Shepherd of Hermas – Robert D. Heaton, Ph.D.

For a recent term paper I needed to catalogue the contents of the 24 extant Greek manuscripts (pre-7th century) of the Shepherd of Hermas. I’ve written here about the Shepherd’s significance from the 2nd to 4th centuries CE before, but for this paper I was interested in all of the Greek manuscripts we have before Latin became the primary language of both the Church at large and the Shepherd’s use/preservation.

cssh-92-4 — The *Shepherd of Hermas* in Codex Sinaiticus (Herm. 92.2 [Mand. 9.15.2]).

I was surprised to find that no one had really put together the relevant information in a list or chart. Appendix 1 in Larry Hurtado’s The Earliest Christian Artifacts: Manuscripts and Christian Origins (pgs. 224-225) only extends to the early fourth century, as he’s primarily interested in the pre-Constantinian period. But he doesn’t list contents of the manuscripts.

I built off the graph Hurtado started, adding all other manuscripts through the 6th century CE with a number of additional fields. For one thing, the Shepherd uniquely uses two systems of numbering/citation—one continuous, with chapters numbering 1 to 114, and another that breaks the chapters up into sections of 5 Visions, 12 Commandments/Mandates, and 10 Similitudes.[1] I list the contents of each manuscript using both notational forms, given that both still carry currency and are recommended by SBL in citation of the Shepherd. I’ve also linked to each manuscript fragment’s entry in the exceedingly useful Leuven Database of Ancient Books, or LDAB, and to each manuscript’s free and open online access, where available. Without further ado (click image for full chart):

shep — Click the image for full chart (PDF).

I hope that this chart is useful to other scholars and interested persons. In the future I may add where the critical editions of each manuscript can be found, but LDAB has much of this information even if it is sometimes difficult to decipher. Let me know if you see any errors that require fixing.

Offline Citations

[1] Because this is so wonky, we who study the Shepherd even get our very own Appendix in the SBL Handbook of Style! See Appendix D of the 2nd ed., pgs. 331-332.

Based on manuscript recoveries alone, the most popular books in second, third and early fourth-century Christianity[1] were as follows: the Book of Psalms, the Gospel according to John, the Gospel according to Matthew, and the non-canonical Shepherd of Hermas.

According to Larry Hurtado’s 2006 book The Earliest Christian Artifacts: Manuscripts and Christian Origins, manuscript recoveries for biblical books in this period are as follows:

Psalms: 16-18 copies[2]
John: 11-15 copies
Matthew: 12 copies
The Shepherd of Hermas: 11 copies
Genesis and Exodus: 8 copies each
Luke and Acts: 7 copies each
Isaiah: 6 copies
Revelation: 5 copies[3]

A number of factors complicate matters here: very few books from the second century in particular survive complete and intact, whether from natural use and deterioration or deliberate destruction by the ruling Roman authorities during periods of local persecutions. Many of the manuscripts that survive are just a handful of pages or no more than a couple of verses or chapters. For the book of John in particular, scholars are unsure whether some of the pieces belonged to the same manuscript or represent distinct manuscripts. And finally, all of our conclusions related to this earliest period must be considered “temporary” or “in progress”: as Hurtado admits, “only about 1% of the estimated 500,000 manuscripts from this period” have been properly identified![4]

The popularity of John and Matthew in this early period is pretty understandable, given that the books were thought to have been written by the disciples of Jesus of those names. The book of Psalms might not have been your first guess at the most popular book in early Christianity, but its popularity is also sensible.

What might throw you off is the early popularity of the Shepherd of Hermas, a book that is now only really known among the scholarly community. Written in Rome in the mid-second century, the Shepherd of Hermas consists mostly of moral instructions delivered in the form of revelations from a shepherd-like character to the author, named Hermas. The book is exceptionally long—if it had made its way into the canon, it would be easily the longest book of the New Testament, and only exceeded in length in by the Old Testament books of Genesis and Jeremiah.[5] In the book, Hermas portrays himself as somewhat of a ditz—he must be given explanations about all of his visions by the shepherd—although he is genuinely interested in moral improvement and proper behavior.

So, why would a book like this eventually be regarded as non-canonical, in spite of its early popularity? In other words, why is the Shepherd not in the Bible? That discussion may feature as a “Random Bible Fact” in the future.

[1] Why this period, from roughly 100-320 CE? This is before Christianity became the entrenched state religion of the Roman Empire, before a rigid orthodoxy developed from the unifying vision of Emperor Constantine.

[2] Two copies of the Pslams retain the Tetragrammaton, or “Yahweh” in Hebrew characters, as the name of God, and therefore these may be Jewish manuscripts rather than Christian. Either way, the book of Psalms wins out as the most popular pre-Constantinian Christian book.

[3] These data come from Hurtado (2006), 19-28 (see especially, 19-21 and 23). His book includes an excellent appendix cataloging and detailing the precise contents of biblical and apocryphal books from this period.

[4] Hurtado (2006), 25.

[5] This is according to the stichometric list inserted into the sixth-century Codex Claromontanus, which lists the Shepherd at 4,000 lines, Genesis at 4,500, and Jeremiah at 4,050.