547
identical with the prevalent Greek text of the Middle Ages was used by Chrysostom and other Antiochian Fathers in the latter part of the fourth century, and thus must have been represented by MSS as old as any MS now surviving. This Antiochian or 'Syrian' text can frequently be recognised as standing out in opposition to the text or texts of most of the definitely ancient documents.
Another great landmark is furnished by the writings of Origen, which carry us to the middle of the third century, and even earlier. They establish the prior existence of at least three types of text, which can be identified through numerous readings distinctively attested by characteristic groups of extant documents. The most clearly marked of these is one that has long been conventionally known as 'Western'. Another, less prominent as being less consistently represented by any single ancient document, may be called 'Alexandrian'. The third holds a middle or neutral position, sometimes simply opposed to Western or to Alexandrian readings, occasionally opposed to Western and to Alexandrian readings alike. On the other hand Origen's writings contain no certain traces of distinctively Syrian readings.
The priority of two at least of the three texts just noticed to the Syrian text is further brought to light by the existence of a certain number of distinctively Syrian readings which prove on close examination to be due to a combination of the Western with the neutral readings. Moreover the use of Western and of neutral readings thus presupposed renders it morally certain that other readings from the same sources were adopted as they stood, sometimes from a Western, sometimes from a neutral text; and the supposition is fully confirmed by an analysis of the distribution of documentary attestation. A similar analysis in other cases shews that Alexandrian readings also were sometimes adopted by the authors of the Syrian text. To the two processes of combination and direct selective adoption must be added a third, selective adoption with modifications. In fact the Syrian text has all the marks of having been carefully constructed out of materials which are accessible to us on other authority, and apparently out of these alone. All the readings which have an exclusively Syrian attestation can be easily accounted for as parts of an editorial revision; and none of them have the stamp of genuineness to attest the use of extraneous and purer sources.
