THE NEW SCHAFF-HERZOG
of the Apostles' Creed, has now been abandoned. The only other author's name is offered by a lost manuscript from St. Gall, printed by Melchior Goldast in 1610, which calls it Euphronii presbyteri expositio. Morin identified this Euphronius with the bishop of Tours of that name (555-572), who was well known to Venantius Fortunatus. Burn is inclined to see its author in Euphronius of Autun, who built the church of St. Symphorian there about 450. But this positive criticism is very hazardous in view of the number of anonymous manuscripts, to say nothing of the frequency of the name Euphronius in Gaul. A more important question is that of its date. An attempt has been made to decide this from the fact that the author explains the words in sceculo in section 31 of the creed (Schaff, Creeds, ii, New York, 1887, 68) by " that is, in the sixth millennium [sextum miliarium] in which we now are." This has been supposed to indicate 799 as the terminus ante quern ; but no stress can be laid on this; people spoke of the sextum miliarium, with Augustine, after 799 as well as before it. Just as little can be made of its supposed dependence on Alcuin for a terminus post quern, as Ommanney has shown. The only sure limit of date might be supposed to be given by the fact that the oldest manuscript (Bodleian. Junius 25) belongs to the ninth century - — probably the beginning — were it not that a whole group of other ancient commentaries allow us to put the terminus ante quern further back. Ommanney has rendered a signal service to the investigation by the discovery of these, and Burn has followed independently. These are, in the order of the dates given by Burn: (2) the Expositio Parisiensis, certainly written between Gregory the Great and 900; (3) the Expositio Trecensis, assigned by Ommanney to the seventh, by Burn to the end of the eighth century; (4) the Expositio Oratorii, found in the same manuscript, dated by Ommanney about 700, by Burn a century later; (5) the Stabulensis, ninth century according to Burn; (6) the Buheriana, based on (4), and written, according to Ommanney, in the first half of the eighth century, to Burn, in the ninth; and (7) the Aurelianensis, first published in 1892 by Cuissard, who attributes it to Theodulf of Orleans, while Burn is for an author of the middle or end of the ninth century. Now, of all these commentaries, only the Expositio Fortunati and the Trecensis (which in its first part is very dependent on the former), do not evidence a knowledge of the entire Quicunque. To be sure, Burn's dates — to say nothing of Ommanney's — are by no means certain. But none the less these commentaries are of great importance as helps to a decision of the difficult problem under discussion. The last-named, one of the latest (because dependent on three or four of the others), is preserved in a manuscript which Delisle assigns to the ninth century; and the Trecensis, used in the compilation of this, presupposes in its turn the Expositio Fortunati. This being so, it is not too bold a conclusion that the latter, everything about which shows it to be the oldest of them all, belongs to the period before 799. If this is granted, one may go a little further, and point out
that since its author says nothing about the approaching end of the sextum miliarium, he did not live very near that date.
Both the Expositio Fortunati and the Expositio Trecensis leave certain verses of the Quicunque without mention. Are we to conclude that the whole of it was not known to their authors ? We have seen how far the testimony of the manuscripts supports the theses of Ffoulkes, Swainson, and Lumby; our Quicunque was definitely in existence
before the end of the eighth century.
4. The But that does not in itself militate
Theory of against the acceptance of the theory of
Two two sources; Harnack considers it
Sources, possible that both halves of our present
creed were found in conjunction in the eighth century, or even earlier. We must therefore look further into that theory. Its main support is the manuscript referred to above as (3), the Codex Colbertinus 784 (now known as Paris. 3836), which all authorities agree to place in the eighth century, Swainson dating it as early as 730. In this manuscript the Christological portion of the Athanasian Creed (though with noteworthy verbal variants) is found under the rubricated caption Hmc invini treveris in uno libro scriptum sic incipiente Domini nostri Jesu Christi fideliter credat et reliqua. Now, assuming that the scribe copied exactly what he found in the Treves manuscript, Swainson, Lumby, and Harnack see in this text, which goes well back into the eighth century (possibly to 730), distinct documentary evidence for the separate existence of the Christological half of the Quicunque. But it does not seem to have been observed that the manuscript will not sustain this contention. The copyist put down in red ink, as his introduction, words which actually form a part of the verse which makes, in the complete creed, the transition from the Trinitarian to the Christological section. The " Treves fragment " is thus really a fragment — part of a whole whose first half stood in the same relation to our Quicunque as the extant second half. There is nothing surprising in this conclusion. That a preacher (and Swainson himself has noticed that this fragment is clearly a fragment of a sermon) should have undertaken to set forth " the faith," and then have spoken only of the Incarnation and not of the Trinity, would have been much more surprising. But the conclusion, if not surprising, is none the less weighty; for it takes both balves of the creed distinctly further back than any of the manuscripts described above. We do not know how old the Treves manuscript was when the writer of Paris. 3836 copied it in 750 or 730; but there is room for a logical train of reasoning which leads to valuable results. It is obviously improbable that a copyist with a complete manuscript before him should copy only the last part, beginning in the middle of a sentence; therefore the Treves manuscript (or its original) must have been defective. This train of thought gains in force when we notice that the " fragment " represents exactly a third of our Quicunque. On the assumption that the two first pages of the original went down to incarnationem quoque, the third beginning