THE APOSTOLIC FATHERS: by the late J.B.Lightfoot DD., DCL., LL.D. Lord Bishop of Durham. Edited and completed by J.R.Harmer MA. Fellow of Corpus Christi College, Cambridge, sometime Chaplain to the Bishop. First published Macmillan & Co. 1891. Prepared for katapi by Paul Ingram, 2014.
HOME | APOSTOLIC FATHERS: contents. | SHEPHERD: intro | setting | date | text auth. | text & translation |


THIS work is entitled in the most ancient notices 'The Shepherd', or 'The Shepherd of Hermas'. Hermas is both the narrator and the hero of the narrative. The Shepherd is the divine teacher, who communicates to Hermas, either by precept or by allegory, the lessons which are to be disseminated for the instruction of the Church. Later confusions, which identify Hermas with the Pastor, find no countenance in the work itself. Hermas' own personal and family history are interwoven from time to time into the narrative, and made subservient to the moral purposes of the work. In this case it resembles the Divina Commedia, though history plays a much less important part here than in Dante's great poem.

The structure of the work is seriously impaired by the common division into three parts or books, Visions, Mandates, and Similitudes, as if they stood on the same level. It may be convenient to use this mode of division for purposes of reference alone; but we must not suffer it to dominate our conception of the work. The Visions are introductory, and the Shepherd does not appear until their close. He delivers his message to Hermas in two parts, (i) Mandates or Precepts, (2) Similitudes or Parables, i.e., moral lessons taught by allegory.

The person first introduced in the book is one Rhoda (Vis. i. x), to whom Hermas had been sold when brought from Rome as a slave. 1 ter part is somewhat the same as Beatrice's in Dante's poem. She appears to him in the heavens as he is on his way to Cumai, and reproaches him with his not altogether blameless passion for her. I laving thus aroused his conscience, she withdraws. Then he sees before him an aged woman whom (considering the place) lit: not unnaturally mistakes for the Sibyl (Vis. ii. 4), but who proves to be the Church. The object of the Visions indeed seems to be to place before the reader the conception of the Church under the guise of an aged woman, whose features become more youthful at each successive appearance. Thus the lessons of a smitten and penitent conscience, of the Church growing and spreading (the Church Militant), lastly, of the Church purified by suffering (the Church Triumphant), and the terrors of the judgment, occupy the four Visions properly so called. Hermas is enjoined to write down all that he hears. One copy of his book he is to send to Clement, who is charged with making it known to foreign cities ; another to Grapte, whose business it is to instruct the widows and orphans, and he himself, together with the presbyters, is to read it to the people of 'this cityi.e., Rome ( Vis. ii. 4).

The fifth Vision is different in kind from the preceding four, and indeed is designated, not a Vision (ὅρασις), but a Revelation (ἀποκάλυψις). Hermas is now in his own house. The appearance is no longer the representation of the Church, but a man of glorious visage in a pastoral habit, who has been sent to dwell with him, and teach him to the end of his days. He is 'the Shepherd, the angel of repentance', who delivers to him certain Mandates and Similitudes, which he is ordered to write down, and which form the two remaining books—the main part of the work.

The teaching of the Shepherd then is contained in the twelve Mandates and the ten Similitudes which follow. But the tenth and last of the latter is not strictly a parable like the rest. It contains a final chapter, summing up the function of the Shepherd and his heavenly associates, in the work of perfecting the instruction of Hermas.


The geographical setting of the narrative has its centre in Rome, where evidently the work itself was written. Hermas' home in the city, the road to Cumse, the Via Campana,—these are the localities mentioned by name. There is one exception. Arcadia is chosen as the subject of a Similitude (Sim. ix.), the last properly so called, because the mountains visible from a central height by their character and position afford a good subject for the concluding parable, the component elements of the Church (see J. A. Robinson, The Athos Codex of the Shepherd of Hermas, p. 30, where the views of Rendel Harris are discussed and further developed and modified). As he was brought to Rome, and sold as a slave there, Arcadia may have been his native place.


The date is uncertain. The work is found in general circulation in the Eastern and Western Churches, soon after the middle of the second century. About this time also it must have been translated into Latin. It is quoted by Irenaeus in Gaul, by Tertullian in Africa, by Clement and Origen in Alexandria. All these fathers—even Tertullian, before he became a Montanist—either cite it as scripture, or assign to it a special authority as in some sense inspired and quasi-canonical. The same inference as to its early influence may be drawn from the denunciation of Tertullian, who—now become a Montanist—rejects it as repulsive to his puritan tendencies (de Pudic. 10), and the author of the Muratorian Canon (c, A.D. 180), who denies it a place among either the prophets or the apostles, though apparently allowing it to be read privately for edification. Its canonicity moreover had been the subject of discussion in more than one council, when Tertullian wrote (I. c., not before A.D. 212).

With the date is closely connected the question of authorship. On this point there are two ancient traditions.

(1) The author of the 'Shepherd' was the same Hermas, who is greeted by S. Paul as a member of the Roman Church, A.D. 58 (Rom.xvi.14). This is the view adopted by Origen (iv. p. 683) in his commentary on the passage, where he speaks of the book as 'a very useful scripture, and in my opinion divinely inspired'; but, as he introduces this view of the authorship with ' ut puto' it is plain that he does not fall back on any historical tradition in support of his opinion. His influence had great weight with subsequent writers.

(2) It was written by one Hermas, the brother of pope Pius I (c. A.D. 140—155) during the episcopate of the latter. This is stated in the Muratorian Canon (c. A.D. 180) 'sedente cathedram urbis Romae ecclesiae Pio episcopo fratre eius'. This statement, however, is not consistent with the mention of Clement as a contemporary. If it be true, either some other Clement is meant, or the original Greek of the Canon, of which only the Latin is extant, cannot have stated that Pius was actually bishop at the time when it was written.

This tradition appears likewise in one or two subsequent writings, which however are perhaps not independent. It is somewhat discredited by the fact that its motive in depreciating the value of the work, as being quite recent and having no claim to be read in the Church like the writings of the Apostles and prophets, appears in the context. These words are illustrated by the fact that (a) in the Codex Sinaiticus (א) the Shepherd (a fragment, see below, p. 295) appears at the end of the volume, following on the Epistle of Barnabas, which again follows the Apocalypse and the books of the Canonical New Testament; (b) in the list appended to the Codex Claromontanus (VIth Cent.) again il follows the New Testament proper, of which the closing books are 'Revelation of John', 'Acts', and is succeeded by the apocryphal 'Acts of Paul', and ' Revelation of Peter'; (c) in several mss of the Latin version it appears in different parts of the Old Testament.

(3) Besides these two traditional views, a third and intermediate Hermas, not otherwise known, is postulated as the author about a.d. 90—100, to meet the difficulty about Clement. This is the view of several recent critics (Zahn, Hirt des Hermas p. 14 sq, followed by Caspari and others). The notices of the Christian ministry, and of the condition of the Church generally, seem to be consistent with either the second or the third view, though they suggest the earlier date rather than the later ( Vis. ii. 2, 4, iii. 5, 9, Sim. ix. 27).

On the whole we may, though not without diffidence, adopt (2) the ancient tradition, which is definite and claims to be almost contemporary, as the safest guide; though confessedly (3) the modern suggestion has stronger support from internal evidence, such as it is.

The Æthiopic version, which identifies the author with S. Paul, ought to be regarded as a blunder, rather than a tradition founded on Acts xiv.12 τὸν δὲ Παῦλοω Ἐρμῆν.


The authorities for the text are as follows :


1. The celebrated Sinaitic MS (א) of the fourth century, where, after a gap caused by the loss of six leaves, the Shepherd follows the Epistle of Barnabas at the end of the volume. Unfortunately, however, only a fragment, roughly speaking the first quarter of the text, survives, the manuscript, after several lacunae, breaking off finally in the middle of Mand. iv. 3.

2. The Athos MS (A), written in a very small and cramped hand of the fourteenth century. This consists of three leaves now in the University Library at Leipsic, and six leaves still remaining in the Monastery of Gregory on Mount Athos. The portion of the manuscript now at Leipsic was in 1855 brought from Mount Athos by the famous forger Simonides, who sold it to the University there, as well as what purported to be a copy of six other leaves of the same document. This copy was subsequently edited by Anger. The existence, however, of the original manuscript was questioned until 1880, when Dr Lambros rediscovered it at Mount Athos. His collation of the readings of these six leaves was in 1888 published by J. A. Robinson (The Athos Codex of the Shepherd of Hermas ). Like the Sinaitic, this manuscript is incomplete, having lost a leaf at the end; but from Mand. iv. 3 to Sim. ix. 30 (where it fails us), that is to say, for nearly three fourths of the whole work, it is our sole Greek authority for the text.

Besides Simonides' apographon mentioned above, another copy was subsequently found among his papers after his arrest, and published by Tischendorf. The publication of Dr Lambros' collation shows us that, whereas the apographon edited by Anger was a forgery, the second apographon was truly described as being a transcript of the Athos MS. In passages therefore where the Athos codex has become damaged and illegible between 1855 and 1880, this apographon (As) has a certain value.


1. Latin Versions. These are two in number, (a) the so-called Old Latin Version (L1), which exists in about twenty manuscripts, the mutual relation of which has not yet been made quite clear. From this version Faber Stapulensis published his editio princeps in 1513. (b) The Palatine Version (L2), found in one manuscript of the fourteenth century, and in 1857 published in full by Dressel. Both these versions give us the text virtually complete.

2. Æthiopic Version (E). This exists in a manuscript discovered in 1847 in the monastery of Guindaguinde by A. d'Abbadie, who procured a transcript, but did not realise the full importance of his discovery. At length at Dillmann's earnest request he published the text with a Latin translation in 1860. This version likewise contains the Pastor complete.

The mutual relations and comparative value of our authorities are matters of considerable dispute; but a comparison of the early chapters, where the Greek of the Sinaitic ms exists, shows us that א generally agrees with L, L2 against AE, the close connexion of this latter pair of authorities being noticeable throughout. Again, within these groups, L2 appears to preserve a purer text than L1 and E than A.


Besides these direct authorities for the text, the Shepherd of Hermas is quoted in the Greek by Clement of Alexandria and Origen, while considerable passages have been incorporated into the texts of Antiochus the Monk and ps-Athanasius.