Paper presented at the XII Encontro de Estudantes de Pós- Graduação em Lingüística da USP, University of São Paulo, Brazil, June 2009. Franks, Steven. 1998. “Clitics in Slavic”. Position paper presented at the Comparative Slavic Morphosyntax Workshop, June 5–7, Bloomington. Han, Chung-Hye. 1999. “Cross-linguistic Variation in the Compatibility of Negation and Imperatives”. Proceedings of the 17th West Coast Conference on Formal Linguistics. 265–279. Han, Chung-Hye. 2001. “Force, Negation and Imperatives”. The Linguistic Review 18. 289–325. Kayne, Richard. 1992. “Italian Negative Infinitival Imperatives and Clitic Climbing”. Hommages à Nicolas Ruwet. Ed. by Liliane Tasmowsky & Anne Zribi-Hertz, 300–312. Ghent: Communication and Cognition.
© 2011. John Benjamins Publishing Company All rights reserved
Negative imperatives in Portuguese and other Romance languages
Lopes Rossi, Maria Aparecida Garcia. 1993. “Estudo Diacrônico sobre as Interrogativas do Português do Brasil”. Português Brasileiro: Uma Viagem Diacrônica. Ed. by Ian Roberts & Mary A. Kato, 263–303. Campinas, Unicamp. Mioto, Carlos. 1992. Negação Sentencial no Português Brasileiro e a Teoria da Gramática. Ph.D. Dissertation. Universidade de Campinas (Brazil). Postma, Gertjam & Wim van der Wurff. 2007. “How to Say No and Don’t: Negative Imperative in Romance and Germanic”. Imperative Clauses in Generative Grammar: Studies Offered to Frits Beukema ed. by Wim van der Wurff, 205–249. Amsterdam: John Benjamins. Rivero, Maria-Luisa. 1994. “Negation, Imperatives and Wackernagel Effects”. Rivista di Linguistica 6. 39–66. Rizzi, Luigi. 1982. Issues in Italian Syntax. Dordrecht: Floris. Sadock, Jerrold M. & Arnold M. Zwicky. (1985). “Speech Act Distinction in Syntax”. Language Typology and Syntactic Description. Ed. by Timothy Shopen, 155–196. Cambridge University Press, Cambridge. Torres Morais, Maria Aparecida C. R. 1993. “Aspectos Diacrônicos do Movimento do Verbo, Estrutura da Frase e Caso Nominativo no Português do Brasil”. Português Brasileiro: Uma Viagem Diacrônica. Ed. by Roberts, Ian & Mary A. Kato, 263–303. Campinas, Unicamp. Zanuttini, Raffaella. 1991. Syntactic Properties of Sentential negation. A Comparative Study of Romance Languages. Ph.D. Dissertation, University of Pennsylvania. Zanuttini, Raffaella. 1995. Reflexes of Clausal Structures in the Syntax of Negation: a Comparative Study of Romance Languages. Georgetown University. Zeijlstra, Hedde H. 2006. “The Ban on True Negative Imperatives”. Empirical Issues in Syntax and Semantics 6. 405–424.
© 2011. John Benjamins Publishing Company All rights reserved
© 2011. John Benjamins Publishing Company All rights reserved
