UNTIL PREFACE. NTIL about thirty years ago the Shepherd of Hermas was known to us only in a Latin Version. In 1857 Dressel edited a second Latin Version from a Palatine manuscript of the fourteenth century preserved in the Vatican. In 1860 an Ethiopic Version was also published. This had been discovered by Antoine d'Abbadie some thirteen years previously: but he was not aware of its importance until he had shewn it to Dillmann, with whose assistance he then published it together with a literal Latin translation1. Meanwhile an extraordinary controversy was raging in Leipsic. In 1855 Constantine Simonides had sold to the Library of the University what he declared was the original Greek text of the whole of the Shepherd, with the exception of a small missing portion at the close. This he produced in a twofold form, consisting of (1) three leaves of a paper MS. from Mount Athos, very closely written in a hand of the fourteenth century, and (2) a copy of six other leaves of the same MS. which he had not been able to bring away with him. The text thus obtained was immediately published by Anger and Dindorf2 who edited it with scrupulous exactness from the three original leaves and the apographon of Simonides. They promised to add a volume of critical materials: but this 'pars 1 Hermae Pastor. Aethiopice primum edidit et aethiopica latine vertit Antonius d'Abbadie. Lipsiae, 1860. It came out as no. 1 of the second series of the Abhandlungen der Deutschen morgenländischen Gesellschaft. It has a strange argument appended at the close to prove that Hermas was S. Paul himin proof of which is quoted the text: " They called Silas Zeus and Paul Hermes". self; 2 Hermae Pastor. Graece primum ediderunt et interpretationem veterem Latinam ex codicibus emendatam addiderunt Rudolphus Anger et Gulielmus secunda' was not destined to appear. For just at that moment Simonides was accused of having forged the 'Uranius'. He was arrested and sent to Berlin: and among his papers was found another copy of the same six leaves of the Hermas Ms. This upon examination was shewn to contain a very different text from that of the apographon which he had sold to the Leipsic Library: it had moreover been largely corrected and modified both in pencil and in ink by Simonides himself. When the two apographa were compared, it was found that the text of the one which Anger had already edited embodied for the most part the corrections of the other. At this point Tischendorf took up the matter, and edited for Dressel the three genuine leaves, together with the more recently discovered apographon, which alone he believed to be of any value at all. At the same time he propounded two theories: first, that the Greek text of the Athos MS. was not really the original Greek of the Shepherd, but had been constructed in the middle ages out of some Latin Version, which was however neither the 'Old Latin' nor the 'Palatine': secondly, that, the apographon, which Anger and Dindorf had used for their edition, was written by Simonides not on Mount Athos at all, but in Leipsic; whereas the other copy was really made on Mount Athos and afterwards corrected and modified by the aid of the Old Latin Version and the Greek quotations in the Fathers. The fact that Simonides had also produced certain pages of Palimpsest of the Shepherd seemed to reveal the motive of this strange patchwork. Tischendorf held that Simonides had kept back his Athos copy so as to have a different text to use in forging his Palimpsest. The first of these theories was challenged at the time, and was finally disposed of by Tischendorf's own discovery of the Codex Sinaiticus. From this great Bible he published in 1863 a fragment of the Shepherd, which comprised roughly speaking the first quarter of the book, and presented a text in substantial agreement with that of the Athos Ms. The second theory met with a more favourable reception. Anger and Dindorf at once admitted that they had been deceived, and that their edition was absolutely worthless. In 1866 Hilgenfeld re-edited the Greek text from the Sinaitic Codex and the Leipsic materials. In his apparatus he distinguishes to some extent between the two apographa, though he relies mainly on the one which Tischendorf had edited. In 1877 appeared the edition of Gebhardt and Harnack, which gives the Palatine Version infull, and provides an apparatus criticus dealing with the Greek texts and the Aethiopic and Old Latin Versions. It is noteworthy that Harnack (note on p. vii of his Prolegomena) absolutely scorns the apographon which Simonides sold to the Leipsic Library, and collates only the other which was subsequently extracted from him by the police. Hilgenfeld however in his latest edition (1887) has taken a very different course. The occasion of this new edition is the recent publication by Draeseke of what claims to be the missing Greek conclusion of the Shepherd. This document was printed by Simonides in 1859 together with some other miscellaneous tracts: but by that time he had so completely lost his character, that no one would even look at anything he produced. Draeseke has now rediscovered it, and he asserts its genuineness. Hilgenfeld follows him, and accordingly publishes for the first time a complete Greek text1. Harnack on the contrary de nounces it as an obvious fraud2. Hilgenfeld, then, in his new edition states his belief that Simonides really made both his apographa on Mt. Athos, as he said; but that he probably made them from different MSS., the readings of which he somewhat capriciously modified partly by collation and partly by conjecture. This agrees substantially with the account of the matter given by Simonides himself, who always spoke with the greatest contempt of the text which he had kept concealed, as being the recent work of one Abraham the Telian, who had depraved it by the introduction of modern Greek forms. This latest edition of Hilgenfeld's is a very laborious but a most unfortunate piece of work. In the first place, Harnack is 1 Hermae Pastor. Graece integrum ambitu primum edidit Adolfus Hilgenfeld. Lipsiae, 1887. 2 Ueber eine in Deutschland bisher unbekannte Fälschung des Simonides. undoubtedly right in saying that the supposed Greek ending is nothing but a loose retranslation from the Latin: and, with regard to the origin and the respective merits of the two apographa, the decision of Tischendorf, which Harnack has always accepted, is now completely confirmed; on the one hand by the evidence presented by Hilgenfeld himself who has collated both the apographa again for his new edition; and on the other hand by the fresh light which has been thrown upon the subject by the discovery of which I must go on to speak. In the Easter Vacation of last year, as I was going to Patmos to collate a MS. of Origen's Philocalia, I made the acquaintance in Athens of Dr Spyr. P. Lambros, who is well known for his labours in cataloguing the MSS, of the numerous monasteries of Mount Athos. He was good enough to give me the proof-sheets of the earlier portion of his Catalogue, and he called my attention especially to his description of a Codex containing portions of the Shepherd of Hermas, which he was at first disposed to regard as spurious, but on which he promised to give me further information. He has since placed in my hands, for translation into English, and for publication, a tract which he has written in German, dealing briefly with the previous history of the Greek text of the Hermas, and containing a full collation of the remaining fragments of the MS. from which Simonides extracted the three leaves which are now at Leipsic. Of the last leaf of all Dr Lambros could find no traces whatever. There can be little doubt that it was torn away before Simonides ever saw the Codex. In order to secure the greatest possible accuracy in the Collation Dr Lambros sent over at my request the transcript of the MS. made for him by Dr Georgandas, who went to Mount Athos for the purpose. I have gone carefully through it in order to verify the proof-sheets, and Dr Lambros has since seen the proof-sheets himself, and has revised them again by a fresh comparison with the transcript. It will be observed that nearly half the corrections are marked with the word (sic). In all these cases the Editors, had already been able by the aid of the Sinaitic Codex or the Versions to restore the true text, which |