CHAPTER I.

TITLES, GROUPING, NUMBER, AND ORDER OF THE BOOKS.

THE Greek Old Testament, as known to us through the few codices which contain it as a whole, and from the lists which appear in the Biblical MSS. or in ancient ecclesiastical writings, differs from the Hebrew Bible in regard to the titles of the books which are common to both, and the principle upon which the books are grouped. The two collections differ yet more materially in the number of the books, the Greek Bible containing several entire writings of which there is no vestige in the Hebrew canon, besides large additions to the contents of more than one of the Hebrew books. These differences are of much interest to the Biblical student, since they express a tradition which, inherited by the Church from the Alexandrian synagogue, has widely influenced Christian opinion upon the extent of the Old Testament Canon, and the character and purpose of the several books.

1. The following tables shew (A) the Hebrew, Greek, and Latin titles of the canonical books of the Old Testament; (B) the order and grouping of the books in (1) lists of Jewish origin, (2) the great uncial MSS. of the Greek Bible, (3) patristic and synodical lists of the (a) Eastern, (b) Western Church.

A. TITLES OF THE BOOKS.

B (1). ORDER OF THE BOOKS IN JEWISH LISTS⁴³⁴⁴³⁴This list has been adapted from Ryle, Canon of the O.T. (table following p. 280)..

B (2). ORDER OF THE BOOKS IN UNCIAL MS. BIBLES.

B (3) (a). ORDER OF THE BOOKS IN PATRISTIC AND SYNODICAL LISTS OF THE EASTERN CHURCH.

7. Amphilochius (ad Seleuc. ap. Greg. Naz. carm. 11. vii. Migne, P.G. xxxvii. 1593).	8. Pseudo-chrysostom (syn. script. sacr. praef.). Migne, P.G. lvi. 513 sqq.
Ἡ πεντάτευχος	Τό ἱστορικόν, ὡς
(Κτίσις, Ἔξοδος, Λευιτικόν, Ἀριθμοί, Δευτερονόμιον)	Ἡ Γένεσις		(ἡ ὀκτάτευχος)
	Ἡ Ἔξοδος
Ἰησοῦς	Τὸ Λευιτικόν
Οἱ Κριταί	Οἱ Ἀριθμοί
Ἡ Ῥούθ	Τό Δευτερονόμιον
Βασιλειῶν αʹ—δʹ	Ἰησοῦς ὁ τοῦ Ναυή
Παραλειπομένων αʹ, βʹ	Οἱ Κριταί
Ἔσδρας αʹ, βʹ	Ῥούθ
Στιχηραί βίβλοι εʹ	Αἱ Βασιλεῖαι αʹ—δʹ
(Ἰώβ, Ψαλμοί, τρεῖς Σολομῶντος—Παροιμίαι, Ἐκκλησιαστής, ᾎσμα ᾀσμάτων)	Ἔσδρας
	Τὸ συμβουλευτικόν, ὡς
	Αἱ Παροιμίαι
Προφῆται οἱ δώδοκα	Ἡ τοῦ Σιρὰχ Σοφία
(Ὡσῆε, Ἀμώς, Μιχαίας, Ἰωήλ, Ἀβδίας, Ἰωνᾶς, Ναούμ, Ἁμβακούμ, Σοφονίας, Ἁγγαῖος, Ζαχαρίας, Μαλαχίας	Ὁ Ἐκκλησιαστής
	Τὰ ᾌσματα τῶν ᾀσμάτων
	Τὸ προφητικόν, ὡς
	Οἱ δεκαὲξ προφηταί
Προφῆται οἱ τέσσαρες	Ῥούθ (?)
(Ἠσαίας, Ἰερεμίας, Ἰεζεκιήλ, Δανιήλ)	Δαυείδ
Τούτοις προσεγρκίνουρι τὴν Ἐσθήρ τινες

9. Σύνοψις ἐν ἐπιτόμῳ ap. Lagarde, Septuagintast., ii. p. 60 f.436436Lagarde, l.c.: "ich widerhole sie, von mir redigiert."	10. Anonymi dial. Timothei et Aquilae.
	αʹ. Γένεσις		Ἡ Μωσαικὴ πεντάτευχος
Τὰ Μωσαϊκά	βʹ. Ἔξοδος
αʹ. Γένεσις	γʹ. Τὸ Λευιτικόν
βʹ. Ἔξοδος	δʹ. Οἱ Ἀριθμοί
γʹ. Λευιτικόν	εʹ. Τὸ Δευτερονόμιον
δʹ. Ἀριθμοί	ςʹ. Ὁ τοῦ Ναυή
εʹ. Δευτερονόμιον	ζʹ. Οἱ Κροταί, μετὰ τῆ Ῥούθ
Τὰ ἕτερα	ηʹ. Τὰ Παραλειπόμενα αʹ, βʹ
ςʹ. Ἰησοῦς ὁ τοῦ Ναυή	θʹ. Τῶν βασιλειῶν αʹ, βʹ
ζʹ. Κριταί	ιʹ. Τῶν βασιλειῶν γʹ, δʹ
ηηʹ. Ῥούθ	ιαʹ. Ἰώβ
Τέλος τῆς ὀκτατεύχου	ιβʹ. Τὸ Ψαλτήριον τοῦ Δαυίδ
Τὸ τετραβασίλειον	ιγʹ. Αἱ Παροιμίαι Τολομῶντος
θʹ. Βασιλειῶν αʹ	ιδʹ. Ὁ Ἐκκλησιαστής, σὺν τοῖς ᾌσμασιν
ιʹ. Βασιλειῶν βʹ.
ιαʹ. Βασιλειῶν γʹ	ιεʹ. Τὸ δωδεκαπρόφητον· Ἠσαίας, Ἰερεμίας, Ἰεζεκιήλ, Δανιήλ, Ἔσδρας
ιβʹ. Βασιλειῶν δʹ
ιγʹ. Παραλειπόμενα αʹ
ιδʹ. Παραλειπόμενα βʹ	καʹ. Ἰουδίθ
ιεʹ. Ἔσδρα αʹ	κβʹ. Ἐσθήρ
ιςʹ. Ἔσδρα βʹ	Ἀπόκρυφα
ιζʹ. Ἐσθήρ	Τοβίας
ιηʹ. Τωβίτ	Ἡ Σοφία Σολομῶντος
ιθʹ. Ἰουδήθ	Ἡ Σοφία Ἰησοῦ υἱοῦ Σιράχ
κʹ. Ἰώβ
Τοῦσολομῶντος
καʹ. Σοφία
κβʹ. Παροιμίαι
κγʹ. Ἐκκλησιαστής
κδʹ. ᾎσμα ᾀσμάτων
Οἱ ιβʹ προφῆται
κεʹ. Ὠσηέ
κςʹ. Ἀμώς
κζʹ. Μιχαίας
κηʹ. Ἰωιλ
κθʹ. Ἀβδιού
λʹ. Ἰωνᾶς
λαʹ. Ναούμ
λβʹ. Ἁββακούμ
λγʹ. Σοφονίας
λδʹ. Ἀγγαῖος
λεʹ. Ζαχαρίας
λςʹ. Μαλαχίας
Οἱ δʹ μεaάλοι προφῆται
λζʹ. Ἡσαΐας
ληʹ. Ἱερεμίας
λθʹ. Ἱεζεκιήλ
μʹ. Δανιήλ
Τέλος τῶν ἓξ καὶ δέκα προφητῶν
μαʹ. Σοφία Ἰησοῦ τοῦ Σιράχ

B (3) (b). ORDER OF THE BOOKS IN PATRISTIC AND SYNODICAL LISTS OF THE WESTERN CHURCH.

Prophetarum xii
(Osee, Ioel, Amos, Abdias, Ionas, Michaeas, Nahum, Habacuc, Sophonias, Aggaeus, Zacharias, Malachias)
Prophetae iv maiorum voluminum
(Isaias, Ieremias, Daniel, Ezechiel)

2. We may now proceed to consider the chief points which these tables illustrate.

(1) THE TITLES OF THE BOOKS. It will be seen that the Hebrew titles fall into three classes. They consist of either (1) the first word or words of the book (Genesis—Deuteronomy, Proverbs, Lamentations); or (2) the name of the hero or supposed author (Joshua, Judges, Samuel, Kings, Isaiah and the other Prophets, Job, Ruth, Esther, Daniel, Ezra); or (3) a description of the contents (Psalms, Song of Songs, Chronicles). Titles of the second and third class are generally reproduced in the Greek; there are some variations, as when Samuel and Kings become 'Kingdoms,' and 'Diaries' (דִּבְי־הֵיָּטִים) is changed into 'Omissions' (Παραλειπόμενα⁴⁴²⁴⁴²Or less correctly Παραλειπόμενοι, 'omitted books,' as in some lists. ), but the system of nomenclature is the same. But titles of the first class disappear in the Greek, and in their place we find descriptive names, suggested in almost every case by words in the version 215itself. Thus Genesis appears to come from Gen. ii. 4 αὕτη ἡ βίβλος γενέσεως οὐραωοῦ καὶ γῆς, Exodus from Ex. xix. 1 τῆς ἐξόδοθ τῶν υἱῶν Ἰσραὴλ ἐκ γῆς Αἰγύπτου, Numbers from Num. i. 2 κατὰ ἀριθμὸν ἐξ ὀνόματος, Deuteronomy from Deut. xvii. 18 γράψει αὐτῷ τὸ δευτερονόμιον τοῦτο εἰς βιβλίον⁴⁴³⁴⁴³On this rendering see Driver, Deuteronomy, p. i. The Massora calls the book מִשְׁגֵה הַתּוֹרָה. , Ecclesiastes from Eccl. i. 1 ῥήmata ἐkklhsiastoῦ.

The Greek titles are probably of Alexandrian origin and pre-Christian use. Not only were they familiar to Origen (Eus. H. E. vi. 25), but they are used in Melito's list, although it came from Palestine. Some of them at least appear to have been known to the writers of the New Testament; cf. Acts ii. 30 ἐν βίβλῳ ψαλμῶν, xiii. 33 ἐν τῷ ψακνῷ τῷ δευτέρῳ, Rom. ix. 25 ἐν τῷ Ὡσῆε λέγει⁴⁴⁴⁴⁴⁴See also Acts xiii. 20, 33, Rom. x. 16, xv. 11, Heb. xi. 22.. Philo⁴⁴⁵⁴⁴⁵See Prof. Ryle's Philo and Holy Scripture, p. xx. ff. uses Γένεσις, Λευιτικὸν or Λευιτικὴ βίβλος, Δεθτερονόμιον, Βασιλεῖαι, Παροιμίαι, but his practice is not quite constant; e.g. he calls Exodus ἡ Ἐξαγωγή⁴⁴⁶⁴⁴⁶So in Cohn-Wendland's edition (iii. 4, 57, 230); in ii. 271 this title is ascribed to Moses, although ἐξαγωγή does not like ἔξοδος occur in the Alexandrian version of the book. Ἡ Ἐξαγωγή was also the title of the Hellenist Ezekiel's poem on the Exodus (see below, p. 371).; Deuteronomy is sometimes ἡ Ἐπινομίς, and Judges ἡ τῶν Κ͓ιμάτων⁴⁴⁷⁴⁴⁷Cf. the change from מְלָּכִים to Βασιλεῖαι. βίβλος, Similar titles occur in the Mishna⁴⁴⁸⁴⁴⁸See Ryle, Canon of the O. T., p. 294., whether suggested by the Alexandrian Greek, or independently coined by the Palestinian Jews; thus Genesis is סֵפֶּר יְצִירָה, Numbers ס׳ מִסְפָּרִים, Proverbs ס׳ חָכְמָה, Lamentations קִינוֹת.

Through the Old Latin version the Greek titles passed into the Latin Bible⁴⁴⁹⁴⁴⁹Sometimes in a simple transliteration, as Genesis &c. Tertullian has Arithmi but in Cyprian the Latin Numeri is already used; see Burkitt, O. L. and Itala, p. 4., and from the Latin Bible into the later versions of Western Christendom. In three instances, however, the influence of Jerome restored the Hebrew titles; 1, 2 Kingdoms 216have become 1, 2 Samuel, and 3, 4 Kingdoms, 1, 2 Kings, whilst 'Chronicles,' representing the Hebrew דִּבְרֵי־הַיָּמִים, has taken the place of Paralipomenon.

Cf Hieron. Prol. Gal.: "tertius sequitur Samuel, quem nos Regnorum primum et secundum dicimus; quartus Malachim, id est Regum, qui tertio et quarto Regnorum volumine continetur . . . septimus Dabre aiamim, id est 'Verba dierum,' quod significantius Chronicon totius divinae historiae possumus appellare."

The Greek titles vary slightly in different codices and lists. Besides the variations of cod. A which appear in Table B (2), the following are mentioned in the apparatus of Holmes and Parsons. Joshua: Ἰησοῦς ὁ Ναυή, ὁ τοῦ Ναυή, Judges: Κριταὶ τοῦ Ἰσραήλ, αἱ τῶν κριτῶν πράξεις. Chronicles: Παραλειπομένων τῶν βασιλειῶν Ἰούδα. Psalms: Δαυὶδ προφήτου καὶ βασιλέως μέλος. When Nehemiah is separated from Ezra its title is: τὰ περὶ Νεεμίου or λόγοι Ν. υἱοῦ Ἁχαλία. A few further forms may be gleaned from the patristic lists. As an alternative for Παραλειπομένων the Apostolic Canons give τοῦ βιβλίου τῶν ἡμερῶν, while Ezra is known to Hilary as sermones dierum Esdrae. The Psalter is sometimes βίβλος Ψαλμῶν, liber Psalmorum, or Ψαλτήριον Δαβιτικόν, Psalmi David regis, Psalterium Daviticum. For ᾎσμα ᾀσμάτων we have occasionally ᾄσματα ᾀσμάτων—a form rejected by Origen (ap. Eus. H.E. vi. 25 οὐ γάρ, ὡς ὑπολαμβάνουσί τινες, ᾌσματα ᾀσμάτων), but used by Pseudo-Chrysostom and John of Damascus, and found in cod. A and in several of the Latin lists⁴⁵⁰⁴⁵⁰The official Vulgate had Canticum, until the plural was adopted by Sixtus V.; see Nestle, ein Jubiläum der Lat. Bibel, p. 18.; cf. the English Article VI. "Cantica, or Songs of Solomon." The lesser Prophets are οἱ δώδεκα or δεκαδύο, τῶν δώδεκα προφητῶν μία βίβλος, τὸ δωδεκαπρόφητον, prophetae xii; the greater, οἱ τέσσαρες, prophetae iv, prophetae iv maiorum voluminum, or simply maiores; when the two collections are merged into one they become οἱ δεκαέξ or οἱ ἑκκαίδεκα, τὸ ἑκκαιδεκαπρόφητον, prophetae xvi.

(2) THE GROUPING OF THE BOOKS. The methods of grouping adopted in the Hebrew and Alexandrian Greek Bibles differ not less widely than the nomenclature of the books. The Hebrew canon is uniformly tripartite, and "the books belonging to one division are never (by the Jews) transferred to another⁴⁵¹⁴⁵¹Driver, Introd., p. xxvii.." Its three groups are known as the Law 217(תּוֹרָה), the Prophets (נְבִאִים), and the Writings (כְּתוּבִים). The Massora recognised, however, certain subdivisions within the second and third groups; the Prophets were classed as Former (רִאשׁוֹנִים), i.e. Joshua, Judges, Samuel, Kings; and Latter (אַחֲרוֹנִים), and among the 'Latter' the Twelve minor Prophets formed a single collection⁴⁵²⁴⁵²So already in Sir. xlix. 10 τῶν ιβʹ πρφητῶν. . Similarly 'the five Rolls' (מְגִלּוׂת), i.e. Ruth, Canticles, Ecclesiastes, Lamentations, Esther, made a subsection among the Kethubim. The tripartite division of the canon was known at Alexandria in the second century B.C., for the writer of the prologue to Sirach refers to it more than once (1 f. τοῦ νόμου καὶ τῶν προφητῶν καὶ τῶν ἄλλων τῶν κατ᾿ αὐτοὺς ἠκολουθηκότων: 6 f. τοῦ νόμου καὶ τῶν προφητῶν καὶ τῶν ἄλλων πατρίων βιβλίων: 14 f. ὁ νόμος καὶ αἱ προφητεῖαι καὶ τὰ λοιπὰ τῶν βιβλίων). It is also recognised in the New Testament, where the Law and the Prophets are mentioned as authoritative collections, and in one passage the 'Writings' are represented by the Psalter (Lc. xxiv. 44 πάντα τὰ γεγραμμένα ἐν τῷ νόμῳ Μωυσέως καὶ τοῖς προφήταις καὶ ψαλμοῖς). But the New Testament has no comprehensive name for the third group, and even Josephus (c. Ap. i. 8) speaks of four poetical books (probably Psalms, Job, Proverbs and Ecclesiastes) as forming with the Law and the Prophets the entire series of sacred books; the rest of the Hagiographa seem to have been counted by him among the Prophets⁴⁵³⁴⁵³See Ryle, Canon of the O.T., p. 165 f.. At Alexandria the later books were probably attached to the canon by a looser bond. The writer of the De vita contemplativa appears to recognise four groups⁴⁵⁴⁴⁵⁴Unless we omit the comma after ὕμνους and regard ὕ. καὶ τὰ ἄλλα as = the Hagiographa; cf. Joseph. c. Ap. as quoted below, p. 220. (§ 3 νόμους, καὶ λόγια θεσπισθέντα διὰ προφητῶν, καὶ ὕμνους, καὶ τὰ ἄλλα οἷς ἐπιστήμη καὶ εὐσέβεια συναύξονται καὶ τελειοῦνται).

Only the first of the three Palestinian groups remains undisturbed⁴⁵⁵⁴⁵⁵Yet even the Torah was not always kept apart in the Greek Bible, as the names Octateuch and Heptateuch witness. 218in the Alexandrian Greek Bible, as it is preserved to us in MSS. and described in Christian lists. When the Law was translated into Greek, it was already a complete collection, hedged round with special sanctions, and in all forms of the Greek Bible it retains its precedence and has resisted any extensive intrusion of foreign matter. It is otherwise with the Prophets and the Hagiographa. Neither of these groups escaped decomposition when it passed into the Greek Bible. The Former Prophets are usually separated from the Latter, the poetical books coming between. The Hagiographa are entirely broken up, the non-poetical books being divided between the histories and the prophets. This distribution is clearly due to the characteristically Alexandrian desire to arrange the books according to their literary character or contents, or their supposed authorship. Histories were made to consort with histories, prophetic and poetical writings with others of their respective kinds. On this principle Daniel is in all Greek codices and catalogues one of the Greater Prophets, while Ruth attaches itself to Judges, and Canticles to Ecclesiastes.

In many of the Greek patristic lists the Alexandrian principle of grouping receives express recognition. Thus Cyril of Jerusalem, Gregory of Nazianzus, and Leontius, divide the books of the Old Testament into (1) historical—12, including the Mosaic Pentateuch; (2) poetical—5; (3) prophetical—5. Epiphanius, followed by John of Damascus, endeavours to combine this grouping with a system of pentateuchs⁴⁵⁶⁴⁵⁶Dr Sanday (in Studia Biblica, iii. p. 240) regards this as Palestinian, identifying it with Cyril's method. But Cyril begins with a dodecad (δωδεκάτη ἡ Ἐσθήρ· καὶ τὰ μὲν Ἱστορικὰ ταῦτα).—(1) legal, (2) poetical, (3) historical⁴⁵⁷⁴⁵⁷The term γραφεῖα (כְּתוּבִים) or ἁγιόγραφα is transferred to this group., (4) prophetical 219—an end which he attains by relegating Ezra and Esther to an appendix. Pseudo-Chrysostom's arrangement is similar, though slightly different in some of its details; according to his view the Bible began with an Octateuch, and the στιχηρά are broken up, the Psalter being placed with the Prophets, and the Salomonic books described as 'hortatory⁴⁵⁸⁴⁵⁸So Leontius τὰ ταραινετικά), but he classed the Psalter among them.' (τὸ συιβουλευτικόν). Even in the eccentric arrangement of Junilius⁴⁵⁹⁴⁵⁹See Kihn, Theodor v. Mopsuestia u. Junilius, p. 356 f. the Greek method of grouping is clearly dominant.

The relative order of the groups in the Greek Bible, being of literary and not historical origin, is to some extent liable to variation. The 'five books of Moses' always claim precedence, and the 'rest of the histories' follow, but the position of the poetical and prophetical books is less certain. Codex B places the poetical books first, whilst in Codd. א and A the prophets precede. But the order of cod. B is supported by the great majority of authorities both Eastern and Western (Melito, Origen, Athanasius, Cyril, Epiphanius (1, 3), Gregory, Amphilochius, the Laodicene and 'Apostolic' canons, Nicephorus, Pseudo-Chrysostom, the Cheltenham list, the African canons of 397, and Augustine). Two reasons may have combined to favour this arrangement. 'David' and 'Solomon' were higher up the stream of time than Hosea and Isaiah. Moreover, it may have seemed fitting that the Prophets should immediately precede the Evangelists.

(3) THE NUMBER OF THE BOOKS. In our printed Hebrew Bibles the books of the Old Testament are 39 (Law, 5; Former Prophets (Joshua—2 Kings), 6; Latter Prophets, 15; Hagiographa, 13). But Samuel, Kings, Ezra-Nehemiah, and 220Chronicles⁴⁶⁰⁴⁶⁰Chronicles-Ezra-Nehemiah appears to have been originally a single book. But while Ezra and Nehemiah are still joined in the Greek Bible, Chronicles stands by itself both in and , and in it follows Nehemiah and forms the last book of the Canon (cf. Mt. xxiii. 35, and see Barnes Chronicles, in the Cambridge Bible, pp. x.—xiii.)., were originally single books⁴⁶¹⁴⁶¹The division probably began in the LXX., and the Minor Prophets were also counted as a single book. Thus the number is reduced to 24 (Law, 5; Former Prophets, 4; Latter Prophets, 4; Hagiographa, 11), and this answers to the prevalent Jewish tradition. On the other hand Josephus expressly limits the books to 22 (Law, 5; Prophets, i3; Hymns and moral pieces, 4). He has probably included the historical Hagiographa among the Prophets, and treated Ruth and Lamentations as appendices to Judges and Jeremiah respectively.

Both traditions were inherited by the Church, but the latter was predominant, especially in the East. In some lists indeed the twenty-two books became twenty-seven, the 'double books' being broken up into their parts (Epiph. 1)⁴⁶²⁴⁶²Jerome, Prol. Gal.: "quinque a plerisque libri duplices aestimantur." As the twenty-two books answered to the twenty-two letters of the Hebrew alphabet, so these 'double books' were thought to correspond to the 'double letters,' i.e. those which had two forms (כ ,מ ,נ ,פ ,צ). The 'double books were not always identical in different lists; see Sanday, op. cit. p. 239.; in some a similar treatment of the Dodecapropheton raised the number to 34 (the 'Sixty Books'), and there are other eccentricities of numeration which need not be mentioned here.

Josephus, c. Ap. i. 8: οὐ μυριάδες βιβλίων εἰσὶ παρ᾿ ἡμῖν ἀσυμφώνων καὶ μαχομένων, δύο μόνα πρὸς τοῖς εἴκοσι βιβλία . . . καὶ τούτων πέντε μέν ἐστι Μωυσέως . . . οἱ μετὰ Μωυσῆν προφῆται . . . συνέγραψαν ἐν τρισὶ καὶ δέκα βιβλίοις· αἱ δὲ λοιπαὶ τέσσαρες ὕμνους εἰς τὸν θεὸν καὶ τοῖς ἀνθρώποις ὑποθήκας τοῦ βίου περιέχουσιν. He is followed by Origen ap. Eus. l.c. οὐκ ἀγνοητέον δ᾿ εἶναι τὰς ἐνδιαθήκους βίβλους ὡς Ἐβραῖοι παραδιδόασιν, ὅσος ὁ ἀριθμὸς τῶν παῤ αὐτοῖς στοιχείων ἐστίν· and Cyril. Hier. catech. iv. 33 ἀναγίνωσκε τὰς θείας γραφάς, τὰς εἴκοςι δύο βίβλους τῆς παλαιᾶς διαθήκης. Similarly Athanasius, ep. fest. 39 (Migne, P.G. xxvi. col. 1437). When another numeration was adopted, efforts were 221made to shew that it did not involve a real departure from the canon of twenty-two; cf. Epiph. haer. i. I. 8, αὗταί εἰσιν αἱ εἴκοσι ἑπτὰ βίβλοι αἱ ἐκ θεοῦ δοθεῖσαι τοῖς Ἰουδαίοις, εἴκοσι δύο δὲ ὡς τὰ παῤ αὐτοῖς στοιχεῖα τῶν Ἐβραικῶν γραμμάτων ἀριθμούμεναι διὰ τὸ διπλοῦσθαι δέκα βίβλους εἰς πέντε λεγομένας· dial. Tim. et Aq. (ed. Conybeare, p. 66), αὗται αἱ βίβλοι αἱ θεόπνευστοι καὶ ἐνδιάθετοι, κςʹ μὲν οὖσαι, κβʹ δὲ ἀριθμούμεναι διὰ τὸ . . . ἐξ αὐτῶν διπλοῦσθαι.

On the other hand the numeration in 4 Esdr. xiv. 44 rests, if nongenti quatuor be the true reading, on a tradition which makes the Hebrew books 24. This tradition is supported by the testimony of the Talmud and the Rabbinical literature⁴⁶³⁴⁶³Cf. Ryle Canon, pp. 157 f., 222, 292; Sanday, op. cit. p. 236 ff., and the Canon is known in Jewish writings by the name כ״ד ספרים, "the Twenty-Four Books." It finds a place in certain Western Christian writers, e.g. Victorinus of Petau comm. in Apoc.: "sunt autem libri V.T. qui accipiuntur viginti quatuor quos in epitome Theodori invenies⁴⁶⁴⁴⁶⁴Zahn offers a suggestion, to which Sanday inclines, that the writer refers to the Excerpta ex Theodoto which are partly preserved in the works of Clement of Alexandria.." Victorinus compares the 24 books to the 24 Elders of Apoc. iv., and the same fancy finds a place in the Cheltenham list ("ut in apocalypsi Iohannis dictum est Vidi XXIIII seniores mittentes coronas suas ante thronum, maiores nostri probant hoc libros esse canonicos"). Jerome knows both traditions, though he favours the former (Prol. Gal. "quomodo igitur viginti duo elementa sunt . . . ita viginti duo volumina supputantur . . . quamquam nonnulli Ruth et Cinoth inter Hagiograpba scriptitent et libros hos in suo putent numero supputandos et per hoc esse priscae legis libros viginti quatuor").

Let us now turn to the ecclesiastical lists and see how far the Hebrew Canon was maintained.

Our earliest Christian list was obtained from Palestine⁴⁶⁵⁴⁶⁵Melito ap. Eus. H.E. iv. 26 ἐπειδὴ μαθεῖν τὴν τῶν παλαιῶν βιβλίων ἐβουλήθης ἀκρίβειαν, πόσα τὸν ἀριθμὸν καὶ ὁποῖα τὴν τάξιν εἶεν . . . ἀνελθὼν εἰς τὴν ἀνατολὴν καὶ ἕως τοῦ τόπου ἔνθα ἐκηρύχθη καὶ ἐπράχθη . . . ἔπεμψά σοι. , and probably represents the contents of the Palestinian Greek Bible. It is an attempt to answer the question, What is the true number and order of the books of the Old Testament? Both the titles and the grouping are obviously Greek, but the books are exclusively those of the Hebrew canon. Esther does not appear, but the number of the books is twenty-two, if we are intended to count 1—4 Regn. as two.

The next list comes from Origen. It belongs to his commentary on the first Psalm, which was written at Alexandria⁴⁶⁶⁴⁶⁶Eus. H.E. vi. 24., i.e. before A.D. 231. The books included in it are expressly said to be the twenty-two of the Hebrew canon εἰσὶ δὲ αἱ εἴκοσι δύο βίβλοι καθ᾿ Ἑβραίους αἵδε). Yet among them are the first book of Esdras⁴⁶⁷⁴⁶⁷Already cited freely by Josephus as an authority for the history of the period. Origen, it should be added, regards 1, 2 Esdras as a single volume (Ἔσδρας πρώτη, δευτέρα ἐν ἑνί). and the Epistle of Jeremiah, which the Jews never recognised. With the addition of Baruch, Origen's list is repeated by Athanasius, Cyril, Epiphanius (1), and in the Laodicean canon; Amphilochius mentions two books of Esdras, and it is at least possible that the Esdras of Gregory of Nazianzus is intended to include both books, and that the Epistle, or Baruch and the Epistle, are to be understood as forming part of Jeremiah in the lists both of Gregory and Amphilochius. Thus it appears that an expansion of the Hebrew canon, which involved no addition to the number of the books, was predominant in the East during the fourth century.

The Eastern lists contain other books, but they are definitely placed outside the Canon. This practice seems to have begun with Origen, who after enumerating the twenty-two books adds, ἔξω δὲ τούτων ἐστὶ τὰ Μακκαβαϊκά. Athanasius takes up the expression, but names other books—the two Wisdoms, Esther⁴⁶⁸⁴⁶⁸Cf. Melito's omission of Esther, and the note appended to the list of Amphilochius., Judith, and Tobit⁴⁶⁹⁴⁶⁹The N.T. members of the same class are the Teaching and the Shepherd.. Palestine was perhaps naturally conservative in this matter; Cyril will not allow his catechumens to go beyond the Canon, and Epiphanius mentions only, and that with some hesitation, the two books of Wisdom (εἰσὶ δὲ καὶ ἄλλαι παῤ αὐτοῖς βίβλοι ἐν ἀμφιλέκτῳ⁴⁷⁰⁴⁷⁰Haer. I. i. 1. . . . 223αὗται χρήσιμοι μέν εἰσι καὶ ὠφέλιμοι, ἀλλ᾿ εἰς ἀριθμὸν ῥητῶν οὐκ ἀναφέρονται)⁴⁷¹⁴⁷¹De mens. et pond. 4.. And this was the prevalent attitude of the East even at a later time. There are exceptions; Pseudo-Chrysostom places Sirach among the Hortatory books of the canon; the Apostolic canons, while excluding Sirach, include three books of Maccabees. But John of Damascus reflects the general opinion of the Greek fathers when, while reckoning both books of Esdras⁴⁷²⁴⁷²Like Origen, he explains that they form together but a single book (τοῦ Ἔσδρα αἱ δύο εἰς μίαν συναπτόμεναι βίβλον). as canonical, he repeats the verdict of Epiphanius upon the two Wisdoms, Ἐηάρετοι μὲν καὶ καλαί, ἀλλ᾿ οὐκ ἀριθμοῦνται⁴⁷³⁴⁷³The non-canonical books (τὰ ἔξω) are however carefully distinguished from real apocrypha when the latter are mentioned; e.g. in the stichometry of Nicephorus, and in the list of the 'Sixty Books.'.

On the other hand the West, further from the home of the Hebrew canon, and knowing the Old Testament chiefly through the Latin version of the LXX., did not scruple to mingle non-canonical books with the canonical. Hilary and Ruffinus⁴⁷⁴⁴⁷⁴In symb. 38 "alii libri sunt qui non canonici sed ecclesiastici a maioribus appellati sunt." were doubtless checked, the one by the influence of Eastern theologians, the other by the scholarship of Jerome; but Hilary mentions that there were those who wished to raise the number of the canonical books to twenty-four by including Tobit and Judith in the canon. From the end of the fourth century the inclusion of the non-canonical books in Western lists is a matter of course. Even Augustine has no scruples on the subject; he makes the books of the Old Testament forty-four (de doctr. Chr. ii. 13 "his xliv libris Testamenti Veteris terminatur auctoritas⁴⁷⁵⁴⁷⁵Cf. Retract. ii. 4."), and among them Tobit, Judith, and two books of Maccabees take rank with the histories; and the two Wisdoms, although he confesses that they were not the work of Solomon, are classed with the 224Prophets. His judgement was that of his Church (Conc. Carth. iii. can. xlvii. "sunt canonicae scripturae Salomonis libri quinque . . . Tobias, Judith . . . Machabaeorum libri duo"). The African Church had probably never known any other canon, and its belief prevailed wherever the Latin Bible was read.

There can be little doubt that, notwithstanding the strict adherence of the Eastern lists to the number of the Hebrew books, the Old Latin canon truly represents the collection of Greek sacred books which came into the hands of the early Christian communities at Antioch, Alexandria, and Rome. When Origen and the Greek fathers who follow him fix the number of the books at twenty-two or twenty-four, they follow, not the earlier tradition of the Church, but the corrected estimate of Christian scholars who had learned it from Jewish teachers. An earlier tradition is represented by the line of Christian writers, beginning with Clement of Rome, who quoted the 'Apocryphal' books apparently without suspecting that they were not part of the Canon. Thus Clement of Rome⁴⁷⁶⁴⁷⁶1 Cor. 55. places the story of Judith side by side with that of Esther; the Wisdom of Sirach is cited by Barnabas⁴⁷⁷⁴⁷⁷c. 19. 9. and the Didache⁴⁷⁸⁴⁷⁸c. 4., and Tobit by Polycarp⁴⁷⁹⁴⁷⁹Philipp. 10.; Clement of Alexandria⁴⁸⁰⁴⁸⁰Strom. i. 10, v. 14. and Origen appeal to Tobit and both the Wisdoms, to which Origen adds Judith⁴⁸¹⁴⁸¹Cf. Westcott in D. C. B. iv. p. 130.. Our earliest MSS. of the Greek Bible confirm the impression derived from the quotations of the earliest Christian writers. Their canon corresponds not with that of the great writers of the age when they were written, but with that of the Old Latin version of the LXX. Codd. B א A contain the two Wisdoms, Tobit, and Judith; 1—2 Maccabees are added in א, and 1—4 Maccabees in A; cod. C still exhibits the two Wisdoms, and when complete may have contained other books of the same class. 225Moreover, the position of the books shews that the scribes of these MSS. or of their archetypes lacked either the power or the will to distinguish them from the books of the Hebrew canon. In the light of the facts already produced, it is clear that the presence of the non-canonical books in Greek Bibles cannot be attributed to the skilled writers of the fourth and fifth centuries. They have but perpetuated an older tradition—a tradition probably inherited from the Alexandrian Jews.

An explanation of the early mixture of non-canonical books with canonical may be found in the form under which the Greek Bible passed into the keeping of the Church. In the first century the material used for literary purposes was still almost exclusively papyrus, and the form was that of the roll⁴⁸²⁴⁸²See Kenyon, Palaeography of Greek papyri, pp. 24, 113 ff.. But rolls of papyrus seldom contained more than a single work, and writings of any length, especially if divided into books, were often transcribed into two or more separate rolls⁴⁸³⁴⁸³Ib. p. 122: "no papyrus roll of Homer hitherto discovered contains more than two books of the Iliad. Three short orations fill the largest roll of Hyperides.". The rolls were kept in boxes (κιβωτοί, κίσται, capsae, sistae)⁴⁸⁴⁴⁸⁴E. M. Thompson, Greek and Latin Palaeography, p. 57., which served not only to preserve them, but to collect them in sets. Now while the sanctity of the five books of Moses would protect the cistae which contained them from the intrusion of foreign rolls, no scruple of this kind would deter the owner of a roll of Esther from placing it in the same box with Judith and Tobit; the Wisdoms in like manner naturally found their way into a Salomonic collection; while in a still larger number of instances the two Greek recensions of Esdras consorted together, and Baruch and the Epistle seemed rightly to claim a place with the roll of Jeremiah. More rarely such a writing as the Psalms of Solomon may have found its way into the company of kindred books of the canon. It is not a serious objection to this hypothesis 226that Philo does not quote the Apocrypha, and has no certain allusion to it⁴⁸⁵⁴⁸⁵Ryle, Philo and Holy Scripture, p. xxxiii.. A great scholar would not be deceived by the mixture of heterogeneous rolls, which might nevertheless seriously mislead ordinary readers, and start a false tradition in an unlettered community such as the Christian society of the first century.

(4) THE INTERNAL ORDER OF THE GROUPS. Even in Jewish lists of the Hebrew Canon there are variations in the internal order of the Prophets and the Hagiographa. The 'Great Prophets' occur in each of the three orders (1) Isaiah, Jeremiah, Ezekiel; (2) Jeremiah, Ezekiel, Isaiah; (3) Jeremiah, Isaiah, Ezekiel⁴⁸⁶⁴⁸⁶See Ryle, Canon, p. 225 ff.. The order of the Hagiographa varies more extensively. In the printed Bibles they are arranged in three subdivisions: (1) Psalms, Proverbs, Job; (2) Canticles, Ruth, Ecclesiastes, Lamentations, Esther (the five Megilloth); (3) Daniel, Ezra, Chronicles. The Talmudic order is as follows: Ruth, Psalms, Job, Proverbs, Ecclesiastes, Canticles, Lamentations, Daniel, Esther, Chronicles. The MSS. vary, many agreeing with the printed Bibles; others, especially those of Spanish provenance, following the order: Chronicles, Psalms, Job, Proverbs, Ruth, Canticles, Ecclesiastes, Lamentations, Esther, Daniel, Ezra⁴⁸⁷⁴⁸⁷Ryle, ib., pp. 229 ff., 281 f..

In the lists of the Greek Bible and the sequence of its MSS. the Law and the 'Former Prophets' generally retain their Hebrew order, with the noteworthy exception that Ruth is always attached to Judges. But there are also minor exceptions which are of some interest. Even in the Pentateuch Melito, Leontius, and the Cheltenham list reverse the common order of Leviticus and Numbers⁴⁸⁸⁴⁸⁸On this see Sanday, Studia Biblica, iii. p. 241.. The sequence is broken in some lists after Ruth (Laod., Epiph. 1), or even after Joshua 227(Epiph. 3⁴⁸⁹⁴⁸⁹Ruth is attached to 1 Regn. in the Cheltenham list, and Augustine inclines to this arrangement (see Sanday, 1.c., p. 242). The result was to create a Heptateuch; for the word cf. J. E. B. Mayor, The Latin Heptateuch, p. xxxvi. R. Peiper's text of the Heptateuchos, to which Prof. Mayor refers (p. xxxiv.), appeared in the Vienna Corpus scr. eccl. lat. vol. xiii. (1895).) or Deuteronomy (Epiph. 1). Occasionally Chronicles, which is an intruder from the Hagiographa, precedes 1—4 Regn. (Epiph. 2, Dial. Tim. et Aq.), or drops out altogether (Ps.-Chrys., Junilius, Cod. Clarom.). All these disturbances of the normal order may be ascribed to local or individual influences, and find no support in the uncial MSS. of the Greek Bible. But it is otherwise when we come to the 'Latter Prophets' and the Hagiographa. With regard to the Prophets, three questions of order arise. (1) There is the relative order of the Twelve and the Four. In the majority of patristic lists the Twelve precede (Ath., Cyr., Epiph., Greg., Amph., &c.), and this is also the order of Codd. A, B, N-V. But Cod. א begins with the Four, and it is supported by other authorities, chiefly Western (Ruff, Chelt., Ps.-Gelasius, Cassiodorius, Nicephorus); whilst in a few the subdivisions are mixed (Melito, Junilius, Ebedjesu⁴⁹⁰⁴⁹⁰For statements by early Mohammedan writers as to the extent of the Jewish and Christian Canons see Margoliouth in Exp. Times, Nov. 1899, p. 91.). (2) The internal order of the δωδεκαπρόφητον in most of the MSS. and catalogues⁴⁹¹⁴⁹¹The chief exceptions are: Cod. v, Hosea, Amos, Joel, Obadiah, Jonah, Micah; Greg. Naz. and Cod. Barocc., Hosea, Amos, Micah, Joel, Jonah, Obadiah; Junilius, Ebedjesu, Augustine, the Hebrew order. where it is stated differs from the Hebrew order in regard to the relative positions of the prophets in the first half of the group; the Hebrew order being Hosea, Joel, Amos, Obadiah, Jonah, Micah, but the Greek, Hosea, Amos, Micah, Joel, Obadiah, Jonah. The dominant Greek order may perhaps be due to "an attempt to secure greater accuracy in the chronological arrangement⁴⁹²⁴⁹²Ryle, Canon, p. 229.." (3) The 228Greek order of the Greater Prophets follows the oldest Hebrew tradition (Isaiah, Jeremiah, Ezekiel), but it appends Lamentations to Jeremiah, and enlarges the group by placing Daniel either before (Melito, Origen, Hilary, Chelt., Augustine), or, more usually, after Ezekiel.

The relative order of the Hagiographa in the LXX. is more perplexing. For Ruth, Lamentations, and Daniel we have already accounted; there remain Chronicles, Job, Psalms, Proverbs, Ecclesiastes, Canticles, Esther, and Ezra. Chronicles, in accordance with the theory enshrined in its Greek name, usually follows Kings. Psalms, Proverbs, Ecclesiastes, Canticles, for the most part hold together in that order, as a group of poetical books; but there are many exceptions. 'David' sometimes goes with the Prophets (Ps.-Chrys., Junilius, Augustine, Isidorus), and the group is then regarded as 'Salomonic,' or 'hortatory.' Lists which admit the two books of Wisdom usually join them to this subdivision (Ebedjesu, Carth., Augustine, Innocent, Cod. Clarom., Ps.-Gelasius, Cassiodorius, Isidorus). The internal order of the Salomonic books varies (Proverbs, Ecclesiastes, Canticles; Ecclesiastes, Canticles, Proverbs; Proverbs, Canticles, Ecclesiastes); the Wisdoms usually follow, but sometimes break the sequence of the three canonical books. Much difficulty seems to have been felt as to the place of Job; the book normally appears in connexion with the poetical books, either last or first, but it is sometimes placed among the histories (Augustine, Innocent, Cod Clarom., Ps.-Gelasius, Cassiodorius), or after the Prophets (Origen). The position of Esdras is not less uncertain; its normal place is after Chronicles, but it is also found before or after the Prophets (Melito, Epiph., John of Damascus, Cod. Barocc.), or in connexion with a group of the apocryphal histories (cod. A, Carth., Augustine, &c.). Esther is still more erratic; sometimes it follows the poetical books, sometimes the Prophets, sometimes the 229histories; not a few lists place it among the antilegomena, or omit it altogether. When admitted to a place in the Canon, it is usually to be found at or near the end (Origen, Epiphanius, Amphilochius, John of Damascus, Hilary, Carth., Cod. Clarom., Ps.-Gelasius, Cassiodorius), and in company with apocryphal books, especially Judith⁴⁹³⁴⁹³The proximity of Esther to Judith in many lists is perhaps due to the circumstance that in both books the central figure is a woman; cf. p. 213 (right-hand column). and Tobit (codd. BאA, Chelt., Carth., Augustine, and the later Latin lists⁴⁹⁴⁴⁹⁴Cf. Ryle, Canon, p. 199 ff.). It seems as if the doubt which the Jewish authorities felt with regard to this book was inherited by many Christians. On the other hand Cyril, who represents the tradition of the Church of Jerusalem, makes it the twelfth of the canonical books, and in the Laodicene list it stands eighth.

Except in cases where an old or well-defined tradition fixed the internal order of groups of books, there was clearly room for every possible variation so long as the books were written on separate rolls. The cista might serve to keep a group together, but it offered no means of fixing the relative order of its contents. In the codex, on the other hand, when it contained more than one writing, the order was necessarily fixed⁴⁹⁵⁴⁹⁵Cf. Sanday, Studia Biblica, iii. p. 233 ff., and the scribe unconsciously created a tradition which was followed by later copyists. The 'transition to vellum,' and the consequent transition from the roll to the codex, does not seem to have been general before the fourth century, although in the case of Biblical MSS. it may have begun a century earlier⁴⁹⁶⁴⁹⁶See Kenyon, Palaeography of papyri, p. 119 f.; Sanday, l.c. Papyrus was freely used for codices in Egypt during the third century; cf. Grenfell and Hunt, Oxyrhynchus Papyri, ii. p. 2.; and thus we may regard our earliest uncial codices as prototypes of the variations in order which mark the mass of later MSS. A single instance may suffice. It has been stated that Esther is frequently found in company 230with Judith and Tobit. But these books occur in varying order in the oldest MSS.; in B we have Esther, Judith, Tobit, but in א A, Esther, Tobit, Judith; a favourite Western order is Tobit, Esther, Judith (Chelt., Augustine, Innocent, Gelasius, Cassiodorius, Isidorus); another, sanctioned at Carthage in 397, is apparently more common in MSS. of the Vulgate, viz., Tobit, Judith, Esther⁴⁹⁷⁴⁹⁷For the order of the books in Latin MS. Bibles see S. Berger, Histoire de la Vulgate, pp. 301—6, 331—9.. Such variations, resting on no obvious principle, are doubtless ultimately due to the judgement or caprice of a few scribes, whose copies supplied the archetypes of the later Greek MSS. and the daughter-versions of the Septuagint.

LITERATURE. On the general subject of this chapter the student may consult C. A. Credner, Gesch. d. N. T. Kanons (ed. Volkmar, Berlin, 1860); Th. Zahn, Gesch. d. N.T. Kanons, ii., p. 143 ff. (Erlangen, 1890); B. F. Westcott, Hist. of the Canon of the N. T.⁶ (Cambridge, 1891); W. Sanday, The Cheltenham List, in Studia Biblica, iii., pp. 226—243 (Oxford, 1891); Buhl, Kanon u. Text des A. T. (Leipzig, 1891); H. E. Ryle, Canon of the O.T. (London, 1892); E. Preuschen, Analecta (Leipzig, 1893); H. L. Strack, art. Kanon des. Alten Testamentes in P.R.E.³ ix. 741—767.