Of any doctrinal bias the early Versions show little trace, though we often lind in the Fathers complaints of falsification, which cannot, however, be maintained.
As possible instances of intentional alteration may b« men- tioned the Nestorian substitution of leavened for iinUaoeiwd bread at 1 Co b^, a tendency towards Encratite views in the Syri.ac version of 1 Co 72-8-7, and more clearly in reference to the virginity of Mar>'. Berger* traces the adaptation of various Latin MSS at 2 Mac 12-*8 in regard to a passage bearing on prayers for the dead. EUicott lands ' a slightly Arian tinge ' in the'Gothic version of Ph 'i^. But these are isolated instances, which must not, however, be ignored.
When we come to compare Versions made by Roman Catholics with other Versions, there is more evidence of a preference for words which will support special ecclesiastical positions or views. Thus, in the French version of de Sacy, elders become pretrfS, in Gn 315 it is la.fetnme who will bruise the serpent's head, St. Paul hopes to be" delivered by U nUrite des pniret ; and other instances might be given.t
(c) Again, it must be remembered that the power and intention of the best translator are limited by the material which he has to use, and that in two ways. In the first place, one language may be incapable of literally reproducing the grammatical idioms of another. Thus there are no distinctions of gender in Armenian, no neuter in Arabic, no passive voice in Bohairic, no article in Latin, and therefore these Versions afford no help where readings involving such points are being dis- cussed. Again, words have to be supplied in a translation which were not required in the original, t Such cases may be indicated in later times by the use of italics, but they are a more or less modern device and not alwaj'S accurately employed. Somewhat akin to the point now being discussed is the ambiguity, which arises as to their evidence, in languages like the Syriac and Arabic, owing to the system of vowel points. Secondly, the translator was hampered not only by grammatical but also by lexical difiiculties, as is the case with the missionary of to-day. § It is true that words could be coined, such as semini- verbius, to represent ffircpnoXiyo! (Ac 17"), camum miftes to translate xriiiuo-ecs (1 Co 9^), or in more modern times (as in Sir John Cheke's version) hundreder for centurion. Transliteration otiers another device, adopted frequently in the case of the oldest Versions, but the result is not an effective or an intelligible translation. Another and more important consideration, which aflects, however, the exegetical rather than the critical use of Versions, is that the words used by the translator must often suggest either more or less than the expression translated. This is a difficulty which is felt, for example, in rendering the NT into Chinese.ll
(d) It must be remembered in connexion with the literary side of translations that a translator will not alwaj's use one word or expression, and one only, to render any particular word or expres- sion of the original. As instances where the Eng- lish Bible shows this freedom in translation we may refer to the equivalents given for irapaxi'irTij in Lk 24'', Jn '20'- ", or the various renderings of TTpaiTupi.oi' : similarly, the word ' dvo/ios is trans- lated in five different ways' in the NT-U Othet Versions will provide a number of instances of a like kind.** The point is important in connexion
• nutoire de la Vulgate, p. 2a
t Revue de TMoloqie, ii. 1, 311.
j See below, p. SoS**.
5 Cf. Life of 11. Caltmcay, bishop of Caffraria, pp. 2-19-2S0, as to the Kaffir and Zulu languages ; and for difflculties in con- nexion with Hindustani see Church ilitnonari/ Gleaner, Oct. 1899. „_„
II See correspondence In the Guardian for 1899 oo ths Chinese rendernig for ' priest.'
U Plummer's St. Luke. p. 606.
•• Thus Westcott. EpMles of St. John, p. xxvii, notices thai Trpir» in three successive verses of the Eptstle is translated >J ob&ervare, custodire, neroarc
VERSIONS
VERSIONS
853
■with the use of Versions for critical purposes, because it reminds us tliat wo cannot argue from a variation in the translation to a similar varia- tion in the ori^'iiial. Versions, therefore, often fail to give assistance where there is a doubt between two words of almost the same meaning, or between two words which the translators may possibly not have dill'erentiated.*
(e) It is jjcrhaps hardly necessary, after what has been said as to necessary precautions, to give a reminder that the evidence of Versions can be used only at lirst hand, and not througli the medium of a translation. Many of the scholars who first used the Oriental Veisicns for purposes of textual criticism had to rely on Latin trans- lations of them, and many misstatements of the evidence have resulted, and may easily be perpetu- ated, even from the apparatus criticus of such an authority as Tischendorfs Sth ed. of his Novum Tcstamentum.i
iii. Uses of the Versions. — (a) It is only if we bear all these points in mind, as of possible im- portance in connexion with tlie evidence of a Version in a particular passage, that we are in a projier position to consider the most important of the uses which ma3' be made, especially of the early translations, viz. their use in textual cnticism. (1) We have three different classes of authorities for determining the text of the Bible, viz. MSS of the original Hebrew or Greek text, Versions, and Patristic Quotations. The importance of the last two is that they enable us to a great extent to date and localize particular readings found in the MSS, and thus provide us with the means by which to reconstruct the history of textual changes in a way which would be quite impossible from the MSS alone. An obvious instance of this may be found in the way in which Versions and Patristic Quotations enable us to trace back the readings of the so-called ' Western ' text of the NT to the 2nd cent., a date nearly 200 years before that to which our oldest MSS of the Greek are assigned. Without their help we might well have said that readings of this kind belonged to a much later date, and might be dismissed as unimportant. From the Versions we also see not only the an- tiquity but the wide prevalence of this so-called ' Western ' text, for its readings are found not only in properly called Western authorities, such as the MSS of the Old Latin Version, but also in the early Syriac Version. We see, therefore, how misleading this term ' Western '• is. On the other hand, caution has to be observed in using Versions to localize a particular text, for the Sanidic and Bohairic, though both connected with Egypt, repre.sent different Greek texts.