• Nu S-Au Găsit Rezultate

THE TOBLER-MUSSAFIA LAW

N/A
N/A
Protected

Academic year: 2022

Share "THE TOBLER-MUSSAFIA LAW "

Copied!
19
0
0

Text complet

(1)

THE TOBLER-MUSSAFIA LAW

ALEXANDRU NICOLAE1, DANA NICULESCU2

Abstract. This study analyzes the position occupied by pronominal clitics in the clause with respect to the verb in old Romanian (OR) on the basis of an extensive corpus analysis of 16th – 18th century texts. The corpus analysis shows that, from the earliest texts, OR pronominal clitics are attested in second, third, fourth, etc. position in the clause, and exceptionally also in first position. Therefore, they do not fully observe the Tobler-Mussafia Law, which was in function in old Romance languages. OR pronominal clitics are IP-clitics, which can be placed both in pre- and in postverbal position (proclisis and enclisis). Gradually, due to the going reduction of V-to-C movement, pronominal proclisis generalizes.

Keywords: pronominal clitics, enclisis, proclisis, old Romanian, the Tobler- Mussafia Law, V-to-C movement.

1. INTRODUCTION

Word order in old Romanian (OR) is generally freer than in modern Romanian (MR).

For instance, in OR, arguments of nouns could be placed not only to the right, but also to the left of their head and pronominal clitics and auxiliaries could be positioned both pre- and postverbally; scrambling is attested in the verbal domain – the displacement of the clitic / auxiliary / clitic plus auxiliary from the lexical verb through the insertion of phrasal constituents (Dragomirescu 2013, 2014) –, as well as in the nominal domain – the adjunction of adverbials and conjunctions to phrasal DP-internal constituents (Nicolae 2015).

Having a descriptively oriented empirical goal, this article investigates OR word order as reflected by (a) the position occupied by dative and accusative pronominal clitics within the clause (first, second, third position, etc.) and (b) their ordering in relation to their morphosyntactic host, the inflectional phrase, aiming to reveal the changes that took place in the transition from old to modern Romanian (MR). The ordering of OR pronominal clitics within the cluster is identical to that of MR, namely the fixed dative–accusative order. However, while in MR pronominal clitics generally occur to the left of the verb (in       

1 “Iorgu Iordan – Al. Rosetti” Institute of Linguistics, Bucharest & University of Bucharest, [email protected]. Alexandru Nicolae acknowledges the support of the Sectorial Operational Programme Human Resources Development (SOP HRD), financed from the European Social Fund and by the Romanian Government under the contract number SOP HRD/159/1.5/S/136077.

2 University of Amsterdam, [email protected].

RRL, LX, 2–3, p. 223–242, Bucureşti, 2015

(2)

 

proclisis), except for gerunds and imperatives, in OR, they are attested both as proclitics and as enclitics with all types of verb forms. These two word orders characterize not only old Romanian, but also the older stages of other Romance languages (de Dardel and de Kok 1996, Fontana 1997, Fischer 2002, Poletto 2014 a.o.). Previous studies on old Romance (some of them including Romanian data) account for the possibility of pronominal enclisis by invoking the Tobler-Mussafia Law, while other analyses focusing especially on Romanian look at the relation between postverbal clitic placement and Wackernagel’s Law.

We aim to determine the extent to which either the Tobler-Mussafia Law or Wackernagel’s Law were in effect in the earliest OR texts. We argue that, as early as the 16th century, OR clitics are not Wackernagel clitics, a conclusion confirming the results reached by Alboiu and Hill (2012), Hill and Alboiu (2016: 61 ff.), whose analysis is based on a later and much smaller corpus of OR texts than the one used here. The fact that OR clitics take the inflectional phrase as their morphosyntactic host raises the question whether they might observe the Tobler-Mussafia law, which supposedly was in effect in Romanian in a stage prior to its first attestations.

In order to establish the place of OR pronominal clitics in the clause and within the IP, we conducted an exhaustive corpus analysis for the first OR period (1500–1640) and an analysis of a large second OR period corpus (1640–1780), using the OR corpus compiled by Emanuela Timotin for The Syntax of Old Romanian (ed. Gabriela Pană Dindelegan, in preparation). It is important to mention that both corpora contain translated, as well as original Romanian texts. Most of the translated texts have an Old Church Slavonic source (a lesser number of texts have a Hungarian or Greek source), which raises the question whether some of the encountered syntactic structures are literally copied from the original.

While this does remain a possibility, we maintain the hypothesis that the syntax of these texts, including the position of clitic pronouns, reflects the real usage of the period to a great extent; therefore, we take the data drawn from translations to be relevant for our investigation.

The novelty of our analysis in relation to previous studies on pronominal clitic word order consists in a systematic (and partly exhaustive) corpus analysis of OR, which provides a precise image of the range of word order possibilities in this period and of the changes taking place from the first to the second OR period.

With respect to the syntax of the clause, we adopt the following generally accepted ideas in the literature: (i) clauses are split into three syntactic domains, the (discourse- related) CP-field, the IP-field (hosting mainly the Mood-Tense-Aspect projections, but also accommodating other types of syntactic material), and the vP-field (the thematic domain), hierarchically structured (CP > IP > VP) and (ii) verbs undergo movement on the clausal spine, movement being always to the left (i.e. raising).

The paper is structured as follows: section 2 briefly discusses the Tobler-Mussafia Law in relation to Wackernagel’s Law and the accounts that were proposed in the literature for the fixed position of pronominal clitics within the clause; section 3 contains the results of the corpus analysis; section 4 brings evidence that the OR pronominal clitics attach at the level of the inflection, instead of the level of the complementizer phrase, and, therefore, are verbal clitics; the fifth section is represented by the conclusions.

(3)

 

2. OLD ROMANIAN CLITICS AND THE TOBLER-MUSSAFIA LAW The Tobler-Mussafia Law (TML) was put forth by Adolf Tobler (1875) and Adolfo Mussafia (1886) with reference to old Romance languages, i.e. old French and old Italian.

In a nutshell, it states that clitic object pronouns cannot occupy the initial position in the clause (Salvi 1993, Benincà 1994, de Dardel and de Kok 1996, Fontana 1997, Fischer 2002, Salvi 2011: 363–365), where the initial position is defined with respect to the minimal clause, without hanging topics or left dislocated elements, which are extrasentential (Salvi 1993, Ledgeway 2012: 161). Thus, pronominal clitics have a strong tendency to occur in the second position of the clause. If the verb phrase occupies the first position, the pronominal clitic will be placed postverbally, a possibility lost by the modern Romance languages, except for Portuguese and Asturo-Leonese varieties (Martins 1995, Salvi 2011:

364).

Wackernagel’s Law (WL), formulated by Jakob Wackernagel (1892), provides a phonological generalization regarding the position occupied by clitics in the clause in Indo- European languages. The law states that inherently unstressed/unstressable words (i.e.

clitics) are generally found in second position in the clause, after the first prosodic word, and they cannot occupy the clause’s first position (Fischer 2002: 26). WL manifests in Indo-European, Latin, Ancient Greek, Old English, Old High German, Modern German, Slavic languages, with the exception of Modern Bulgarian and Macedonian (Cardinaletti and Roberts 2002, Bošković 1999, Tomić 2004, Pancheva 2005).

The main characteristics of Wackernagel (W) clitics are: (a) they do not need to be adjacent to the verb, as in Old Bulgarian (Pancheva 2005), and (b) they can occur at the left of negation, as in Serbo-Croatian (Rivero 1997). In opposition to them, Tobler-Mussafia (TM) clitics must be placed adjacent to the inflection and at the right of sentential negation.

This different syntactic behaviour reflects the fact that Wackernagel clitics are Complementizer-oriented (clausal clitics), while TM clitics are Inflection-oriented (verbal clitics; Rivero 1997). However, there is a relation between TML and WL: studies on old Romance languages agree that TML is the Romance variant of Wackernagel’s law (Wanner 1987, Rivero 1991, Salvi 1993, Martins 1995, de Dardel and de Kok 1996, Fischer 2003, Ledgeway 2011, 2012 a.o.). It should be noted that these two laws actually describe tendencies, not unbreakable laws, since there are exceptions to second position clitic placement, as shown, for example, by Taylor (1990) (apud Fontana 1997) and Halpern (1995).

Halpern (1995) makes a typology of second position clitics in general, distinguishing between Second Word (2W) and Second Daughter (2D) languages according to the type of constituent which acts as the host of encliticization. In 2W languages, clitics are placed in second position after the first prosodic word of the clause, even if this means placing them inside a complex constituent (e.g. Serbo-Croatian), while in 2D languages, clitics occur in second position after the first phrase of the clause (e.g. old Romance languages, see (1)).

(1) a. [Toutes ces choses] te presta Noste Sires (old French) all these things CL.DAT.2SG lent our Lord

‘All these things our Lord lent you’

(de Kok 1985: 74, in Cardinaletti and Roberts 2002)

(4)

 

b. [esto] les mandaua el rey cuydando que… (old Spanish) this CL.DAT.3PLordered the king thinking that

‘The king ordered them to do this, thinking that…’(GEI.002,in Fontana 1997: 228)

The possibility of placing pronominal clitics to the right of the verb (2) is attested from the 13th century up to the 15th century in Italian, up to the 16th century in French, and up to the end of the 17th century in Spanish and Portuguese; in Portuguese, the verb–clitic word order occurs again from the 19th century. Across Romance, the TML stopped being consistently observed between the 13th and the 17th centuries, with French as the first Romance language where the TM Law started to erode (13th c.) and Portuguese the last (17th c., Martins 1995, Fischer 2002: 56).

(2) Doné-li terme (old Piemontese) gave.3sg=CL.DAT baths

‘He gave them baths’ (Sermones Subalpini.12th–13thc., in Benincà 1994)

Both syntactic (Benincà 1994, Rivero 1991, 1997, Halpern 1995, Hirschbühler and Labelle 2000, Labelle and Hirschbühler 2005, Cardinaletti and Roberts 2002, Fischer 2003, Pancheva 2005, Ledgeway 2008, 2012: 162) and phonological analyses (Fontana 1993, Bošković 1999, 2001) have been put forth to account for the word order restrictions displayed by second position clitics (i.e. Wackernagel and Tobler-Mussafia clitics).

Syntactic accounts place second-position clitics in the head or specifier of a phrase that is the complement of C and, on its turn, takes the verb’s inflection as its complement.

The phrase is labelled Agr1P (Cardinaletti and Roberts 2002), Wackernagel P / Tobler- Mussafia P (Rivero 1997), or ΣP (Fischer 2003). Alternatively, it has been proposed that second position Romance clitics adjoin at IP/TP (Fischer 2003, Pancheva 2005). The ban on the clitic’s first position was analysed as a verb-second effect, since old Romance languages are verb-second languages (Adams 1987, Benincà 1995, Manzini 1994, Fontana 1997, Hirschbühler and Labelle 2000, Salvi 2004, Ledgeway 2007, 2008, 2012: 162, Poletto 2014; but see Rivero 1997 for arguments that old Spanish is not verb-second). In verb-second languages, verb movement (to the Finiteness Phrase of the C-domain, Ledgeway 2008, 2012: 161) and the presence or absence of a phrasal constituent XP in the CP (Theme / FocusP) determine the pre- or postverbal position of the clitic. When a constituent is in the Theme / FocusP, the verb is in second position, therefore the XP – clitic – verb word order is obtained, since the clitic does not count as a prosodic word. When the Theme / FocusP is empty, the verb raises to C giving rise to a verb-first (V1) structure, and the verb – clitic word order is obtained (Salvi 2004, Labelle and Hirschbühler 2005, Benincà 2006, Ledgeway 2008). In this paper, we shall also assume that the pre- and postverbal placement of pronominal clitics is the consequence of the level of verb movement (V-to-I or V-to-C).

Halpern (1995) puts forth a syntactic analysis that includes a PF operation, i.e.

movement in the phonological component, which is considered responsible for the placement of the clitic in second position, in order to avoid starting a clause with an unstressed constituent.

(5)

 

There are also phonological accounts (Fontana 1993, Bošković 1999) arguing that clitic placement takes place in syntax, and no further movement operations occur in the phonological component; however, phonology filters out those word orders that are grammatical, but do not conform to the clitics’ requirement to attach enclitically to their host.

3. THE EVOLUTION OF OLD ROMANIAN CLITIC PLACEMENT FROM A ROMANCE PERSPECTIVE

The main observation about OR clitics of previous studies is that old Romanian had second-position clitics. In the earliest old Romanian texts, dating from the 16th century, first position clitics are generally avoided. This was explicitly or implicitly put down to the action of either the Tobler-Mussafia Law or of Wackernagel’s Law (Frâncu 1980, 1997:

172, de Dardel and de Kok 1996, Reinheimer Rîpeanu 2002, Alboiu and Hill 2012).

Although Romanian is a Romance language, Frâncu (2009) argues that Wackernagel’s Law could have been in effect in OR as a consequence of the Old Slav(on)ic influence.

Counterarguments to this claim are brought by Alboiu and Hill (2012), who show that 17th– 18th century old Romanian did not have Wackernagel clitics. Furthermore, Alboiu, Hill and Sitaridou (2014) agree with Benincà (1983/4, apud Alboiu, Hill and Sitaridou 2014) that Romanian may have been a V2 language in a period for which there are no attested texts;

however, as early as the first Romanian documents of the 16th c., the alleged V2 rule was no longer active.

3.1. What counts as first position in Romanian and in the early Romance clause?

Before we investigate the position occupied by OR pronominal clitics in the clause, it is incumbent on us to establish what counts as the first position of a clause and whether there are differences between the data contained in the old Romanian corpus and the data provided by existing studies on medieval Romance languages.

3.1.1. Clitic placement after a coordinating conjunction

Romance continuators of the Latin coordinating conjunctions ET ‘and’, AUT ‘or’ and

MAGIS,originally an adverb, but having the same behaviour as the other conjunctions in old French, where it came to have the meaning‘but’ (de Dardel and de Kok 1996: 126),are argued not to function as first prosodic words in early old Romance, when the TML was strictly applied. The consequence is that clitics were not allowed to attach enclitically to these conjunctions (Renzi 1992: 279, de Dardel and de Kok 1996: 113, Wanner 1987), so that the word order was and/but/or – XP/verb – clitic. Hirschbühler and Labelle (2000) call this stage the strict Tobler-Mussafia phase (3)–(4).

(3) […] e mist la al lit David (old French) and placed CL.ACC on.the bed David

‘and placed it on David’s bed’ (QLR 2 R 8.27, in Hirschbühler and Labelle 2000)

(6)

 

(4) Ma dicoti (old Italian)

but tell=CL.DAT.2SG

‘But I tell you’(Nov.141, 7, in Renzi and Salvi 1992: 279)

As early as the end of the 12th century, French clitics start to occur immediately to the right of et ‘and’ (Hirschbühler and Labelle 2000), as shown in (5), which means that the coordinating conjunction was reanalyzed as the first word of the clause.

(5) Au soir dist Lancelos a la dame…, et la mercia moult at.the evening said Lancelot to the lady and CL.DAT thanked much

‘In the evening, Lancelot said to the lady..., and thanked her very much’

(Artu 56,2, in Benincà 1994)

As far as OR is concerned, the corpus analysis has shown that in 16th century texts there are only attestations of the e (< ET)–(XP)–verb – clitic word order (6). Considering that first position clitics are exceptional in 16th century texts, the postverbal placement of the clitic with this conjunction can be due to the fact that, in the earliest documents, the Romanian conjunction e, now obsolete, does not function as a first position in the clause.

(6) e pleca-te-veri (CC1.1567: 165v)

and humble.INF=CL.REFL.ACC.2SG=AUX.FUT.2SG

‘and you will humble yourself’

From the 17th century onwards, the e – clitic – verb ordering is attested in Romanian texts (7). This coincides with the increase in frequency of first-position pronominal clitics;

therefore we cannot decide whether e ‘and’ is reanalyzed as a first prosodic word of the clause at this point in the history of Romanian or the pronominal clitic itself is in first position.

(7) e să boteadză (CazV.1643: 413) and CL.REFL.3SG baptize.PRES.3SG

‘and he gets baptized’

Differently from the status of e ‘and’ in the 16th century, the coordinating conjunctions şi ‘and’, ci/ce ‘but’, însă ‘but’, and au ‘or’ function as clause-initial elements, as shown by the fact that clitics frequently occur to their right, in second position (8)–(10).

A possible explanation for the different behaviour of the two synonymous conjunctions, e and şi ‘and’, is that the former has its etymological source in a Latin conjunction (ET), while the latter continues the Latin adverb SIC ‘so’,which always functioned as a first prosodic word of the clause. The şi / ci/ce / însă / au – verb – clitic word order is, however, not excluded in OR (11).

(8) şi veselesc (CP1.1577: 50v) and CL.REFL.ACC.1SG rejoice.PRES.1SG

‘and I rejoice’

(9) nimini de rudele meale nu m-a grijit, nobody of relatives.DEF my not CL.ACC.1SG=AUX.PERF take.care.PPLE

(7)

 

ci m-au căutat Dragnea (DÎ.1600: VIII) but CL.ACC.1SG=AUX.PERF look.after.PPLE Dragnea

‘none of my relatives took care of me, but Dragnea looked after me’

(10) când intra în cortul adevărăturei au se apropiia

when enter.IMPERF in tent.DEF truth.DEF.GEN or CL.REFL.3SG draw.near.IMPERF

cătră oltariu (PO.1582: 317) towards altar

‘when he entered the tent of truth or he came near the altar’

(11) şi lăuda-se-voru (PH.1500–10: 4r) and boast=CL.REFL.3PL=AUX.FUT.3PL

‘and they will boast’

One special context in which pronominal clitics are in preverbal position after a coordinating conjunction is represented by verbs in the imperative. Attestations of the clitic–imperative word order are found both in OR and in MR, such as in (12), in the second conjunct of a coordination.

(12) ia patul tău şi te du în

casa ta

take.IMP bed.DEF your and CL.ACC.2SG go.IMP in house.DEF your

‘Take your bed and go into your house!’ (CC2.1581: 55)

3.1.2. The sentential negator nu (‘not’)

In the strict TML phase, the sentential negator counts as the first prosodic word of the clause (de Dardel and de Kok 1996: 115). Our corpus analysis has shown that the situation of 16th century OR is similar to that of other Romance languages. There are many attestations of the negative adverbial nu (‘not’) in first position in the clause, followed by the pronominal clitic in preverbal position (13)–(14).

(13) Nu ţi se cade să o iai ea not CL.DAT.2SG CL.REFL.ACC ought SUBJ CL.ACC.3SG take.SUBJ her

‘You ought to not take her’ (CT.1561: 30r)

(14) Nu se va muta cătră tine hicleanulu not CL.REFL.ACC.3SG AUX.FUT.3SG come.INF towards you evil.DEF

‘The evil one will not come to you’ (PH.1500–10: 3v)

3.2. Old Romanian second-position clitics

Our extensive 16th century corpus analysis has shown that second-position clitics occur both in main (15)–(25) and subordinate clauses (26)–(27). In main clauses, proclitic second-position clitics are attested in declaratives (15), in yes-no interrogatives (16), as well as in wh-interrogatives (17), with an adverbial (15) or DP (16)–(17) occupying the first position in the clause.

(8)

 

(15) Aşa mă rog domnivoastră (DÎ.1599–600: XXII) so CL.REFL.1SG ask.PRES.1SG you.POL.DAT

‘so I ask you’

(16) Frate drag, voia ţi-e această fată în leagea lu Dumnezeu brother.VOC dear will.DEF CL.DAT.2SG=is this girl in law LUI God să o iai la tine, să-ţi fie căsătorie?

SUBJ CL.ACC.3SG take.SUBJ at you SUBJ=CL.DAT.2SG be.SUBJ.3SG wife

‘Dear brother, do you want to take this girl in God’s law to be your wife?’

(CM.1567–8: 258v)

(17) de cine mă tem? (CP1.1577: 43v) of who CL.REFL.ACC.1SG be.afraid.PRES.1SG

‘Who am I afraid of?’

The corpus analysis has shown that enclisis in non-imperative clauses, with the clitic in second (but also in third, fourth, etc.) position in the clause, occurs in Romanian from earliest texts until the end of the OR period (the end of the 18th century). Second-position clitics occur as enclitics in declaratives (18)–(21), both with synthetic (18), (19), (21) and analytic (20) verb forms, in yes-no interrogatives (22), (23), and in clauses with imperative force, whose finite verb is in the subjunctive (25) or in the imperative (26).

(18) Rogu-mă măriei tale (DÎ.1598: XVII)

ask=CL.REFL.SG higheness.DEF.DAT your

‘I ask your highness’

(19) Sârguiia-se această uşă să treacă (CC2.1581: 179) try.IMPERF=CL.REFL.3SG this door SUBJ pass.SUBJ

‘He was trying to pass beyond this door’

(20) Priimitu-mi-au în jele lacrămi ovilite (DPV.1673: 57) received=CL.DAT.1SG=AUX.PERF in sadness tears poor

‘He received my poor tears when I was sad’

(21) Mulţumescu-ţi Ţie (ITM.1750: 571) thank.PRES.1SG=CL.DAT.2SG you.DAT

‘I thank You’

(22) Lepădaşi-te de satana? (Mol.1689: 215) reject.PS.2SG=CL.REFL.2SG of Satan

‘Have you rejected Satan?’

(23) Cunoşti-mă pre mine, au ba? (NL.~1750–66: 199v) recognize.PRES.2SG=CL.ACC.1SG DOM me or not

‘Do you recognize me or not?’

(24) Hie-ţ, Doamne, mila ta cea be.SUBJ.3SG=CL.DAT.2SG God mercy.DEF your CEL.F zvântă preste noi

holy over us

‘May your holy mercy descend over us’ (DPV.1673: 229) (25) Ascundeţi-vă comoara (CC2.1581: 44)

hide.IMP.2PL=CL.REFL.DAT.2PL treasure.DEF

‘Hide your treasure’

(9)

 

Imperative clauses and, to a more limited extent, imprecatory conditionals are the only ones in which the verb-clitic word order is still in use in standard MR. Pronominal clitics in enclisis are still attested dialectally in declaratives and in yes-no interrogatives, for instance in Crişana and Maramureş (TDR 1984: 309, 337). The preservation of this archaic syntactic characteristic in contemporary Romanian dialects constitutes evidence that the verb−clitic word order in OR texts cannot be explained as a loan translation from Old Church Slavonic sources.

Pronominal proclisis in subordinate clauses is very frequent. It is also the expected word order if the second-position rule was in effect in OR, because the wh-phrase (26) or the complementizer (27) occupies the first position of the subordinate.

(26) cându ne aducem aminte (CCat.1560: 9v) when CL.REFL.1PL bring.PRES.1PL in-mind

‘when we remember’

(27) că se teame de noi că-l vomu that CL.REFL.3S be-afraid.PRES.3SG of us that=CL.ACC.3SG AUX.FUT.1PL prinde (DÎ.1599: XVIII)

catch.INF

‘because he is afraid that we’ll catch him’

3.3. Deviations from the TML: Clitics in 3rd / 4th / 5th positions

Third position clitics can be pre- or postverbal, but, usually, studies on old Romance relate the clitic’s 3rd position to its postverbal placement, which results in the XP–verb–

clitic word order. This yields enclisis in contexts in which the verb does not occupy the first position in the clause (28).

(28) Et Ewruins ot en gran dol (old French)

and Ewruins had CL.ADV great sadness

‘And Ebroïn was very sad’ (Vie de saint Leger, in de Dardel and de Kok 1996: 34) Examples like (28) constitute deviations from the Tobler-Mussafia Law. They were explained either through the co-existence of archaic and/or innovative word orders during each dominant phase in the evolution of clitic placement in Romance (de Dardel and de Kok 1996: 122), or, with specific reference to old Romanian, through the action of discourse factors, namely of movement to focus (Alboiu and Hill 2012, Alboiu, Hill and Sitaridou 2014). The XP–V–clitic word order is attested in French as early as the 11th century (de Dardel and de Kok 1996: 35) and in Spanish, Italian, and Portuguese, from the 13th century. After the 17th century, in all Romance languages with the exception of Portuguese, this word order is no longer attested (see section 2 above; Salvi 1993, Fischer 2002: 86ff.). Our corpus analysis has shown that Romanian kept this archaic word order longer than the other Romance languages (with the exception of Portuguese), namely until the end of the OR period.

In OR, both pre- and postverbal clitics could occur in 3rd, 4th, 5th position in the clause, not only in main (29)–(38), but also in subordinate clauses (39)–(44) (see also Alboiu and Hill 2012, Alboiu, Hill and Sitaridou 2014). In main clauses, preverbal (39) or

(10)

 

postverbal (30)–(35) clitics in third (29)–(31), (36)–(38), fourth (32)–(33) and fifth (34)–

(35) position are attested in declaratives, in yes-no interrogative clauses (36), as well as in wh-interrogatives (37), (38). In wh-interrogatives, the word order wh-phrase–clitic–verb is expected, since the wh-phrase can act as the clitic’s phonological host, but the corpus analysis has shown that the wh-phrase–verb–clitic order is also possible, although less frequent. We found no occurrences of encliticization in wh-interrogatives in original texts, where interrogatives are scarce.

(29) Deci iară rog domilor voastre

so again CL.REFL.ACC.1SG ask.PRES.1SG lordships.DEF.DAT your

‘So I ask your lordships again’ (DÎ.1600: CXIV) (30) Domnul râde-şi lui (CP1.1577: 65r)

God laugh=CL.DAT him

‘God is laughing at him’

(31) Într-aciia vreme tâmplatu-s-au (NL.~1750–66: 215r) in=that time happened=CL.REFL=AUX.PERF

‘In that time it happened’

(32) Şi toţi oamenii aflămu-ne ca într-o corabie întraţi (CC2.1581: 55) and all people.DEF find=CL.REFL.1PL like in=a ship enter.PPLE

‘and we, all the people, are as if we had entered a ship’

(33) Şi cu adevăratu cade-i-să acestu nume

and with true ought=CL.DAT.3SG=CL.REFL.3SG this name

‘And he really deserves this name’ (CLM.1700–50: 178r)

(34) Şi a mulţi oameni, cu hitlenşugul său, de avuţie i-au golit and at many people with slyness.DEF his of wealth CL.ACC.3PL=AUX.PERF emptied

‘And has robbed many people of their wealth with his slyness’ (DÎ.1593:

LXXXIX)

(35) Iară Iuda în veaci sălăşlui-să-va (DPar.1683: I.6v) and Judas in ages live=CL.REFL=AUX.FUT

‘And Judas will live [there] forever’

(36) Doară ispovedi-ţi-se ţărâna? (CP1.1577: 49r) Q confess=CL.DAT.2SG=CL.REFL.ACC.3SG ground.DEF

‘Will the ground confess to you?’

(37) întru iadu [cine spovedi-se-va]? (PH.1500–10: 4r) in hell who confess=CL.REFL=AUX.FUT

‘Who will confess in hell?’

(38) Până când rădică-se dracul mieu spre mine? (CP1.1577: 18r) till when rises=CL.REFL devil.DEF my towards me

‘How long will my devil keep coming to me?’

Both proclisis (39)–(40) and enclisis of pronominal clitics (41)–(44) in third position are attested in subordinate clauses. In the first OR period, postverbal clitics occur in subordinate clauses both in translations and in original texts (with wh-relatives, postverbal clitics occur only in translations). This shows that old Romanian preserves an ancient word order that disappeared much earlier from the other Romance languages. In the evolution of

(11)

 

Romance, the verb–clitic word order was first lost in embedded clauses (Fischer 2003) and, for this reason, in medieval Romance languages (with the exception of old Catalan and old Neapolitan) clitics do not occur postverbally in subordinate clauses (Benincà 1994, Martins 1995, Fischer 2003, Ledgeway 2007).

(39) că foarte-mi trebuiescu (DÎ.1599–600: XXII) that much=CL.DAT.1SG need.PRES.3PL

‘because I need them a lot’

(40) câţ Dumnezeu i va da

how.many God CL.DAT.3SG AUX.FUT give.INF

‘how many God will give him’ (DÎ.1595–6: XIII) (41) Că sculară-se (CP1.1577: 60r)

that get.up.PS.3PL=CL.REFL

‘that they got up’

(42) Pentru că acest satu şi aceşti rumâni fostu-i-am

because this village and these serfs be.PPLE=CL.ACC.3PL=AUX.PERF.1SG cumpărat eu (DRH.B.XXX.1645: 248)

buy.PPLE I.NOM

‘Because I had bought this village and these serfs’

(43) Ferice de cei [ce lăsară-se fără-legile]

happy of those that gave.up=CL.REFL bad.deeds

‘Those that gave up bad deeds are happy’ (CP1.1577: 53r) (44) Că ca iarba curând usucă-se (CP1.1577: 64r)

that like grass.DEF soon dries=CL.REFL

‘that it dries as quickly as grass’

3.4. The rise of first position clitics

Pronominal clitics begin to be infrequently attested in first position starting with the late 12th century in French (45) and Italian, and from the 13th century in Spanish (Adams 1987: 160, de Dardel and de Kok 1996: 54 for French; Benincà 1994 for Italian; Fischer 2002: 87 for French, Spanish and Italian). Portuguese is late in allowing first position clitics, from the 17th (and up to the 19th century, Martins 1995). French allowed first- position clitics till the end of the 16th century, when the realization of subject pronouns becomes obligatory (Fischer 2002: 87).

(45) Ba! me connissies vos? fait Aucassins. (old French) ah CL.ACC.1SG recognize you says Aucassins

‘Ah! Do you recognize me? Aucassins said’ (Auc.XXIV, 34, in Adams 1987: 161) In OR, first position clitics occur sporadically in the 16th century and more often in the first half of the 17th century (van Eeden 1985.1: 404). Our extensive corpus analysis revealed extremely few examples of first position clitics in both original (46)–(47) and translated 16th century texts (48). There is a higher number of occurrences of clitics in first

(12)

 

position in original texts (although they represent a much lower number of pages of text), which shows that their syntax is more innovative than that of translations.

(46) rog domniia-voastră să căutaţ

CL.REFL ask.PRES.1SG lordship.DEF=your SUBJ search.SUBJ.2PL

‘I ask your lordship to search [...]’ (DÎ.1599–600: XXV)

(47) S-a scris aceasta când s-a ferecat

CL.REFL=AUX.PERF written this when CL.REFL=AUX.PERF.3SG bound

‘This was written when it was bound’ (ITM.16thc.: 83)

(48) luo pâine întru sfânta a sa şi nevinovata mână dulce-har deade, take.PS.3SG bread in holy.DEF AL.F his and innocent hand grace give.PS.3SG

binecuvântă, o sfinţi, o frâmse bless.PS.3SG CL.ACC.3SG consecrate.PS.3SG CL.ACC.3SG break.PS.3SG

‘he took some bread in his holy and innocent hand, he gave us grace, he blessed it, he consecrated it, he broke it’ (CL.1570: 32v)

The number of first-position clitics increases in the 17th and 18th century texts (49)–(50), inversely proportional with the frequency of pronominal enclisis.

(49) minunez, cu adevărat (AIP.1705: 353)

CL.REFL.1SG wonder with true

‘I truly wonder’

(50) Te râd pe tine că eşti un

CL.ACC.2SG laugh.PRES.3PL DOM you.ACC that be.PRES.2SG a dobitoc (Bert.1774: 20r)

fool

‘They are laughing at you because you are a fool’

4. OLD ROMANIAN CLITICS ARE VERBAL (IP-ORIENTED) CLITICS Alboiu and Hill (2012) show that 17th – 18th century old Romanian clitics do not obey Wackernagel’s Law. Our 16th century corpus analysis confirms their conclusions and, furthermore, it brings evidence that, as early as the first OR documents, pronominal clitics took the inflectional phrase as their morphosyntactic host, and did not raise to the complementizer phrase.

4.1. The OR clitics’ position relative to săSUBJ and aINF

The first evidence for the fact that OR clitics are verb-oriented is provided by their position relative to the subjunctive marker să and to the infinitive marker a. In modern Romanian, să and a are analyzed as complementizers, more precisely as Finiteness heads (Ledgeway 2004, 2012: 170, Nicolae 2013, Hill 2013). SăSUBJ had the same status in OR, and the same can be argued for aINF, at least in those contexts in which it clearly no longer has a prepositional value, for instance, when the a-infinitive is clausal and occurs as a subject or as a direct object.

(13)

 

In the 16th century corpus, clitics are only attested at the right of săSUBJ (51)–(52) and of aINF (53)–(54),even if they are postverbally placed (52), (54), this proving that they do not leave the Inflectional domain.

(51) cum să se întoarne

how SUBJ CL.REFL.ACC return.SUBJ.3SG

‘how to return’ (PO.1582: 200)

(52) şi hitlenşug înnaintea ta feciu, ca să and slyness before your do.PS.1SG thatSUBJ

dereptezi-te întru cuvintele tale straighten.SUBJ=CL.REFL.2SG in words.DEF your

‘and I did this sly thing in front of you to make you talk justly’(CL.1570: 131) (53) Nu se cade împrejur a se tăia

not CL.REFL.3SG ought.PRES.3SG around AINF CL.REFL.3SG cut.INF limbile

people.PL.DEF

‘People ought to not be circumcized’ (CPr.1566: 217)

(54) Şi deca fu a descăra-se noao

and if be.PS.3SG AINF leave=CL.REFL.3SG us.DAT

‘And if it happened had we left’ (CB.1559–60: 168)

4.2. The OR clitics’ position relative to the sentential negation

Ever since the publication of Zanuttini’s (1997) study on negation it is generally accepted that, in Romance, sentential negation takes the Inflection Phrase as its complement, closing off the IP field. The consequence is that verbal clitics, being part of the Inflectional field, can only occur to the right of negation (Salvi 1993, Rivero 1997 for TM clitics). This is confirmed by our corpus analysis: in OR there are no attestations of the clitic raising above negation (55)–(57). Very rare examples such as (56), (57), which are only encountered in translations and possibly imitate the word order of the original text, show the verb raising above the pronominal clitic, but not above the negator nu ‘not’.

(55) Nu mă voiu rădica în veac (CP1.1577: 49r) not CL.REFL.ACC.1SG AUX.FUT.1SG raise.INF in age

‘I shall not raise in ages’

(56) ceia ce nu înşală-se de mine (CT.1560–1: 52) those that not cheat=CL.REFL.ACC.3PL by me.ACC

‘those that are not cheated by me’

(57) Derept aceaia nu spăimântămu-ne (CP1.1577: 85v) for that not get-scared.PRES.1PL=CL.REFL.ACC.1PL

‘for this reason we do not get scared’

(14)

 

4.3. Clitic – verb (non-)adjacency

In modern Romanian, like in the other Romance languages, the strict adjacency between the pronominal clitic and the verb is obligatory, as the clitic takes the verb as its support (Dobrovie-Sorin 1994, Avram 1999 a.o.). Adverbial clitics are the only constituents that can break up the clitic–verb complex (Îl (mai) / (*acum) văd ‘I can (still) see him (now)’).

Old Romance clitics can be non-adjacent to the verb (58) (see Wanner 1987, Benincà 1995 for old Romance; Fontana 1993 for old Spanish, Martins 1995 for old Portuguese, Dragomirescu 2013, 2014 for old Romanian). The possibility to insert a phrase between the clitic and the verb was analysed as a relic of Wackernagel’s Law (de Dardel and de Kok 1996: 129, for old French; Rivero 1997, for old Spanish).

(58) otro dia queles [este buen mandado] dixo Moysen (old Spanish) other day that= CL.DAT.3PL this good order told Moses

‘The day after Moses had given them directions’ (GE-I.216v, in Fontana 1997: 229) OR has attestations of clitic–verb non-adjacency such as (59)–(60), decreasing in frequency from the 16th to the 18th century, and occurring both in translated (59) and original texts (60). The occurrence of such word orders in original texts indicates that their presence is not accidental or due to the close imitation of the word order of the original text.

We argue that OR clitics that are non-adjacent to the verb are adverbal, and not Wackernagel clitics and we follow Dragomirescu (2013, 2014) in analysing non-adjacency as a consequence of scrambling.

(59) ce se [pururea] pomeneaşte (Prav.1581: 258r) what CL.REFL always remember.PRES.3SG

‘which is always remembered’

(60) numele nu [nice] povesteşte (CLM.1700–50: 166v) name.DEF not CL.REFL even tell.PRES.3SG

‘his name is not even told’

4.4. Explanations for the loss of TM clitics

The first accounts of the loss of Tobler-Mussafia clitics by Romance languages are phonological in nature (Meyer Lübke 1897, apud Benincà 1994). With reference to French, Adams (1987: 165, 198) puts forth the idea that a change in accentuation by which all words lost their original stress to the new oxytone stress led to the weakening of the rule of placing a stressed word at the head of a clause.

Frâncu (1980) considers that phonological factors are responsible for the rise of first position clitics and the generalization of procliticisation in OR. The author claims that proclisis arises as a consequence of the phonological strengthening of the clitic through prothetic î- (e.g. mi > îmi ‘me.DAT’) in the 16th century (see Alboiu and Hill 2012, Nicolae and Niculescu 2016 for arguments against this hypothesis).

We shall adopt a syntactic stand, according to which the weakening and eventual loss of the Tobler-Mussafia law falls out from the changes that took place in the domain of verb

(15)

 

movement in Romanian (and Romance more generally, cf. Poletto 2014 and references therein on the V2 grammar3 of certain old Romance varieties). More precisely, the restriction and specialization of V-to-C movement (very extensive in OR, cf. Dragomirescu 2013, 2014, Zafiu 2014) gradually gives raise to pronominal proclisis; in other words, the pronominal clitic is also not constrained to the second position any longer (see, for Romance, Adams 1987, Benincà 1995, Manzini 1994, Hirschbühler and Labelle 2000, Ledgeway 2008, 2012: 162). The loss of the V-to-C (and of V2) is caused by the rise of configurational structure in the Romance languages, in their evolution from Latin, which had a non-configurational system. The complementizer field of the Romance clause becomes further articulated, which leads to the possibility of projecting multiple discourse- related phrases in the Left Periphery. As more landing sites become available in the CP-field for information-structurally marked DPs, the verb starts to occupy the third, fourth etc. position of the clause. The ultimate shift taking place in the evolution of the Romance languages and Romanian (Cornilescu 2000) is the fact that the verb raises only as far as inflection in declarative clauses, which means that the pronominal clitics, which occupy a high position in the inflectional domain, preceding the MTA projections (Nicolae 2014), remain proclitic. Verb movement to the complementizer domain is triggered by the imperative and interrogative illocutionary forces (Labelle and Hirschbühler 2005, Ledgeway 2008, Ledgeway 2012: 179). As for Romanian, in the evolution from OR to MR finite verb movement to C is lost in most types of clauses, except for imperatives and conditional imprecations, where we still witness pronominal enclisis (for the possibility of proclisis with Romanian imperatives, see Niculescu 2015).

5. CONCLUSIONS

Ever since the 16th century, Romanian pronominal clitics are adverbal, even when non-adjacent to the verb; this means that Wackernagel’s law does not function in OR. The sentential negator nu and the Finiteness complementizers să and a always precede pronominal clitics, proving that they attach higher in the syntactic derivation than clitics.

Adverbal pronominal clitics are confined to the inflectional domain of the clause.

Old Romanian pronominal clitics do occur in second position, but not exclusively.

Third- or fourth-position clitics are also attested in the earliest OR texts, and, sporadically, first-position clitics; this indicates that the Tobler-Mussafia Law is no longer (fully) functional in 16th century old Romanian.

The main consequence of the fact that the Tobler-Mussafia Law (which, in its turn, is a verb-second effect) stopped being observed in OR is the loss of pronominal enclisis in relation to the morphosyntactic host of the clitic, namely the verb (more precisely, the verbal inflection) in declarative and interrogative clauses, a process that took place in the       

3 Poletto (2014: 33–34) briefly highlights the differences between the V2 grammar of old Romance varieties and Germanic (German, Dutch, mainland Scandinavian) V2, showing that in the former group of languages V2 is not a strict rule, but rather translates as V-to-C, with the verb occupying different positions in the C-field, not necessary a constraint on the occurrence of the verb in strict clause-second position.

(16)

 

evolution from OR to MR. In (standard) MR, enclisis is reduced to imperatives, conditional imprecations, and gerunds.

CORPUS

AIP.1705 Antim Ivireanul, Învăţătură pentru taina pocăinţii. Ed.: Antim Ivireanul, Opere, ed. G. Ştrempel, Bucureşti, Minerva, 1972, 347–351.

Bert.1774 Bertoldo. Ed. Magdalena Georgescu, Bucureşti, Minerva, 1999 (Cele mai vechi cărţi populare în literatura română, 3), 157–239.

CazV.1643 Varlaam, Cazania, ed. J. Byck, Bucureşti: Editura Academiei, [s.a.], 1–506.

CB.1559–60 Codicele popii Bratul. Ed. Al. Gafton: http://media.lit.uaic.ro/gafton.

CC2.1581 Coresi, Evanghelie cu învăţătură. Ed. S. Puşcariu, Al. Procopovici: Diaconul Coresi, Carte cu învăţătură (1581), vol. I, Textul, Bucureşti: Socec, 1914.

CCat.1560 Coresi, Catehism. Ed. Al. Roman-Moraru, in I. Gheţie (coord.), Texte româneşti din secolul al XVI-lea. I. Catehismul lui Coresi; II. Pravila lui Coresi; III.

Fragmentul Todorescu; IV. Glosele Bogdan; V. Prefeţe şi Epiloguri, Bucureşti:

Editura Acadmiei Române, 1982, 101–5.

CC1.1567 Coresi, Tâlcul Evangheliilor. Ed.: Coresi, Tâlcul evangheliilor şi molitvenic românesc, ed. V. Drimba, Bucureşti: Editura Academiei Române, 1998, 31–187.

CL.1570 Coresi, Liturghier. Ed. Al. Mareş, Bucureşti: Editura Academiei, 1969, 127–148.

CLM.1700–50 Miron Costin, Letopiseţul Ţărâi Moldovei. Ed.: M. Costin, Opere, ed.

P.P. Panaitescu, Bucureşti, Editura de Stat pentru Literatură şi Artă, 1958, 41–201.

CM.1567 Coresi, Molitvenic. Ed.: Coresi, Tâlcul evangheliilor şi molitvenic românesc, ed.

V. Drimba, Bucureşti: Editura Academiei Române, 1998, 189–211.

CP1.1577 Coresi, Psaltire slavo-română. Ed.: Coresi, Psaltirea slavo-română (1577) în comparaţie cu psaltirile coresiene din 1570 şi din 1589, ed. S Toma, Bucureşti:

Editura Academiei, 1976, 35–662.

CPr.1566 Coresi, Apostol. Ed. I. Bianu, Texte de limbă din secolul XVI, IV, Lucrul apostolesc tipărit de diaconul Coresi la 1563, Bucureşti, Cultura Naţională, 1930.

CT.1560–1 Coresi, Tetraevanghel. Ed.: Tetraevanghelul tipărit de Coresi. Braşov 1560 – 1561, comparat cu Evangheliarul lui Radu de la Măniceşti. 1574, ed.

F. Dimitrescu, Bucureşti: Editura Academiei, 1963.

Documente şi însemnări româneşti din secolul al XVI-lea, text stabilit şi indice de Gh. Chivu, M. Georgescu, M. Ioniţă, Al. Mareş, Al. Roman-Moraru, Bucureşti, Editura Academiei, 1979.

DPar.1683 Dosoftei, Parimiile preste an, 1683, ed. M. Ungureanu, Jassy: Editura Universităţii „Al. I. Cuza”, 2012, 95–356.

DRH.B.XXX Documenta Romaniae Historica. B. Ţara Românească, Bucharest: Editura Academiei Române, 1998: vol. XXX (1645).

DPV.1673 Dosoftei, Psaltirea în versuri. Ed.: Dosoftei, Opere, 1, Versuri, ed. N. A. Ursu, Iaşi, Mitropolia Moldovei şi a Sucevei, 1974, 3–1065.

ITM Însemnări pe de pe manuscrise şi cărţi vechi din Ţara Moldovei, ed. I. Caproşu and E. Chiaburu, Iaşi, Demiurg, 2008, vol. I (1429–1750), 130–582; vol. II (1751–1795), 5–325.

Mol.1689 Molităvnic. Ed. A. Dumitran, A.-M. Gherman, A. Vanca, Alba Iulia, Reîntregirea, 2009, 163–1075.

NL.~1750–66 Ion Neculce, Letopiseţul. Ed.: Ion Neculce, Letopiseţul Ţării Moldovei şi O samă de cuvinte, ed. I. Iordan, Bucureşti: Editura de Stat pentru Literatură şi Artă, ed. a II-a, 1959, 31–388.

(17)

 

PH.1500–10 Psaltirea Hurmuzaki, ed. I. Gheţie and M. Teodorescu, Bucureşti, Editura Academiei Române, 2005.

PO.1582 Palia de la Orăştie. Ed. V. Pamfil, Bucureşti, Editura Academiei, 1968.

Prav.1581 Pravila ritorului Lucaci. Ed. I. Rizescu, Bucureşti, Editura Academiei, 1971, 161–183.

REFERENCES

Adams, M., 1987, Old French, Null Subjects and Verb Second Phenomena, PhD dissertation, UCLA.

Alboiu, G., V. Hill, 2012, “Early Modern Romanian and Wackernagel’s Law”, SKY Journal of Linguistics, 25, 7–28.

Alboiu, G., V. Hill, I. Sitaridou 2014, “Discourse Driven V-to-C in Early Modern Romanian”, Natural Language & Linguistic Theory, DOI 10.1007/s11049-014-9270-8.

Avram, L., 1999, Auxiliaries and the Structure of Language, Bucureşti, Editura Universităţii din Bucureşti.

Benincà, P., 1994. “La sintassi dei clitici complemento nelle lingue romanze medievali”, in:

P. Benincà (ed.), La variazione sintattica, Bologna, Il Mulino, 213–245.

Benincà, P., 1995, “Complement Clitics in Medieval Romance: the Tobler-Mussafia Law”, in:

A. Battye, I. Roberts (eds), Clause Structure and Language Change, Oxford, Oxford University Press, 325–344.

Benincà, P., 2006, “A Detailed Map of The Left Periphery of Medieval Romance”, in: R. Zanuttini, H. Campos, E. Herberger, P. Portner (eds), Crosslinguistic Research in Syntax and Semantics: Negation, Tense, and Clausal Architecture, Washington, Georgetown University Press, 53–86.

Bošković, Ž., 1999, “Second Position Cliticization: Syntax and/or Phonology?”, in: F. Beukema, M. den Dikken (eds), Clitic Phenomena in European Languages, Amsterdam/Philadelphia:

John Benjamins, 71–120.

Bošković, Ž., 2001, On the Nature of the Syntax-Phonology Interface: Cliticization and Related Phenomena, Amsterdam, North-Holland.

Cardinaletti, A., I. Roberts, 2002, “Clause Structure and X-Second”, in: G. Cinque (ed.), Functional Structure in DP and IP: the Cartography of Syntactic Structures, volume I, Oxford, Oxford University Press, 123–166.

Cornilescu, A., 2000, “The Double Subject Construction in Romanian”, in: V. Motapanyane (ed.), Comparative Studies in Romanian Syntax, Amsterdam / Lausanne / New York / Oxford / Shannon / Singapore / Tokyo, Elsevier, 83–133.

de Dardel, R., A. de Kok, 1996, La position des pronoms régimes atones, personnels et adverbiaux, en protoroman: avec une considération spéciale de ses prolongements en français, Genève, Droz.

Dobrovie-Sorin, C., 1994, The Syntax of Romanian, Berlin, Mouton de Gruyter.

Dragomirescu, A., 2013, “O schimbare parametrică de la româna veche la româna modernă în sintaxa formelor verbale compuse cu auxiliar”, Limba română, LXII, 2, 225–239.

Dragomirescu, A., 2014. “The Syntax of Compound Tenses and Scrambling in Old Romanian”, paper presented at the SyntaxLab, University of Cambridge, February 2014.

Fischer, S., 2002, The Catalan Clitic System: A Diachronic Perspective on its Syntax and Phonology, Berlin / New York, Mouton de Gruyter.

Fischer, S., 2003, “Rethinking the Tobler-Mussafia Law”, Diachronica, 20, 259–288.

Fontana, J., 1997, “On the Integration of Second Position Phenomena”, in: A. van Kemenade, N. Vincent (eds), Parameters of Morphosyntactic Change, Cambridge, Cambridge University Press, 207–250.

(18)

 

Frâncu, C., 1980, “Formele verbale şi pronominale cu î- protetic în limba română”, Dacoromania, 5, 80–93.

Frâncu, C., 2009, Gramatica limbii române vechi (1521–1780), Iaşi, Demiurg.

Gheţie, I. (ed.), 1997, Istoria limbii române literare: epoca veche (1532–1780), Bucureşti, Editura Academiei Române.

Halpern, A., 1995, On the Placement and Morphology of Clitics, Stanford, Center for the Study of Language and Information.

Hill, V., 2013, “The Emergence of the Romanian Subjunctive”, The Linguistic Review, 30, 547–583.

Hill,V., G. Alboiu, 2016, Verb Movement and Clause Structure in Old Romanian, Oxford, Oxford University Press.

Hirschbühler, P., M. Labelle, 2000, “Evolving Tobler-Mussafia Effects in the Placement of French Clitics”, in: S. N. Dworkin, D. Wanner (eds), New Approaches to Old Problems. Issues in Romance Historical Linguistics, Amsterdam/Philadelphia, John Benjamins, 165–182.

Labelle, M., P. Hirschbühler, 2005, “Changes in Clausal Organization and the Position of Clitics in Old French”, in: M. Batllori, M.-L. Hernanz, C. Picallo, F. Roca (eds), Grammaticalization and Parametric Change, Oxford, Oxford University Press, 60–71.

Ledgeway, A., 2004, “Il sistema completivo dei dialetti meridionali: La doppia serie di complementatori”, Rivista italiana di dialettologia, 27, 89–147.

Ledgeway, A., 2007, “Old Neapolitan Word Order: Some Initial Observations”, in: A.L. Lepschy, A. Tosi (eds), Languages of Italy: Histories and dictionaries, Ravena, Longo, 119–146.

Ledgeway, A., 2008, “Satisfying V2 in Early Romance: Merge vs Move”, Journal of Linguistics, 44, 437–470.

Ledgeway, A., 2011, “Syntactic and Morphosyntactic Typology and Change”, in: M. Maiden, J.C. Smith, A. Ledgeway (eds), The Cambridge History of the Romance Languages, Cambridge, Cambridge University Press, 382–471.

Ledgeway, A., 2012, From Latin to Romance. Morphosyntactic Typology and Change, Oxford, Oxford University Press.

Manzini, M. R., 1994, “Triggers for Verb-Second: Germanic and Romance”, The Linguistic Review, 11, 299–314.

Martins, A. M., 1995, “Clitic Placement from Old to Modern European Portuguese”, in: H. Andersen (ed.), Historical Linguistics 1993: Selected Papers from the 11th International Conference on Historical Linguistics, Los Angeles, 16–20 August 1993, Amsterdam/Philadelphia, John Benjamins, 295–307.

Mussafia, A, 1886, “Una particolarità sintattica della lingua italiana dei primeri secoli”, in: G. I. Ascoli et al. (eds), Miscellanea di filologia e linguistica in memoria di Napoleone Caix e Ugo Angelo Canello, Florence, Le Monni, 255–261.

Nicolae, A., 2013, Types of Ellipsis in Romanian, PhD dissertation, University of Bucharest &

University of Cambridge.

Nicolae, A., 2014, “On Verb Movement in Romanian”, paper presented at the SyntaxLab, University of Cambridge, February 2014.

Nicolae, A., 2015, “Word Order and Configurationality”, in: G. Pană Dindelegan (ed.), The Syntax of Old Romanian, Oxford, Oxford University Press, in preparation.

Nicolae, A., D. Niculescu, 2015, “Pronominal Clitics. Clitic Ordering, Clitic Clusters”, in: G. Pană Dindelegan (ed.), The Syntax of Old Romanian, Oxford, Oxford University Press, in preparation.

Niculescu, D. 2015, “Clitics and verb movement in Romanian. Evidence from the imperative”, paper presented at The 17th Annual Conference of the English Department, 4–6 Jume, University of Bucharest.

Pancheva, R., 2005, “The Rise and Fall of Second-Position Clitics”, Natural Language & Linguistic Theory, 23, 103–167.

Poletto, C., 2014, Word Order in Old Italian, Oxford, Oxford University Press.

(19)

 

Reinheimer Rîpeanu, S., 2002, “Locul pronumelor neaccentuate din limba română veche în perspectivă romanică”, in: G. Pană Dindelegan (ed.), Actele Colocviului Catedrei de limba română 22–23 nov. 2001. Perspective actuale în studiul limbii române, Bucureşti, Editura Universităţii din Bucureşti, 199–212.

Rivero, M. L., 1991, “Clitic and NP Climbing in Old Spanish”, in: H. Campos, F. Martínez-Gil (eds), Current Studies in Spanish Linguistics, Washington, Georgetown University Press, 241–282.

Rivero, M. L., 1993, “Long Head Movement vs. V2, and Null Subjects in Old Romance”, Lingua, 89, 217–245.

Rivero, M. L., 1997, “On Two Locations for Complement Clitic Pronouns: Serbo-Croatian, Bulgarian and Old Spanish”, in: A. van Kemenade, N. Vincent (eds), Parameters of Morphosyntactic Change, Cambridge, Cambridge University Press, 170–206.

Salvi, G., 1993, “Difesa e illustrazione della legge di Wackernagel applicata alle lingue romanze antiche: La posizione delle forme pronominali clitiche”, Cad.Est.Ling Campinas, 24, 111–130.

Salvi, G., 2004, La formazione della struttura di frase romanza: Ordine delle parole e clitici dal latino alle lingue romanze antiche, Tübingen, Niemeyer.

Salvi, G., 2011, “Morphosyntactic Persistence”, in: M. Maiden, J.C. Smith, A. Ledgeway (eds), The Cambridge History of the Romance Languages, Cambridge, Cambridge University Press, 318–381.

TDR – Rusu, V., 1984, Tratat de dialectologie românească, Craiova, Scrisul Românesc.

Tobler, A., 1875, “Besprechung von J. Le Coultre, De l’ordre des mots dans Chrétien de Troye”, Vermische Beiträge zur französischen Grammatik, 5, 395–414.

Tomić, O. Mišeska, 2004, “The Balkan Sprachbund properties: An introduction”, in: O. Mišeska Tomić (ed.), Balkan Syntax and Semantics, Amsterdam, John Benjamins, 1-55.

van Eeden, W., 1985, Învăţături preste toate zilele (1642): Édition et étude linguistique, vol 1., Amsterdam, Rodopi.

Wackernagel, J., 1892, “Über ein Gesetz der indogermanischen Wortstellung”, Indogermanische Forschungen, 1, 333–435.

Wanner, D., 1987, The Development of Romance Clitic Pronouns: From Latin to Old Romance, Berlin, Mouton de Gruyter.

Zafiu, R., 2014, “Auxiliary Encliticization in 16th Century Romanian: Restrictions and Regularities”, Linguistica Atlantica, 33, 71–86.

Zanuttini, R. 1997, Negation and Clausal Structure: A Comparative Study of Romance Languages, New York, Oxford University Press.

Referințe

DOCUMENTE SIMILARE

The reason is roughly that under the high voltage, the electric field strength is very strong, the electromigration speed of the O 2- ion is very large, the O 2- ions in the

Its presence as a second-order desire (cf. For example, my will to finish this article is revealed by the fact that I stifled.. my concurrent urge to go to bed despite the

The comparison of the properties of WO 3 with respect to the used substrate material (FTO glass, quartz glass and graphite substrates) shows that there is a dependency on

Energy by Position Distribution function of electron beam energy (5-30 kV), corresponding to the multilayer structures having a total thickness of 38 nm (left) and.. 1220

Section 3 further examines the discourse-related role of dative clitic doubling: after closely examining the degree of optionality of clitic doubling in Romance (subsections

The same conclusion is suggested by example (13a) above, where in the sequence 1 st +2 nd person dative clitics+ 3 rd person accusative clitic, it is possible to have an ED (the

In this article we investigate duration patterns of internal (adjacent vowels within the word) and external hiatus (adjacent vowels across word boundary) in Romanian as a function

In particular, there exist a number of properties setting EDs aside from other HDs: EDs are ‘non-actantial’ datives, since they are not part of the valency of the verb but have