RRL, LXV, 2, p. 135–144, Bucureşti, 2020
A VENI ‘COME’ AS A COPULATIVE VERB IN ROMANIAN
1ADNANA BOIOC APINTEI2, ŞTEFANIA COSTEA3
Abstract: Most of the attention in the literature has been drawn to the grammaticalization of the verbs of movement as temporal and aspectual markers in the Romance languages (see Heine and Kuteva 2002); however, Romanian and Italian seem to hold a third value for this kind of verbs (e.g., for a veni ‘come’): the copulative one (van Peteghem 1991; Dragomirescu 2012; Dragomirescu and Nicolae 2014). What rested unnoticed was the mechanism that triggers the copulative use in Romanian. Up to our preliminary observations, a veni ‘come’ as a copula is chosen when a third participant is implied in the process. Thus, this article aims to offer (i) a descriptive account of the contexts in which a veni ‘come’ is selected (on the basis of questionnaires given to native speakers, and of corpora study, i.e. old Romanian texts and dialectal texts), and (ii) a new insight into the syntactic mechanism that triggers the copulative use of a veni ‘come’.
Keywords: verbs of movement, copulative verbs, grammaticalization.
1. INTRODUCTION
In this paper, we will focus on the use of a veni ‘come’ as a copulative verb in Romanian; this research is part of a larger project. The main objectives of the project are: to establish the full inventory of motion verbs which have undergone this type of reanalysis (e.g., common paths of grammaticalization, such as [motion > tense/aspect]; rarer paths of grammaticalization, such as motion > copula] and [motion > passive]); to analyse the processes by which different motion verbs grammaticalized as aspectual and inceptive verbs (cf. a (se) apuca de ‘catch > begin, be about to’; a se opri din ‘stop (somewhere) >
cease’; a se porni pe ‘depart > start’; a prinde a/să ‘catch > start’; a se pune pe ‘sit > start’;
a urma ‘follow > be about to’; a sta să ‘sit, lie > be about to’ a-i veni să ‘come > feel like’;
1 This work was supported by a grant of the Ministery of Research and Innovation, CNCS- UEFISCDI, project number PN-III-P1-1.1-TE-2016-0341, within PNCDI III.
2 “Iorgu Iordan – Alexandru Rosetti” Institute of Linguistics of the Romanian Academy/
University of Bucharest, Bucharest, [email protected].
3 “Iorgu Iordan – Alexandru Rosetti” Institute of Linguistics of the Romanian Academy/
University of Cambridge, [email protected].
see Guţu Romalo 1961, GALR I: 457–459), copula verbs (cf. a ajunge ‘arrive > become’;
a ieşi ‘exit > become’; a se prinde ‘catch > become’; a rămâne ‘stay > remain (in a state)’;
a trece de ‘pass > be considered’; a se ţine ‘hold, follow > be’; a veni ‘come > be’; see GBLR: 479f.), and passive auxiliaries (cf. a se afla ‘be found > be’; a veni ‘come > be’; see Iordan 1950; GALR II: 136f.; Dragomirescu şi Nicolae 2014); to identify switch contexts that have favoured the reanalysis from motion to aspect/inceptive, copula or passive.
2. MOTION VERBS: THEORETICAL BACKGROUND
The reference literature has mostly focused on the grammaticalization of verbs of movement as temporal and aspectual markers in the Romance languages (cf. Heine and Kuteva 2002), this being a common path, given the space–time transfer (Bybee, Perkins, Pagliuka 1994:269, Stolova 2005). But Romanian presents paths of grammaticalization that have not been studied so far, such as the [motion verb > copula] path of grammaticalization (not mentioned by Heine and Kuteva (2002)), and the [motion > passive] path of grammaticalization (mentioned within the Romance context only for the Italian verbs venire ‘come’ and andare ‘go’).
These changes within the verbal domain, i.e. passing from a fully lexical verb to a (more) grammatical verb, can be explained through the process of grammaticalization, which implies four main mechanisms (i.e., desemanticization (loss of meaning content), extension (the use of the element in a new context), decategorization (loss of morphosyntactic properties characterizing the source form), and finally erosion (loss in phonetic substance)). As noticed by Roberts and Roussou (2003: 20ff.), the grammaticalization generally implies an upward movement on the clausal spine, from the lexical area to the functional domain (and subsequent direct merger in the functional domain). Apart from grammaticalization, reanalysis plays a central role, inasmuch as it implies changes within the underlying structure of a syntactic pattern (with consequences in the semantics of the relevant verb), without involving any change in its surface manifestation (cf. Harris and Campbell 1995:50). This process accounts for the changes from motion verbs to aspectual auxiliaries, as well as from motion verbs to copula verbs.
All in all, it must be noted that the loss of the thematic structure of the verbs undergoing the change is the only feature that is common for all three processes mentioned earlier (cf.
Roberts 2013).
As for the first path mentioned above, i.e. [motion verb > copula], copulas originating in motion verbs are attested in other (Romance) languages (Van Peteghem 1991:158f.) (cf. (1) below), although they have not been analysed from the perspective of their grammaticalization.
(1) Fr. tomber, retomber, passer, rester, demeurer, apparaître, demeurer Sp. tornarse, mantenerse, quedar, mostrarse
It. tornare, venire, restare, rimanere
The prima facie impression is that the system of modern standard Romanian copulas is simple and symmetric, i.e. there is a neutral copula (a fi ‘be’); an inchoative copula (a deveni ‘become’ – attested only starting with the 18th century); a continuative copula
(a rămâne ‘be, remain’); and a terminative one (a ajunge ‘become’). In short, we have:
verbs originally expressing motion/change of location (cf. a ajunge ‘arrive > become’;
a ieşi ‘exit > become’; a trece de/drept ‘pass > be considered’; a veni ‘come > be’); verbs originally expressing location (cf. a rămâne ‘stay > remain (in a state)’); and verbs originally expressing a type of movement without motion/change of location (cf. a se prinde ‘catch > become’; a se ţine ‘hold, follow > be’).
However, Dragomirescu (2016: 151–163) convincingly argues that the system of both old and modern spoken Romanian copulas is, in fact, more complicated, allowing for broad synonymy, with minor semantic differences; nevertheless, it must be noted that many of the values expressed by different copulas in old and dialectal Romanian were taken over in standard Romanian by the pan-Romanic verb a deveni, which entered the language in the 18th century. Thus, there is a series of verbs that were preserved in the passing from old to modern Romanian, e.g. a rămâne ‘remain’; a ajunge ‘arrive’; a ieşi ‘exit’: a se prinde
‘catch’; a se ţine ‘hold’; a veni ‘come’. Of these, only a ajunge and a rămâne have been preserved as copulas in the present-day standard language, with a ieşi, a se prinde, a veni, and a se ţine surviving only in the colloquial language. Finally, there are also modern Romanian verbs that have lost their copula nature (typical of old Romanian); cf. a se afla
‘be placed’; a sta ‘stay’; a intra ‘enter’; a purcede ‘proceed’; a se ridica ‘raise’; a sosi
‘arrive’; a se aşeza ‘settle’; a se pune ‘sit down’.
The motion verbs grammaticalized as copulas (such as a veni ‘come’, a ajunge
‘arrive’, a sosi ‘arrive’) preserve to some extent the deictic meaning related to motion. That is to say, unlike the neuter copula BE, they add an inchoative information to the predicate (cf. a se prinde, a veni, a purcede, a intra); a continuative information to the predicate (cf. a rămâne, a se ţine, a se afla, a sta); or a terminative information to the predicate, i.e.
which introduces a property as a final stage of a dynamic process (cf. a ajunge, a ieşi, a sosi).
Another striking feature of Romanian copulas derived from motion verbs concerns their status; that is to say, they are not auxiliaries (cf. Italian), and their grammaticalization process does not involve a double verbal construction (as it is the case for tense, aspect and passive auxiliaries). Instead, the context favouring grammaticalization is usually represented by the verb followed by a kind-denoting or property-denoting noun or by an adjective. Hence, the [motion > copula] path certainly involves a reanalysis process, but it cannot be described along the lines of Heine’s (1993) stages of grammaticalization (with the notable exception that all motion verbs grammaticalized as aspectual verbs, passive auxiliaries and copulas are generally unable to assign theta-roles), inasmuch as the complement they take is not a subordinate or a non-finite clause.
3. CASE STUDY: A VENI ‘COME’
3.1. Methodology
In order to correctly assess the status of the selected verb (employing both a diachronic and a synchronic perspective), we used the old Romanian corpus made for The Syntax of Old Romanian (Oxford, 2016); the academic dictionary of Romanian (DA/DLR);
a dialectal corpus survey; an online questionnaire, where we asked native speakers to provide grammaticality judgements about utterances containing a veni ‘come’ as a copula (these examples will be marked with a ‘Q’).
The above mentioned questionnaire was designed to identify different usages of a veni ‘come’, e.g. in structures such as îmi vine să plâng (CL.DAT.1SG comes SĂ.SUBJ
cry.SUBJ.PRES.1SG) ‘I feel like crying’; cartea vine aşezată pe masă (book.the comes put.PPLE on table) ‘the book needs to be put on the table’; Maria îmi vine cumnată (Maria
CL.DAT.1SG comes sister-in-law) ‘Maria is my sister-in-law’. The questions asked were formulated in such a manner, that native speakers were able to express grammaticality judgements for specific contexts, e.g. Let us suppose that you’ve recently got married, and your significant other has a brother – Mircea. Would you naturally say ‘Mircea îmi vine cumnat’ (Mircea CL.DAT.1SG comes brother-in-law ‘Mircea is my cousin-in-law’)?; In your day-to-day life, would you use the following expression: ‘Ana îmi vine mamă.’ (Ana
CL.DAT.1SG comes mother ‘Ana is my mother’)?.
3.2. Literature overview
Dragomirescu and Nicolae (2014) observed that there are multiple grammaticalization processes in Romanian regarding the verb a veni ‘come’.
Briefly, such process concerns the passing from expressing a change of location to expressing a change of state, conceptualized as inception (i.e. the aspectual value).
Presumably, the first step consisted in the (old Romanian) nominal phrase being interpreted as a complement (cf. (2) below). The following step consisted in the [nominal phrase + subjunctive] being interpreted as a complement (cf. (3) below). Finally, the last stage (which is to be found in present-day Romanian) consists in the nominal being dropped, leaving a veni ‘come’ to be followed by a subjunctive complement (cf. (4) below).
(2) De greşaşte omul, nu-i vine luiş [...] foame if makes.mistakes man.DEFNEG=CL.DAT.3SG comes he.DAT hunger
‘If humans make mistakes they don’t get hungry (as they would normally do)’
(old Romanian, CC2.1581, in DLR, s.v. veni) (3) Dacă-i vine aşa o nebuneală sergentului
if=CL.DAT.3SG comes such a craziness sergeant.DAT
să spuie că el e stăpânul averii…
să.SUBJ say that he is owner.DEF wealth.GEN
‘If the sergeant starts acting crazy by saying that he is the owner of the wealth…’
(modern Romanian, Caragiale, in DLR, s.v. veni)
(4) Îmi vine să plâng.
CL.DAT.1SG comes SĂ.SUBJ cry ‘I feel like crying.’
(present-day Romanian) Another process is related to the grammaticalization of a veni ‘come’ as a passive auxiliary in Romanian (with the verb raising from the lexical domain to the functional domain, precisely in the VoiceP projection; cf. (5) and (6) below). In this case, the switch context for reanalysis is [a veni ‘come’ + past participle], while the root modal meaning is most probably derived from its iterative/habitual or generic meaning. In contrast to the regular BE-passive, which is static, the a veni ‘come’-passive is dynamic and, in contrast to the reflexive passive, the a veni ‘come’-passive contributes a stronger deontic or iterative
value to the verbal event. In the imperfect, the passive auxiliary a veni ‘come’ yields a habitual, dispositional reading of the verbal event (deontic and imperfective are incompatible).
(5) Blagoslovit vine, în numele Domnului, împăratul izraililor
blessed comes in name God emperor Israeli
‘The Israeli’s emperor comes and is blessed in the name of God’ / ‘Blessed is in the name of God the Israeli’s emperor’ (old Romanian, CC2.1581) (6) Casa aceea vine aşezată aici.
house that comes placed here
‘The house is placed / will be placed / should be placed here’
(present-day Romanian) The final value of a veni ‘come’ is the copulative one. In this particular case, we have a transfer from [change of location] to [marker of indirect kinship relation] (cf. (7) below).
(7) Ion îmi vine cumnat.
Ion CL.DAT.1SG comes brother-in-law
‘Ion is my brother-in-law.’ (present-day Romanian)
In the old language, we identify a recurring combination of a veni ‘come’ with bare, person-denoting nouns, which have a kind-level denotation (cf. (8) below). Later, in the 20th century, a veni ‘come’ started to be followed by kinship nouns denoting an indirect kinship relation (e.g., cumnat ‘brother-in-law’, socru ‘father-in-law’, cuscru ‘father of a son- /daughter-in-law’, văr ‘cousin’, etc.). Nevertheless, a rudiment of the initial motion semantics has been preserved by a veni ‘come’ in this structure, inasmuch as nouns denoting direct kinship cannot be used in this structure.
(8) Mircea-vodă iar au venit domn
Mircea-vodă again AUX.PERF.3SG come.PPLE king al treilea rând.
for.the.third.time
‘Mircea-vodă (be)came again king for the third time.’
(old Romanian, AC.1650-90) (9) Îmi vine nepot / *tată.
CL.DAT.1SG comes nephew/ father
‘He is my nephew / *father.’ (present-day Romanian)
3.3. A veni ‘come’ as a copulative verb in Romanian. More than one story?
Although examples (8) and (9) above are superficially similar, we argue that the use of a veni ‘come’ is triggered by different semantic mechanisms when it comes to indirect kinship and kind-level terms.
3.3.1. Kind-level terms: old and present-day Romanian
For a better diachronic understanding of the phenomenon, it is important to mention that old Romanian only shows contexts in which a veni ‘come’ precedes person-denoting nouns, e.g. domn ‘king’ (cf. (10) below).
(10) a. Dacă au auzit boerimea că if AUX.PERF.3SG heard.PPLE boyars.the that vine măria sa domn, cei mai mulţi
comeshighness=his king most.of.them
n-au mai aşteptat venirea
NEG=AUX.PERF.3PL more wait.PPLE arrival
‘When the boyars found out his highness would become king, most of them did not wait for his arrival’ (old Romanian, AC.1650-90)
b. pentru dânşii am venit la domnie
for them AUX.PERF.1SG come.PPLE to reign ca aceasta
like that
‘For them I came to be a king like that’ (old Romanian, VRC.1645)
c. au venit domn în ţară Alexandru
AUX.PERF.3SG come.PPLE king in country Alexandru vodă, feciorul Radului vodă celui Mare
the.king son Radu.GEN king the great
‘Alexandru, the son of Radu the Great, became king’
(old Romanian, CLM.1700-50) d. lăsă boiari să păzească scaunul,
let boyars SĂ.SUBJ defend throne.the
până le va veni alt domnu
until CL.ACC.3PL will come another king de la împărăţie
from kingdom
‘(He) left the boyars to defend the throne until another king is to be sent from
the kingdom’ (old Romanian, ULM.1725)
e. după ce au venit samoderjeţii
afterwards that AUX.PERF.3PL come.PPLE autocrats.the împăraţ
kings
‘After the coronation of autocrats’ (old Romanian, CIst.1700–50)
f. Să să ştie de când au
SĂ.SUBJ CL.REFL.3SG know since.when AUX.PERF.3SG
vinit Petre împărat în Moldova
come.PPLE Petre emperor in Moldova
‘Let it be known as a sign of Petre becoming the emperor of Moldova’
(old Romanian, ITM.1710-1)
g. şi a venit domn nou în scaun
and AUX.PERF.3SG come.PPLE king new in throne Neculae voievod
Neculae voivode
‘And Neculae voivode became the new king’ (old Romanian, ITM.1711) It seems plausible to argue for a use of a veni ‘come’ specialized for expressing repeated changes, in the sense that public dignities terms preceded by a veni ‘come’ are
always associated with the idea of a limited period of time (i.e., kings usually did not lead the country their whole life; rather, they had short – sometimes multiple – periods in which they held this title). The idea of a continuous change at the top of the country’s hierarchy (which is by itself associated with movement) eventually triggered a semantic reinterpretation [change of location] > [marker of (short) time-limited functions] (i.e., the king-to-be came to the throne from outside the top hierarchy).
This observation seems to hold even in present-day Romanian, as native speakers that completed our questionnaire rejected examples as in (11) below as ungrammatical, whereas utterances as in (12) were generally accepted by our informants.
(11) *Ana vine asistentă medicală / profesor universitar /
Ana comes nurse professor university
inginer / farmacista mea.
Engineer pharmacist my
‘Ana is a nurse / professor / engineer / my pharmacist.’
(Q.; present-day Romanian) (12) a. Dacă vine iarăşi Ana director executiv, e de rău.
if comes again Ana director executive is of bad
‘If Ana is to be again the executive director, it will be bad.’
(Q.; present-day Romanian) b. Moscviciov vine primar la Sangeorgiu de Mureş.
Moscviciov comes mayor to Sangeorgiu.de.Mureş
‘Moscviciov will come mayor to Sangeorgiu de Mureş.’
(present-day Romanian) The explanation lies, we argue, in the fact that, once Ana acquires a title such as a nurse, professor, or engineer (cf. (11) above), it will not be taken away from her. Thus, these are not (short) time-limited functions. On the other hand, in the case of utterances as in (12), the presented functions are temporary, in the sense that both executive director and mayor imply mandates.
An interesting tendency revealed by our online questionnaire regards the preference of native speakers to employ a veni ‘come’ (to the detriment of copulative BE) when emphasising the repetition of a certain event, i.e. when a person occupies a time-limited function more than one time (cf. (13) and (14) below).
(13) Ne era teamă că va veni
CL.ACC.1PL be.IMPF.3SG fear that AUX.FUT.3SG come tot el director.
again he director
‘We were afraid that he would become our director again.’
(Q.; present-day Romanian)
(14) I-am ameninţat că o să venim noi
CL.ACC.3SG=AUX.PERF.1PL threat that will.come we consilieri în mandatul următor.
counsellors in mandate next
‘We threatened them we would be counsellors next mandate.’
(Q.; present-day Romanian)
Last but not least, what examples like (12)–(14) above show is that, when preceding short-term functions, a veni ‘come’ may express different tenses, e.g. indicative present in (12), indicative future in (13), and may take a variety of (singular and plural) subjects.
3.3.2. Indirect kinship terms: old and present-day Romanian
Now, the second value of a veni ‘come’ that interests us regards its usage to express indirect kinship relations (not attested in the old Romanian texts we have consulted), e.g.
mamă vitregă ‘stepmother’, cumnat ‘brother-in-law’ (cf. (15) below). Interestingly, our questionnaire revealed a strong tendency of native speakers to use the third person singular form of the present indicative, irrespective of the temporal value of the larger contexts, for this specific usage of the verb.
(15) a. Mircea îmi vine cumnat.
Mircea CL.DAT.1SG comes brother-in-law
‘Mircea is my brother-in-law.’
(Q.; present-day Romanian) b. Dacă tu eşti verişoară bună cu acea
if you are cousin good with that
fată, [prietenul tău] îi vine cumnat girl boyfriend your CL.DAT.3SG come brother-in-law după verişoară.
after cousin
‘If you are that girl’s cousin, [your boyfriend] is her brother-in-law.’
(present-day Romanian)
Nevertheless, utterances in which a veni ‘come’ takes a different subject (while still having its present indicative form) are still to be found in present-day Romanian, albeit rarer (cf. (16) below).
(16) Acum, eu îi vin mătuşă Mariei.
now I CL.DAT.3SG come aunt Maria.GEN
‘Now, I am Maria’s aunt.’
(Q.; present-day Romanian) For this specific usage, we argue that a somewhat different reinterpretation comparing to the one presented in §3.3.1. above is at play (albeit in modern stages of Romanian). That is to say, we take contexts as in (17) below, where it is clear that we were a family when the baby came (i.e. (s)he was the one who ‘entered’/ ‘came into’ the family), to be triggering the relevant copulative value.
(17) Copilul a venit când ni-l
baby.DEF AUX.PERF.3SG come.PPLE when CL.DAT.1PL=CL.ACC.M.3SG
doream cel mai mult.
wish.IMPF.1PL the.most
‘We had the baby when we most wanted it.’ (Q.; present-day Romanian)
In time, the verb started to precede indirect kinship relations, whereby a person from outside the family comes into the family through an actual member (cf. (18) below, where Ana becomes my stepmother by marrying my father (an actual member of my family); thus, she comes into our family through my father).
(18) Ana îmi vine mamă vitregă.
Ana CL.DAT.1SG comes stepmother
‘Ana is my stepmother.’ (Q.; present-day Romanian) It should be noted, however, that here the roles played in (15), (16), and (18) above are not short-term ones but (somewhat) lifetime ones (cf. the discussion in §3.3.1., whereby we argue lifelong (official) functions cannot be preceded by a veni ‘come’). Thus, what our questionnaire shed light on is that contexts containing indirect kinship terms preceded by a veni ‘come’ do not completely pattern with their kind-level counterparts, though some similarities are still to be found. For example, a veni ‘come’ preceding both kind-level and indirect kinship terms generally implies a person (an ‘outsider’) coming into a certain group, be it an official one or a family.
4. CONCLUSIONS
Despite highlighting a series of semantic differences between kind-level and indirect kinship terms felicitously preceded by a veni ‘come’, such as the temporal limitation compulsory for the former, and the temporal illimitation required by the latter, our study showed that copulative usages of this specific verb should always derive from the idea of movement from the outside (the hierarchy, the family, etc.) to the inside.
Importantly, a comparative view of the contexts employing kind-level terms from both old and present-day Romanian, and of context employing indirect kinship terms (only) from present-day Romanian allows us to observe a paradox of grammaticalization processes. That is to say, when preceding time-limited functions (i.e., kind-level terms), the stage reached is extension (i.e., the motion verb is used in a new context, losing its basic motion semantics), while, when preceding indirect kinship terms, the stage reached is decategorization (i.e., the motion verb is used in a new context, its basic semantic being lost and it has started losing the morphosyntactic properties of the source form).
Unexpectedly, although the second use is more recent, it has reached a higher grammaticalizaton stage.
CORPUS
AC.1650-90 = Cronicari munteni, ediţie îngrijită de Mihail Gregorian, studiu introductiv de Eugen Stănescu, I, Stolnicul Constantin Cantacuzino, Anonimul Cantacuzinesc, Radu Popescu, Bucharest, Editura pentru Literatură, 1961.
CC2.1581 = Coresi, Evanghelie cu învăţătură, ed. S. Puşcariu, Al. Procopovici: Diaconul Coresi, Carte cu învăţătură (1581), vol. I, Textul, Bucharest, Socec, 1914.
CIst.1700–50 = Constantin Cantacuzino, Istoria Ţării Româneşti, Ed.: Istoria Ţărâi Rumâneşti atribuită stolnicului Constantin Cantacuzino, ed. O. Dragomir, Bucharest, Editura Academiei Române, 2006.
CLM.1700–50 = Miron Costin, Letopiseţul Ţărâi Moldovei, ed. M. Costin, Opere, ed. P. P. Panaitescu, Bucharest, Editura de Stat pentru Literatură şi Artă, 1958.
DA = Dicţionarul limbii române, tomurile I-II, Bucharest, 1913–1948.
DLR = Dicţionarul limbii române, serie noua, tomul XIII, partea I, litera V, V-veni, Bucharest, Editura Academiei Române, 1997.
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.
ULM.~1725 = Grigore Ureche, Letopiseţul Ţării Moldovei, ed. P. P. Panaitescu, Bucharest, Editura de Stat pentru Literatură şi Artă, 1955, 57–210.
VRC.1645 = Varlaam, Răspunsul împotriva catihismusului calvinesc, ed. Varlaam, Opere, Răspunsul împotriva catihismusului calvinesc, ed. M. Teodorescu, Bucharest, Minerva, 1984, 143–230.
The electronic database of Old Romanian texts of ‘Iorgu Iordan - Al. Rosetti’ Institute of Linguistics, Bucharest.
REFERENCES
Bybee, J., R. Perkins, W. Pagliuka, 1994, The Evolution of Grammar. Tense, Aspect, and Modality in the Languages of the World, Chicago, London, The University of Chicago Press.
Dragomirescu, A., 2016, “The subject(ive) predicative complement”, in SOR: 151–163.
Dragomirescu, A., A. Nicolae, 2014, “The multiple grammaticalization of Romanian veni ‘come’.
Focusing of the passive construction”, in: M. Devos, J. van der Wal (eds), Come and go off the beaten of grammaticalization path, Berlin, Mouton de Gruyter, 69–100.
GALR I, II – Gramatica limbii române, 2008, ed. V. Guţu Romalo, Bucureşti, Editura Academiei Române.
GBLR – Gramatica de bază a limbii române, ed. G. Pană Dindelegan, Bucureşti, Univers Enciclopedic.
Guţu Romalo, V., 1961, “Semiauxiliare de aspect?”, Limba română, X, 1, 3–15.
Harris, A. C., L. Campbell, 1995, Historical Syntax in Cross-Linguistic Perspective, Cambridge, Cambridge University Press.
Heine, B., 1993, Auxiliaries. Cognitive Forces and Grammaticalization, New York/Oxford, Oxford University Press.
Heine, B., T. Kuteva, 2002, World Lexicon of Grammaticalization, Cambridge, Cambridge University Press.
Iordan, I., 1950, ‘Note sintactice’, Studii şi cercetări lingvistice, I, 2, 269–279.
Roberts, I., 2013, ‘Some speculations on the development of the Romance periphrastic perfect’, Revue roumaine de linguistique, LVIII, 1, 3–30.
Roberts, I., A. Roussou, 2003, Syntactic Change, Cambridge, Cambridge University Press.
SOR – G. Pană Dindelegan (ed.), The Syntax of Old Romanian, Oxford, Oxford University Press.
Stolova, N.I., 2005, “Expressing Time trought Space: A Comparative Diachronic Perspective on Romance Periphrastic Past and Future”, in: M. Coene, L. Tasmowski (eds), On Space and Time in Language, Cluj-Napoca, Clusium, 193−207.
Van Peteghem, M., 1991, Les phrases copulatives dans les langues romanes, Wilhelmsfeld, Gottfried Egert.
