Bryn Mawr Classical Review 97.12.19


Dirk Obbink (ed.), Philodemus, On Piety Part 1. Oxford: Clarendon Press, 1996. Pp. xii, 676; 4 figures, 8 plates. $145.00. ISBN 0-19-815008-3.


Reviewed by Lee T. Pearcy, The Episcopal Academy, lpearcy@ea.pvt.k12.pa.us, Bryn Mawr College, lpearcy@brynmawr.edu.

Here is an account of the nature of virtue:

How this is ... will be mad[e ...
b]y the following consideration
of the specific ... that is
seven lines missing
neither too much ... not one, nor
the same for all ... ten ... and two is
few, six ... of the object;
for it exceeds ... amount; this
is intermediate ... proportion.
Given this document, which is in fact a mutilitation of Ross's translation of Nichomachean Ethics 1106a24ff, in this condition, could we reconstruct Aristotle's doctrine of virtue as the mean between two extremes? If all of Nic. Eth. was in a similar or worse condition, what could we know about Aristotle's ethics?

We might, as Dirk Obbink suggests (p. vi), know as much about it as we know about a dinosaur from a reconstruction of its skeleton. We would have the outlines and general dimensions of the beast, but its colors and behaviors would remain unknowable. The living, breathing animal would always elude us.

Obbink's latest contribution to our knowledge of Philodemus of Gadara presents the skeleton of a philosophical text at least as densely argued as Nic. Eth. The Greek in which it is written, according to Albert Henrichs, "is so esoteric" that it would be difficult to edit and translate even if De Pietate had come down from antiquity in mint condition (p. v). Much of Philodemus' treatise has been handed down in a condition not unlike that of the mutilated Aristotle quoted above. The rest is in even worse shape. Obbink's work in reconstructing this shattered, scattered text, even in its skeletal form, deserves to be numbered among the greatest philological achievements of our generation. A description of the evidence with which he worked will give some indication of what he has accomplished.

The papyrus roll containing the first part of De Pietate was among those covered by the eruption of Vesuvius that buried Herculaneum, with its Villa of the Papyri, in A.D. 79. These carbonized scrolls were so distorted and damaged that the original excavators mistook them for lumps of charcoal. In 1787 the roll containing De Pietate was opened. The best practice of the time, the so-called "method of Paderni," prescribed cutting through the charred outer layers of the roll until the pliable, un-carbonized inner layers came to light. Although this inner portion, the midollo, presented formidable challenges, it could often be read as a series of fragmentary papyri, and sometimes even as a partial roll of several columns, like P.Herc. 1428 which contains the end of the second part of De Pietate.

The outer halves, on the other hand, had to be destroyed in order to be read. These scorze, or "bark sections," consisted of fused layers of carbonized papyrus. Draftsmen, who were selected for their skill in drawing exactly what they saw and for their ignorance of Greek, made faithful copies of each layer before peeling it away to expose the next. In nearly every case, the act of revealing one layer destroyed its predecessor. In the end, the sole evidence for most of the outer layers of the first part of consisted of a series of these copies or apographs.

Enter now the malign gods of chance, muddle, and scholarship. When the first part of De Pietate was cut open, the two halves were for some reason given different catalog numbers and transcribed separately, with an interval of nearly two decades separating the transcriptions. One half, designated P.Herc. 1077, was drawn on two separate occasions between 1787 and 1809; shortly afterwards, the Rev. John Hayter absconded with several fragments of papyrus and the complete set of apographs, which are now in the Bodleian. The Bourbon government of Naples had another set of drawings made from the surviving fragments, but not before the edges of the outer half had been somehow severely damaged. Not until 1825 did the Italian technicians move on to P.Herc. 1098, the other half of the original roll, now believed to be a separate document.

Several editions of all or part of De Pietate appeared or were proposed during the nineteenth century, and some editors (most notably the first, Bernardo Quaranta, whose edition became a casualty of the collapse of the Bourbon Kingdom of Naples and remains to this day in manuscript) even suggested that P.Herc. 1077 and 1098 might belong to the same treatise. But because the Rev. Mr. Hayter had removed the evidence for one half of the treatise to Oxford, some early editors did not have important pieces of the text, while other, later workers who had all the pieces allowed themselves to become prisoners of the assumption that P.Herc. 1077 and 1098 were two independent series of columns.

Yet the evidence was there. Just before the knife that sliced through Philodemus' charred text in 1787 reached the soft midollo, it cut through a column (col. 54 of Obbink's edition) in which Philodemus, defending Epicurus against the charge of atheism, points out that Epicurus managed to live in Athens without becoming the butt of comedy or the target of lawsuits -- sure evidence of a life lived in accordance with the pieties of the time. Most of this column became part of P.Herc.1098 fr. 2 and remained in Naples; the left-hand margin of it, along with roughly every second preceding column, became 1077 fr. 1 and went to Oxford with the Rev. Mr. Hayter. In his edition of 1866, Th. Gomperz restored the sundered beginnings of the lines of the column in P.Herc. 1098 fr. 2 but did not recognize that his restorations matched the text of a partial column recorded on the Oxford apograph of P.Herc. 1077 fr. 1.

Obbink's edition, then, is the first to recognize and take advantage of the fact that the Oxford apograph of P.Herc. 1077 fr. 1 and the Naples apograph of P.Herc. 1098 fr. 2, together with Gomperz' restorations and the stichometric indications in the papyrus itself, give incontrovertible proof of a physical join at column 54 between the separately numbered papyri and allow the reconstruction of the original roll and its order of columns. Nearly every column of P.Herc. 1077 can now be matched with its successor and predecessor in P.Herc. 1098. Other bits of equally brilliant and meticulous detective work create joins between other fragments (p. 53), and it seems likely that the fundamental outlines of De Pietate and the order of the surviving fragments have now been established.

The story of this single papyrus roll is a complicated one, and roughly half of Obbink's 104-page Introduction, as well as substantial sections of the commentary on each column of the text, is devoted to it. Other sections of the Introduction treat the charge of atheism leveled against Epicurus and its history in the ancient controversy over Epicureanism, the treatise De Pietate itself, and the way in which it is presented to the reader in this edition. Obbink has designed a format full of information. On each left-hand page appears a diplomatic edition of one column of the original papyrus, with full apparatus criticus. On the top part of the facing right-hand page appears a continuous text of the same column for readers, with the spelling of the papyrus normalized -- e.g. E)KI/NEI for E)KEI/NEI -- and brief explanatory notes; below that is a literal translation, which makes no attempt to smooth over the awkwardness and difficulty of Philodemus' language. De Pietate must from now on be cited according to this edition, by Obbink's column and line number.

Detailed critical and exegetical notes follow the text. A bibliography, a concordance of this edition with previous ones, an index verborum, a general index, and an index locorum complete the repertory of tools for the study of this text. Even if this edition were not ground-breaking in its restoration of the true shape of this beast's skeleton, it would deserve praise for its complete, useful gathering of the bones. Its publishers, also, deserve gratitude for their willingness to undertake a complicated project for a market that must necessarily be limited to libraries and a few specialized scholars, and for bringing it out at what seems to me a very reasonable price. Typographical errors are few and easily recognized; only "roman" for "arabic" on p.99 will cause even momentary hesitation.

Obbink's philological labor and acumen have gained for us an important piece of evidence for the kind of intellectual debate that constituted Hellenistic philosophy in the wake of the founders of the Academy, the Lyceum, and the great schools of Stoicism, Epicureanism, Cynicism, and Scepticism. Students of intellectual life in the late Roman Republic will find that study of Philodemus' treatise adds to their understanding of that important period. When Horace recalled that as a young man he learned curvo dinoscere rectum, / atque inter silvas Academi quaerere verum (Ep. II.2, 44-45), he was looking back on debates like those fossilized in Philodemus' De Pietate. Philodemus argues that Epicurus believed in, or at least accepted, the existence of gods, understood as recurring, existent states of matter, producing and in some sense created by the mental concepts corresponding to them. This position and those against which Philodemus argues must lie behind Lucretius, whose alma Venus now appears less as an incongrous, epic introduction to an atheistic treatise than as a poetic assertion of Epicurean theism. Horace, Odes I.34 also now stands out against a rich background of polemic not only between Stoics and Epicureans, but perhaps within first-century Epicureanism itself.

Philodemus' treatise also reminds us of the subtlety, sophistication, and rigor with which ancient philosophers approached the fundamental questions of their subject, and of the continuing interest that their approaches and lines of reasoning can have today. Epicurus' arguments as reported by Philodemus often take a nuanced, relativistic line that seems in tune with modern and post-modern thinking. Epicurus, it seems (Obbink col. 12 and p.321), believed that the gods did not have a constitution or SU/STASIS like ordinary objects, which have a one-to-one correspondence with themselves. (An apple is an apple, because every bit of matter in the apple, no matter how small, exists in a defined relationship both to itself and to every other bit.) Instead, the gods are a relationship between certain recurring states of matter, rather like thoughts in some current models of human cognition, or action in a movie, or, to use Obbink's own image, a waterfall.

This treatise, recovered with such toil and learning from the ruins of Herculaneum, poses for a reader the question of its own SU/STASIS. What is the ontological status of its parts and the whole they form, of the bones and the dinosaur? All our classical texts are to a greater or lesser extent reconstructions, and we usually think that we know the answer to these questions. Aristomenes of Tauromenium (let us imagine) wrote three books on the history of the Crimean Anthropophagoi, and we have two and one-half books, transmitted in five manuscripts that can be classified in two groups, with an additional thread of occasional quotation in later ethnographers and anthologists. In cases like this we have an author, physical evidence, and a text to read. A clear chain of inference links author to evidence, and evidence to text. The editor, in this model, is a kind of textual palaeontologist, whose goal is an objective reconstruction that can be linked at as many points as possible to the evidence on which it is based. Even without the editor, the text is in some sense out there, objective and available to anyone who has the evidence for it.

No account of Philodemus' De Pietate, in contrast, can have this kind of relationship to the evidence on which it is based. Instead, like Epicurus' gods, Philodemus' De Pietate exists as a relationship between its readers, its editors, its transcribers, and its author. Joseph Farrell's description of the Vergil Project's indeterminate text applies equally well to the De Pietate: "This text is not a thing, but a system, and the user bears some responsibility for creating the text that he or she wants to use." Most of the physical evidence for Philodemus and his text crumbled as it was transcribed. The attribution of this treatise to Philodemus rests on a single letter phi preserved in the subscriptio (p.88), and it remains possible that "Philodemus" on this volume's title page should be "Phaedrus," the Epicurean whose work "On the Gods" or "On Piety" Cicero requested from Atticus in 45 B.C. (p.96). What remains of the text itself, often only the last layer of papyrus to have been transcribed, continues to change, deteriorate, and even create new evidence. Obbink's evocative description of this process (p.25) gives some indication of the Borgesian indeterminacy of this text's physical foundation:

Sometimes additional, unremoved layers (sottoposti lie intact underneath the last exposed layer. Not infrequently in the case of such fragments, the uppermost layers become separated from the lower layers over the passage of time. The phenomenon may have been caused by the fluctuation in climatological conditions in the Biblioteca Nazionale, geo-seismic factors, unauthorized interference with the surface of the scorza by readers, or even simply by moving the cassetto in order to facilitate access, so that scorze sometimes multiply over time in the boxes (cassetti) in which they are conserved. Sometimes once visible texts simply vanish, because the exposed layer of the extremely thin papyrus disintegrates, leaving in its place sometimes a mysterious heap of dust, sometimes a new text on the underlying layers where twenty years before there had been a different one.
Philodemus' De Pietate, like the gods of Epicurus, seems to proceed from a series of EI)/DWLA, thin, filmy emanations -- apographs, fragments, conjectures and rearrangements. As Obbink explains (p. 322), "Each EI)/DWLON constituting a god is different from any of its other EI)/DWLA, though they are similar in certain respects, and regularly and consistently coalesce in the human mind to form the appearance of an individuated entity." So the De Pietate, a text constituted in the best post-modern fashion by its users, consists of a series of images that have regularly and consistently coalesced into a text. Obbink's great achievement is to give us an EI)/DWLON more vivid and persuasive than its predecessors. In his hands, the dry bones of this skeletal reconstruction give off light.