1 Herculaneum and the Villa of the Papyri
In the Bay of Naples, buried between about twenty meters of volcanic material and debris, lie the Roman cities of Pompeii and Herculaneum. Ever since the eruption of the Vesuvius in 79 the two cities have functioned almost like a time capsule; preserving a snapshot of daily life in Roman times within a thick cover of dust and ash. The rediscovery of Herculaneum commenced in 1709, when news spread that local well-diggers discovered ancient art and statues buried in the ground, presumably from Herculaneum’s ancient theatre.1 The excavations of the area started with Prince Emmanuel d’Elboeuf taking over from the well-diggers and starting a search for more sculptures and other artefacts that European elites could acquire to decorate their houses.
For our purposes here, 1750 is even more significant, because in that year Karl Weber, the Swiss army engineer and architect that had joined the excavations the year before, investigated a marble floor that turned out to be part of the structure that is now known as the Villa of the Papyri, a large villa similar to the other suburban residences in Herculaneum, Pompeii, and Stabiae where Roman elites came to escape the commotion of the capital. The exploration of the Villa was largely motivated by the potential to acquire bronze and marble statues, but Weber explored the site very systematically and the measurements he took in the excavation tunnels allow us to get some idea of the size of the structure.2 The construction of the Villa is, with some variety, generally dated to the second half of the first century BCE.3 The main building was arranged according to the plan of a classical Roman atrium, and surrounded by a very large garden of 94 × 32 m, and a smaller one of 21 × 21 m.4 Antonio De Simone characterized the Villa as “one of the clearest and most significant examples of this phenomenon of regal residences,” which is supported by the presence of significant collections of about 90 statues, wall paintings, metal objects.5 The clearest indication of the extraordinary status of the Villa however is the large papyri collection it would be named after.
2 Deposition Context and ‘Lived’ Context
Our main literary source for the destruction of Herculaneum is found in two letters Pliny the Younger wrote to Tacitus in which he provides us with an eyewitness account of the eruption of the Vesuvius. He had witnessed the disaster from Misenum in the Bay of Naples, where his uncle Pliny the Elder was stationed.6 Though Pliny’s account mainly focuses on the heroics of his uncle, it at the same time evokes a scene of complete terror and bewilderment. From one moment to the next the roughly 20,000 inhabitants of Pompeii and Herculaneum were struck by an event they could not understand. The tragic circumstance that people were trapped during their daily lives provides scholars today the opportunity to access a first century Roman site that is undisturbed by later events. The unique circumstance of discovering an ancient urban centre unhampered by centuries of occupation is an important characteristic of the site of Herculaneum that contributes to the significance of the Villa of the Papyri.
The discovery of the Villa of the Papyri is of importance for the study of book collections in antiquity and for understanding the ways in which ancient scribes and readers engaged with their texts. Here, I wish to focus on the relevance of deposition and lived contexts. In his analyses of the Dead Sea Scrolls and other Judaean Desert manuscript collections, Popović developed the distinction between the manuscripts’ deposition context, which he defines as “the evidence that caused manuscripts being deposited in the places where they were found,” and their ‘lived’ context, “the use and the owners of the Dead Sea Scrolls before they were deposited or left behind.”7 As Popović notes, this distinction is to a certain degree arbitrary because the archaeological records in general only preserve the deposition context, leaving the challenge to researchers in order to reason back to the lived context.8 Nevertheless, it serves as a point of departure and heuristic tool by allowing scholars to arrive at a more sustained analysis of their evidence by separating the interpretation of these two contexts in order to think through the different scenarios in which the manuscripts were used and disseminated.
It is important to first get to a careful description of the deposition context in which historical artefacts were discovered. This is not self-evident. Archaeologists have traditionally distinguished between ‘primary’ and ‘secondary’ contexts to refer to the degree of disturbance since objects were deposited.9 Artefacts that have not been disturbed or transformed after their deposition are said to be in their primary context. However, many different factors can affect objects after their original abandonment. Jürgen Zangenberg refers to these factors as post-depositional processes and points to the many natural (such as erosion or collapse), animal-made (silverfish or bat guano), and human (looting or demolition) causes that impact the condition in which a site survives before modern archaeologists first inspect it.10 Artefacts that have been affected by such post-depositional processes are said to be in their secondary context. The distinction between these different contexts is to a certain extent a scholarly construct. A completely primary context seems utopian, because objects under normal conditions are continuously influenced by their surroundings. However, the degree to which these post-depositional processes take place is very much context-dependent. Pompeii and Herculaneum are very unusual in this regard: the thick layer of volcanic ash has allowed both sites to be preserved in extraordinary conditions, and to be much less affected by post-depositional processes than otherwise would have been the case. This is evidenced by the fact that Herculaneum is one of the very few locations where a sizable papyrus concentration has survived outside Egypt (where natural conditions are much more favourable for the preservation of organic material). The absence of significant post-depositional interference in Herculaneum allows us to move much closer to the primary depositional context. In the case of the Villa of the Papyri this constitutes the situational context in which the manuscripts were left behind.
Subsequently, when the depositional context is understood as best as possible, one can move forward and try to think through the ‘lived’ context of the site, the circumstances under which our artefacts were deposited, and part of the social reality of the people that were involved. As we will see, the deposition context can provide different direct and indirect forms of evidence that can help us contemplate the lived context.
Of course, in most cases only the deposition context has been preserved to base our enquiries on. The deposition context of the Oxyrhynchus papyri is a trash dump where the manuscripts were discarded together with other rubbish ranging from keys and textile to broken pottery.11 This does not mean that the archaeological context is of no value: which types of manuscripts were disposed in such a prosaic fashion can tell us more about how particular manuscripts were perceived at a certain moment. However, the deposition context of Oxyrhynchus does not provide a clear window on how the papyri were used before they were deposed, which manuscripts were collected together, and which types of reading events the people behind the papyri organized around them.
Nevertheless, regarding the Dead Sea Scrolls, Popović argues that under certain circumstances it is “possible to reason back from the deposition context to the lived context.”12 To provide an example of scrolls that were possibly discovered in their original, lived context Popović points to the Psalms manuscript Mas 1f (MasPsalms-b) and Ben Sira manuscript Mas 1h (MasSir) that were found at Masada.13 These manuscripts were discovered in rooms where individual families resided during the years of the Jewish Revolt, which indicates that they were part of the property of a small family rather than for example a library or institution.14 Such evidence is important to consider when thinking about the dissemination of book possession. One can safely infer, for example, that book possession was not restricted to libraries or other institutions, but that individuals or families could own at least a limited corpus of manuscripts as well.15 More specifically, the deposition context of these Masada manuscripts indicates that such book possession extended not only to legal documents, but also to literary manuscripts. These privately owned manuscripts were apparently considered to be of such value that refugees took them with them on their search for safety.
3 The Archaeology of the Villa of the Papyri
The Villa of the Papyri seems to present us with the rare occasion of both deposition and lived context virtually overlapping, because the manuscripts were discovered inside the structure in which they were housed during at least the period leading up to the eruption of the Vesuvius. Before we hypothesize about the use of the manuscripts and their relation to the building, we should first try to describe their deposition context as closely as we can.
The exact locations of discovery are difficult to recover, because the earliest archaeological records from the 18th century do not present us with a fully reliable account.16 As was addressed above, during the first decades after the rediscovery of Herculaneum explorers were mostly occupied with the search for statues and other treasures to add to their antiquarian collections. The location of discovery of these artefacts is therefore usually unknown, because excavation reports were hardly managed and excavation tunnels would sometimes run right through stratigraphic layers.17 Nevertheless, it is possible to reconstruct where the papyri were discovered with some degree of certainty.
In October 1752 excavations were going on inside the tablinum of the building when workers discovered carbonized wooden shelves against the wall with what seemed to be charcoal objects that were thrown aside in search for valuable items.18 Karl Weber, the excavation leader, identified writing on some of them and discovered that they actually were papyri carrying ancient texts. Apart from the tablinum, scrolls were found in at least three other rooms:19
-
The majority of manuscripts were unearthed in a small room which in Weber’s floor plan is numbered “V.” The room is located close to the living quarters of the Villa and offers a view on the courtyard. The scrolls in this room were found both in cabinets against the wall, and piled up on the floor. David Sider reconstructs a table here on which the papyri could have been read.20 Scholars sometimes refer to this room as the library of the Villa.21
-
A room of about half the size of the ‘library,’ which Weber numbered “XVI.” This room is located next to the dining room and has an entry into the peristylium-garden. The floor was decorated with a mosaic pavement. Here scrolls were found both in piles on the ground and in capsae, a kind of bucket-shaped leather case that was used to carry scrolls.22 It has been suggested that these leather cases were filled with papyri by people who wanted to bring them in safety before the Villa would be destroyed by ash and pumice from the volcano eruption.23
-
Next to the doorway that connects Weber’s room “XVI” to a small peristyle, an uncovered porch. Also, here papyri were discovered in both capsae and in piles on the ground.
The exact number of manuscripts uncovered at the Villa is unclear. Many scrolls were probably disposed in the 18th century, or were otherwise destroyed upon unrolling.24 Marcello Gigante’s overview of all papyri lists 1826, but this number is probably too high because it includes many smaller fragments of scrolls as separate manuscripts.25 The total collection is now generally accepted to number between 800 and 1100 scrolls.26
The excavators in the 18th century unfortunately did not record which manuscript was found at which findspot, but we do know that the bulk of the corpus was found in Weber’s room V, or the ‘library.’ The evidence is unclear concerning the further arrangement of this room, especially because apart from the scrolls other items such as furniture did not stand the test of time. However, later sources do provide some anecdotal evidence. Johann Winckelmann, who had visited Herculaneum several times, describes in a letter from 1764 a shelf-like structure or table in the middle of the room “designed to hold texts on either side so that one could walk around it. The wood of these shelves was burnt to charcoal, and, as one can easily imagine, they collapsed upon being touched.”27 Unfortunately Winckelmann’s report does not offer further details about its dimensions or traces of use. If one cautiously accepts this brief testimony by Winckelmann, one can surmise that the table appears to have functioned as a desk, on which manuscripts could be read, studied, and maybe even copied.28 Thin metal sheets were discovered in and around this room, which originally would have carried an inscription that may have been used as a label for the bookshelves on which the papyri were stored.29
The collection was multilingual: the majority of manuscripts contained texts in Greek, but also a sizable collection of about sixty to eighty Latin papyri has been discovered.30 The issue of multilingualism in manuscript collections will be discussed in some detail below. With regard to the archaeology of the site it is important to note that there is some evidence that Roman libraries tended to store Greek and Roman papyri in different departments.31 It is possible that a different library with Latin works was housed in a part of the Villa that has not yet been excavated.32 However, Greek and Latin papyri were discovered next to each other in room V, the ‘library’, and probably in the tablinum as well, which would suggest that works in both languages were stored side by side in the Villa.33
4 The Collection of Papyri
The association of the Villa papyri with the Epicurean philosopher and poet Philodemus of Gadara and Epicurean philosophy in general goes back to the first decades after their discovery.34 The first type of evidence that pointed historians in this direction, the three sculptures of Epicurus and two of the Epicurean philosopher Hermarchus that were unearthed in the Villa, was shortly followed by the discovery of papyri containing works by Philodemus.35 This categorization is still commonplace in scholarship and probably rightly so. However, some footnotes need to be added here. First, this qualification primarily applies to the Greek texts. The much smaller Latin corpus is less uniform and cannot straightforwardly be associated with Epicurean philosophy, though it is possible that it included at least one copy of Lucretius’ Epicurean poem De rerum natura.36 Correspondingly, one cannot exclude the possibility that non-philosophical literature was stored in parts of the Villa that have not yet been explored. Finally, we should note that according to Houston only about half of the titles from the collection have been identified thus far.37 To summarize the challenge: our current assessment could be based on only part of the collection, and even that part is not completely understood.
Despite these reservations, with the evidence that is available to us there is good reason to characterize at least the Greek papyri as an Epicurean collection. All works are related to Hellenistic philosophy and the majority to Epicureanism with Philodemus as the most eye-catching name.38 On the basis of the papyri’s subscriptiones, the title and/or author of the composition given at the end of the manuscript, Houston estimates that about 58 % of the manuscripts contain compositions by Philodemus.39 Among these, many of his philosophical treatises have been preserved that are dedicated to a particular topic, such as On Music, On Death, On Rhetoric, and On Poems. Eighteen of his texts survived in multiple copies in the Villa, such as his large work History of the Philosophers, which we will return to later.40
Due to the considerable representation of his works among the papyri, Philodemus has become the focal point of many studies on the Villa collection. A recent article by Houston foregrounds the manuscripts by authors other than Philodemus. Unsurprisingly these include a number of Epicurean writers starting with Epicurus himself. Seven books from his 37-volume work On Nature are present among the papyri (some in two or three copies), but interestingly no other of his titles even though Diogenes Laertius attributes about 40 works to Epicurus.41 At least six other Epicurean writers are represented: Carneiscus, Colotes, Demetrius Laco, Metrodorus of Lampsacus, Polystratus, and Zeno of Sidon. Their writings fall under the genre of philosophical treatises, but deal with a wide range of topics, such as ethics, textual criticism, and geometry. Houston observes that if his estimation that 58 % of manuscripts of the collection are indeed by Philodemus is correct, then the collection as it survived today could not possibly have included all works by the major Epicurean writers that are mentioned in Diogenes Laertius’ Lives of the Philosophers.42 The Villa collection therefore either was far from complete, or deliberately only preserves a selection of Epicurean philosophy.
There are also a few works that ostensibly are not related to Epicurean philosophy. The clear outliers are at least three and perhaps up to five manuscripts by the Stoic philosopher Chrysippus represented in the collection.43 One explanation offered is that this composition would serve to find arguments against Stoicism.44 This is certainly possible, but remains speculation. It seems to me that the broad interest in Hellenistic philosophy of the people behind the Villa collection is enough to warrant the presence of Stoic works. It should be noted in this regard that the Villa also housed many statues of non-Epicureans. So if we characterize the Villa papyri as Epicurean, we have to admit that it’s not only an Epicurean collection, but also to some degree a reflection of other philosophical trends of the time.
As noted above, the about sixty to eighty Latin scrolls form a less coherent corpus. They survived mostly in a very fragmentary state. The most complete papyrus is the anonymous epic hexameter poem Carmen de bello Actiaco (P.Herc. 817), which concerns a retelling of the Battle of Actium.45 Three other papyri seemingly contain speeches, while Knut Kleve’s identification of works by Ennius, Caecilius Statius, and Lucretius remains tentative.46
5 Diachronicity and Dynamism
One aspect that has not yet been addressed is the issue of diachronicity. In order to date the Villa-papyri researchers rely on the assessment of scribal hands, which results in a chronology ranging from the third century BCE to the first century CE, within which the individual manuscripts can be placed.47 These palaeographic datings can help us think through the practice of collection building in antiquity. The underlying assumption is that the older manuscripts entered the collection earlier, while the younger papyri were added later perhaps with the intention to enrich the library, or to reflect the changing interest of the collectionneurs. One cannot take this sequence of events for granted, but such a scenario is worthwhile to think through.
Many of the scribal hands of the Villa-papyri have been analysed individually, but the most extensive synthesis of all manuscripts together has been provided by Guglielmo Cavallo in his Libri scritture scribi a Ercolano.48 Cavallo subdivided the scripts of the papyri into 17 different groups and furthermore identified at least 34 scribal hands that would have copied two or more different manuscripts. Cavallo is interested not only in palaeographic typologies, but also in how they can help us understand the evolution of the collection. In this way, he reconstructs a history of the collection that can be divided in three different stages (see table 8).49
Cavallo’s reconstruction starts with a core cluster of thematically related compositions, around which the wider collection was shaped. This core cluster with the oldest manuscripts consists of works associated with Epicureanism.50 We have to be careful not to overstate the evidence, but the inclusion of non-Epicurean works in the later stage of the collection could indicate that Epicureanism lost popularity in the region, or perhaps that the Villa and the collection ended up in different hands that did not share the exclusive interest of the former owners in Epicureanism.51
Table 8
Stages of the Villa-papyri according to Cavallo
|
Stage 1 |
Early manuscripts (from the third century until the early first century BCE) that consist of a core collection of basic Epicurean works; including manuscripts by Epicurus, Demetrius Laco, Carneiscus, Polystratus, and Metrodorus. This core collection would have consisted of up to 200 scrolls.52 |
|
Stage 2 |
The majority of the manuscripts that were made in the first century BCE, in particular the works by Philodemus. Main body of papyri dates from this period. |
|
Stage 3 |
The manuscripts that were dated after Philodemus’s death; consists of Epicurean works, but also the majority of Latin works and the compositions by Chrysippus.53 |
However, this scenario also poses a number of problems. First, there is the aforementioned uncertainty regarding palaeographic datings in general, which can only provide estimations. Next to that, it is important to reiterate that such a palaeographic date is not the same as the date that the scroll became part of the collection. It is possible that these were unrelated events. We cannot exclude the possibility that the collectionneurs behind the Villa-papyri acquired manuscripts that, at the time, already were several decades or perhaps even centuries old.54 Additionally, Houston has rightly noted in this context that it is unknown to us which manuscripts were removed from the collection in the several centuries of its existence, as “texts might have been damaged and discarded, given away as presents, lent and not returned, or sold.”55 One therefore has to be careful when attempting to make definite statements about the nature of the collection in earlier phases. Second, Cavallo’s thesis is centred around the assumption that the collection was shaped by Philodemus personally while working in the Villa. Whether or not the death of Philodemus really inaugurated a new phase in the evolution of the collection remains doubtful. As we will see in the next section, we have no decisive evidence that demonstrates this connection of Philodemus to the Villa.56
Nevertheless, and despite these caveats, this periodization presents us with an impression of dynamism. The character of the papyri collection changes over the course of time, and interests sharpen out. The worldviews and socio-religious beliefs of the people behind the Villa of the Papyri were not stable and unchanging, but developing. To provide just one example: all the six manuscripts carrying compositions by the Greek Stoic philosopher Chrysippus are dated to the latest stage of the collection (from the late first century BCE onwards).57 These manuscripts are among the Villa’s very few non-Epicurean Greek papyri, which is interesting, especially because from an Epicurean perspective Stoicism was regarded to be a competing ideology. It is possible that the people that engaged with the collection and tried to expand it around the turn of the millennium, sought to broaden their horizon by including works from other philosophic traditions. Alternatively, one can imagine that they wanted to intensify their polemics with Stoicism through the study of their intellectual opposition. Whatever the reason, on the basis of our evidence we can conclude that something changed: over time the collection displays a growing heterogeneity, allowing manuscripts to be included that in earlier stages did not have a place. In this way, scholars become increasingly aware of the broader dynamism of ancient manuscript collections. Ancient manuscript collections are selective and the result of ongoing adaptation, for example, to the changing interests of the owner, or perhaps to what was available on the market. The pre-Villa-collection in the second century BCE is something different from the Villa-collection in the first century CE. One may perceive ‘collections’ as stable entities from our present-day perspective, but we have to be open to the possibility that in reality we are dealing with continuously changing constellations. Herculaneum is a wonderful case that reminds us of these dynamics and presents us with the opportunity to ask particular questions concerning, for example, the formation of manuscript collections or the scribal setting, use, and potential circulation of manuscripts. We take these questions with us when we consider other corpora and are, in this way, challenged to examine our material from a new perspective. However, before we take this exploration to a next level, we return to the unfinished task of gauging the lived context of the Villa-papyri.
6 Herculaneum as a Private, Scholarly Collection
In the preceding sections we have looked at both the deposition context and the corpus of Villa-papyri itself. The question I would like to ask here is how we should understand the lived context in which these texts were meaningful to people? What do the characteristics of the Herculaneum scrolls and their archaeological context, as described above, tell us about the people behind the collection that used the Villa and read, studied, and collected the manuscripts?
There are a few points that need recapitulation. We have concluded that the papyri were discovered around the living quarters of a typical Roman residential villa; this means that they were found in what we can safely characterize as a private setting. As Houston has argued as well, ‘public’ and ‘private’ are difficult concepts to apply to the ancient world, because they are joined by a set of modern connotations that cannot easily be translated to the distant past (such as the premise that public institutions are funded through public taxes, which most probably was not the case).58 When considering the Villa of the Papyri, however, there is no reason to suggest that we are dealing with a building that served a public function. The archaeology of the site can safely be identified as a classical atrium house that fits very well with other regional country houses occupied by Roman elites. Its residential character is evidenced through the two peristyle gardens, the elaborate mosaic pavements, and the sizable collection of bronze and marble statues. Furthermore, excavations in the 21st century have demonstrated that the Villa complex is larger than previously assumed: on the lower terrace outside the main structure a smaller seaside pavilion was discovered that included a pool and a monumental hall.59 Though this villa is unusually big, it probably was not beyond the financial means of one of Rome’s ruling families. The chamber that housed the majority of papyri is not a storage room or an archive, but a room where the manuscripts could be read, studied, and perhaps even composed or copied.60 The fact that Greek and Roman literatures were kept together (which is unusual for Roman public libraries) confirms this. When we zoom in on the papyri themselves we are struck by the fact that the corpus is not a random intersection of classical literature of the time or the type of manuscript corpus that we associate with a school. Rather we are dealing with a specialized, philosophical collection that displays a general interest in Epicureanism. The Bay of Naples is known to have been the location of many Epicurean centres, so the presence of such a collection would certainly fit a broader regional pattern.61
In conclusion, these aspects of the archaeological evidence of the Villa lead us to suggest for the papyri that we are dealing with a private, scholarly collection dedicated to Epicurean philosophy. We may characterize the manuscript corpus as a ‘living’ collection: the many scribal practices visible in the manuscript evidence testify to a continuous engagement with texts on a high level. We find examples of opisthographs, papyri with scribal interventions such as notes and marginal annotations, and manuscripts consisting of quotations or extracts.62 A particular strong indication that we are dealing with a scholarly collection is the fact that some of Philodemus’s texts that survived among the Villa-papyri are based on his notes on lectures by his teacher Zeno of Sidon: his On Frank Criticism, which survived on P.Herc. 1471, contains the phrase
The manuscripts were meaningful to the people who owned them. It is in this regard very relevant that a number of papyri were discovered in capsae, portable leather boxes, in what seems to be a last-minute attempt to save them from destruction. We do not know which and how many manuscripts were stored there, but the very act of deposition in those boxes demonstrates how meaningful the texts were for the people behind the collection. If the capsae were indeed intended to rescue the papyri from the volcanic eruption, then it evokes a possible resemblance to Qumran. It is widely assumed that the Dead Sea Scrolls were deposited in the caves around the settlement of Qumran in an attempt to safeguard them from Roman soldiers.64 In this way, both sites demonstrate the significance of manuscripts: texts belong to the essential set of belongings that people would try to bring into safety even in life-threatening circumstances.
The palaeographical datings of the Herculaneum papyri range to a vast period from roughly the third century BCE to the early first century CE. The most likely explanation is that the people behind the collection had a continued interest in Epicurean philosophy that extended beyond the lifetime of one single collectionneur. However, we cannot exclude the possibility that all manuscripts were acquired at one moment in time (though that would mean that papyri that at that time were already 300 years old would have been exchanged).
The question to whom both the Villa and the manuscript collection belonged is difficult to answer, but Lucius Calpurnius Piso Caesoninus (ca. 100–ca. 43 BCE) is generally regarded to be the most likely candidate. Piso Caesoninus was known to be a follower of Epicureanism, which would explain his interest in the manuscripts. Furthermore, being both a Roman senator and the father-in-law of Julius Caesar, he was in a position that would have presented him with the financial resources to own and maintain the Villa.65 As we have seen, scholarly estimations propose that over half of the manuscripts from the collection would have contained compositions by Philodemus. This, together with the fact that Piso Caesoninus is known to have been his patron, makes it appealing to locate Philodemus in Herculaneum as well and to consider the manuscript collection as his working library. It should be noted that there is only circumstantial evidence that points in this direction. We do not have any data that can concretely connect Philodemus to the Villa. However, we do have other evidence that links Philodemus more closely to the manuscript collection, as there are indications that the working draft of at least one of the texts attributed to Philodemus was discovered in Herculaneum. Aside from this connection to Philodemus, this particular manuscript can contribute to our questions regarding the different collecting and scribal practices behind the Villa-papyri from another perspective as well. As we are concerned here with different forms of textual engagement, a working draft can provide a window onto compositional strategies that are concealed in a final work. Could such a manuscript help us to get closer to the reading events that were concentrated around these texts, and subsequently to the different ways in which a manuscript could be experienced? With this question in mind, I will discuss this Herculaneum papyrus strongly associated with Philodemus, the opisthograph P.Herc. 1021.
7 Opisthograph P.Herc. 1021 as an ‘Opisthographic Collection’
The Greek manuscript Papyrus Herculaneum 1021 (P.Herc. 1021) is one of the two exemplars of the History of the Academy by Philodemus, whose work is, as we have seen, very much associated with the Villa collection.66
This composition deals with the history of Plato’s academy from its foundation until the Platonic philosopher Antiochos of Ascalon (125–68 BCE). P.Herc. 1021 is the most complete manuscript and can be characterized by its informal and hastily written script.67 It is generally taken to be either Philodemus’s own autograph, or otherwise dictated by him. We find evidence for active scribal involvement such as additions, corrections, and transposition signs made by what seem to be two different hands.68 A number of marginal annotations and additions were placed both above and below the columns, which were to be processed into a final version. This final edition is most probably preserved in P.Herc. 164, a very fragmentary and palaeographically younger manuscript of the same composition that takes the revisions of P.Herc. 1021 into account.
P.Herc. 1021 is an opisthograph. The text on the verso consists of several columns that were probably intended to supplement or replace the text on the recto.69 In this way, the opisthograph can provide an insight into the compositional strategies of ancient scribes. We observe for example that P.Herc. 1021 does not preserve a full narrative, but rather presents extracts from many different authors that wrote on Plato and the Academy, such as Hermippus of Smyrna.70 Many of these extracts follow the literal wording of their base texts, but sometimes they are reworked. The practice of note-taking and collecting of extracts for personal use is well established in classical literature, such as in the works of for example Plutarch and Clement of Alexandria, who reworked notes and excerpts when they composed their treatises.71 Luc Van der Stockt argued that Plutarch for his De tranquillitate animi had his hypómnema (notes, annotations, and comments) within reach and consulted them when necessary, a working method he probably also used for other treatises.72 We have earlier looked at Pliny the Younger’s letter to Baebius Macer and how he would have used opisthographs to copy or mark down relevant passages, resulting in a scroll that consists of a collection of excerpts from different compositions. We have seen in the previous part that there is also evidence for these practices in a number of Dead Sea Scrolls.
P.Herc. 1021 seems to correspond very well to these types of textual practices. Tiziano Dorandi sees “a sample of how Pliny commentarii could look like in P.Herc. 1021, a roll treasuring a collection of excerpta taken from different sources, brought together by Philodemus in prevision of his work on Academics philosophers included in Collection of the Philosophers.”73 Dorandi hypothesizes how Philodemus would have read a range of sources during which he (or an assistant) would have noted down the passages that caught his interest on a separate manuscript.74 The recto of P.Herc. 1021 would have been the first result of such an investigation, additional excerpts were later added on the verso, including many parallel passages with the front side. P.Herc. 1021 demonstrates how an opisthograph could function as (1) a collection of texts from many different sources, and (2) as a rough draft of a new composition. Additionally, this opisthograph ties Philodemus more closely to the Villa-papyri, especially when we consider P.Herc. 1021 in relation to P.Herc. 164. This allows us to establish that the papyri do not just constitute a collection of someone with a particular interest in Epicureanism and Philodemus in particular, but that they are directly associated with Philodemus’ working library. The most likely explanation seems that it concerns Philodemus’ own autograph or a text dictated by him directly. But if this was not the case than at least the papyrus must have originated from the same circle of scribes in which he was working.
However, the main point here is that P.Herc. 1021 presents us with a striking parallel for the collection of extracts from different texts on one opisthograph with the intention to compose a new text. For this type of scrolls, manuscripts that collect parts of texts from different sources on both recto and verso, I would like to introduce the term ‘opisthographic collection.’ I think this term could be useful, because it acknowledges both the codicology of the manuscript and the nature of the (extracts of the) compositions it carries. It is in this context relevant to think back of book list 16 from Rosa Otranto’s Antiche Liste di Libri su Papiro, which we discussed above. This list raised our attention, because it mentions three different papyri that would have carried multiple works of Plato’s dialogues on one manuscript, which indicates that we might be dealing with an example of a similar ‘opisthographic collection.’ Whether or not these dialogues were partly or fully copied is unknown, though it seems unlikely that one papyrus would have had the length to contain three full compositions by Plato.
Returning to P.Herc. 1021, we notice that the diversity of sources collected is also meaningful. The Villa of the Papyri in general, and Philodemus’ works in particular, are associated with Epicureanism. But P.Herc. 1021 presents a text that offers a perspective on a different intellectual tradition—that of Plato’s Academy, and as such draws on a variety of philosophical sources in order to create a new text. We will return to this in the final chapter of this part.
Up to this point we have considered Graeco-Roman ‘collections’ from a variety of perspectives and aimed to be mindful of different types of evidence: literary sources, papyri, and in the context of the Villa of the Papyri, the archaeological record. Furthermore, we have moved from the formation of the collection on a macrolevel to the analysis of an individual opisthograph on a microlevel. In the final part I will explore in which ways a comparison with the Villa of the Papyri could (or could not) be meaningful for how we think about the Dead Sea Scrolls as a collection or in terms of collecting practices.
The rediscovery of Pompeii, Herculaneum, and other sites in the region is described in detail in Christopher C. Parslow, Rediscovering Antiquity: Karl Weber and the Excavation of Herculaneum, Pompeii, and Stabiae (Cambridge: Cambridge University Press, 1995); overviews of more recent archaeological research are provided in Mantha Zarmakoupi, ed., The Villa of the Papyri at Herculaneum: Archaeology, Reception, and Digital Reconstruction, Sozomena 1 (Berlin: de Gruyter, 2010).
David Sider, The Library of the Villa dei Papiri at Herculaneum (Los Angeles: Getty Publications, 2005), 4, 18.
Houston, Inside Roman Libraries, 88; Antonio De Simone argues on the basis of a comparison with other elite residences in the area that the Villa was built around the middle of the first century BCE: “Rediscovering the Villa of the Papyri,” in The Villa of the Papyri at Herculaneum: Archaeology, Reception, and Digital Reconstruction, 1–20, at 11. A study of the pavements and wall paintings by Maria P. Guidobaldi and Domenico Esposito dates the first phase of the Villa to the third quarter of the first century BCE: “New Archaeological Research at the Villa of the Papyri in Herculaneum,” in The Villa of the Papyri at Herculaneum: Archaeology, Reception, and Digital Reconstruction, 21–61, at 40–41.
Sider, The Library of the Villa dei Papiri, 4.
De Simone, “Rediscovering the Villa of the Papyri,” 18.
Pliny the Younger, Epistulae, 6.16 and 6.20: Pliny the Younger, Letters, Volume I: Books 1–7, Translated by Betty Radice, LCL 55 (Cambridge, MA: Harvard University Press, 1969).
Popović, “Qumran as Scroll Storehouse in Times of Crisis?” 553; idem, “The Manuscript Collections: An Overview,” 44.
Popović, “Qumran as Scroll Storehouse in Times of Crisis?” 553.
See for example Colin Renfrew and Paul G. Bahn, Archaeology: Theories, Methods and Practice, 6th ed. (London: Thames & Hudson, 2012), 49–72.
Jürgen Zangenberg, “The Functions of the Caves and the Settlement of Qumran: Reflections on a New Chapter of Qumran Research,” in The Caves of Qumran, 195–209.
Many of the objects other than manuscripts that were discovered at Oxyrhynchus are listed in Donald M. Bailey, “Objects from Oxyrhynchus in the British Museum and the Victoria and Albert Museum,” in Oxyrhynchus: A City and Its Texts, ed. Alan K. Bowman et al. (London: Egypt Exploration Society, 2007), 369–381.
Popović, “Qumran as Scroll Storehouse in Times of Crisis?” 553.
Ehud Netzer, Masada III: The Yigael Yadin Excavations 1963–1965 Final Reports: the Buildings, Stratigraphy and Architecture (Jerusalem: Israel Exploration Society, 1991), 554–556, 564–565.
Popović, “Qumran as Scroll Storehouse in Times of Crisis?” 568.
Studies of ancient literacy and manuscript circulation tend to be concentrated on either institutions or on priestly and scribal elites, while families or individuals are often not considered. However, Michael Wise has recently argued that Roman Judaea might have had a reasonably large group of relatively wealthy householders that were literate: Michael O. Wise, Language and Literacy in Roman Judaea: A Study of the Bar Kokhba Documents (New Haven: Yale University Press, 2015), 279–355.
Some accounts and logbooks were published in Michele Ruggiero, Storia degli scavi di Ercolano ricomposta su’ documenti superstiti (Naples, 1885).
Sider, The Library of the Villa dei Papiri, 17.
Sandra Sider, “Herculaneum’s Library in 79 A.D.: The Villa of the Papyri,” Libraries and Culture 25 (1990): 534–542, at 536.
Sider, “Herculaneum’s Library in 79 A.D.,” 537; Houston, Inside Roman Libraries, 88; David Sider provides a map on which the findspots are indicated: The Library of the Villa dei Papiri, 62.
Sider, The Library of the Villa dei Papiri, 62, figure 64.
Carlo Gallavotti, “La libreria di una villa romana ercolanese,” Bollettino istituto di patologia del libro 3 (1941): 129–145. This room was immediately after its discovery in 1754 identified as a library. Camillo Paderni mentions it in a letter to his English friend Thomas Hollis that is dated 27 April 1754: CDP, 242.
Menahem Haran, “Book-Scrolls at the Beginning of the Second Temple Period,” 116 n. 9.
Francesca Longo Auricchio and Mario Capasso, “I rotoli della Villa ercolanese: Dislocazione e ritrovamento,” CErc 17 (1987): 37–47.
Sider, The Library of the Villa dei Papiri, 64.
Marcello Gigante, Catalogo dei papiri ercolanesi (Naples: Bibliopolis, 1979); Houston, Inside Roman Libraries, 99.
Richard Janko, Philodemus: “On Poems,” Book 1. Edited with Introduction, Translation and Commentary (Oxford: Oxford University Press, 2000), 4; idem, “The Herculaneum Library: Some Recent Developments,” Estudios Clásicos 121 (2002): 25–41; Sider, The Library of the Villa dei Papiri, 64.
Carol C. Mattusch, Johann Joachim Winckelmann: Letter of Report on the Discoveries at Herculaneum (Los Angeles: The J. Paul Getty Museum, 2011), 117. To my knowledge, Winckelmann’s letter offers the only description of this table.
This would, however, be a unique circumstance: there is no archaeological evidence for the presence of a table or desk in Graeco-Roman libraries. See the discussion by Jocelyn P. Small, Wax Tablets of the Mind: Cognitive Studies of Memory and Literacy in Classical Antiquity (London: Routledge, 1997), 133–137, 143. There is, however, iconographic evidence: the relief of an orator from the 4th century CE preserved in the Museo Ostiense (Inv. 130). This relief depicts two scribes sitting at tables and writing on what seems to be tablets, while a figure on a stage is orating in the middle: Lucio Del Corso, “L’Insegnamento alla luce delle testimonianze iconografiche,” in L’Enseignement supérieur dans les mondes antiques et médiévaux, ed. Henri Huggonard-Roche, Textes et traditions 16 (Paris: Vrin, 2008), 307–331, at 327.
Sider, “Herculaneum’s Library in 79 A.D.,” 538.
Houston, Inside Roman Libraries, 97.
Greek and Latin texts were stored in separate locations within the library of the Baths of Caracalla. Furthermore, Suetonius tells us that Augustus added a Latin and a Greek library to the Temple of Apollo: Suetonius, Augustus, 29.3. Cf. T. Keith Dix, “ ‘Public Libraries’ at Rome: Ideology and Reality,” Libraries and Culture 29 (1994): 282–296; Casson, Libraries in the Ancient World, 81; Daniel Stökl Ben Ezra, “Old Caves and Young Caves: A Statistical Reevaluation of a Qumran Consensus,” DSD 14 (2007): 313–333, at 323. Arguments against this suggestion are provided in Popović, “Qumran as Scroll Storehouse in Times of Crisis?” 571.
Sider, The Library of the Villa dei Papiri, 43.
Houston, Inside Roman Libraries, 88 n. 4.
David Sider designates Camillo Paderni, curator of the Museum Herculanense in Portici, to be the first to report about the philosophical nature of the collection: Sider, The Library of the Villa dei Papiri, 20–23, 60.
Overview of the sculptures: Carol C. Mattusch, The Villa dei Papiri at Herculaneum: Life and Afterlife of a Sculpture Collection (Los Angeles: The J. Paul Getty Museum, 2005), 289–296.
According to Winckelmann’s overview of discovery from 1762 the first four rolls that were fully unrolled all belonged to Philodemus: Johann J. Winckelmann, Critical account of the situation and destruction by the first eruptions of Mount Vesuvius of Herculaneum, Pompeii, and Stabia (London, 1771). Translation of Lettre de M. L’Abbé Winckelmann, à Monsieur le Comte de Brühl, Sur les Découvertes d’Herculanum (Paris, 1764). By the early 19th century the Epicurean nature of the collection was evident: David Sider, “The Books of the Villa of the Papyri,” in The Villa of the Papyri at Herculaneum: Archaeology, Reception, and Digital Reconstruction, 115–127, at 122.
Identified by Knut Kleve in “Lucretius in Herculaneum,” CErc 19 (1989): 5–27. This identification remains tentative however: Houston, Inside Roman Libraries, 98 n. 57.
Houston, Inside Roman Libraries, 99–100.
Sider, The Library of the Villa dei Papiri, 73.
This is based on the fact that Philodemus is mentioned as the author for 44 of the surviving 75 subscriptiones: George W. Houston, “The non-Philodemus Book Collection in the Villa of the Papyri,” in Ancient Libraries, 183–208, at 184–185.
More detailed overviews of Philodemus’ works found in the Villa are provided by Gigante, David Sider and Houston: Marcello Gigante, Philodemus in Italy: the Books from Herculaneum, trans. Dirk Obbink (Ann Arbor: University of Michigan Press, 1995); Sider, The Library of the Villa dei Papiri, 78–95; Houston, Inside Roman Libraries, 92–96.
Houston, “The non-Philodemus Book Collection in the Villa of the Papyri,” 186.
Houston, Inside Roman Libraries, 100.
Houston, Inside Roman Libraries, 90. Houston also attributes the anonymous P.Herc. 1158 to a Stoic author: ibid. 97.
Philodemus does attack Stoics on several occasions and often polemically refers to Chrysippus and other Stoic philosophers, but also to Plato and Aristotle. Janko, “The Herculaneum Library: Some Recent Developments,” 32; Sider, The Library of the Villa dei Papiri, 79–81, 87. For mentions of works by other authors in Philodemus’ work: Daniel Delattre, “Les mentions de titres d’œuvres dans les livres de Philodème,” CErc 26 (1996): 143–168.
Edition: Giovanni Garuti, C. Rabirius, Bellum Actiacum e papyro Herculanensi 817 (Bologna: N. Zanichelli, 1958).
Houston, Inside Roman Libraries, 98.
It should be noted that palaeographic datings by their very nature offer approximations. In the most favourable scenario, they will still give us a timeframe of at least several decades. One should therefore be careful not to attach too much importance to the dating of individual papyri. However, the palaeographic analysis of the scribal hands of a large manuscript corpus over a significant period of time can give us a good indication of tendencies in datings and stylistic developments; and can allow us to identify different palaeographic periods.
Cavallo, Libri scritture scribi a Ercolano. Cavallo draws on Eric Turner’s study of Greek papyri from Egypt as a comparative model: Turner, Greek Manuscripts of the Ancient World; idem, “Ptolemaic Bookhands and Lille Stesichorus,” Scrittura e Civiltà 4 (1980): 19–40.
Cavallo’s analysis of the Villa-papyri has contributed significantly to our understanding of Greek palaeography, but his typology should nevertheless be treated with some caution. His classification is still the only full treatment of the palaeography of the Herculaneum papyri and has never been critically assessed in its entirety.
Cavallo, Libri scritture scribi a Ercolano, 28–46, 58–66.
More manuscripts with works by other authors than Philodemus were dated to the second century BCE than to any other century; Houston, “The non-Philodemus Book Collection in the Villa of the Papyri,” 188.
Declining interest in Epicureanism in the region: Houston, Inside Roman Libraries, 123; Villa changed hands: Mario Capasso, Manuale di papirologia ercolanese (Galatina: Congedo, 1991), 198.
This reconstruction suggests that part of the Villa-collection precedes the Villa of the Papyri and perhaps also Philodemus. Cavallo in fact hypothesizes that the older manuscripts were produced in Greece or the eastern Mediterranean and later ended up in the hands of Philodemus (perhaps as part of an inheritance); Guglielmo Cavallo, “I rotoli di Ercolano come prodotti scritti: Quattro riflessioni,” Scrittura e Civiltà 8 (1984): 5–30, at 5–12; Gigante, Philodemus in Italy, 18.
The Latin works that certainly date to this stage include P.Herc. 817 and P.Herc. 1067. P.Herc. 817, mentioned above, concerns a poem on the Battle of Actium which took place in 31 BCE. P.Herc. 1067 is a historical work that contains references to one of the emperors and was recently associated with Seneca the Elder: Valeria Piano, “Il PHerc. 1067 latino: Il rotolo, il testo, l’autore,” CErc 47 (2017): 163–250.
Paolo Radiciotti has dated a number of the other Latin manuscripts palaeographically to the first century BCE, which means that they could also belong to stage 2: Paolo Radiciotti, “Osservazioni paleografiche sui papiri latini di Ercolano,” Scrittura e Civiltà 22 (1998): 353–370, at 365–370. As far as I know no Latin papyri have been dated to stage 1.
The palaeographic dates of the papyri that are grouped in stage 1 by Cavallo predate the Villa by at least several decades. The exact moment that these manuscripts were brought together into one collection is impossible to reconstruct. Gigante argues that “the most ancient core of the library was formed outside Campania. It was probably brought from Athens to Herculaneum by Philodemus himself, where he had formed it or inherited it from his master or masters.” Gigante, Philodemus in Italy, 18.
Houston, Inside Roman Libraries, 123.
Houston, Inside Roman Libraries, 124–125. Houston also discusses a number of alternative reconstructions.
Houston, “The non-Philodemus Book Collection in the Villa of the Papyri,” 198.
Houston, Inside Roman Libraries, 3–4.
Guidobaldi and Esposito, “New Archaeological Research at the Villa of the Papyri in Herculaneum.” The remainders of at least four pieces of furniture were discovered here as well.
As discussed above, according to Winckelmann’s testimony there would have been a working desk in this room.
John H. D’Arms, Romans on the Bay of Naples and Other Essays on Roman Campania (Bari: Edipuglia, 2003), 64–67.
Overview of the opisthographs found among the Villa-papyri: Mario Capasso, “I papiri ercolanesi opistografi,” in Atti del V convegno nazionale di egittologia e papirologia, Firenze, 10–12 dicembre 1999, ed. Simona Russo (Florence: Istituto Papirologico G. Vitalli, 2000), 5–25. On marginal markings: Houston, “The non-Philodemus Book Collection in the Villa of the Papyri,” 191. P.Herc. 118a is an example of a manuscript consisting of extracts: Cesira Militello, Filodemo. Memorie epicuree (PHerc. 1418 e 310): Edizione, traduzione e commento (Naples: Bibliopolis, 1997), 82–83.
L. Michael White, “A Measure of Parrhesia: The State of the Manuscript of Pherc. 1471,” in Philodemus and the New Testament World, ed. John T. Fitzgerald, Dirk Obbink, and Glenn S. Holland, NovTSup 111 (Leiden: Brill, 2004), 103–130. This phrase also occurs on P.Herc. 1003 and P.Herc. 1389.
An elaborate reconstruction of this scenario can be found in Stegemann, Die Essener, Qumran, Johannes der Täufer und Jesus, 93–101. Many variations on this view have been proposed. For example, among others Dimant and Daniel Stökl Ben Ezra have argued that Cave 4 served as a library to the people living at the settlement (a suggestion that goes back to Cross); Cross, The Ancient Library of Qumrân, 26; Dimant, “Qumran Manuscripts: Contents and Significance,” 36; Stökl Ben Ezra, “Old Caves and Young Caves.” Cave 7 was characterized by Crawford as a “residential cave” that would have belonged to a single inhabitant with a particular interest in Greek manuscripts: Crawford, “Qumran: Caves, Scrolls and Buildings,” 259, 265. However, Popović has pushed back against this assumption: “Qumran as Scroll Storehouse in Times of Crisis?” 571–572. More recently Crawford proposed an alternative explanation and suggested that the scrolls from Cave 7 “came from the Greek section of the originating library”; Crawford, Scribes and Scrolls at Qumran, 147.
Sider extensively discusses the main candidates and concludes that Piso Caesoninus remains the most likely owner of the Villa: Sider, The Library of the Villa dei Papiri, 5–8.
Editions: Siegfried Mekler, Academicorum philosophorum index herculanensis (Berlin: Weidmann, 1902), Tiziano Dorandi, Storia dei filosofi: Platone e l’Academia: (PHerc. 1021 e 164) (Naples: Bibliopolis, 1991), and Konrad Gaiser, Philodems Academica, Supplementum Platonicum I (Stuttgart: Frommann-Holzboog, 1988). Gaiser’s edition is confined to the parts that concern the Old Academy, which is 26 of the 49 columns of P.Herc. 1021. I am aware of the problems concerning the title. Though it is commonly referred to as Index Academicorum, Gaiser prefers the title Academica.
Cavallo, Libri scritture scribi a Ercolano, 33; Dorandi, Storia dei filosofi: Platone e l’Academia, 104.
Cavallo, Libri scritture scribi a Ercolano, 61.
Unfortunately, it is not possible to read the verso directly from the manuscript, because the backside was glued unto cardboard (‘cartoncino’), which was a common preservation method in the 18th century. Before this was done however drawings were made of the 12 columns on the verso, which are now stored in the Bodleian Library in Oxford: Aurélie Tournié et al., “Ancient Greek Text Concealed on the Back of Unrolled Papyrus Revealed through Shortwave-infrared Hyperspectral Imaging,” Science Advances 5/10 (2019),
Hermippus is being referred to in column 10:40–12:41. Other authors include Dicaearchus of Messana, Antigonus of Carystus, and Apollodorus of Athens. See Gaiser for an overview of Philodemus’ sources: Philodems Academica, 97–133. Kilian Fleischer, “Melanthios von Rhodos in Apollodors Chronik (P.Herc. 1021, col. XXXI),” Phil 162 (2018): 15–24.
Annewies van den Hoek, “Techniques of Quotation in Clement of Alexandria: A View of Ancient Literary Working Methods,” VC 50 (1996): 223–243; Luc van der Stockt, “A Plutarchan Hypomnema on Self-Love,” AJP 120 (1999): 575–599.
Van der Stockt, “A Plutarchan Hypomnema on Self-Love,” 596.
Tiziano Dorandi, “Notebooks and Collections of Excerpts: Moments of ars excerpendi in the Greco-Roman World,” in Forgetting Machines: Knowledge Management Evolution in Early Modern Europe, ed. Alberto Cevolini, LWW 53 (Leiden: Brill, 2016), 35–57, at 42.
Based on this manuscript Dorandi proposes a schematic reconstruction of Philodemus’ working method: Dorandi, “Notebooks and Collections of Excerpts,” 49.