The Old Serbian Version of the Antiochene Recension of Samuel-Kings: Some Preliminary Issues in Textual Criticism

Alessandro Maria Bruni La versione serba antica della recensione antiochena dei libri di Samuele e dei Re: alcune questioni preliminari di critica del testo Il contributo è dedicato all’antica traduzione slava meridionale dei libri biblici di Samuele e dei Re, preservata in due testimoni serbi dei secoli XV-XVI. Questa versione è un testimone indiretto della cosiddetta recensione antiochena o lucianea della Septuaginta, pervenutaci in soli cinque codici manoscritti greci di epoca bizantina. Il presente lavoro si configura come un primo tentativo di studio comparato del testo slavo con il suo originale greco. Particolare attenzione è prestata all’individuazione delle cosiddette lezioni protolucianee e all’analisi di alcune caratteristiche testuali del tutto uniche che sono rinvenibili nella tradizione serba. Алессандро Мария Бруни Древнесербский перевод “Антиохийской” редакции книг Царств: предварительные текстологические заметки Доклад посвящен южнославянскому переводу книг Царств, сохранившемуся в двух сербских рукописях XV-XVI вв. Этот текст восходит к так называемой “Антиохийской” или “Лукиановской” редакции Септуагинты, полностью известной лишь в пяти византийских списках. В работе совершается первая попытка сравнительного изучения славянской версии и греческого оригинала. Особое внимание уделяется выделению “протолукиановских” чтений, а также анализу ряда уникальных текстологических особенностей сербской традиции.


The LXX L of Samuel-Kings and the Old Greek text
The books of 1-2 Samuel and 1-2 Kings (1-4 Kgdms) pose serious challenges for biblical scholars, given the major textual differences between the 1 According to the colophon of Slav O (ff. 332-334), this version was carried out in 1416 (Popruženko 1894: 1-4). Authorship has been attributed to Constantine of Kostenets (ca. 1380Kostenets (ca. -after 1427 or to his contemporary, the monk Gabriel of Hilandar, who is known for having rendered from Greek the Catena in Job (Thomson 1998: 762-763). In Nikolova's view, the translation was however undertaken in the fourteenth century by a representative member of the Tarnovo Literary School (Nikolova 1995: 62). 2 Močul'skij 1890: 5-6; Popruženko 1894; Kopylenko et al. 1960: 550;Korol'kova et al. 1963. 3 Viktorov 1879: 3-4. This manuscript was discovered by Grigorovič in Ohrid (Grigorovič 1877: 184). Folios ff. 210-373 are thought to have been written by the Hieromonk Vissarion of Debar (See: Nikolova 1995: 62 and1996: 363-402;Turilov 2004: 545). existing testimonies. Within this framework, a crucial, yet unsolved, issue lies in the appraisal of LXX L , a text-type found in only five Byzantine minuscule codices (N° 19, 108, 82, 93, 127; previous sigla of the first four: b, o, c 2 , e 2 ) 4 . This group significantly deviates from the rest of the Greek tradition, but finds parallels in some Latin, Syriac and Armenian sources. The denomination Antiochene or Lucianic recension is to be understood conventionally: the redaction was shown to be composed of different layers, the earliest of which was named proto-Lucianic, since its characterizing readings are to be found in several sources preceding the historical Lucian, namely the Qumran scrolls (Q), Josephus (J), thе Vetus latina (VT) and the writings of some Church Fathers 5 .
Several scholars assume that LXX L constituted (or probably constituted) the Old Greek (OG) text of LXX (namely its earliest textual stratum), which was translated from a Hebrew source, differing from the Masoretic text (MT). The analysis of LXX L is especially relevant in the case of 1 Sam, since it has been suggested that the former derives from the same archetype of 4QSam a . In a different opinion, however, LXX L is not believed to plainly represent the OG. Doubts have been mainly voiced regarding those parts of the translation that, in virtue of a hebraizing revision, are commonly known as the kaige-sections. Moreover, the claim has been made that OG readings must be assessed on a case-by-case basis, since neither LXX nor LXX L directly represent the original translation. Recently, Kreuzer's studies have brought new arguments in support of the identification of LXX L with the OG, but the question seems to be still open, especially because the process of the preparation of the Göttingen editions of LXX Samuel-Kings is in progress 6 .

Overview of past research on Slav-LXX L
Over more than a century, Slavicists very rarely addressed the question of the textual analysis of Slav-LXX L . The two major contributions on this topic date from the late eighteenth century. In 1894 M.G. Popruženko published a short monograph on Slav O . Along with the edition of excerpts from the biblical text and from the marginal notes, which include readings from "the Three" (Aquila, Symmachus, Theodotion) 7 and from J, he presented a brief sketch of the grammatical features and of the orthography of the manuscript. Some years later 4 Reference critical edition: Fernández Marcos et al. 1989-1992. 5 Fischer 1951Spottorno 1995;Tov 1999;Piquer et al. 2008;Torijano Morales 2012. 6 For reference bibliography see (at least): Barthélemy 1963;Brock 1996;Cross 1964;Fernández Marcos 1994;Hugo 2010 and2013;Kauhanen 2012;Kreuzer 2015;Rahlfs 1911;Tov 1999;Taylor 1992Taylor -1993Ulrich 1978. 7 In Slav-LXX L the number of the available readings from "the Three" is consistently higher than assumed by Popruženko (1894: 123-129), at least if looking at 1 Sam in manuscript Slav M (see: Bruni 2016b: 442-443). This new material awaits editing and S.M. Kul'bakin undertook a comparative textual examination of some passages of 1 Sam (1 Kgdms) that he carried out by taking as a basis selected South and East Slavic sources dating from the fourteenth-sixteenth centuries. According to his conclusions 8 , Slav O would preserve a text closer to the original than Slav M , since in 1 Sam 5: 6,9; 6:4,18; 9:8, 23, 27; 10:2; 17:43; 19:4; 24:4 the latter would display textual contamination with the previously existing Slavonic version of the book, dating back to the Old Church Slavonic (Old Bulgarian) period (late ninth-early tenth century) 9 . Moreover, the Russian scholar was the first to establish the reliance of Slav-LXX L on a Lucianic model 10 .
In more recent times, a number of other studies have also been produced. On the one hand, several readings from Slav O were included by D. Dunkov in his edition of the Old Church Slavonic (supposedly Glagolitic) version of 1-4 Kgdms 11 , even though the manuscript is a witness to Slav-LXX L and should therefore not have been used for such a purpose 12 . On the other hand, S. Nikolova expressed the opinion that Slav O and Slav M derive from a common, untraced, exemplar of middle Bulgarian and not Serbian origin 13 . Finally, R.V. Bulatova published a paper on the accentual system of Slav O14 .

The Crucial Issue: The Nature of the Lucianic Text in Slav-LXX L
As of mid 2017, the Slav-LXX L remains unpublished and still awaits to be studied in detail: nowadays no systematic collation of its two testimonies, Slav O and Slav M , is available. Moreover, this tradition has not yet been investigated in the light of the apparatus of the reference edition of LXX L , whose authors were not aware of the existence of a Slavonic text 15 . This last point is not surprising since this secondary tradition is usually not even mentioned in studies dealing with the textual history of Samuel-Kings 16 . A rare exception is Tov's to be thoroughly collated with evidence found in LXX and other secondary sources such as the Armenian version (Ibidem). 8 Kul'bakin 1901: 23-25, 43. 9 Some scholars ascribe this earliest translation to Methodius, while others to Gregory the Presbyter (see: Thomson 1998: 758;Alekseev 1999: 120-122;Bruni 2016b: 437). 10 Kul 'bakin 1901'bakin : 23, 44. 11 Dunkov 1995'bakin -1996 Slav-LXX L is a new translation based on a different Greek prototype. 13 With the aim of substantiating the hypothesis of a Bulgarian origin of the translation, Nikolova (1995: 62) refers to Lavrov's (1914: 305-306) remarks concerning the use of nasal vowels in Slav M . These characters are, however, to be found in the first part of Slav M only, in which the Old Church Slavonic (Old Bulgarian) Octateuch is preserved (ff. 1-209; see, e.g., ff. 116v-117). 14 Bulatova 1995. 15 Fernández Marcos et al. 1989Marcos et al. -1992 See above bibliography (footnotes 5 and 6). 1972 (1999) paper on the proto-Lucianic problem in Samuel-Kings, in which reference is expressly made to the Slavonic version. In this contribution, the fundamental question was raised as to whether Slav-LXX L is based on the ancient textual layer of the five Greek testimonies of LXX L only, or whether, being chronological post-Lucianic, it reflects LXX L as a whole 17 . This crucial issue, which was left unanswered by Slavicists 18 , currently remains at the very core of research into this tradition 19 .
The analysis of Slav M proves that this unique feature is not due to scribal mistakes, but instead intentionally appears to replicate a lost prototype. In this regard the following observations can be made. On the one hand, before 3 Kgdms-LXX L 11:42 codex Slav M inserts a heading informing the reader that 4 Kgdms begins at the reign of Rehoboam 25 . On the other one, a very interesting marginal comment is to be found alongside the translation of 4 Kgdms 1:1 (333v) 26 . This note tells us that in the original the copyist had before his eyes 4 Kgdms started precisely at this point, while in other testimonies at the reign of Rehoboam 27 . Whether such an alternative structure was Slavonic or Greek, is unfortunately not specified by the Serbian glossator. Regardless of this, it is however evident that the author of Slav M , or of its archetype, deliberately orientated his work towards a different LXX L tradition, known today thanks to a single secondary witness.

Proto-Lucianic Readings in Slav M
A first text-internal comparative analysis of Slav M with LXX L has produced the following results. This source includes several readings that belong to the ancient textual layer of the Antiochene recension. Accordingly, a positive response to Tov's question 28 may now be given: the Serbian tradition represents a new witness not only to LXX L as a whole, but also to the proto-Lucianic textual stratum. With an aim to providing an initial illustration of this crucial textual feature, an edition of selected passages of Slav M is offered below. Slav M (fol. 223v): и тече самѹиль и поеть его ѿонѹд. и ста саѹль посрѣд людїи, и въꙁнесе се ѿ въсѣх людїи паче подобїа и повыше. LXX L : καὶ ἔδραμε Σαμουὴλ καὶ ἔλαβεν αὐτὸν ἐκεῖθεν. καὶ κατέστη Σαοὺλ ἐν μέσῳ τοῦ λαοῦ, καὶ ὑψώθη ὑπὲρ πάντα τὸν λαὸν ὑπερωμίαν καὶ ἐπάνω 32 .

Conclusions
Research presented in this paper represents a first attempt towards a comprehensive analysis of the textual features of Slav-LXX L against the background of the Greek testimonies of the Antiochene recension of Samuel-Kings. Despite being preliminary, the undertaken work indeed looks to be promising.
The most notable implication of the current study is that Slav M was shown to be a new witness to the ancient textual layer of the Lucianic recension, namely to the proto-Lucianic stratum. Moreover, the analysis has furthermore provided indirect evidence of the existence of a lost edition of LXX L 1-2 Kings that featured an alternative subdivision of the books, according to which 4Kgdms consisted of 3 Kgdms-LXX L 11:42-22:54 + 4 Kgdms-LXX L 1:1-25:30. The simultaneous presence in the Serbian version of text-internal proto-Lucianic elements makes it very likely that this arrangement dates back to Late antiquity and not merely to the Middle Ages. Consequently, the hypothesis may be advanced that this Slav translation provides scholars with a new window into the textual history of Antiochene recension of Samuel-Kings and, ultimately, into the OG text of these books. Accordingly, this Serbian tradition can safely be placed at the very center of the debate surrounding one of the most complex issues facing contemporary biblical scholarship.