The other ancient versions are of minor im portance. The Armenian seems to have been made in the Fifth Century, the Georgian a little later. The Gothic, the work of second bishop of the Goths in Mcesia, dates from the latter half of the Fourth Century. The Ethiopic, used in the churches of Abyssinia, was made in the Fifth Century. The Persian and the Arabic are several centuries later. The earliest Slaric version dates from the Ninth Century. The Prankish of the Ninth and the of the Tenth Century were made from the Latin.
A third class of witnesses to the text of the New Testament consists of the patristic citations. Among the Fathers, 'remelts.. Tertullian, and Cy prian are the main representatives of'the West ern text, Clement of Alexandria and Origen of the Alexandrian, while the post-Nicene fathers, especially of the East, generally represent the Syrian text.
(b) The Principles of Textual Criticism.— The first printed Greek Testaments were uncriti cal and based on few and late MSS. With the ap pearance of Mill's New Testament in A.D. 1707 it was seen that the number of witnesses to be consulted was very large, and that the variant readings were to be counted by the thousands. The first attempts to decide between readings were based. naturally, either on mere individual preference or on the comparative number and age of the opposing authorities. It was soon found, however, that such simple rules were not suffi cient. For very often the reading that commends itself on grounds of intrinsic worth and proba bility is absent from the majority of MSS., in cluding some of the oldest. It is well known that difficult readings in RR autograph are quite likely to be changed in the course of transcription to easier or smoother ones. The reverse is not the ease. Hence we have the celebrated canon of Bengel, "proclivm scriptioni prxstat ardua," i.e. the more difficult reading is to be preferred to the easier. Other principles or canons are, "that reading is to be preferred which seems to have been likely to have been the source of the others," and, "the shorter reading is to be preferred."
Such canons are the result of close observation and are necessary to prevent the too free use of mere conjecture. But a textual critic deals not merely with readings alone. Readings are con tained in documents, and the quality and inter relationship of the documents must be considered. If nine out of ten manuscripts are all copies of one and the same manuscript, their united testi mony amounts only to the testimony of the one manuscript from which they are derived. If that one was a poor manuscript, the text of the nine copies can be no better. Therefore, the testimony of one manuscript may be of more weight than that of a number of opponents. That a majority of witnesses favor a reading is not a necessary indication of its correctness. The witness of a manuscript which is carefully executed or seems to have been exceptionally fortunate in its ances try should be allowed great weight. The general value of a manuscript is ascertained by a care ful examination of each of its various readings. Hence the canon adopted by Westeott and Hort, "knowledge of documents should precede judg ment on readings." Another canon. formulated by the same critics, is drawn from the fact that the relation between the different manuscripts of a work is like that of the several branches of a genealogical tree. With any new copy or set of copies serious changes may be introduced. Such copies may become, in their turn, parents of large numbers of others, all of which will exhibit the same characteristic changes. These readings will be absent from manuscripts belonging to another line of descent. So we have the canon, "all trustworthy restoration of corrupted texts is founded upon a study of their history—that is, of the relations of descent or affinity which connect the several documents." The most consistent example of the application of such principles is Westeott and Ilort's Greek New Testament, of ISS1. Other critical editors. while recognizing the value of these principles, have allowed their individual preferences fuller play.