Textual Evidence

Metzger on the
'Lucianic Recension'

Excerpt from: Bruce Metzger, Chapters in the History of NT TC, (Eerdmans, 1963)

Page Index

Review: Stuff:
    Introduction The Lucianic Recension of the Greek Bible
    I. Ancient Testimonies others on Lucian
    II. Research on Lucian a survey of recent work
       A. Old Testament - Lagarde, Rahlfs
       B. New Testament - Bengel, Hug, Scholz, Streeter
       C. Characteristics - of the Lucianic Text
    III. The Antocian Text - Influence outside the Greek Church
    IV. Critical Value - of the Lucianic Recension
       A. Old Testament - Driver, Rahlfs, Fischer, Staerk
       B. New Testament - Burkitt
    V.Problems - current questions relating to the Lucianic Recension

Return to Index

Metzger on the
Lucianic Recension

Taken for review purposes from:
Bruce Metzger,
Chapters in the History of NT TC, (Eerdmans, 1963)

Headings have been added for clarity and navigation.



Among the several scholars of the ancient Church who occupied themselves with the textual criticism of the Bible, one of the most influential was Lucian of Antioch. Though not as learned or as productive in a literary way as either Origen or Jerome, Lucian's work on the text of the Greek Bible proved to be of significance both in his own day and, to an even greater extent, during the centuries following. In fact, his recension of the text of the New Testament, with only minor modifications, continued to be used widely down to the nineteenth century, and still lives on in the so-called Ecclesiastical text of the Eastern Orthodox Church.

Little is known of the life of Lucian of Antioch. Born probably at Samosata in Syria about the middle of the third century, he was educated at Edessa under a certain Macarius, who, according to Suidas, was a learned expounder of holy Scripture. After a period when he may have studied at Caesarea, Lucian transferred to the famous theological school of Antioch, of which he and Chrysostom, Diodorus, Theodoret, and Theodore of Mopsuestia were to be some of the more distinguished alumni.

Apparently Lucian was in sympathy with the theological views of his fellow townsman, Paul of Samosata, and when Paul was deposed for Christological heresy in A.D. 268 (or 270), he too with drew from the Church. During his later years Lucian seems to have become more orthodox, and under the episcopate of Cyril of Antioch (A.D. 283-304) he was restored to ecclesiastical fellowship. He died in the peace of the Church, suffering martyrdom for the faith at Nicomedia, Bithynia, probably on January 7, 312.

Many are the historical and theological problems connected with the person and influence of Lucian of Antioch. The question has even been raised as to whether Lucian the excommunicated heretic was the same person as Lucian the martyr and Biblical scholar. 2

1 The substance of this chapter was presented as a lecture at the Sym posium on Antioch of Syria held during May, 1959, at the Dumbarton Oaks Research Library and Collection, Washington, D.C,

2 This view, suggested earlier by Ceillier, Fleury, De Broglie, and Oikonomos, has been revived by D. S Balanos in Πρακτικα της Ακαδημιας Αθηνων, VII (1932), 306-311, and by Gustave Bardy, Recherches sur saint Lucien d'Antioche et son ecole (Paris, 1936).

Lucianic Recension p.2

There is no need, however, to suppose the existence of two Lucians, one orthodox and one heretical; the somewhat conflicting reports can easily be reconciled by the asssumption that Lucian was a critical scholar whose views on the Trinity and on Christology differed from what was later defined at Nicea as the orthodox position, but that he wiped out all stains of doctrinal aberrations by his heroic confession and martyrdom.1 It is quite understandable that during his connection with the school at Antioch he exerted a pervasive influence upon the theological views of those who came to adopt Arian theology. Indeed, Arius himself, a former pupil of Lucian's, declared (in a letter to Eusebius of Nicomedia) that he was merely following Lucian's views regarding Christology. Along with Arius, other sympathizers of Lucian's point of view -- such as Eusebius of Nicodemia, Maris of Chalcedon, Leontius of Antioch, Eudoxius, Theognis of Nicaea, Asterius -- became a closely-knit group that were dubbed "Collucianists".

It is, however, not the vicissitude of doctrinal disputes 2 that is our concern here, but the part that Lucian played in editin gthe text of the Greek Bible. Though not a little has been writtern on the subject, it is unfortunate that, with only a few exceptions, 3 scholars have confined their attention either to problesm relating to his recension of the Greek OT or to those relating to the NT. Such restrictions of interest have worked to the disadvantage of both groups of scholars. Just as the grammarian and the lexicographer of the NT can learn much from an examination of the language of the Septuagint, so too the tex-

1. So, for example, Albert Ehrhard, Die Kirche der Martyer (Munich, 1932), pp. 304ff. and Adhemar d'Ales, "Autour de Lucien d'Antioche,", Melanges de l'universite Saint Joseph (Beyrouth), XXI (1937), 185-199, who point out that it is extremely unlikely that two persons of the same name should have played important roles in Antioch at the same time without leaving in the sources a trace of their differentiation from each other.

2. For discussions of Lucian from a doctrinal point of view, E. Buonaiuti, "Luciano martire, la sua dottrina e la sua scuola" Rivista storico-critica delle scienze teologiche 1908, pp.830-36, 909-923, 1909, pp. 104-118; Friedrich Loofs, Das Bekenntnis Lucians, des Martyrers (Berlin, 1915), pp.576-603; Adolf von Harnack Dogmengeschichte 5te Aufl. II (Tub. 1931), 187-190, and esp. Bardy, op.cit.

3 E.g. M. Spanneut's recent study, "La Bible d'Eustathe d'Antioche Contribution & 1'histoire de la 'version lucianique.' " Studia Patristica, ed. F. L. Cross, iv (TU, LXXIX [Berlin, 1961]), 171-190.

Lucianic Recension p.3

tual critic of the New Testament will profit from considering the problems and tasks of Septuaginta-Forschung. As an exploration in methodology, therefore, the aim of the present chapter is to shed light upon the text of the New Testament by giving attention to one specific type of text common to both Old and New Testaments, the Lucianic text.

The following pages will present, first, a r6sum6 of ancient testi monies relating to Lucian and his work as textual critic; second, a survey of research on the Lucianic or Antiochian text of the Greek Bible ; third, the influence of this text outside the Greek Church ; and fourth, a critical evaluation of the Lucianic recension. The chapter will conclude with a list of some of the problems that remain to be solved.


The earliest references to Lucian are two brief and highly favor able estimates which Eusebius includes in his Church History. Here Lucian is described as a presbyter of Antioch, "whose entire life was most excellent (αριστος)" (vm.xiii.2), and as "a most excellent (αριστος) man in every respect, temperate in life and well-versed in sacred learning" (ix.vi.3).

Later in the fourth century Jerome makes three references to Lucian which differ considerably in temper and appreciation of his work. The differences are no doubt to be accounted for by consider ing the several contexts and Jerome's immediate purpose in refer ring to Lucian. On the one hand, when Jerome is comparing his own work as reviser of the Old Latin text with similar work by others in Greek, he is rather severe in his judgment of Lucian. Thus in his Preface to the Four Gospels, which takes the form of an open letter addressed to Pope Damasus and which was composed perhaps about the year 383, he refers somewhat contemptuously to the "manuscripts which are associated with the names of Lucian and Hesychius, the authority of which is perversely maintained by a few disputatious persons." Continuing in the same vein Jerome condemns the work of Lucian and Hesychius as infelicitous:

"It is obvious that these writers could not emend anything in the Old Testament after the labors of the Seventy; and it was useless to correct the New, for versions of Scripture already exist in the lan guages of many nations which show that their additions are false." 1


Praetermitto eos codices quos a Luciano et Hesychio nuncupates pau corum hominum adserit peruersa contentio: quibus utique nee in ueteri instrumento post septuaginta interpretes emendare quid licuit nee in nouo pro fuit emendasse, cum multarum gentium linguis scriptura ante translata doceat falsa esse quae addita sunt (edita sunt, ms. E; John Wordsworth, and H. J. White, Novum Testamentum Domini Nostri lesu Christi Latine, I [Oxford, 1889], 2).

There has been a curious reluctance among many scholars to admit that Jerome here refers to any more than the Lucianic text of the Old Testament. But, as B. H. Streeter pointed out,

"seeing that Jerome is writing a careful and considered Preface to a revised version of the Four Gospels, and that he only mentions the Lucianic and Hesychian versions in order to contrast their inferior text with that of the 'ancient codices' he himself has used, I simply cannot understand why some scholars have raised doubts as to whether the Lucianic and Hesychian recensions included the New Testament as well as the Old" (The Four Gospels [London, 1936], p. 59*)

As regards the much more nebulous figure of Hesychius, whom no Greek author mentions, the situation is different. Despite the popularity of WilhelmBousset's suggestion that the so-called "Neutral" text is to be attributed to Hesychius, most scholars today are inclined to agree with Sir Frederic G. Kenyon, who concluded his study of "Hesychius and the Text of the New Testament" (Memorial Lagrange [Paris, 1940], pp. 245-250) with the words: "The title of 'Hesychius' rests in fact upon what is little more than a shadow of a shade."

Lucianic Recension p.4

Subsequently, in the Preface to his translation of the books of Chronicles, Jerome makes a more temperate allusion to the work of Lucian and other Biblical scholars. In referring to the diversity of the editions of the Greek Old Testament, he declares that three are current in various parts of the Empire: "Alexandria and Egypt in their [copies of the] Septuagint praise Hesychius as author; Con stantinople to Antioch approves the copies [containing the text] of Lucian the martyr; the middle provinces between these read the Palestinian codices edited by Origen, which Eusebius and Pam philus published." 1

In his valuable Lives of Illustrious Men, written soon after A.D. 392, Jerome is still more generous in his description of Lucian. Here, in a biographical sketch devoted to the martyr from Antioch, he charac terizes him as "a man of great talent'' and "so diligent in the study of the Scriptures that even now certain copies of the Scriptures bear the name of Lucian." 2 What is of special importance is the declara tion that copies of the Scriptures (and not just of the Septuagint, as Jerome is sometimes quoted) passed under the name oiLucianea,

1 Alexandria et Aegyptus in LXX suis Hesychium laudat auctorem; Con stantinopolis usque Antiochiam Luciani martyris exemplaria probat; mediae inter has provinciae Palaestinos codices legunt, quos ab Ongene elaborates Eusebius et Pamphilus vulgaverunt (Migne, PL, xxvin, 1392 A, and Fried rich Stummer, Einfuhrung in die lateinische Bibel [Paderborn, 1928], p. 239).

2 Lucianus, vir disertissimus, Antiochenae ecclesiae presbyter, tantum in Scriptuarum studio laborat, ut usque nunc quaedam exemplaria Scrip turarum Lucianea nuncupentur (de Viris inlustribus, 77 [TU, xiv, pp. 4 if., ed, E. C. Richardson] ),

Lucianic Recension p.5

Information of the widespread use of Lucian's recension of the Psalter is contained in Jerome's letter to Sunnias and Fretela (about A.D. 403). These two Gothic churchmen had inquired of Jerome why his own Latin Psalter (the so-called Roman Psalter) differed so frequently from the Septuagint. In his reply Jerome points out that they have been misled by their edition of the Septuagint, which varied widely from the critical text of Origen given in the Hexapla and used by himself. Jerome writes : e You must know that there is one edition which Origen and Eusebius of Caesarea and all the Greek commentators call κοινη, that is common and widespread, and is by most people now called Lucianic; and there is another, that of the Septuagint, which is found in the manuscripts of the Hexapla, and has been faithfully translated by us into Latin." 1 Here Jerome distinguishes the Lucianic text from that of the Hexapla, and indicates that the former met with such uni versal acceptance that it received the name of the Vulgate or common text.

Later testimonies refer to Lucian's competence in Hebrew. For example, Suidas and Simeon Metaphrastes (in the Passio S. Luciani martyris) assert that "he translated [literally, renewed] them all [i.e. the books of the Old Testament] again from the Hebrew language, of which he had a very accurate knowledge, spending much labor on the work." 2 Though Lucian may have consulted the Hebrew in connection with his revision of the Septuagint, this statement is obviously exaggerated in the manner of hagiographers. More sober, and doubtless nearer to the truth of what Lucian attempted to do, is the description of pseudo-Athanasius in his Synopsis sacrae scripturae:

1 Illud breuiter admoneo, ut sciatis aliam esse editionem, quam Origenes et Caesariensis Eusebius omnesque Graeciae tractatores κοινη id est communem appellant atque uulgatam et a plerisque nunc Λουκιανειος dicitur, aliam septuaginta interpretum, quae et in εξαπλοις codicibus repperi tur et a nobis in Latinum sennonen fideliter uersa est (Epist. 106, 2, 2 [CSEL, vol. LV, p. 248, ed. Hilberg] ).

2 Both Suidas and Simeon, who here agree (except in inconsequential details) in their accounts of Lucian, depend upon earlier hagiographical sources. The variant επανεσωσατο, which Adler adopts into her text of Suidas, is clearly the inferior reading; either Simeon's ανενεωσατο / επανενεωσατο is to be preferred. For the text of both see Joseph Bidez's ed. of Philostorgius, Anhang vi ( Griechische christliche Schriftsteller. 1913, P. 187).

Lucianic Recension p.6

"Using the earlier editions [i.e. of Aquila, Theodotion, and Symmachus] and the Hebrew, and having accurately surveyed the expressions which fell short of or went beyond the truth, and having corrected them in their proper places, he published them for the Christian brethren." 1

Among testimonia of uncertain origin there is an unequivocal statement that Lucian concerned himself with the New Testament as well as the Old. Under the date of October 15, the Menaeon of the Greek Church (this is a liturgical volume which includes sHort accounts of saints and martyrs to be read on their festivals) states that Lucian made a copy with his own hand of both the Old and New Testaments, written in three columns, which afterwards belonged to the Church in Nicomedia. 2 Substantially the same in formation in a more extended hagiographical context is contained in the Synaxarium ecclesiae Constantinopolitanae for October 15. 3

This list of testimonies may be brought to a close 4 with a reference to the condemnation of Lucian in the so-called Decretum Gelasianum, where mention is made of Evangelia quae falsavit Lucianus, apocrypha, Evangelia quae falsavit Hesychius, apocrypha (v, iii, 8-9). It is generally agreed that this statement rests upon a misunderstanding of the critical remarks of Jerome. 5

By way of summarizing ancient testimonies concerning Lucian's textual work, we find that his contemporaries generally regarded him as an able scholar, entirely competent to undertake such a recension. As a native Syrian he could, of course, have consulted the Syriac version; he also appears to have had some acquaintance with Hebrew. As would have been expected, he made use of pre vious Greek translations of the Old Testament, and sought to adjust the Greek to the underlying Hebrew text. But we are told nothing as to the amount of revision which he undertook in either Old or

1 Theodor Zahn dated this document not later than the fifth or sixth century; see his Geschichte dev neutestamentlichen Kanons, n (Erlangen, 1890), 311. The text is printed in Migne, PG, vol. xxvin, col. 433, see also H. Dome's discussion of the textual transmission of this passage in his article, "Zur Geschichte der Septuagmta im Jahrhundert Konstantins," ZNW, xxxix (1940), 79-87.

2 The relevant passage is a follows: εις καλλος γραφειν επισταμενος βιβλιον κατελιπε τη Νικομηδεων εκκλησια, γεγραμμενον σελισι τρισσαις (εις τρεις στηλας διηρημενης της σελιδος), περιεχον πασαν παλαιαν τε και την νεαν διαθηκην.

3 Edited by Hippolyte Delehaye, Propylaeum ad Acta Sanctorum, Novem bns [vol. LXI], 1902, pp. i38fif.

* For several other ancient testimonies to Lucian, see M, J. Routh, Reli quiae sacrae, 2nd ed,, iv (Oxford, 1846), 5-10.

5 See Ernst von Dobschiitz, Das Decretum Gelasianum ( = T7, xxxvm, 4) , DD. a and 202.

Lucianic Recension p.7

New Testament text, the nature of the manuscripts which he consulted, the relation of his work to the Hexapla, and other similar matters. For information bearing on such problems, we must turn to the manuscripts which have been thought to contain the Lucian recension.


A. The Old Testament

Our account begins with the first printed edition of the entire Greek Old Testament; namely, that contained in the famous Com plutensian Polyglot Bible sponsored by the Spanish Cardinal Francisco Ximenes de Cisneros (1437-1517), Archbishop of Toledo, and published at Complutum (now Alcald de Henares). The four folio volumes containing the Old Testament in Hebrew, Latin, and Greek were printed between 1514 and 1517. As it happened, one 1 of the two manuscripts of the Greek Old Testament which Pope Leo X sent from the Vatican Library for the use of the editors of the Polyglot, and which forms the basis of a large part of their text, contains the type of text now thought to be Lucianic, at least in Samuel and Kings. In other parts of the Old Testament, however, this manuscript departs from the Antiochian type of text. Further more, the Spanish editors frequently adopted readings from several other Greek witnesses, and occasionally even conformed the Greek to the Hebrew without any manuscript authority. This Compluten sian text was followed on the whole in subsequent Polyglot Bibles (those published at Antwerp, 1569-72; Heidelberg, 1586-1616; Hamburg, 1596; and Paris, 1645).

During the seventeenth and eighteenth centuries scholars laid the basis for subsequent investigations by collecting variant read ings from manuscripts of the Greek Old Testament, from the early versions, and from the quotations of the Fathers. Brian Walton's Polyglot Bible (London, 1654-57), Humphrey Hody's valuable researches into the text of the Septuagint (Oxford, 1703), and the collection by Bernard de Montfaucon of the remains of Origen's Hexapla (Paris, 1713) were climaxed by the publication of the monumental Vetus Testamentum Graecum cum variis lectionibus edited by Robert Holmes and James Parsons (5 vols,, Oxford, 1798-1827). The variant readings of about three hundred separate

1 Namely, Cod. Vat. gr. 330 (= Holmes 108). New Testament Tools and Studies IV

Lucianic Recension p.8

codices, of which twenty are uncial, are given. In addition to patris tic citations, evidence is also supplied from the Old Latin, the Coptic, Arabic, Slavonic, Armenian, and Georgian versions, obtained partly from manuscripts, partly from printed texts. This immense apparatus now made it possible to group manuscripts by families; indeed, it became necessary to do so if only to bring some kind of order out of the chaotic mass of evidence.

During the latter part of the nineteenth century, through the researches of Ceriani, Field, and that polymathic scholar, Lagarde, 1 a beginning was made in the assigning of extant manuscript wit nesses to each of the major recensions of antiquity. In the case of the Lucianic text, two touchstones were available for identifying the paternity of variant readings. One was the frequent agreement between this text and the quotations of the Fathers of the fourth and fifth centuries who almost certainly used the Antiochian Bible, in particular Chrysostom and Theodoret. A second means of identi fying certain individual readings was supplied by the presence of the siglum και λ which is found prefixed to marginal readings in several Greek manuscripts, as well as the letter lomadh (ל) marking variants in certain Syriac manuscripts. Although in some instances the Greek siglum is to be interpreted as meaning και λοιποι, most scholars are agreed that in other instances it is to be resolved as και Λουκιανος. The key to this resolution of the siglum was discovered in the 19th century in a note 2 prefixed to the Arabic translation of the Syro-Hexaplar, which states,

"Lucian compared with the greatest care these Hebrew copies, and if he found anything different or superfluous he restored it to its place, prefixing to the part which he emended the initial letter L."

In the same passage reference is made to the marks by which the readings of Aquila,

1 The literary activities of this scholar were immense. He published books in no less than ten languages. As the Prorector of the University said at his funeral, probably no one of his colleagues could spell out the alphabets of all the languages in which Lagarde had edited texts; see p. 170 of the address delivered by George Foot Moore, entitled "Paul Anton de Lagarde," on the occasion of the opening of the Lagarde Library in the University of the City of New York, April 29, 1893 (The University Quarterly, July, 1893, pp. 166-179). An (incomplete) bibliography of his publications, compiled by R. J. H. Gottheil (Proceedings of the American Oriental Society, 1892, pp. ccxi-ccxxix), includes 297 major publications.

2 A Latin translation of this Arabic note is found in Field, Origenis Hexa plarum quae super sunt ... fragmenta (Oxonii, 1875), pp. Ixxxiv sq.; cf. Giovanni Card. Mercati, "Di alcune testimonianze antiche sulle cure bibliche di San Luciano," B t xxiv (i943), 1-17, especially pp. 7ff.

Lucianic Recension p.9

Symmachus, Theodotion, and the other Greek versions are denoted; and then the writer proceeds:

but if [the letters] elif , vaw, ra [א ר ו] are used, these are Origen's readings; if the letter lomadh, that is Lucian's."

Making use of this aid in identifying Lucianic readings, scholars were able to isolate and classify witnesses to the Antiochian recen sion. Thus, for the historical books of the Old Testament Ceriani l and Field 2 discovered that the text of the cursive Greek manuscripts 19, 82, 93, 108 agrees frequently with the form of text quoted by the Antiochian Fathers, and that these same manuscripts contain readings marked as Lucianic in the Syro-Hexaplar. Working in dependently Lagarde had come to almost the same results, and on the basis of evidence from these four manuscripts plus another, codex 118, he reconstructed the Lucianic recension of about one half of the Old Testament (Genesis to II Esdras, and a double text of Esther). 3 Unfortunately, except for the Book of Esther and, in another publication, the first fourteen chapters of Genesis, Lagarde provided no apparatus criticus. Thus, the scholar who wishes to check the variants in order to evaluate Lagarde's judgment must still go to Holmes and Parsons' thesaurus of variant readings.

In the Prophets another group of manuscripts has been found to contain the text of Lucian. Field satisfied himself that codices 22, 36, 48, 51, 62, 90, 93, 144, 147, 233, 308 offer in more or less pure form the Antiochian text. Later scholars, however, have criticized Field's grouping, and some of the manuscripts have been removed from his list of Lucianic witnesses. Thus, Cornill 4 struck out four (62, 90, 147, 233), and in this he was supported by Lagarde. In the Minor Prophets, the doctoral research of a young Dutch scholar, Schuurmans Stekhoven, indicated a slightly different grouping of manuscripts (22, 36, 42, 51, 62, 86, 95, 147, 153, 185, 238, 240, and in Zech. ch. 13 also 231) , 5 He also pointed out that they do not all supply the Lucianic text in an equally pure form.

During the first decade of the twentieth century a group of scholars in Germany, many of them under the leadership of Alfred

1 Monumenta sacra et profana, n, 2 (1864), PP7 6 > 9 8 > IO2 > Rendiconti del R. IstiMo Lombardo, Ser. 2, vol. xix (1886), 206 flf.

2 Op. cit., p. Ixxxvii.

3 Librorum Veteris Testamenti canonicorum pars prior graece (Gottingen, 1883).

4 C. H. Cornill, Das Buck des Propheten Ezechiel (Leipzig, 1886), pp. 65-66.

5 J. Z. Schuurmans Stekhoven, De alexandrijsche vertaling van het Dode kapropheton (Leiden, 1887), p. 44.

Lucianic Recension p.10

Rahlfs, Lagarde's successor at Gottingen, began a systematic investigation of the Lucianic text within certain books of the Bible. For example, Johannes Dahse attempted a classification of the manuscripts in Genesis, and assigned codices 53, 56, 129 to Lucian. 1 His reasons for this assignment, however, rested on a narrow selec tion of evidence, and subsequent scholars have indicated their dissatisfaction with his judgment. On the basis of a large induction of evidence, Ernst Hautsch found that, for the Book of Joshua, the quotations of the Antiochian Fathers agree with codices 44, 54, 75, 76, 84, i6, 134, while in the Book of Judges, the Bible of Theodoret (and Chrysostom) is most purely represented in codices 54, 59, 75, with which group 44, 82, 84, 106, 134 are frequently jointed. 2 For the books of the Pentateuch, Hautsch found no over whelmingly clear distribution of manuscripts that had previously been considered to be Lucianic; indeed, the most assured conclusion he could draw is that certain manuscripts which had been previously thought to be Lucianic (e.g., b, w, 108) do not in fact represent this recension. Procksch continued the research on the history of the Septuagint text of the Prophets. 3 He concluded that in Isaiah the Lucianic text is represented in manuscripts 22, 36, 48, 51, 93, 144, 308; that in Jeremiah it is in manuscripts 22, 36, 48, 51, 96, 144, 229, 231; that in Ezekiel it is in manuscripts 36, 48, 51, 231; and that in the Minor Prophets it is in 22, 36, 48, 51, 93, 95, 9 6 > **4> I 30> 153, 185, 240, 308, 311.

The most vigorous and thorough-going investigation of the Sep tuagint text during the twentieth century was that undertaken by Alfred Rahlfs. His aim, like that of his master, Lagarde, was to distinguish among the mass of manuscripts the three principal recensions (those of Origen, Hesychius, and Lucian), and, from the agreements among these, to recover the original pre-hexaplaric Septuagint text. In the first of his Septuaginta-StuMen he examines in minute detail Theodoret's quotations from the Books of Kings and from II Chronicles. 4 Though in general his findings confirm the

1 "Textkritische Studien," Zeitschriftfur die alttestamentliche Wissenschaft, xxvm (1908), 1-21, 161-173.

2 Der Lukiantext des Oktateuch [ = Mitteilungen des Septuaginta-Unter nehmens der koniglichen Gesellschaft der Wissenschaften zu Gottingen, vol. 11 (Berlin, 1909)

3 O. Procksch, Studien zurGescUchte der Septuagint a: Die Pvopheten (Leip zig, 1910).

4 Septuaginda-Studien] Heft I, Studien zu den Kdnigsbuchem (Gottingen. 1904).

Lucianic Recension p.11

view of previous critics (namely, that Theodoret commonly quoted a text of the type represented in Lagarde's edition), Rahlfs dis covered that in a considerable number of passages Theodoret's text does not agree with that of Lucian. In answer to the question of the nature of the text underlying the recension of Lucian, he found that it stands closest to Vaticanus (13) and to the Ethiopic version in the older and purer form represented by Dillmann's codices S and A. Thus Lucian is sometimes, especially in I Kings, an important witness to this old, "pre-hexaplaric" form of text.

Rahlfs gave attention next to an examination of the text of the Greek Psalter. 1 With rigorous and scrupulously careful weighing of evidence, his research is a model of patient and exact scholarship. In the chapter devoted to Lucian's text, 2 beginning with the hint given in Jerome's letter to the Gothic churchmen Sunnias and Fretela, 3 Rahlfs examines the quotations from the Psalter in Jerome, Theodoret, and Chrysostom, and concludes that the Lucianic Psalter was widely used throughout the East, where it, indeed, had obtained the status of the "official" text of the Greek Church. This text also circulated, in more or less pure form, even in the West, and at Milan influenced a revision of the Old Latin Psalter. As it happens, however, no manuscript is extant today which contains a pure Lucianic text of the Psalter (codex Alexandrinus, for example, presents a mixed text).

In a subsequent part of Septuaginta-Studien, Rahlfs criticizes Lagarde's edition of the Lucianic text of the books of Samuel and Kings. 4 Nowhere in his edition did Lagarde set forth the principles which he had followed in constructing the text. From Lagarde's Nachlese, Rahlfs shows that in the Books of Kings Lagarde depended primarily on codex 93, but here and there introduced into a trans cript of this manuscript readings from other witnesses, besides making certain changes in proper names and in grammatical details without support in any manuscript whatever. Moreover, several inadvertent and arbitrary departures from the tradition also found their way into his text. The value of Lagarde's edition, therefore, as

1 Septuaginta-Studien; Heft n, Der Text des Septuaginta-Psalters (Gfcttin gen, 1907).

2 Op. cit., pp. 169-182. 8 See p. 5 above.

4 Septuaginta-Studien; Heft in, Lucian' 's Rezension der Konigsbucher (Gottingen, 1911) ; the substance of this monograph was awarded a prize by the Gottingen Academy.

Lucianic Recension p.12

Rahlfs points out, is that it gives in convenient form a general view of the character of the recension which it represents, but for the detailed study of the Lucianic text it is quite inadequate.

Using Lagarde's collations, checked by Holmes and Parsons, Rahlfs found that in the Books of Kings four manuscripts, which fall into two sub-groups, preserve the Lucianic text; they are codices 19, 108, and 82, 93. Tested by internal probability of readings, the second of these sub-groups proves itself to be markedly superior to the other group. (It will be recalled that 108 played a considerable part in the construction of the Greek text in the Com plutensian Polyglot, while Lagarde's text was based largely on 93). Rahlfs added as a major Lucianic witness codex 127. Related manuscripts are 56, 158, 245.

In subsequent publications Rahlfs continued to investigate the tangled history of the transmission of the Septuagint. In what is one of the most searching contributions ever made to the textual criticism of the Septuagint, in 1922 Rahlfs published 117 pages devoted to the textual history of the little Book of Ruth. 1 In the same year he issued in pamphlet form a Probe-Ausgabe of the Greek text of Ruth, 2 which opened a new era in the century-long work on the Septuagint. Rahlfs found the Lucianic recension to be preserved in codices 54, 59, 75, 82, 93, 314, and (from 4.11 to the end of the book) in 19 and 108.

Four years later the text of Genesis was published, being the first of a proposed sixteen-volume edition of the Septuagint. 3 Here the enormous mass of material and, more particularly, the lack of distinct lines of text-type prevented the editor from assigning in clear-cut fashion any codices to the Lucianic recension. The most that he was able to say is that in Genesis codex 75 is a representative of the text of Lucian, but that it contains strands of other text-types as well. 4 Rahlfs turned next to the task of editing the Psalms, and

1 Studie uber den griechischen Text des Buches Ruth ( = Mitteilungen des Septuaginta-Unternehmens, m, 2-3).

2 Das Buck Ruth griechisch, als Probe einev kritischen Handausgabe der Septuaginta (Stuttgart, 1922).

3 Septuaginta; Societatis Scientiarum Gottingensis auctoritate edidit Alfred Rahlfs; i, Genesis (Stuttgart, 1926). For a discerning essay on the proposed edition see P. L. Hedley in HTR, xxvi (1933), 57-72.

4 It is instructive that in his Preface to the edition of Genesis Rahlfs makes the statement that, though Lagarde's program of first reconstructing the three great ancient recensions of the Septuagint is correct in principle, yet in practice the enormous magnitude of the task prevents the attainment of

Lucianic Recension p.13

in 1931, as part X of the large edition, his Psalmi cum Odis appeared. The volume contains an extensive introduction, in which the author supplements and modifies the second part of his Septuaginta Studien (1907). The great bulk of witnesses fall into the category of the Lucianic recension, which, as has been mentioned earlier, was extremely widespread and which, in fact, became the authori tative text of the Psalter for the Greek Church.

Subsequent fascicles of the Gottingen Septuagint were prepared by Werner Kappler, Joseph Ziegler, and Robert Hanhart. The manuscripts which they found to be Lucianic in the several books thus far edited are as follows:

In I Maccabees (ed. Kappler, 1936) the Lucianic manuscripts are 64, 236, 381, 536, 728. A sub-group of Lucianic manuscripts includes 19, 62, 93, 542.

In the XII Prophets (ed. Ziegler, 1943) the Lucianic manuscripts are 22, 36, 48, 51, 231 (only a fragment), 719, 763. Two sub groups of Lucianic manuscripts are: (i) 62, 147, and (2) 46, 86, 711.

In Isaiah (ed. Ziegler, 1939) the Lucianic manuscripts are 22, 48, 51, 231, 763. Three sub-groups of Lucianic manuscripts are (i) 62, 142; (2) 90, 130, 311; (3) 36, 93, 96. Other manuscripts show sporadic Lucianic readings, e.g. 46, 233, 456, and 926.

In Ezekiel (ed. Ziegler, 1952) the Lucianic manuscripts are 22, 36, 48, 51, 96, 231, 763. Two sub-groups of Lucianic manuscripts are (i) 311, 538, and (2) V, 46, 449.

In Susanna, Daniel, Bel and the Dragon (ed. Ziegler, 1954), the Lucianic manuscripts are 22, 36, 48, 51, 96, 231, 763. Two sub-groups of Lucianic manuscripts are (i) 311, 538, and (2)88,449.

In Jeremiah, Baruch, Lamentations, and the Epistle of Jeremiah (ed. Ziegler, 1957), the Lucianic manuscripts are 22, 36, 48, 51, 96, 231, 311, 763. A sub-group of Lucianic manuscripts includes 62, 198, 407, 449.

In II Maccabees (ed. Kappler and Hanhart, 1959), the Lucianic

that ideal. For discussions of the problems by two scholars who are critical of the Lagardian program, see Alexander Sperber's "Probleme einer Edition der Septuaginta," in Studien zur Geschichte und JKultur . . . Paul Kahle (Leiden, 1935), pp. 39-46; the same author's study of "The Problems of the Septuagint Recensions/' JBL, LIV (1935), 73-92; as well as Paul Kahle in The Cairo Geniza (London, 1947), pp. 154 ff.; and ed. (1959), pp. 231 if. For a criticism of Kahle's position, see the article by Peter Katz, "Septuagintal Studies in the Mid-Century," in The Background of the New Testament and its Eschatology, edited by W. D. Davies and D. Daube (Cambridge, 1956), especially pp. 205-208.

Lucianic Recension p.14

manuscripts are 64, 236, 381, 534, 728. A sub-group of Lucianic

manuscripts includes 19, 62, 93, 542. In III Maccabees (ed. Hanhart, 1960) the Lucianic manuscripts are 64, 236, 381, 534, 728. A sub-group of Lucianic manuscripts includes 19, 62, 93, 347 (1.1-2.19), 542.

The above survey of scholarly attempts to identify and study the Lucianic text suggests something of the magnitude and com plexity of the problem. It will be understood that the Septuagint is not a unified version of the Old Testament, but a collection of independent translations of the several books or groups of books made at different times and places. Of some books there was more than one translation, and even in the case of individual books the hand of more than one translator can be discerned. 1 It was inevitable that during the centuries these translations should have been corrected, one by another, and all of them occasionally by the Hebrew which may or may not have been the same form of Hebrew text as that from which the book was originally translated. Possi bilities for additional contamination were accelerated by the publi cation in the second Christian century of three new Jewish trans lations, those of Aquila, Theodotion, and Symmachus. 2 Further more, the three recensions of the Septuagint prepared by Christian scholars Origen, Lucian, and Hesychius in the third century, so far from putting an end to the confusion, gave it a new impulse, It is therefore not surprising that today the manuscripts of the Greek Old Testament present a mixed form of text. Nor should the investigator imagine that it will be possible in every case to distin guish neatly ordered families of witnesses ; in his search for the Luci anic text he must be prepared to acknowledge that for some of the books of the Old Testament it has left no recognizable trace among extant manuscripts. 3

1 See, e.g., H. St. J. Thackeray in JTS, iv (1902/03), 245-266, 398-411, 578-585; vm (1906/07), 262-278; ix (1908), 88-98.

2 According to Samuel Krauss two other Jewish translations into Greek were prepared by Ben La'ana and by Ben Tilga (see "Two Hitherto Un known Bible Versions in Greek," Bulletin of the John Rylands Library, xxvn [1942-43], 97-105).

8 For the difficulties involved in the contamination of recensional and non recensional manuscripts, see the incisive comments of Heinrich Dorrie in his monograph, "Zur Geschichte der Septuaginta im Jahrhundert Konstantins," ZNW, xxxix (1940), 57-110. Raymond Thornhill, in his discussion of "Six or Seven Nations; a Pointer to the Lucianic Text in the Heptateuch with Special Reference to the Old Latin Version" (JTS, N.S. x [i959], 233-246), finds that on the whole the Lucianic text of the Pentateuch has been pre served in relatively few manuscripts.

Lucianic Recension p.15

B. The New Testament

The first scholar to propound the critical principle of classifying New Testament manuscripts by families was the eighteenth century Pietist, Johann Albrecht Bengel, Superintendent of the Evangel ical Lutheran Church of Wiirttemberg.With the view of reducing extant witnesses to the text into "companies, families, tribes, and nations" and thus to simplify the task of evaluating the merits of variant readings, he divided the documents first into two "nations"; these were, he held, the Asiatic, chiefly those written in Constan tinople and its neighborhood, and the African, which were more ancient and therefore more valuable than those of the Asiatic group, which he tended to disparage. 1

Apparently the first scholar to use the term "recensions" in referring to groups of New Testament manuscripts, as well as the first to identify one of these recensions with the work of Lucian of Antioch, was Johann Salomo Semler, the pioneer of Biblical criti cism at Halle. Though his work today is chiefly remembered for his part in pursuing the free investigation of the Canon of the Scrip tures, it was Semler who, taking up Bengel's ideas, classified New Testament manuscripts into two recensions, which he called the Oriental, or that of Lucian, and the Western or Egypto-Palestinian, which Origen produced. 2 A modification of this division of manu scripts was proposed by Sender's pupil, Johann Jakob Griesbach. In what Westcott and Hort acknowledged to be an important antecedent of their own position, Griesbach divided the witnesses of the New Testament into the Western, Alexandrian, and Constan tinopolitan groups.

At the beginning of the nineteenth century Johann Leonhard Hug, a Roman Catholic Biblical scholar at Freiburg im Breisgau, elaborated the theory that Lucian was responsible for one of the early recensions of the New Testament. 3

1 J. A. Bengel, "Prodromus Novi Testament! Graeci recte cauteque adornandi," which is prefaced to his edition of Chrysostom's de Sacerdotio hbri sex graece et latine (Stuttgart, 1725), and in which he sets forth his pro gram for his edition which appeared at Tubingen in 1734.

2 See Semler's edition of Joh. Jac. Wetstenii libetti ad crism atque interpre tationem Novi Testamenti (1766), to which he appended (pp. 167-206) his own "Spicilegium observationum de variantibus Novi Testamenti lectionibus in quo praecipua etiam ex Joh. Alb. Bengelii . . . recensentur."

3. J. L. Hug, Einleitung ^n die Schriften des Neiien Testaments (Stuttgart und Tubingen, 1808), 38.

Lucianic Recension p.16

According to Hug, by the middle of the third century the text of the New Testament in the general mass of codices had degenerated into the form exhibited by codex Bezae, the Old Latin, minuscules i, 13, 69, 124, the Sahidic, and several other witnesses. This uncorrected text, to which he gave the name κοινη εκδοσις ("common edition"), was then separately revised, he thought, by three scholars of the ancient Church, Origen, Hesychius, and Lucian.

The Lucianic recension spread from Antioch to Constantinople, and from there to many other places, including Thrace, where it was quoted by Theophylact.

According to Hug's textual investigations in the Gospels, Lucian's recension is contained in the later uncial manuscripts E F G H S V and in lectionaries 47 and 50, as well as in the great majority of minuscule manuscripts.

Making a comparison in Mark 4 of the readings of the Lucianic recension with the readings of the Hesychian recension (represented chiefly by manuscripts B C L ), Hug showed the distinctive nature of each of these two recensions, as well as the relationship of the Lucianic text with the Peshitta Syriac.

In the Acts of the Apostles Hug found that the Antiochian text is less close to the Peshitta than in the Gospels, and is contained in codices 101, 102, 103, 241, 464, and la 59, as well as in many other minuscules.

In the Pauline and Catholic Epistles, besides D E F G the principal manuscript of this class is the beautifully written uncial codex K, along with many minuscules.

In the Apocalypse this recension is found in codices 29, 241, 242, 2023, 2039, 2040.

Though Hug's enthusiasm for pressing the evidence into the neat categories of three recensions led him to make statements which a more cautious scholar could challenge (as was done, for example, by S. P. Tregelles 1), it remains true that in certain respects Hug was ahead of his time, perceiving, as he did, the antiquity of many readings preserved in the later minuscule codices 1, 13, 69, 124.

The next stage, which was partly one of retrogression, was dominated by the work of Johann Martin Augustin Scholz, 2 the

1 In vol. iv of T. H. Horne's Introduction to the ... critical Study and Knowledge of the Holy Scriptures, 15th edition (London, 1872), pp. 78 f.

2 J. M. A. Scholz, Novum Testamentumgraece ... i (Leipzig, 1830), pp. xv-xxi. Unlike Streeter, however, Scholz placed his chief emphasis upon certain external signs of provenance, such as details of palaeography, iconography, marginal notes, colophons, and evidence regarding local saints who were honored in menologia.

Lucianic Recension p.17

Roman Catholic Dean of Theology at Bonn and a former pupil of Hug. It must be said to his credit that he was the first to stress the mportance of seeking to ascertain the geographical localities represented by the several manuscripts, a point which B. H. Streeter was to elaborate in the twentieth century in his theory of 'local texts'.

After some tentative attempts at classifying manuscripts, Scholz came to adopt essentially Bengel's classification into two families, which he called the Alexandrian and the Constantinopolitan.

Being impressed by the large number of manuscripts which preserve the Constantinopolitan text, he unwisely preferred this to the earlier Alexandrian type of text. In his discussion of the textual scholars in the early Church, Scholz sought to neutralize the force of the condemnation of Lucian's works, mentioned in the Decretum Gelasianum, by observing that only such books are put on the Roman Index of Prohibited Books as have circulated far and wide (longe lateque), and that therefore this adverse notice of Lucian's work is in reality a testimony to the wide diffusion of his critical labors on the New Testament text. 1

Later in the nineteenth century the distinguished English textual critics, B. F. Westcott and F. J. A. Hort, adopted and slightly modified Griesbach's theory of three major ancient types of text of the New Testament. As is well known, they gave the name "Syrian" to the type of text which was formed about the beginning of the 4th century, and they tentatively associated Lucian with the production of this recension. 2

Since the Syrian text displays signs of having been produced by conflation of earlier texts, Westcott & Hort rejected its testimony in seeking to recover the earliest form of the text of the New Testament. 3 Though they did not supply lists of manuscripts which, in their judgment, present more or less purely the Antiochian type of text, they discussed the method of deciding whether a reading is pre-Syrian or not. 4

1 Op. cit., vol. i, p. xxiv.

2 The New Testament in the Original Greek, [Vol. ii], Introduction \and] Appendix (Cambridge, 1881), pp. 137 ff.

3 It is significant that Bishop John W. Burgon, who ardently and, at times, acrimoniously opposed Hort's estimate of the secondary character of the Syrian text, nevertheless acknowledged that Lucian revised the text of the New Testament; see his volume, The Revision Revised (London, 1883), p. 29.

4 Op. cit. t pp. 163 1

Lucianic Recension p.18

Readings which have little or no support in such uncials as א B C D L P Q R T Z (A in Mark) E (also 33) for the Gospels (and similar lists for other parts of the NT), are probably Syrian, and therefore are to be rejected. The isolation and rejection of this type of text, according to the judgment of James Hardy Ropes, "constitute the most important abiding result of nineteenth century textual criticism." 1

During the twentieth century by far the most thorough investigation of the Antiochian text was undertaken by Hermann von Soden. Designating it the K (Koine) text, Von Soden divided it into about 17 groups which present a progressive modification of the text in minor details, mostly stylistic in nature. The chief groups representing the gradual corruption of the #-text, according to Von Soden, are:

1) K1, which is found in 2 S V and about fifty minuscule codices.

2) Ki (that is, K with a mixture of the I-form of text), which is found in E F G H, etc., etc.

(3) Kx, which is found in the great mass of minuscule codices of the 11th and 12th centuries.

(4) Kr the latest form of the K-text, an officially prepared text, which is found especially in manuscripts of the 13th and later centuries, and which is furnished with a full apparatus for lectionary purposes.

In addition to these principal groups, Von Soden also distin guished a Ka-text, which is preserved in more than one hundred manuscripts, the oldest of which is Codex Alexandrinus. Characteristic readings of this text are often found in various codices of the I-text; 2 it is related to the texts used by the translators of the Peshitta and by Chrysostom, the Cappadocians, and Theodoret of Cyros. 3

Von Soden had no hesitation in identifying the K1 group, which is the oldest form of the K-text, with the recension of Lucian. 4 He also isolated the manuscripts which are provided with commentaries written by members of the Antiochian school. 5

1 The Text of Acts (=The Beginnings of Christianity, ed. J. F. Foakes Jackson and Kirsopp Lake, vol. in; London, 1926), p. cclxxvi.

2 Die Schnften des Neuen Testaments in ihrer altesten eweichbaren Text gestalt, r (Berlin, 1902), 850-893, 1160-1170, and 1460-1468. Von Soden believed that the X-text was used by Victor of Antioch in his commentary on Mark.

3 Ibid., p. 1471. For a detailed list of its characteristics, see pp. 1456-1459.

4 Ibid., pp. 718-721 and 765-774; cf. 1459-1469.

5 Ibid., pp. 249-257; cf. 535 ft.

Lucianic Recension p.19

Though von Soden assigned several small fragments which date from the (5th or) 6th century to the K-type of text, the earliest codex which contains this text complete dates from the 8th century (codex Ω). 1

As is well known, in his textual evaluation of variant readings Von Soden gave equal weight to evidence from each of his three main text types (the I H K texts), and therefore his critical edition is closer to the Textus Receptus than is true of the previous great critical texts, such as those of Tregelles, Tischendorf, Westcott /Hort, Weiss.

To a greater or less extent, Von Soden's methodology has influenced the textual labors of several Roman Catholic editors of the Greek NT, notably H. J. Vogels, M. J. Bover, and August Merk. 2

In the field of textual criticism the name of B. H. Streeter is linked to the development of the theory of "local texts" of the NT.

Though not the first to suggest the idea (Scholz, as was mentioned above, had thrown out suggestions which may be regarded as anticipatory in certain respects, and early in the 20th century Kirsopp Lake 3 had specifically called attention to the desirability of such investigations) , Streeter was the first who systematically collected evidence for the localizing of text-types at the great Sees of the ancient church, and was the first also to isolate the so-called Caesarean text. 4 In connection with his discussion of "The Revised Versions of Antiquity" 5 Streeter declared that:

1 IUd. t p. 713.

2 For an analysis of the textual complexion of these editions, see K. W. Clark's contribution to The Background of the New Testament and its Eschatol ogy, edited by W. D. Davies and D. Daube (Cambridge, 1956), pp. 29-42, and Kurt Aland's discussion of "The Position of New Testament Textual Criticism," in Studia Evangehca (= TU, LXXIII; Berlin, 1959), PP7*7~73*> especially pp. 719 ff.

3 Referring to the use of Von Soden's apparatus cnticus, Lake wrote that it would be profitable "to see whether it does not really point to the existence of several local texts in the second and third centuries" (Review of Theology and Philosophy, iv [1908/09], 295) ; see also Lake's remarks about the possi bility that fam. i and fam. 13 were "one ancient local text, which has suffered different degrees of corruption from mixture with the Antiochian text," Codex i of the Gospels and Its Allies (= Texts and Studies, vn; Cambridge, 1902), p. li.

4 Though here again he was anticipated as to terminology by Burkitt, who pointed out that "for St. Mark, at least, Dr. Hort's categories are too few We have ... an Alexandrian-Caesarean group" etc. ; so F. C, Burkitt in Edward A. Button's Atlas of Textual Criticism (Cambridge, 1911), P65.

5 Burnett Hillman Streeter, The Four Gospels, a Study of Origins (London, 1924; revised, 1930), pp. 111-127.

Lucianic Recension p.20

"nothing that has been discovered since [Westcott & Hort's work] appears to me to have weakened their case, so far as the main issue is concerned/' namely, that "the Byzantine text is an essentially revised text following sometimes one, sometimes another of the earlier texts." 1

Streeter, however, pointed out that there are some important respects in which Hort's view of the constituent elements in the Lucianic revision must be modified in the light of subsequent discovery. The newly found Sinaitic Syriac manuscript of the Gospels, as F. C Burkitt observed in the Additional Notes which he contributed to the second edition (1896) of Hort's volume of Introduction [and] Appendix 2 reveals still more plainly than did the Curetonian Syriac manuscript that there once existed an old text of Antioch, related to the Western type of text, and that therefore agreements between the Sy s and the Byzantine text deserve very careful scrutiny as to the possibility of their being original (a notable example is the reading ευδοκια (goodwill toward men) of Luke 2.14, which Burkitt, Streeter, and Ropes held to be preferable to ευδοκιας (to men of goodwill) of B א D). 3

In an Excursus to the notable monograph on "The Caesarean Text of the Gospel of Mark" 4 Kirsopp Lake gave attention to the Ecclesiastical Text of the NT. On the basis of a collation of 19 manuscripts, a test was made of the validity of von Soden's analysis of the sub-groups within the K-text. Among Lake's findings the following are of primary importance:

"We cannot at present distinguish anything which can be identified with Von Soden's K r, nor do we feel any confidence in his K 1 text as a really distinct text," 5

"More important is the case of Von Soden's Ka -family .... His general position is that Ka represents a revision of an I-text by K1. The reviser chose always the variant which had a parallel in the gospels. In details of spelling and the like he followed K1" 6

"The most important observation made by Von Soden is that the Ka-text was used by Theodoret, and in Luke and John by Chrysostom. He thinks that in Matthew and Mark Chrysostom preferred a K1-text, though he is inclined to ascribe to him the creation of the Ka-text... If it be true that Ka is the text of Theodoret, there is a presumption that it, as distinct from the Ecclesiastical, is the Antiochian text of the 5th century." 7

1 Ibid., p. 114.

2 p. 330.

3 Burkitt, op. tit.; Streeter, op. cit., p. 115; Ropes, HTR, x (1917), 52-56 (see also p. 36 below).

4 HTR t xxi (1928), 238-257.

5 Ibid., p. 341.

6 Ibid., p. 342.

7 Ibid., pp. 344-3452 JWd., PP347~34 8

Lucianic Recension p.21

The results of these investigations, are, as Lake observes, both instructive and disconcerting,

"because they substantiate fully only two of Von Soden's subdivisions of the K-text. Of course, the others may prove to be realities, but one would have supposed that the differences between K 1 , K x, and K r would have been plain in a collation of over one hundred MSS for any chapter. It is impossible to repress a doubt whether Von Soden's analysis of the K-text is not as incorrect as his classification of the I-codices." 2

The most significant of Lake's findings, so far as the present survey is concerned, 3 is that K a is the Lucianic recension and was used as the chief basis of the Ecclesiastical text (thus reversing Von Soden's theory).

The most important uncial witnesses of the K a group, according to Von Soden, are codices A K Π. Using the consensus of 2 out of these 3 manuscripts to determine roughly the readings of the K a -text where it differs from the Ecclesiastical text and the Textus Receptus, Lake found that the following manuscripts present a relatively pure K a -text: 1318, 1313, 1219, 1220, 1223, 1346 (these are given in descending order of purity of text). 4

It has often been stated by textual scholars that Chrysostom was one of the first Fathers to use the Antiochian text. 5 This opinion was examined by Jacob Geerlings and Silva New in a study based on evidence which, in default of a critical edition, was taken from Migne's edition of Chrysostom's opera* Their conclusions are that:

3 One of the general conclusions which Lake draws is concerned with the almost total absence of evidence of direct genealogy in extant codices. Except for Family i and Family 13, "there seem to be no groups of manuscripts which are conceivably descendants of a single lost codex. There are cognate groups families of distant cousins but the manuscripts which we have are almost all orphan children without brothers or sisters ... It is hard to resist the conclusion that the scribes usually destroyed their exemplars when they had copied the sacred books/ 1 ibid., pp. 348-349.

4 Ibid., p. 344.

6 For example, Alexander Souter, The Text and Canon of the NT (New York, 1913), p. 85, "Chrysostom was the first great writer to use the fully developed ecclesiastical text" (this sentence is retained unchanged in the second edition, revised by Williams, p. 77) .

6 "Chrysostom's Text of the Gospel of Mark" HTR, xxiv (1931), 121-142.

Lucianic Recension p.22

"Chrysostom's text of Mark is not that of any group of manuscripts so far discovered and classified -- His text of Mark, or rather the text which can faintly be perceived through his quotations, is a 'mixed text', combining some of the elements of each of the types which had flourished before the end of the fourth century" 1

Recognizing that this statement is open to misunderstanding, the authors emphasize that Chrysostom must not be linked with any one manuscript which is said to have a "mixed text."

"The only similarity is the fact of being mixed, not the mixture which results." 2 In other words, "Chrysostom's text of Mark is first of all one peculiar to himself and full of unattested variants."3

In another of the Gospels, that according to Matthew, Chrysostom seems to have followed a type of text which is preserved in the K* group of witnesses. 4

In the Pauline Epistles an earlier study by C. K. Gifford revealed that Chrysostom preserves not a few Western readings. 5

More recently Joachim Forster, on the basis of an examination of 1043 quotations found in 88 of Chrysostom's Homilies, found that they involve "eine nicht kleine Zahl von Lesarten des westlichen Textes und der Koridethi-Familie ()/' 6

In a subsequent study of one part of the Byzantine text, Mrs. Lake 7 found that K a (i.e., Family Π) does not include codex Alexandrinus, as Von Soden thought, but that both had

"...a common ancestor which differed very little from the text which is found today in Π, rather more from that of A." She continues: 'The reconstructed text of Family Π, therefore, represents a manuscript older than the Codex Alexandrinus and affords another witness to a text which must have existed in the early part of the 5th century, if not before. Moreover, both the text of Family Π and the Codex Alexandrinus were elements in the formation of the Ecclesiastical text, the more or less standard text of the Middle Ages, since it differs from each about equally and to the same extent that H differs from A." 8

1 Ibid., 141-142.

2 Ibid., 142,

3 Ibid. The authors, as will have been observed, are careful in each case to indicate that their conclusions refer only to Chrysostom's text of Mark. Though Von Soden thought that in Matthew Chrysostom had used the K 1 - text (Ω V etc.), he could not find material sufficient for defining the type of K-text used in Mark (op. tit., i, ii, 1460 f.).

4 Claude D. Dicks, "The Matthean Text of Chrysostom in his Homilies on Matthew," JBL, LXVII (1948), 365-376. This corrects Von Soden's view that Chrysostom used a K1-text in Matthew.

5 C. K. Gifford, Pauli epistolas qua forma legent Johannes Chrysostomus (Halle Diss., 1902), p. 77.

6 Joachim Forster, "Gerechtigkeit fur Lucian und die antiochenischer Text," Monatschrift fur Pastor altheologie, XLV (1956), 267-272.

7 Family Π and the Codex Alexandrinus, the Text according to Mark (SD, vol. v; London, 1937).

8 Ibid., p. ix.

Lucianic Recension p.23

What is the nature of the text represented in A and (Family) Π? Mrs. Lake indicates that perhaps the most important result of her investi gation is the emphasis which it lays on the process of mixture.

"The Caesarean text was described as a 'mixed' text, primarily a mixture of Neutral and Western. A and II are now found to be mixed texts. The mixture varies in detail but is about the same in kind." 2

The most immediate descendents of codex Π, she found, are minuscules 114, 1079, 1219, 1500; while a descendent of 1219 was an ancestor of 489, 1346, 1816. 3

What was the provenance of the common ancestor of Family II and A ? Without feeling any great confidence, Mrs. Lake hazarded the conjecture that this, and not the K 1 -text (as Von Soden thought), is to be identified with the Lucianic recension. 4

Several other scholars have given serious attention to the great amount of mixture of text types in the Byzantine text. In an article entitled, "The Complex Character of the Late Byzantine Text of the Gospels," Ernest Cadman Colwell analyzes the text of a 13th century codex, Gregory 574, named the Four Gospels of Karahissar.

2 Ibid., p. 71. Mrs. Lake continues as follows: "The question becomes: What texts are not mixed ? The answer must be, none of those now known. The degree of mixture varies. The mixture was made at different periods . . . The whole question is : which manuscript influenced which ? The solution rests upon patristic evidence if there be any. The only certainly unmixed text of each Gospel was its original text. There must once have been such an original, but only the vaguest guesses can be made as to its character."

3 Ibid., p. 29. It will be recalled that Wilhelm Bousset discovered the close relationship of K II and 489; see his Textkritische Studien zum Neuen Testament ( = TU, xi, 4; Leipzig, 1894), p. 134.

4 Fam. Π, pp. ix-x. In a subsequent article, Kirsopp and Silva Lake indulge in several other conjectures, which are stimulating in themselves, but for which there is no proof. Regarding the origin of the K 1 -text,

"We suggest that it may be linked with the renaissance of calligraphy which either started in, or found one of its chief centers in, the Monastery of the Stoudion [at Constantinople] in the ninth and tenth centuries ... If our suggestion has any validity, the K 1 -text text might perhaps be called the 'Macedonian' text, for Basil the Macedonian and his successors were the great encouragers of the Stoudion and of the movement which it inaugurated.

In the same way, K x may have been the fashionable text of the Comneni, and K r that of the Palaeologi" ("The Byzantine Text of the Gospels," Memorial Lagrange [Paris, 1940], p. 255).

For a study of the K r text, see also David O. Voss, "Is Von Soden's K r a Distinct Type of Text ?" JBL, LVII (1938), 311-316. Voss answers the question in the title of his article with a decided affirmative.

Lucianic Recension p.24

Colwell speaks of

"...the almost incredibly confused relationships of this one MS of the Gospels. The most striking feature of the text of 574 is variety. A very close relationship has been established between its text and that of more than a dozen other MSS. But it never agrees with all twelve at once, nor does it agree with any one MS for more than half a gospel." 1

According to the author,

"the two most astounding results of this study were (i) the remarkably large number of variants from the textus receptus, and (2) the absence of one clear-cut dominant group." 2

Two other studies of the Greek text in late witnesses also suggest its heterogeneous character. Ernest W. Saunders discovered half a dozen singular variants in a thirteenth century tetraevangelium at Duke University which agree with readings in the Commentaries of Theophylact of Bulgaria. 3

In a series of articles on the text used by Photius, the Patriarch of Constantinople, J. N. Birdsall refers to a most unexpected textual situation:

"[Photius] knew and habitually quoted . . . texts akin to those isolated by Von Soden and classified in his various /-groups, and denominated 'mixed texts' by Ropes in his study of the text of Acts/' 4

Indeed, Birdsall goes so far as to assert that

"in Photius' time the Byzantine text was not the domi nant text-type in Greek Christendom, and it was either unknown to him or not approved by him." 5

1 JBL, LIV (I935) 2I 4

2 Ibid., p. 213.

3 Ernest W, Saunders, "Studies in Doctrinal Influences on the Byzantine Text of the Gospels/' JBL, LXXI (1952), 85-92.

4 J. N, Birdsall, "The Text of the Acts and the Epistles in Photius." JTS, n.s. ix (1958), 278-291. The quotation is from p. 290.

5 Idem, "The Text of the Gospels in Photius." JTS, n.s. vii (1956), 42-56, 190-198, The quotation is from p. 198.

Lucianic Recension p.25

C. Characteristics of the Lucianic Text

Since the time of Field it has been customary for Septuagint scholars to characterize Lucian's work as follows: 6

(1) Lucian filled up omissions in the Septuagint, showing a certain amount of freedom in handling the text. For example, in Isaiah 40.7-8 he supplied the deficiency in the Septuagint with the rendering of Symmachus and Theodotion, but with certain small changes. In Jeremiah 44.18, he inserted material from Aquila, again with small verbal changes in the Greek.

(2) He produced a large number of double, or conflate, readings; i.e. while retaining the Septuagintal reading he added a rendering that presumably expresses more closely the Hebrew text current in Antioch, which may well have differed from that current in Alexandria several centuries earlier. Thus, in I Sam. 12.2 the Hebrew has "I am old and grayheaded," which the Septuagint, adopting another pointing of the same consonants, renders "I am old and will sit down" (γεγηρακα και καθησομαι). Here Lucian kept the Septuagint and inserted (perhaps from Aquila) και πεπολιωμαι to render the Hebrew. Another instance is found in Isaiah 24.23 where the Hebrew has "the moon shall be confounded, and the sun ashamed," while the Septuagint renders "the brick shall waste away and the wall shall fall." Lucian kept the latter reading and added from Symmachus a literal translation of the Hebrew. Again, in II Kings 20.3, where the Septuagint renders the phrase

with εν αληθεια, Lucian, retaining this rendering, added και εν πιστει, which is probably to be regarded as an alternative translation oi the pregnant Hebrew phrase.

(3) He introduced a number of interpolations to serve as explana tions (e.g. proper names are substituted for pronouns), or to make the connection clearer, or to smooth out instances of grammatical harshness. Thus, in II Samuel 12.1 Nathan begins his parable to David simply, "There were two men," but Lucian inserted the words "Tell me this judgment: there were two men." Again in I Kings 15.23 the Septuagint (with the Hebrew) says of Asa, "in the time of his old age, he was diseased in his feet," which in Lucian appears in the form, "in the time of his old age Asa did wickedly and was diseased in his feet." In the Book of Ruth, Lucian added the proper name Nos^eCv in 1.7; 2.2, 3, 10, 17; 3.6, 7, 14, 16.

(4) He substituted synonyms for many words employed by the Septuagint. In some of these cases it is difficult to discover the reason for the alteration, as
φρονησις for σοφια,
εγενετο for ην,
διεβη for παρηλθεν,
δουλοι for παιδες,
εξειλατο for ερυσατο,
etc., etc.

In other cases it appears that, acting under the influence of the Atticizing tendency of grammarians of the time, Lucian replaced Hellenistic forms of the
Septuagint (such as ελαβοσαν, ειπαν, το ελεος, εγενηθη) with those of
Attic usage (ελαβον, ειπον, ο ελεος, εγενετο). 1

The critical principles and methods which Lucian followed in making his recension of the Old Testament are plainly observable in the Antiochian text of the NT. Indeed, Ropes declares, 'There is not one of the well-known characteristics of the Antiochian New Testament which cannot be illustrated from the Old Testament of Lucian." 2 Hort's comprehensive and elegant summary of these characteristics is a classic description:

"The qualities which the authors of the Syrian text seem to have most desired to impress on it are lucidity and completeness. They were evidently anxious to remove all stumbling-blocks out of the way of the ordinary reader, so far as this could be done without recourse to violent measures. They were apparently equally desirous that he should have the benefit of instructive matter contained in all the existing texts, provided it did not confuse the context or introduce seeming contradictions. New omissions accordingly are rare, and where they occur are usually found to contribute to apparent simplicity. New interpolations on the other hand are abundant, most of them being due to harmonistic or other assimilation, fortunately capricious and incomplete. Both in matter and in diction the Syrian, text is conspicuously a full text. It delights in pronouns, conjuctions, and expletives and supplied links of all kinds, as well as in more considerable additions.

As distinguished from the bold vigour of the 'Western' scribes, and the refined scholarship of the Alexandrians, the spirit of its own corrections is at once sensible and feeble. Entirely blameless on either literary or religious grounds as regards vulgarised or unworthy diction, yet shewing no marks of either critical or spiritual insight, it presents the NT in a form smooth and attractive, but appreciably impoverished in sense and force, more fitted for cursory perusal or recitation than for repeated and diligent study." 3

6 Field, op. cit , pp. bcxxix if. ; for other discussions of the characteristics of the Lucianic recension of various Old Testament books, see W. O. E. Oester ley, Studies in the Greek and Latin Versions of the Book of Amos (London, 1902), pp. 61-67; Rahlfs, Der Text des Septuaginta-Psalters (Gottingen, (1907), p. 231 ;C.C.TorcQy, Ezra Studies (Chicago, 1910), pp. 106-109; O. Procksch, Studien zur Geschichte der Septuaginta: Die Propheten (Leipzig, 1910), pp. 79-87; Rahlfs, Lucians Rezension der Konigsbucher (Gottingen, 1911)* pp. 171-183, 239-288, 294; Rahlfs, Studie fiber den griechischen Text des Bwches Ruth (Berlin, 1922), pp. 83-90; and J. Ziegler, "Hat Lukian den griechischen Sirach rezensiert?" B, XL (1959), 210-229, especially p. 229.

1 Rahlfs, Lucians Rezension der Ko'nigsbucher (Gottingen, 1911), p. 294; nd Das Buck Ruth griechisch (1922), p. 13.

2 James Hardy Ropes, The Text of Acts (Cambridge, 1926), p. cclxxxiiL

3 Westcott and Hort, Introduction & Appendix, pp. 134 f.

Lucianic Recension p.26

It is scarcely necessary to set forth here examples which illustrate these types of editoral revision; Von Soden assembled a large number in order to describe the Koine text. 1

It will be sufficient to observe that, to judge by the pains which the reviser or the revisers obviously took in performing their task, their aim was not merely to bring an old manuscript up-to-date, but to prepare an exemplar from which semi-official copies could be made.


The influence of Lucian of Antioch as textual editor was felt far beyond his native country of Syria. As has been indicated in the previous pages, his recension of the New Testament was adopted at Constantinople and from there it spread widely throughout Greek speaking lands. Portions of his work on the Old Testament (notably the Psalter) became the official text of the Orthodox Church, and manuscripts of his recension of the rest of the Old Testament circulated alongside other forms of the Septuagint. But even beyond the limits of the Greek Orthodox Church, Lucian influenced the form of the New Testament and parts of the Old Testament which were used, and are still used, by millions who never heard of his name.

The first translation of the Bible into a Teutonic language dates from the second half of the 4th century and was made, as is well known, by Ulfilas, the apostle to the Goths. When one considers that it was probably at Antioch in Syria that he was consecrated bishop (about A.D. 341), 2 and that in A.D. 360 he was present at the synod of Constantinople, 3 one is not surprised that the basis of his version should have been the Antiochian-Constantinopolitan type of text.

This was proved by Kauffmann, who, in a thorough textual analysis of the very fragmentary remains of the Gothic Old Testament, showed that Ulfilas followed a Greek text which at some points was in close agreement with Lucian's recension of the Septuagint text. 4

1 Op. cit., I, ii, 1456-1459 (cf. 1361-1400), 1784-1787.

2 See C. A. A. Scott, Ulfilas, the Apostle of the Goths (Cambridge, 1885), pp. 47 f., and 115.

3 See Lagarde, Mitteilungen, iv (Gottingen, 1891), 21-23; and Librorum Veteris Testamenti canonicorum pars prior (Gottingen, 1883), pp. viii fi.

4 Friederich Kauffmann, "Beitrage zur Quellenkritik der gotischen Bibel ubersetzung." Zeitschrift fur deutsche Philologie, xxix (1897), 306-337.

Lucianic Recension p.28

Thus in Neh. 7.24 and 35 the Gothic version agrees with two manuscripts of the Lucianic recension (Holmes and Parsons 93, 108) in the names 'Ασσομ and 'Ηιραμ, agreements found in no other known manuscripts of the Septuagint. 1

On the other hand Kauffmann also collected instances in which the Gothic departs from Lucian and agrees with the B-text of the Septuagint. He concluded, therefore, that the Gothic does not preserve the original Lucianic text, but one derived from it. This conclusion, however, was modified in 1912, when through the cleaning of certain Gothic fragments it became possible to read more precisely the script of a fragment (codex D) which preserves some verses of Neh. 7. It was discovered that no fewer than sixteen new readings could be deciphered, and that all of them agree with the Lucianic text.

Thus, both as to number and significance, this new evidence resulted in a reassessment of Kauffmann's evaluation of the textual character of the Gothic fragments of the Old Testament, which are now seen to be more thoroughly Lucianic than had been previously suspected. 2

As regards the New Testament, scholars generally agree that the type of text represented in the Gothic version is basically the Antiochian or Syrian form of text with a certain number of Western and non-Western readings embedded in it. 3

Von Soden was unable to find in Greek a precisely similar mixture of K and I readings, but he observed that its affinities seemed closest to his K a-type. 4 The Gothic version, therefore, appears to be the oldest extant representative of the Lucianic recension, despite subsequent infiltration of readings from the Old Latin version. 5

1 KaufEmann thought that the text of the brief fragments from Genesis in the Gothic version were also Lucianic, but the subsequent research by Hautsch (see p. 10 above) proving that Lagarde was in error in his recon struction of the Lucianic recension of the book of Genesis necessitates that a new analysis be made of the Gothic version for this book.

2 Wilhelm Streitberg, Die gotische Bibel, 3te aufl., i (Heidelberg, 1950), pp. xxxiv f .

8 E.g. Westcott & Hort, op. cit., p. 158; Friederich Kauffmann, "Bei trage zur Quellenkritik der gotischen Ubersetzung," Zeitschrift fur deutsche Philologie, xxx (1898), 144-183; xxxi (1898-99), 178-194; Paul Odefey, Das gotische Lucas-Evangelium. Ein Beitrag zur Quellenkritik und Textgeschichte (Flensburg, 1908); G. W. S. Friedrichsen, The Gothic Version of the Gospels, a Study of its Style and Textual History (Oxford, 1939) ; Streitberg, op. cit., pp. xxxv-xl; and H, Steubing, "Miscellen zur gotischer Bibelubersetzung des Ulfilas," Zeitschrift fur Kirchengeschichte, LXIV (1952-53), 137-165. On the difficulties of analyzing the Gothic text, see Friedrichsen, "The Gothic Text and the Fourth Century Byzantine Text/ 1 JTS, xxxix (1938), 42-44.

4 Von Soden, op. cit., I, ii, 1469.

5 See, now, Friedrichsen, Gothic Studies (Oxford, 1961), pp. 63 f. and 67 ff.

Lucianic Recension p.29

The Slavonic Version

Besides influencing the Gothic version, the Lucianic recension made itself felt also on the first translation of the Scriptures into a Slavic language, that of Sts. Cyril and Methodius in Old Church Slavonic. In view of the contacts which these two apostles to the Slavs had with Constantinople, such dependence is not unexpected. 1 Those who have examined the question as regards the Old Testa ment report that the amount of Lucianic influence on the Slavonic text varies in different books. On the basis of a limited number of passages, the author of an unsigned article in the Church Quarterly Review 2 found that in the books of Kings the version commonly agrees with Lucian against manuscripts A and B, while for other books it agrees with A and B against Lucian.

A definitive examination of the textual complexion of the Old Slavonic Psalter was made by Josef Vajs. On the basis of a critical analysis of 529 pas sages Vajs discovered that in 449 instances the Old Slavonic version is purely Lucianic in character, while the remaining 80 variants show influence from the Latin Vulgate text (as is not uncommon in Croatian manuscripts). 3

In the case of the NT, several scholars have examined the textual relations of the Old Slavonic version. 4 Though they differ somewhat as to details, the consensus of their findings seems to be that the predominant character of the Old Slavonic New Testament is derived from the Antiochian or Byzantine text, with a certain number of readings from other families of texts. On the basis of an examination of about 2500 variant readings in the Slavonic tetra evangelium, Vajs concluded that almost one half of them belong to the Antiochian recension, about a fifth to the Western, and an even smaller proportion to the Alexandrian.

The Old Church Slavonic Bible formed, as is well known, the basis for other translations into several Slavic languages of the

1 Lagarde remarks in the preface to his edition of Lucian, "Ni omnia fallunt, Slavos nihil aliud vertit nisi Luciani recensionem," p. xv, and in his Mitteilungen, n (Gottingen, 1887), 53, he expresses the need for further examination of the relation of the Slavonic version to Lucian.

2 "Lucian's Recension of the Septuagint," CQR, LI, (1900-1901), 388 f. 8 "Kter6 recense byla fecka pfedloha staroslovSnsk6ho pf ekladu 2altafe,"

(with a French resum.6), Byz., vni (1939-1946), 55-86. In the case of the Book of Ruth, however, Vajs discovered that its text goes back, not to the Lucianic recension but apparently to the Hesychian; Josef Vajs, Kniha, Rut v preklade staroslovanskem ( = Kriticke studie staroslovanskeho texta biblickeho, II ; Prague, 1926) ; cf. Alfons Margulies in Archiv fur slavische Philologie, XLII (1928), 52. 4 For a survey of this research, see pp. 79 ff. below.

Lucianic Recension p.30

past and present. 1 Thus in modified form the Antiochian or Constantinopolitan recension has formed the basis of the NT and the Psalter for millions of Slavic peoples.

As regards the history of the printed form of the Greek New Testament, the so-called Textus Receptus, which was based chiefly on manuscripts of the Antiochian recension, has been reprinted, with only minor modifications, in almost one thousand editions from 1514 down to the 20th century. 2

When one considers how many translations into the vernaculars of Europe, Asia, Africa, and South America have been based on the Greek Textus Receptus, or on a translation which in turn was rendered from the Textus Receptus of the New Testament (such as the King James version or Luther's translation), it will be appreciated how enormous has been the influence of Lucian's recension, made in Antioch about the turn of the 3rd and 4th centuries of the Christian era.


Westcott and Hort's view that the Syrian text of the New Testa ment is worthless for the recovery of the original text has left its mark, not only on their own edition, but on much subsequent textual analysis. As regards the Old Testament Eberhard Nestle declared flatly, "The recension of Lucian is quite the most useless for those objects for which we use and need the Septuagint most." 3

1 For a list of these, see T. H. Darlow and H. F. Moule, Historical Catalogue of the Printed Editions of Holy Scripture in the Library of the British and Foreign Bible Society, 2 vols in 4 parts (London, 1903-1908), and British Museum General Catalogue of Printed Books, vols, xvi-xvin (London, 1936-3?)

2 For a classified list of printed editions of the Greek New Testament, see Eduard Reuss, Bibliotheca Novi Testamenti Graeci (Brunsvigae, 1872). This list is supplemented by Isaac H. Hall in Philip Schaff, A Companion to the Greek New Testament and the English Version, 3rd ed. (New York, 1889), pp. 497-524. For a scathing indictment of British scholarship which ac quiesced to using an outdated Greek text, see Eberhard Nestle, "The Present Greek Testaments of the Clarendon Press, Oxford," JTS, v (1904), 274-279, with a weak defense by W[illiam] S[anday], ibid., pp. 279 f. For a summary of a series of articles against modern critical editions of the New Testament and in favor of the Ecclesiastical Greek text, written by Prof. Ivanov of the Moscow Theological Academy and published in Zhurnal Moskovskoi Patriarchii (1954-1956), see Robert P. Casey, "A Russian Orthodox View of New Testament Textual Criticism," Theology, LX (i957) 5-54

8 Nestle made the statement first in his Septuagintastudien, I (Ulm, 1886), 9. He repeated it (ibid., n [Ulm, 1896], 12) with a comment of approval from Franz Delitzsch, and explained his meaning more fully in "Zur Rekonstruk tion der Septuaginta," Philologus, LVIII (1899), 121-131.

Lucianic Recension p.31

It appears to the present writer, however, that these unfavorable estimates of the value of the Antiochian text must be at least partially revised in the light of critical study of what may be called (for the want of a better name) the Ur-Lucianic text. Let us begin first with the Old Testament.

A. The Old Testament

It is a curious fact that certain readings which have been generally regarded as typical of Lucian's recension of the Greek Old Testament occur in texts and authors that are earlier than Lucian. The following is a list of seven such pre-Lucianic witnesses to a form of text which, at least in part, resembles the Lucianic recension.

(i) Although not all scholars who have investigated the subject are in agreement as to how to explain the data, there seems to be substantial evidence to prove that various parts of the Old Latin version of the Old Testament contain Lucianic readings.

This was noticed first by Ceriani in connection with Lamentations, 1 and then by Vercellone, 2 who observed that when the glosses in the margin of the Leon manuscript depart from the ordinary Septuagint text they agree with the readings of the Lucianic group 19-82-93-108. Other scholars have called attention to the same type of text in other parts of the Old Latin Bible. Thus, Burkitt found that:

"the Old Latin in the Prophets sometimes supports 'Lucianic' readings. This fact proves that among the constituents of the eclectic text most used by the Antiochene Fathers of the fourth century there was an ancient element akin to the Old Latin, but quite indepen dent of our leading MSS codd. A א B ." 3

For the Books of Samuel, S. R. Driver wrote,

"The Old Latin is a version made, or revised, on the basis of manuscripts agreeing closely with those which were followed by Lucian in framing his recension. The Old Latin must date from the second century A.D. ; hence it cannot be based on the recension of Lucian as such; its peculiar interest lies in the fact that it affords independent evidence of the existence of MSS. containing Lucian's characteristic readings (or renderings), considerably before the time of Lucian himself." 4

1 A. M. Ceriani, Monumenta sacra et profana, I, i (Milan, 1861), p. xvi (addenda).

2 Carlo Vercellone, Variae lectiones Vulgatae latinae Bibliorum editionis, II (Rome, 1864), xxi-xxii, 179; cf. i, pp. xciii-xcv.

3 F. C. Burkitt, The Book of Rules of Tyconius ( = TS, III, i ; Cambridge, ) p. cxvii.

4 S. R. Driver, Notes on the Hebrew Text and the Topography of the Books of Samuel, 2nd ed. (Oxford, 1931), p. Ixxvi.

Lucianic Recension p.32

The Belgian scholar, Dieu, sought to explain these parallels in terms of scribal activity in replacing the original form of the Old Latin quotations with a form that resembles the text of Lucian. 2 Although a certain number of Lucianic glosses may have been introduced here and there into one or more Old Latin witnesses, the wide variety of evidence makes it difficult to explain all the data in this way. According to the investigation of Montgomery, 3 the Old Latin text of Daniel likewise displays Lucianic readings, and Haupert, one of Montgomery's students, found that the situation is similar for the Books of Kings. 4

These phenomena in Old Latin manuscripts appear to be corro borated by quotations of the Bible made by Latin authors who lived prior to Lucian. Although Rahlfs had concluded that no Latin author before Lucifer of Cagliari (died 371) cited characteristic Lucianic readings in the Books of Kings, 5 Capelle, in a monograph on the text of the Latin Psalter in Africa, discovered that both Tertullian and Cyprian show a certain amount of acquaintance with a pre-Lucianic form of text of the Psalms. 6 In a recent analysis of Cyprian's citations from the four Books of Kings, Fischer 7 found that Cyprian agrees with Lucian in those readings which are linguistic corrections or which otherwise improve the Greek text according to the Hebrew, but that, quite understandably, he does not agree with Lucian when the latter takes over hexaplaric variants. Furthermore, in many cases where Cyprian agrees with Lucian, other Old Latin witnesses are corrected to the Septuagint.

2 L. Dieu, "Retouches Lucianiques sur quelques textes de la vieille version latine (I et II Samuel)," RB, n.s., xvi (1919), 372-403.

3 James A. Montgomery, A Critical and Exegetical Commentary on the Book of Daniel (New York, 1927), pp. 54-55.

4 R. S. Haupert, The Relation of Codex Vattcanus and the Lucianic Text in the Books of the Kings from the Viewpoint of the Old Latin and the Ethiopic Versions (Univ. of Penna. Diss., 1930), pp. 36 f. Whether these data prove that the Old Latin Version was made originally at Antioch in Syria, as some have thought, need not be examined here; for a classic discussion of the problem, see H. A. A. Kennedy, "Latin Versions, the Old," Hastings' Dic tionary of the Bible, in (New York, 1900), p. 54.

5 Lucians Rezension der Konigsbiicher (Gottingen, 1911), pp. 158 ff.

6 Paul Capelle, Le texte du Psautier latin en Afrique ( = Collectanae b^bl^ca latina t iv; Rome, 1913), p. 204.

7 Bonifatius Fischer, "Lukian-Lesarten in der Vetus Latina der vier KQnigsbucher/ 1 Miscellanea biblica et orientalia R. P. Athanasio Miller . . . oblata (= Studia Anselmiana, xxvii-xxviii; Rome, 1951), 169-177.

Lucianic Recension p.32

The wide distribution of Old Latin evidence and the general consen sus among scholars that the origin of the Old Latin version of the Old Testament dates from about the second century A.D. make the conclusion inevitable that the Greek text lying behind the Old Latin was one element, and perhaps one of the more important elements, from which the composite Lucianic text was constructed,

(2) The Peshitta version of the Old Testament also exhibits numerous readings which find a parallel in the recension associated with Lucian. At the end of the last century Stockmayer 1 found more than a score of readings in I Samuel where Lucian agrees with the Peshitta against the Masoretic text and the current Septuagint text. Although the exact date of the translation of the Old Testament Peshitta is not known, most scholars believe that it was made in the second or third century of the Christian era. Thus, at least in the Books of Samuel, it too affords evidence of Ur-Lucianic readings.

One is not limited, however, to versional evidence testifying to the existence of an Ur-Lutianic text. Several pieces of Greek evidence point in the same direction; they are the following:

(3) A papyrus fragment (Rahlfs' no. 2054) , 2 dating from the third (or possibly the second) Christian century, contains the Greek text of Psalm 77.1-18 in a form which exhibits several significant agreements with the Lucianic text, some of which are against aU other witnesses cited by Rahlfs in his edition of Psalmi cum Odis. In other words, fully half a century before Lucian made his recension in Syria, a Greek text circulated in Egypt which anticipated certain of Lucian's characteristic readings.

(4) At the middle of the second century Justin Martyr cited the Old Testament in a form which Bousset found to agree frequently with the Lucianic recension; moreover, this agreement, Bousset declared, is "nicht nur in einzelnen Stellen, sondern in weiterem Umfange." 3

1 Theodor Stockmayer, "Hat Lucian zu seiner Septuagintrevision die Peschito beniitzt?" Zeitschrift fur die alttestamentliche Wissenschafi, xi (1892), 218-223.

2 The fragment, which is 240 in the Greek and Roman Museum in Alexan dria, was edited by M. Norsa in Bulletin de la SocidU royale d'archdologii d'Alexandrie, xxn (1926), 162-164.

3 Wilhelm Bousset, Die Evangeliencitate Justins des Martyrers in ihren Wertfur die Evangelienkritik (GSttingen, 1891), p. 20.

Lucianic Recension p.34

Puzzled as to how to explain the apparent anachronism, Bousset felt compelled to suppose that scribes in transmitting Justin's works brought his quotations into harmony with the prevailing Antiochian text of the Old Testament. Schiirer, however, in his review of Bousset's monograph pointed out that not every reading which is found in Lucianic manuscripts is later than Justin. 1

(5) At the end of the 1st Christian century Josephus had before him a Greek copy of the Books of Samuel which, according to the research of Mez, diverged widely from codices A and B and habitually agreed with the text of Lucian, following this text even against the Hebrew. 2

Rahlfs re-examined with great care the evidence presented by Mez, extending the scope of the investigation to Josephus's text of the Books of Kings. His conclusion was that Mez exaggerated the measure and significance of the agreement between Josephus and Lucian, but that, particularly in the Books of Samuel, some readings which appear in the Lucianic recension were current at a much earlier time. 3 Going far beyond this cautious evaluation, Thackeray (who showed no acquaintance with Rahlfs's monograph) asserted roundly: "The Josephan Biblical text is uniformly of this Lucianic type from I Samuel to I Maccabees." 4

(6) In an analysis of the Old Testament quotations in the New Testament, Staerk discovered that here and there New Testament citations diverge from all the major codices of the Septuagint and agree with the Lucianic text. 5

It must be admitted, however, that this evidence is not conclusive, for we can never be sure on which side the borrowing may lie (i.e., the Byzantine scribe of the Lucianic codices may have conformed the Old Testament text to the Antiochian form with which he was familiar in the NT quotations).

1 Theologische Liter aturzeitung, xvi (1891), 67.

2 Adam Mez, Die Bibel des Josephus untersucht fur Buch V-VII der Archd ologie (Basel, 1895), P80.

3 Rahlfs, Lucians Rezension der Kdnigsbucher, pp. 80-111.

4 Henry St. John Thackeray, Josephus , the Man and the Historian (New York, 1929), P85. In the Preface to the Cambridge edition of the Septuagint text of the Books of Samuel, Thackeray declares that from I Sam. 8 onwards "Josephus becomes a witness of first-importance for the text of the Greek Bible . . . His main source is a Greek Bible containing a text closely allied to that of the 'Lucianic' group of MSS,, but anterior by more than two centuries to the date of Lucian" (p. ix) .

5 W. Staerk, "Die alttestamentlichen Citate bei den Schriftstellern des Neuen Testaments," Zeitschrift fur wissenschaftliche Theologie, xxxv (1892), 464-485; xxxvi, r (1893), 70-93.

Lucianic Recension p.35

Nevertheless, it is pertinent to observe that the very widespread interpolation of the Pauline catena in Romans 3.13-18 after Psalm 13 (14). 3, an addition which Jerome says 1 was current in the widely-used text of his day and which Rahlfs even prints as part of the Septuagint text, did not appear in the Lucianic recension, nor did the scribes of these manuscripts succumb to the temptation to add it.

(7) In the John Rylands Library at Manchester there are fragmentary remains of a papyrus scroll containing Deuteronomy 23.25; 25.2-3; and 26.18 in Greek. 2 The text of these tiny fragments, which date from about the middle of the second century B.C., appears to be related to the Lucianic form of the Greek Bible. 3

From these seven items it can be seen that various texts and authors earlier than Lucian of Antioch present readings which agree with what is believed to be the Lucianic recension of the Greek Old Testament. 4 The conclusion which one must draw is that, despite the numerous secondary features which Lucian introduced into his recension of the Old Testament, one may expect to find here and there in it certain readings, not extant in the other forms of the Septuagint, which will be useful in ascertaining the most ancient form of the Hebrew text.

1 For a discussion of the correct text of Jerome's comment on this inter polation, see E. F. Sutcliffe, 'The koine, 'diversa 1 or 'dispersa' ? St. Jerome, P.L. 24, 548 B," B, xxxv (1955), 213-222. It is curious that elsewhere Jerome calls the Lucianic recension the 'koine' text; see p. 5 above.

2 It was edited by C. H. Roberts, Two Biblical Papyri in the John Rylands Library, Manchester (Manchester, 1936).

3 So Albert Vaccari, "Fragmentum Biblicum saeculi II ante Christum," B, xvn (1936), 501-504; compare P. E. Kahle, "Problems of the Septuagint," in Studia Patristica, ed. by Kurt Aland and F. L. Cross, i ( = Texte und Unter suchungen, LXIII; Berlin, 1957), 328-338.

4 It may be mentioned here that Paul Wendland's careful examination of the Old Testament quotations in one of Philo's tractates and his conclusion that in a large proportion of cases the text of Philo agrees with Lucian and seldom joins other manuscripts against Lucian, cannot be accepted without being re-examined ("Zu Philos Schrift de posteritate Caini. Nebst Bemerk ungen zur Rekonstruktion der Septuaginta," Philologies, LVII [1898], 248 288). Wendland naturally made use of Lagarde's edition of the Lucianic text, but since this begins to be Lucianic only on p. 259, line 3, with Ruth 4.11 (so Rahlfs, Studie . . . Ruth, pp. 77 f.), a comparison of Philo's Pente teuchal quotations with this edition counts for nothing. See Peter Katz, "Das Problem des Urtextes der Septuaginta," Theologische Zeitschrift, v (1949), 19 f., and his monograph, Philo's Bible (Cambridge, 1950), p. 12, note i.

Lucianic Recension p.36

B. The New Testament

In evaluating the critical worth of the Antiochian text of the New Testament, one finds a general disposition among scholars (except a few who have been influenced by Von Soden's methodology) to disregard readings in the NT which are supported solely or chiefly by this recension. The poor opinion which Westcott & Hort had of the Syrian text is shared by many, and doubtless there is much to justify such an evaluation. On the other hand, however, what was said above regarding the presence of ancient readings in the Lucianic recension of the Old Testament ought to make one cautious about rejecting off-handedly and as a matter of course every Antiochian reading in the New Testament.

In fact, since the time of Westcott & Hort, the acquisition of several new witnesses has tended to put the matter in a new light. Already in the second edition of Westcott & Hort's volume of Introduction & Appendix (1896), F. C. Burkitt proposed several minor modifications of their estimate of the Syrian text. In characterizing the text of the newly discovered Sinaitic manuscript, Burkitt wrote:

"This Eastern text, which does not survive in an approximately pure form in any known Greek MS, has thus affinities with both the great PreAntiochian groups headed by א B D respectively; it also stands in places against א B D united, entering not unfrequently as an independent constituent element into the Antiochian (Syrian) text. Most of the Antiochian readings which contain interesting matter and which are witnessed by neither א B nor D latt are found in the Old Syriac; and while there is no trace in the Old Syriac of the distinctively Antiochian conflations, there are several instances where the Antiochian text has been composed out of the mutually exclusive variants of א B D and the Latins on the one hand, and the Old Syriac, supported perhaps by a few cursives, on the other." 1

As examples of Antiochian readings which contain interesting material, and are not supported by א B or D latt, Burkitt cited the omission of ουδε ο υοις in Matthew 24.36;
the addition of και ταραχαι in Mark 13.8;
ευδοκια in Luke 2.14; and
εδει substituted for γεγραπται in Luke 24.46 2

While perhaps Burkitt would not have gone so far as to accept the Antiochian form of all of these examples as the original text, he saw clearly that one must not reject a Syrian reading out of hand merely because it is not supported by either א B or D latt.

1 Westcott and Hort, The New Testament in the Original Greek, [vol. ii] Introduction Appendix, 2nd ed. (London, 1896), p. 330. 2 Ibid.

Lucianic Recension p.37

A few years later, in his discussion of the relation of the Old Syriac and the Antiochian text, Burkitt again pointed out that, while in general the Textus Receptus "has but little affinity with the Old Syriac Version," yet "in a few cases, some of them of considerable importance, the Old Syriac does agree with the later Greek MSS against the early Western and Alexandrian evidence." 1

He cited three examples to illustrate his point,
ευδοκια of Luke 2.14,
αριστον in Luke 14.15, and
αναβοησας of Mark 15.8. Of these three, the last seemed to him to have the greatest possibility of being the original.

Since, however, the Syr s and Syr c are both incomplete, and since they would supply, even if they were complete, but slender evidence for the Ur-Lucianic text at Antioch, it is not unreasonable to suppose that some of the readings of the Lucianic text which do not appear in the Old Syriac may have been derived from the old text of Antioch. Concerning the possibility of detecting pre-Lucianic elements, Streeter wrote:

"We have no means of identifying those readings of the old text of Antioch, which survive in the Byzantine text, but which do not happen to occur in the Old Syriac, except internal probability. That criterion is, as a matter of fact, unfavour able to most characteristically Byzantine readings; but there are some few which I think are deserving of more serious consideration than was accorded by Hort. For the old Alexandrian text we have MS. evidence not substantially inferior to that possessed by Lucian, and we know how to use it better; but for the various types of Eastern text Lucian must have had MSS. of a greater variety and better quality than any we possess. Hence, though the principles on which he made use of them may have been the reverse of critical, to say offhand that he has never preserved an ancient reading for which we have no other authority seems over-bold." 2

1 Evangelion da-Mepharreshe, n (Cambridge, 1904), 224 f.

2 B. H. Streeter, The Four Gospels (London, 1924), p. 119.

Lucianic Recension p.38

In a similar vein, and with characteristic caution, Ropes declared:

"We may assume that the revisers worked, in part at least, on the basis of Greek MSS. preserved at Antioch that represented such a text as had long been used in this great, rich, and active church, but no literary monuments from Antioch earlier than the time of Lucian are capable of aiding our inquiry. It may well happen, therefore, that readings now found only in the Antiochian recension, or in texts dependent upon it, had been current in Antioch from the earliest times. Any reading, however, which is to be accepted as of this sort, must possess very strong internal credentials of genuineness." 1

During the past decades several papyri have come to light which tend to increase one's uneasiness over Hort's reluctance to acknowledge the possibility that an ancient reading may have been preserved in the Antiochian text even though it be absent from all the great uncial manuscripts. Since the discovery of the Chester Beatty Papyri (particularly P 45 and P 46 ) and the Bodmer Papyrus II (P 66 ), proof is available that occasionally the later Byzantine text preserves a reading that dates from the 2nd or 3rd century and for which there had been no other early witness.

A few examples selected from a large number will serve to illustrate this changed situation in the textual evaluation of the NT (the first variant is that adopted by Westcott & Hort, while the second variant is that found in the TR) :

Luke 11.33 for φως in א B D Θ fam 1 fam 13 pm φεγγος is read by P45 Koine 33 al.

John 10.29 for ο...μειζον in B latt bo, ος...μειζων is read by P66 Koine fam 1 fam 13 al.

John 11.32 for προς in א B C* D L X, εις is read by P66 Koine pm.

John 13.26 for βαψας in א B C L X 33, και εμβαψας is read by P66c A Θ al.

Acts 17.13 παρασσοντες is omitted by P45 Koine E al.

I Cor. 9.7 for καρπον in א* A B D* G P, εκ του καρπου is read by P46 Koine pl.

Eph. 5.9 for φωτος in א A B D* G P, πνευματος is read by P46 Koine pm.

Though this list could be expanded, 2 enough examples have been cited to suggest that some of the roots of the Antiochian text go back to a very early date, antedating Lucian by several generations. It does not follow, of course, that the TR should be rehabilitated en bloc 3 or even that in the examples cited above the Antiochian text is necessarily the original text.

1 Ropes, The Text of Acts, 1926, pp. cclxxxiv f. Referring to the text of Acts, Ropes says, "Apart from the 'Western.' readings found in the Antiochian recension, the Old Uncial base which the revisers used was evidently an ex cellent text," ibid., p. cclxxxvii.

2 For further examples of distinctively Byzantine readings which are also found in P 86 , consult Jn. 1.32; 3.24; 4.14, 51; 5.8; 6.10, 57; 7.3, 39; 8.41, 51. 55; 9-23; 10.38; 12.36; 14.17.

3 Despite Edward F. Hills's valiant attempt to do so in his essay "Dean Burgon in the Light of Recent Research" prefixed to the 1959 reprinting of Burgon's The Last Twelve Verses of the Gospel according to S. Mark, pp. 44-67.

Lucianic Recension p.39

The lesson to be drawn from such evidence, however, is that the general neglect of the Antiochian readings which has been so common among many textual critics is quite unjustified. 1 It is equally unsatisfactory to utilize the evidence of the Koine text in a purely mechanical fashion, as Von Soden did. On the contrary, the only proper methodology is to examine the evidence for each variant impartially, with no pre dilections for or against any one type of text. In the case of the Antiochian recension, very many readings will no doubt continue to be judged to be the result of the editorial labors of Lucian and those who shared his preference for a smooth and often composite reading, but here and there a discriminating criticism will discover ancient and perhaps original readings which the Antiochian revisers took from the texts on which they worked. The possibility should even be left open that a reading which happens to be preserved in only the Lucianic recension may commend itself as the original. 2

1 Compare G. Zuntz, "The Byzantine Text in New Testament Criticism," JTS, XLIII (1942), 25-30; and The Text of the Epistles; a Disquisition upon the Corpus Paulinum (London, 1953), PP49-57 and 150 f.

2 For several examples of readings peculiar to Lucian, which nevertheless so sober a critic as Ropes was disposed to accept as original, see his discussion in The Text of Acts, p. cclxxxv. For an attempt to prove the originality of six Antiochian variants in Matthew, see J. M. Bover, "Variantes semlticas del texto antioqueno en san Mateo," in Miscellanea biblicaB. Ubach, curante Dom Romualdo M. a Dias (Scripta et documenta, i; Montisserrati, 1953), pp. 323-327. H. Greeven argues cogently for the originality of ιερεις in Lk. 20. i. which is supported by the Koine text alone (NTS, vi [1960], 295 f.).

Lucianic Recension p.40



In the course of his discussion of "The Antiochian Recension of the Septuagint," George Foot Moore declared, "Every serious bit of investigation in any spot in the Greek Bible reveals in some new way the immense variety and baffling complexity of the problems it presents." 1a

A few of the problems and tasks which clamor for attention from both Old Testament and New Testament scholars include the following:

(1) The text of Codex Y (Cod. Macedonianus, Gregory 034, Von Soden ε 073), a manuscript dating from the 9th century and containing the Four Gospels (with lacunae), deserves to be studied more thorougly than has hitherto been the case. 1

It is inadequately cited by Tischendorf, while a collation made by Gregory is buried in the "Nachtrage" to his Textkritik. 2

According to Von Soden, the manuscript belongs to his K a -text. 3

Mrs. Lake found that this manuscript shares with Fam. Π and A some readings not preserved elsewhere. 4

(2) According to Mrs. Lake, in both Testaments codex Alexandrinus contains a large number of misspellings or itacisms of a consistent character. A thorough study of these by a Greek philologist would no doubt lead to worthwhile and interesting results.

(3) It is generally believed that John of Damascus used the Ecclesiastical text, but this has never been either proved or refuted.

(4) Though it is commonly said that Chrysostom's NT text was Antiochian, partial studies of the problem suggest that further analysis of his text is much to be desired.

(5) It is not to the credit of textual critics of the Greek Bible that they have been so slow in utilizing information derived from the iconography of Byzantine MSS in determining their date, provenance, and textual relationships. 5

(6) What principles, if any, controlled the formation of the texts of those manuscripts (such as B א A etc.) which contain both Old and New Testament?

(7) Is it possible on palaeographic or iconographical grounds to prove that this or that NT MS belongs to this or that OT MS of Lucianic derivation and that both were originally one and the same complete Bible ? (It does not necessarily follow, of course, that the textual complexion of both Testaments would be the same or even similar.)

(8) In view of the research published by the Lakes on certain parts of the Byzantine text, precisely how far is Von Soden's classification of the Koine text in need of correction ?

(9) Why did the Lucianic O.T. fail to gain the same acceptance as the corresponding Antiochian text of the New Testament ?

(10) What precisely was the textual basis of the Lucianic recension, and to what extent can readings of that recension be accepted as probably inherited, and not produced, by Lucian and his fellow-workers ?

1a American Journal of Semitic Languages and Literatures, xxix (1912-1913), 50

1 For relatively brief accounts of this manuscript see W. C. Braithwaite, "A New Uncial of the Gospels/' Expository Times, xm (1901-02), 114-117, and "The Lection-System of the Codex Macedonianus," JT$, v (1904), 265-274.

2 Vol. in (Leipzig, 1909), pp. 1028-1037.

3 Op. c^t., i, ii, 1161.

4 Family II, p. 57, note 5.

5 Cf. Kurt Weitzmann, "The Relation between Text Criticism and Pic ture Criticism," in Illustrations in Roll and Codex, a Study of the Origin and Method of Text-Illustration (Princeton, 1947), pp. 182 if., and also "Die Illustration der Septuaginta," Munohner Jahrbuch der bildenden Kwnst, Dritte Folge, m/iv (1952/53), 96-120, especially pp. 113-114.

Return to Index