CANON OF THE NEW TESTAMENT

should be read in church and what books should be appealed to in discussion. The former was the primary question. The boolss used at their services by the churches, and therefore admitted by them as having a right to be so employed, were the books to be appealed to in controversy. The testing fact was church usage. Canonical books were the books read at public worship. How it came about that certain books were so used and others not is by no means clear. Prof. Harnack's theory would solve the problem it we could be sure it was valid. Apart from this, (1) traditional usage and (2) assurance of Apostolic authorship appear to have been two grounds relied upon.

Turning to the East, we find Clement of Alexandria (A.D. 165-220) acknowledging the 4 Gospels and Acts, and 14 Epistles of Paul (Hebrews being included), and quoting 1 and 2 John, 1 Peter, Jude, and the Apocalypse. He makes no reference to James, 2 Peter, or 3 John, any of which he may perhaps have known, as we have no list of NT books from his hand, for he does not name these books to reject them. Still, the probabiUty as regards some, if not all, of them is that he did not know them. In the true Alexandrian spirit, Clement has a wide and comprehensive idea of inspiration, and therefore no very definite conception of Scriptural exclusiveness or fixed boundaries to the Canon. Thus he quotes Barnabas, Clement of Rome, Hermas, the Preaching of Peter, the Apocalypse of Peter, and the Sibylline Wri-tings as in some way authoritative. He was a literary eclectic who deUghted to welcome Christian truth in un-expected places. Still he had a NT in two volumes which he knew respectively as 'The Gospel' and 'The Apostle' (see Euseb. HE vi. 14). Origen (a.d. 184^253), who was a more critical scholar, treated questions of canonicity more scientifically. He acknowledged our books of the OT and some parts of the Apocrypha, such as 1 Mac; and in the NT the 4 Gospels, Acts, 13 Epistles of Paul, Hebrews (though the latter as of doubtful authorship; nevertheless in his homily on Joshua he seems to include it among St. Paul's works, since he makes them 14, when he writes that 'God, thundering on the 14 trumpets of his [i.e. Paul's] Epistles, threw down even the walls of Jericho, that is all the in-struments of idolatry and the doctrines of the philoso-phers'), 1 Peter, 1 John, Revelation. He does not directly mention the Epistles of James or Jude, although he seems to refer to them once in a rhetori-cal way, classing Peter, James, and Jude with the 4 Evangelists as represented by Isaac's servants — if we are to trust Rufiuus' version. He mentions 2 Peter and 2 and 3 John as of disputed genuineness, and refers to the Gospel of the Hebrews in an apologetic tone, the Gospels of Peter and James, and the Acts of Paul, and quotes Hermas and Barnabas as ' Scripture,' while he admits that, though widely circulated, Hermas was not accepted by all. It is a significant fact, how-ever, that he wrote no commentaries on any of those books that are not included in our NT.

3. The Settlement of the Canon in the Fourth and Fifth Centuries. — An important step towards the settle-ment of the Canon on historical and scientific lines was taken by Eusebius, who, with his wide reading and the great library of Paraphilus to resort to, also brought a fair and judicious mind to face the problems involved. Eusebius saw clearly that it is not always possible to give a definite affirmative or negative answer to the question whether a certain book should be in the Canon. Therefore he drew up three Usts of books — (1) The books that are admitted by all, (2) the books which he is disposed to admit although there are some who reject them, (3) the books that he regards as spurious. A fourth class, which really does not come into the com-petition for a place in the Canon, consists of heretical works which 'are to be rejected as altogether absurd and impious' (HE iii. 25). The Ivret class, consisting of the books universally acknowledged, contains the 4

CANON OF THE NEW TESTAMENT

Gospels; Acts; the Epistles of Paul — which in one place (iii. 3) are reckoned to be 14, and therefore to include Hebrews, although in another place (vi. 14) Hebrews is placed in the second class, among the dis-puted books; 1 Peter; 1 John; and Revelation (doubt-fully). The second class, consisting of books widely accepted, though disputed by some (but apparently all admitted by Eusebius himself), contains James; Jude; 2 Peter — regarded in another place (iii. 3) as spurious; 2 and 3 John. The third class, consisting of spurious works, contains the Acts of Paul; the Shepherd of Hermas; the Apocalypse of Peter; the Didache; and perhaps, according to some, the Revelation. Under the orders of Constantine, Eusebius had 50 copies of the Scriptures sumptuously produced on vellum for use in the churches of Constantinople. Of course these would correspond to his own Canon and so help to fix it and spread its Influence. Alter this the fluctuations that we meet with are very slight. Athanasius in one of his Festal Letters (a.d. 365) undertakes to set forth in order the books that are canonical and handed down and believed to be Divine. His NT exactly agrees with our Canon, as does the NT of Epiphanius (c. a.d. 403). Cyril of Jerusalem (who died a.d. 386) gives a list of 'Divine Scriptures' which contains all the NT except the Revelation; and Amphilochius of Iconium (a.d. 395) has a versified catalogue of the Biblical books, in which also all our NT books appear except the Revelation, which he regards as spurious; Amphilochius refers to doubts concerning Hebrews and to a question as to whether the number of Catholic Epistles is 7 or 3. Even Chrysostora (who died a.d. 405) never alludes to the Revelation or the last 4 CathoUc Epistles. But then he gives no list of the Canon. One of the Apostolical Canons (No. 85), which stand as an appendix to the 8th book of the Apostolical Constitutions (85), and cannot be dated earlier than the 4th cent, in their present form, gives a list of the books of Scripture. Sirach is here placed between the OT and the NT with a special recommendation to ' take care that your young persons learn the wisdom of the very learned Sirach.' Then follow the NT books — the 4 Gospels, 14 Epistles of Paul (Hebrews therefore included in this category), 2 Epistles of Peter, 3 of John, James, Jude, 2 Epistles of Clement, the 8 books of the Constitutions, Acts. Thus, while Clement and even the Apostolical Constitutions are included, the Revelation is left out, after a common custom in the East. Manifestly this is an erratic Canon.

Returning to the West, at this later period we have an elaborate discussion on the Canon by Augustine (a.d. 430), who lays down rules by which the canonicity of the several books claimed for the NT may be deter-mined. (1) There are the books received and acknowl-edged by ail the churches, which should therefore be treated as canonical. (2) There are some books not yet universally accepted. With regard to these, two tests are to be applied : (a) such asare received by the majority of the churches are to be acknowledged, and (b) such as are received by the Apostolic churches are to be preferred to those received only by a smaller number of churches and these of less authority, i.e. not having been founded by Apostles. In case (o) and (fi) conflict, Augustine considers that ' the authority on the two sides is to be looked upon as equal' {Christian Doctrine, ii. viii. 12). Thus the tests are simply Church reception, though with discrimination as to the respective authority of the several churches. The application of these tests gives Augustine just our NT.

Jerome (a.d. 420) also accepts our NT, saying con-cerning Hebrews and the Revelation that he adopts both on the authority of ancient writers, not on that of present custom. He is aware that James has been questioned; but he states that Utile by little in course of time it has obtained authority. Jude was even rejected by most people because it contained quotations from

115