A Romanian word attested in the Xth century: cuvânt "Word, Council, Speech"

English abstract Romanian full text (for common public) Résumé français

A Romanian word attested in the X^th century: cuvânt "Word, Council, Speech"

First Roumanian texts date from the XV^th century. Some Roumanian words and names are known much earlier, from the XII^th century. There are historians of my language who believe that the words "torna, torna, fratre" by Theophilaktos Simokatta are the first testimony of spoken Roumanian, which is very doubtful. These are few reasons why finding a Roumanian word in the X^th century could be seen as an important discovery and my present article is ment to persuade you that this is indeed the case.

The source I am relying on is the Byzantine chronographer Leon Diakonos, whose "Histories" present the events taken place between 959-976, as they were lived by our author. Not only a contemporary of the facts, but also their eye witness, Leon Diakonos can be considered as a highly trustful source.

One of the episodes described in his work is that of the expedition made at the south of the Danube by Svyatoslav, the knjaz of Kijev, between 969-971. After a short series of victories of the Russians, the recent emperor Ioannes Tzimiskes starts a vigorous counter-offensive, reconquers the town of Preslav and succeeds to push Svyatoslav back to the Danube. The Knjaz, retired in the fortress of the town of Durostorum, still hopes to change the sort of the war, but the Byzantines get into the City and, after some bloody fights, finally defeat the Russians. The Knjaz has to decide if to fight the Greeks to the end or to cross the Danube back home with the remains of his army. He calls, in this purpose, his officers in command to a council. Let us see how does Leon Diakonos present this moment (XI, 7): “Τότε δὲ ἤδη διανισχούσης ἡμέρας βουλὴν ὁ Σφενδοσθλάβος τῶν ἀρίστων ἠκάθιζεν, ἥν καὶ κομέντον τῇ σφετέρῳ διαλέκτῳ φασί ν.“ [And then, when it was about dawn, Svyatoslav set the council of the nobles, which they call in their language also komentos (or komenton)].

The subject of my work is this very gloss κομέντον. The sentence I just quoted seems to show, clearly enough, that this word, on one hand, belonged to the language of those ἄριστοι - which are obviously the Russian military aristocracy, no need for other speculations - and, on the other hand, that its meaning was that of the Greek βουλή "council, gathering". This must be the reason why nobody doubted until now Leo's statement. But a closer look at the facts raises some interesting problems which my article will try, if not to solve, at least to bring them into the scholars' debate.

Let as notice first of all a logical vice of this sentence (or at least of its actual form). If the author wanted to offer to us the Russian correspondent of the Greek βουλή, he would simply have said ... βουλή...,ἥν κομέντον...φασί ν. By introducing the conjunction καί, the chronographer seems to let us understand that, beside komenton, there was a second term having the same meaning: council…they call alsokomenton (and how else?). This inadvertence may simply be a logical mistake of the author or maybe the text was truncated in one of the later copies.

We might also admit, and let's do so, that we have here a simple redundance with no other consequences and that the author's intention was indeed to inform us that in Russian the word κομέντον ment βουλή. In this case we have to deal with another error, this time etymological, because κομέντον is not to be found neither in Russian, nor in any other Slavic language. There is no word in Slavic not even close to this one in form and fitted in meaning. For this notion, Slavic languages use the common sŭborŭ (rus, pol., mac. sobor, bg. soborъ, cech soubor etc. ) or other regional terms: cech.,pol. rada, rus, bg., mac. sovet a.s.o., none of them comparable with komenton.

On the contrary, an almost perfect match can be found in the Latin conventus/conventum, which covers most of the meanings of the Greek βουλή. It is obvious that we can nomore speak about official Latin in Moesia in the X^th century, since it was abandonned for good after the death of Justinian. The only reasonable explanation is that the word in cause belonged to the spoken Danubian Latin, that is Roumanian. Conventus - probably through the military terminology - was already spread in all the east Latin speaking regions of the Haemus Paeninsula, that is the two Moesiae, Scythia Minor and Dardania, as proven by the modern languages of this area. The Albanian, a descendant of a Daco-Moesian dialect (after Vl. Georgiev) strongly influenced by Latin, got kuvënt (pl. kuvënde, verb kuvëndonj) "conversation, talk" and "to talk", borrowed afterwards in Greek as κουβεντιάζω "to speak, to talk". Also from Albanian, the Aroumanian (the main south-danubian Roumanian dialect) has cuvendă "word, speech". On the other side, all Roumanian dialects preserved, directly from the Latin conventus and conventum, the word cuvăntu (pl. cuvinte) "word, council, advice" with quite a large family (drom. cuvântez, arom. acuvintedzu "I speak, say", cuvântare "speech", cuvântător "(adj.) speaking (= not mute)" a.s.o. ). In the mind of ancient Balkanic people, the notion "gathering, council" became synonim with that of "speech, word", maybe under the influence of a substratum word having both meanings. This is why, while western Romance languages semantically moved the Latin "gathering" towards the "religious gathering" - and the word finally denominated the monastery, as in fr. couvent, it. convento, sp. conventual (surely under the permanent influence of the literary language) -, in the east this term ment both "gathering" and "speech, word". Thus, when Aroumanian borrowed the panslavic s(o)borŭ, which means only "gathering" in Slavic, it transferred upon this fresh term the old double meaning and so it created a new root: zbor "word", zburăscu "I speak", so successful that it was afterwards taken back by Slavic languages, as Macedonian and Bulgarian (zbor "word", zboruva "to speak") and borrowed also by the Greek: ζμπόρους "word".

I think that there are enough proves that this word is old and genuine in the Balkanic Paeninsula, and that there can be no doubt about its late Latin = Roumanian origin. In order to explain what exact stage of this evolution is witnessed by our chronographer in his κομέντον, we must first discuss some difficulties raised by the etymology of Roumanian cuvânt. Its direct etymon cannot be the class. lat. conuentus(-m), but a vulgar lat. *couentu. Though not attested, the existance of this vulgar form can be very well postulated if we take into account that there are other cases in which the prefix con- (com-) was reduced to co-. Indeed lat. con-venire > vulg.lat. *co-uenire > rom. a (se) cuveni and lat. com-pre(he)ndere > vulg.lat.*co-prendere > rom. cuprinde. Densuşianu (Hist.de la Lang.Roum., p.413-414) explains this phenomenon by the status of the accent: he says that there where the prefix bears no accent, as in com-pr(eh)éndo, it lost its nasal consonant, while otherwise, as in cóm-paro or cóm-pleo (>rom. cúmpăr, cúmplu), the nasal survived. To avoid entering the details of Densuşianu's explanation, which I shall discuss somewhere else, let's take note that this phenomenon is known to the histories of the Roumanian and, therefore, our reconstruction *co-uentu is fully legitimate. More than that, the French couvent (from *co-ventus, like couvrir from co-operire) seems to indicate that this tendency of reducing con- to co- was even earlier, before the separation of the Danubian Latin from the rest of Romania.

Once accepted the proposed etymon as legitimate, it remains to clear its phonetical problems.

In conformity to the phonetic laws which governed the apparition of the Roumanian, latin intervocalic -u- (-w-) disappeared. So, clavem > *cl'ae > cheie "key", pluuiam > ploia > ploaie "rain" a.s.o. The reason why *couentu preserved its intervocalic -u- is that it is a compound. As long as speakers felt this word as such, they treated the -u- as if it were in initial strong position (as in vânt < ventum, vin < uinum etc.). This is not a speculation of mine: Densuşianu (Hist.de la Lang.Roum., ib.) affirms it and supports it with other examples, such as primăvară (from prima+uera) or a (se) cuveni (from co+uenire). We must accept that this analogy functionned at least until consonantic -u- (= w, spelled at that time bilabial, most likely as b in modern Spanish) got its labio-dental innovative pronounciation: v, as in modern Roumanian (= English v).

Going further, the next question to be asked is if we can admit the persistence of the sound o until the X^th century, that is until after the Roumanian dialectal community broke away, given that all Roumanian dialects closed it to u - in unstressed position - and that we have examples of this closing early in the VI^th century. We must say that these examples, such as Γεμελλομούντις or Καλβομούντις, with –muntis instead of -montis, testify the mutation of o to u in nasal context only. The numerous hesitations and inconsequencies of this phonetic change, as well as the frequent restorations met especially in the Daco-Roumanian dialect, make me admit, together with ILR II (p. 198) that "there has always been a variation between o and u". I am persuaded that even long after the two main dialects split, the mutation of o to u remained only a tendency in progress and therefore the pronounciation o can be accepted as normal in the X^th century at Durostorum.

The following e in komenton (which corresponds to the rom. â in cuvânt) has a similar explanation. Normally, e in nasal position closes to i, as in bine < bene, linte < lentem or vintre < ventrem. But, though there are several causes which concurred in this case to preserve the old phonetism, it is enough to mention here only that precisely this form, with e, cuvente (and also mente, cene) instead of cuvinte (minte, cine), can be found in much later texts, even from the XVI^th century.

The lines above tried to demonatrate that even after Roumanian dialects split, a form *couentu (or even*couăntu) was possible. In this case, the gloss κομέντον would be its very transcription (rather approximate, I agree). I cannot know now why we find an μ instead of the expected β (the normal Greek transcription of *couentu would be at this time *κοβέντον). Maybe this is the way Leon Diakonos heard the word (especially if the nasal wasn't completely disappeared at that time and the word was in fact pronounced *cõuentu) or maybe letters were changed in a later copy, since the graphic confusion μ/β is not at all rare in Greek manuscripts. Whatever the cause might be, I think it cannot be invoked as a serious argument against my hypothesis.

The spot where the Byzantine historian places the action comes to support my conclusions. Durostorum, a celtic settlement as its name shows, quickly became one of the most important centers of the Danubian Romanity. It was already so in the first century BC: here was the castrum of one of the two (later three) legions on the lower course of the river, Marcus Aurelius made it a municipium, and from the VI^th century we constantly find it on the lists of bishop sees in Moesia Inferior. Its people was, from the Roman conquest of Moesia until today, mostly Romanic. Its Roumanian name, Dârstor, inherited directly from Durostorum, was borrowed by the Greeks in the form Dristra and only from here the Bulgarians got their Silistra, the modern official name of the town where Roumanians are still majoritar.

We must now clear a last problem: if indeed κομέντον transcribes the rom. *couentu, then how do we explain its attribution to the Russian? I must say from the very beginning that I don't think we could use this gloss as an argument for the presence of the Roumanian element among those ἄριστοι mentioned by the chronographer. It is very possible that Svyatoslav have had north-Danubian Slavo-Roumanian formations as allies, it is also very possible that local population was on his side along the hostilities, but our gloss remain silent about these speculations. In my opinion, there is another, more simple and more natural explanation for this. Reading his "Histories" one could easily realize that Leon Diakonos saw with his own eyes the theater of the events, and therefore he must have visited those places. So it seems to me very possible that he found out the details of the episode of the council from local people: a habitant of Durostorum, obviously a Roumanian, told him the story and also mentioned *couentu as the equivalent in his own language of the Greek βουλή. Twenty years later, when he wrote down his work (in 992), the historian found this word in his notes and, taking it for Russian, considered it interesting enough to be quoted.

In the end, compressing the contents of this article in a final phrase, I am persuaded that since the word κομέντον (=βουλή) cannot be Russian, nor even Slavic, since it is almost a perfect correspondent of lat. conventum and taking into account the place of the events, it can be nothing else but a stage of the history of the Roumanian word cuvânt, a stage that I reconstructed as *couentu.

Bibliography

Densuşianu HistLR = Ovid Densuşianu, Histoire de la langue roumaine, Ed. “Grai şi Suflet-Cultura Naţională”, Bucureşti, 1997

ILR II = Istoria Limbii Române[The History of the Roumanian Language], Ed. Academiei, vol II, Bucureşti 1969.

DLR = Dicţionarul Limbii Române [The Dictionary of the Roumanian Language] edited by the Roumanian Academy Română

Notes

1. Γεμελλομούντις by Procopius Caesariensis, De Aedificiis, IV p.148 row 14 (Ed. Haury) and Καλβομούντις byTheophylactos Simokatta 2, 15, 3.

2. cf. Densuşianu HistLR p. 426 şi p. 434.

English abstract Romanian full text (for common public) Résumé français