Lucian's translation of the Bible

koot van wyk (DLitt et Phil; ThD)

Kyungbook National University

Sangju Campus

South Korea

Conjoint lecturer of Avondale College

Australia

17 March 2010

Lucian was a scholar from Antioch and he made a translation in Antioch in 311 CE. Of the supposedly original translation, nothing survived and we are obliged in modern times to work with a assemblage of supposed copies, re-workings and adaptations that are all labelled Lucian.  The patristic fathers were aware of his translation.

In 1867 attempts were made by O. F. Fritzsche to identify the manuscripts of Lucian and in 1900, further attempts towards the same goal was undertaken by G. Moore. In 1926, Otto Pretzl worked with the same goal. Max-Margolis (a scholar for whom Johann Erbes greatly respected) worked between 1910-1926 with the same objectives and in 1922 he wrote an article on this topic. Further studies were than of A. V. Billen in 1943 and W. Bodine in 1980. An article by E. Tisserant in 1911 explained some of the problems involved in identifying the text of Lucian. Scholars became convinced that the works of John Chrysostom and Theodoret of Antioch are two sources that assist in identifying the Lucian text (Orlinsky 1941: 82; Pretzl 1926: 244-245). Lucifer of Cagliari, who favored the Vetus Latina and made copies of it, lived in Antioch when John Chrysostom was a young man. If one then finds correlations between the so-called Lucian manuscripts, the Old Latin translation (Vetus Latina) of Lucifer of Cagliari and the works of John of Chrysostom and Theodoret in Antioch, one should not be surprised. All these men were contemporaries connected in time and geographical sphere.

If anyone wish to see the Greek manuscripts allocated by scholars as Lucian, then one should obtain Brooke-McLean (LXX edition) of 1917 and look in the register below.

Lucian manuscripts

Scholars have allocated the following manuscripts as Lucian:

1. Z which dates from 450 CE and which is a Syriac text known as codex Zugninensis, currently in the Vatican.

2. K which is a Greek text from 650 CE and currently in the Leipzig University library.

3. v which dates from 1050 CE which has many hands working on it and there are hexaplaric notes in the margins.

4. t which dates from 1050 CE and which is in Florence in the Biblioteca Mediceo-Laurenziana.

5. n which dates from 1125 CE (1126 CE by Robert 1900: xxii). It is in Oxford.

6. g which dates from 1300 CE and which was also written by different hands. It is in Paris in the Bibliothèque Nationale. This document also contains Aristeas, as synopsis of the Octateuch and Theodoret's book on Questions in the Octateuch (incomplete).

7. p which dates from 1350 CE and is in Ferrara in the Museo Schifanoia. It also contains portions of the New Testament.

8. w which dates from 1250 CE and which is in Athens in the National Library.

9. l which dates from 1450 CE and which is in the Glasgow University Library.

10. d which dates from 1450 CE and which is in Zittau in the Stadtbibliothek, also with portions of the New Testament. 

This researcher's result working with these manuscripts are that the Lucian manuscripts listed above are diversified and fluid (see also Margolis) heterogenic (Tisserant), mixed (Marcos) with conflated readings (Billen). This researcher tend to disagree with the observation of W. Bodine that these manuscripts represent the "most consistent of Old Greek" (Bodine 1980: 134). There is not consistency or otherwise a consistency in problems of elaborations, misspellings, dictation problems etc.

Some Lucian characteristics in the translation

a. In Judges 6:3 all the Lucian manuscripts agreed with an inclusion of two extra words with the Syro-hexapla and the Vetus Latina but not with Qumran text 4QJudg^a. Only one Lucian manuscript did not read this inclusion and that is manuscript n from 1125 CE. The manuscript n thus omitted this inclusion of the preposition and the suffix.

b. In Judges 6:11 the Lucian spelling is different from the spelling of the consonantal text of the Masoretic tradition or the Word of God, as we see it. This reading of the Lucian manuscripts are shared also by Qumran in 4QJudg^a. Especially n, w, g, and l shares this spelling with 4QJudg^a.

c. In Judges 6:5 the Lucianic group of Greek manuscripts v, t, n, w, g, p, l, and d, all have a word order that could be constructed as a parallel of 4QJudg^a. However, this variant is not separate of what we find with the versions of the Vetus Latina (Lucifer and codex Lugdunensis) and also what we find in Origen's Syro-hexapla.

There are diversity between the Lucianic group of manuscripts and one may think that the confusion is due to misreadings of a notebook that was used and shared for translation. It was either copied or misread or misinterpreted by subsequent readers who dictated to their translators in that way. Therefore the inner diversity.

d. The Lucian manuscripts ordered the word camels in Judges 6:5 earlier in the verse than the Word of God or consonantal text of the Masoretic tradition (Hebrew) had it. In this the defective change of the word camels earlier in the sentence follows also the defective manuscript used for the Vetus Latina that had it earlier, as one also find in the defective copy at Qumran in 4QJudg^a. It is n of the Lucian group that contains minor differences with 4QJudg^a.

e. All the Lucian manuscripts entered a misreading at the end of Judges 6:5 "sons of Israel" which is not shared by 4QJudg^a.

f. The Lucian manuscript n used the same defective Hebrew manuscript or notebook as was used by the Coptic K^b in Judges 6:3. The reader to the scribe of the Hebrew manuscript, misread a word and then corrected himself but correction. This degenerated manuscript then served the Coptic translator of K^b and also the Lucian manuscript n. Further explanatory notes in the supposed notebook margin may explain more variants in the Coptic.

g. An investigation of omissions by the Lucianic group of manuscripts indicated that it was especially manuscript d of the Lucianic group that was the most difficult to explain. The conclusion was that the omissions of d in Judges had some other purpose or reason than mere errors due to transmission. In Judges 1:19 an error of dittography was committed by the scribe of d. Omissions in Lucianic manuscript d can be in the beginning, middle or end of the verse. There are no redactional reasons for the omissions by d.

h. An experimental probe of Judges chapters 3-11 in the Lucian tradition, convinced this reader that there are more than one n scribe and more than one w scribe.

1. n1 took care of Judges 4:6-6:33 and also 8:17-9:20.

2. n2 took care of Judges 6:34, 7:1-8:6.

3. n3 took care of Judges 8:7-8:16 and 9:21-9:49.

4. w1 took care of Judges 3:21-7:14 and 9:5-9:54.

5. w2 took care of Judges 8:1-9:4 and 10:1-11:19.

6. n1 considers the Ethiopic and Armenian translations.

7. n2 considers the Ethiopic, Armenian and Targum.

8. n3 cannot spell since his spelling practices are erratic.

9. n has 76 individual spellings.

10. n consulted the Vetus Latina in Judges 5:17; 6:17; 7:24; 9:52; 10:1; 11:17; 11:19; 8:33.

11. n has 43 omissions.

12. w1 considers the Targum and considers the Ethiopic more than w2.

13. w1 uses mimation whereas w2 uses nunation.

14. w1 uses shin for status constructus (genetive) for example in Judges 9:31 e.g. sabed.

15. w2 divide words.

16. w2 do not consult the Vetus Latina.

17. w has 47 individual spellings.

18. w uses n in 6:19; 8:35 and twice in 9:29.

19. w has 20 omissions.

20. The dating of n is 1125 CE and the dating of w is 1250 CE.

21. d consulted n and w. However, in Judges 5-8 and 10-11, seemingly d does not follow n. d followed n in Judges 4:16; 9:26; 9:32; 9:28; 9:39.

22. d has the most omissions and it follows p (1350 CE) at times.

23. l (1450 CE) considers n for Judges 4 and 5 but not for 6-10. l does not consider w for Judges 6-10, but in Judges 4-5 it considers w.

24. There is evidence in Judges 4:13 and 8:13 that Lucianic manuscript g consulted manuscript w. There is also evidence in Judges 8:6 and 9:28 that Lucianic manuscript g consulted n.

25. d followed w in Judges 4:8; 4:14; 5:2; 5:24; 9:20; 9:28; 9:50.

Scholars (Trebolle Barrera et al) are saying the the Lucian manuscripts are witnesses of the Old Greek and Old Latin forms of the Bible. Although they are in general correct, specifically not. There are to many divers features and individual readings by n for example. Yes, there are connections. It appears as if defective copies of Qumran material fell into the hands of the Christians who translated the Vetus Latina and also later into the hands of men like Lucian of Antioch so that those errors were transmitted by cross-mutation. Each generation added to the errors and individual readings due to the modus operandi of scribal practices in those days.

Is it possible that the Qumran caves and its defective copies fell into the hands of Origen in his days and made their way to Antioch's library so that Lucian could translate with it? Was it some of these copies that then came into the hands of Jerome in 392-402 CE who also used defective copies of the Hebrew and shared errors with Lucian, Vetus Latina, Targum and with Qumran?

Margolis and Van Wyk Notes

1. Margolis studied the Lucian manuscripts and concluded that p and d indulged "in textual omissions and contractions" (Margolis 1927: 312). This was for the book of Joshua but it is also for the book of Judges.

2. Margolis found that in Joshua, Lucian manuscripts t, p and d contains a mixed text. This is also true of the book of Judges (Van Wyk 2008).

3. Margolis felt that for the book of Joshua, g, n, w, and the Vetus Latina is generally a purer text. What Margolis means by purer is a semantics that need clarification. Defective is a better description if one compares it with the consonantal text of the Masoretic tradition or Word of God. The manuscripts n is not so promising for the book of Judges since it has many individual spellings and omissions.

4. Margolis found that w sometimes conflates its texts or consult the Vetus Latina in the book of Joshua. For the book of Judges, this researcher has found that w1 will consult the Vetus Latina but w2 will not.

5. Margolis felt that Lucian did not indulge in doublets as a general practice. However, Lucian's translation in our opinion was also based upon a defective Hebrew manuscript that contained both the error and its correction at times.

6. Margolis felt that n has a "propensity to abbreviating the text" (Margolis 1927: 312).

7. Margolis characterizes the text of Lucian as a common text "but a distinct form as the personal names show" (Margolis 1927: 313). Van Wyk has problems to identify Lucian since there is no standard Lucian text that can be used as measure to shift the data. Identifying the Lucian text is a constant elusive aim.

For the books of Genesis and Ruth, N. Marcos found that Theodoret of Antioch contains a mixed text which means that this Antiochene father could not assist in the establishing of the Lucian text for these books (Marcos 1990: 220, footnote 7).

Saiz has found that the Greek manuscripts that one can allocate to the text of Lucian in the book of Ruth differs in Ruth 1:1-4:11 from those in Ruth 4:11 to the end (Margolis 1927: 309). Margolis also described problems with the Lucianic manuscripts in Joshua 2:18c to the end.

For the book of Job, N. Marcos felt in 1990 that the Codex Alexandrinus is the most ancient testimony of the Antiochene text (Marcos 1990: 223). For the book of Ruth, Marcos felt that the "textual relationship between Theodoret and mss. gn(lw) + dpt of Brooke-McLean tends to become closer as we reach the end of the Octateuch" (Marcos 1990: 222).

For the book of Judges, Trebolle Barrera has identified the proto-Lucianic manuscripts as g, l, n, w and for the Lucianic as v, t, p, and d.

Margolis' method or modus operandi of analysis was to identify the genetic relationship of each manuscript according to the spelling of personal and geographic names.