Rice University logo
 
 

Wastepaper trade in ancient Egypt, some programming to do

October 20th, 2013 by gwa1

In my last post, I introduced a project that I have in mind to study the reuse of papyrus manuscripts in ancient Egypt during the Ptolemaic and Roman periods, and I linked to my initial map showing the locations of some reused papyri from some of the covers of the Nag Hammadi Codices. In this post, I want to say more about the overall project and about Tristmegistos.org, which among other things provides metadata for 209,616 ancient manuscripts.

So my big question is whether there was a wastepaper trade or not. I expect to be able to answer the question by finding or not finding patterns in the reuse of papyri. If most papyri were used and then reused in the same village or city or region, then it could be that people just reused what they had at hand or could pick up from the local dump. But if many or at least some papyri during the Ptolemaic and Romans periods were used in one place and then reused in another place more than a day’s walking/riding distance away, it would be hard to explain the movement without a wastepaper trade.

Besides distance, I also want to keep track of the amount of time between use and reuse, plus the kinds of papyri being reused, what were they reused for, and who was reusing them. It is not always possible to get at this information. But there is a lot of it on Trisgmesitos.org.

As I mentioned, Trismegistos provides metadata for over 200K ancient manuscripts. Here is the full list. If you open a record, you’ll see that there are three fields dealing with reuse: reuse type, reuse detail, reuse note. Not all ancient manuscripts were reused. I want to isolate those that were, and in particular the reused papyri. Among the search options on the site, it looks like the only way to do that is a Google Custom Search. (On the advanced search form, it is also possible to search the general ‘note’ field, but not the ‘reuse note’ field, and that does get at a bit of the information I want.) A Google Custom Search for ‘reuse’ turns up 26,300 results.  That gives me an idea of the size of the data set, but the results are not very useful as far as I can tell. To click on each result and put the information into a CSV file would not be practical, to say the least.

The other week, Caleb McDaneil demonstrated how to use Python to go into a list of metadata records and pull out certain information. He demonstrated this on the text records at Internet Archive. But the process would be more or less the same for what I want to do with Trismegistos. Since I have not done anything like that before, it might take me a number of hours or even several days to complete the courses he recommended on The Programming Historian, and to write the program to do what I want to. Still it would be much more efficient than going through 26K Google results.

Once I have a CSV file, I could map the reused manuscripts on ArcGIS and (start to) answer my question.

Wastepaper trade in ancient Egypt, initial map

October 13th, 2013 by gwa1

One of the research projects that I have in mind is a study of the reuse of papyrus manuscripts in ancient Egypt during the Ptolemaic and Roman periods. There has been some debate in scholarship as to whether there was such a wastepaper trade. I would like to find out. I would  like to find out especially because I am interested in where the thirteen Nag Hammadi Codices (NHC) were produced: in a Christian monastery or not.

The Nag Hammadi Codices were produced somewhere in Egypt during the late 300s or early 400s and then buried together for some reason in a cemetery (you can see images of the site here, searching the subject index under the heading of ‘caves’ for instance). The codices contain copies of non-canonical Christian texts like the Gospel Thomas as well as some pagan texts like the Hermetic Discourse on the Eighth and Ninth. Since the time of their discovery near the city of Nag Hammadi in the 1940s, scholars have debated who produced and read the codices, where, and why.

Perhaps the most common scholarly view is that the codices were produced in a Christian monastery or two or three. Several scholars go on to speculate that Christian monks were reading  the non-canonical and pagan texts in these codices for devotional purposes alongside canonical Christian texts, until a church authority banned non-canonical texts, and so the monks took the codices out of the monastery and hid them where they would not be destroyed (but never went back for them?).

Another view is that the codices were not produced or read in a Christian monastery. They were produced and read within what might be called an esoteric community or network of communities whose members may or may not have been Christians of one kind or another. These esotericists collected and read Hermetic texts alongside non-canonical Christian texts, and they may have been Hermetists themselves, one of whom was buried with the esoteric library of the Nag Hammadi Codices. Admittedly the sociological identity of such a community or network of communities is debatable, but it is striking that there are no canonical texts in the codices; if Christian monks were reading both non-canonical and canonical texts for devotional purposes, it is difficult to explain why they did not put a single canonical text in any of these thirteen codices.

The only canonical text in the Nag Hammadi Codices is a fragment of Genesis that was not copied there and read for devotional purposes; it was used as wastepaper.  When the codices were produced, their leather covers were reinforced with reused papyri. A total of 172 of these reused papyri survives, including the fragment of Genesis. Some of the reused papyri are indeed monastic documents. However, many others are not monastic at all:  accounts, lists, deeds, loans, etc.

In order to begin to evaluate the significance of these reused papyri that are monastic and those that are not, it is necessary to understand how papyri were reused generally in ancient Egypt during the Ptolemaic and Roman periods, which is something that is not well understood in scholarship. What kinds of papyri were reused? What were they reused for? Who reused them? Where did people in ancient Egypt get used papyri? Were papyri often used and then reused in the same village or city or region? Were they sometimes used and then reused many miles away? Was there a wastepaper trade?

There are a number of digital humanities resources and tools available to get at answers to these questions. I’ll be exploring and describing some of them in a few posts. Here I want to link to a map that I created in class this week using ArcGIS. It is just an initial map. The only contents so far are the  seven reused papyri from the Nag Hammadi Codices that can be located geographically: nos 1, 22, 23, 28, 31, 64, 153. Nos 22, 23, 28, and 31 can be located to the same place, so there are not seven points on the map but only four. Along with these four orange points, you’ll see a yellow arrow marking the approximate find spot of the Nag Hammadi Codices.

For each of the four points, in the popup box you’ll see a link to the entry in the Trismegistos database of places related to the ancient world. Trismegistos is where I got the coordinates for each location. I’ll say more about it in another post. The coordinates on Trimegistos are degrees minutes. In the CSV file that I used to make my map on ArcGIS I needed to convert them to decimal degrees. I used this calculator to do that (thanks for the help Wright!).

Before I interpret the map, I should mention that the locations of these seven reused papyri may not be exactly where they came from. No 1 involves a contract that workers from a certain village (name missing) have with the city of Diospolis Parva, which is in the same district as the village. Nos  22, 23, 28, and 31 probably come from the city of Antinoopolis, according to the scholar who was in charge of editing them.  No 64 is a loan between a borrower from the little-known village of Techthy (no coordinates on Trismegistos) and a lender from the city of Tentyris (he used to be magistrate there anyway). No 153 is a letter from the city of Chenoboskia, modern-day Nag Hammadi, the largest city near where the thirteen codices were found. The letter was written from Chenoboskia. It is unknown where it was written and delivered to before being reused to strengthen the covers of the Nag Hammadi Codices.

The measuring tool on ArcGIS gives  me the following distances between locations:

~3 miles between Nag Hammadi (reused papyrus no 153 from NHC XI) and Diospolis Parva (reused papyrus no 1 from NHC I)

~25 miles between Nag Hammadi and Tentyris (reused papyrus no 64 from NHC VII)

~148 miles between Nag Hammadi and Antinoopolis (reused papyri nos 22, 23, 28, 31 from NHC V)

The most interesting datum is the last one. It could be that nos 22, 23, 28, and 31 were brought some 150 miles from Antinoopolis to the area of Nag Hammadi where they were used to reinforce the cover of NHC V as it was produced and then buried with the rest of the thirteen codices.  But a better interpretation might be that NHC V was produced in the area of Antinoopolis itself, while perhaps other codices were produced elsewhere, such as in the area of Nag Hammadi. Scholars have grouped the codices together based in part on the handwriting of the non-canonical and pagan texts that were copied onto their pages. The same scribes copied the texts of NHC I, VII, and XI (one group); and the covers of these codices were all reinforced with papyri from locations within 25 miles of the eventual find spot. But the texts of NHC V (from another group) were copied by a different scribe, who may have been working in the area of Antinoopolis, 150 miles away.

If it is assumed that NHC I, VII, and XI were produced in about the same place as well as by the same people near Nag Hammadi, it would still need to be explained how used papyri from as far away as 25 miles were brought together to strengthen the covers of these codices. Was there a local wastepaper trade that sold used papyri, both monastic and non-monastic? Alternatively, did monks go out and get wastepaper from the closest trash heaps whenever they needed to supplement what they had at hand in the monastery?  An explanation is also needed, if NHC V was produced 150 miles away by someone else. Did monks in different areas produce the codices and then gather them together?

An additional layer to have on my map would be the locations of Christian monasteries in Egypt. However, the question remains: why would Christian monks not put a single canonical text in the Nag Hammadi Codices?

I also want to map the reuse of papyri generally, which I’ll post more about next time.

 

Machine-actionable data and the future of academic publishing

September 19th, 2013 by gwa1

Gregory Crane, editor-in-chief of the Perseus Digital Library, recently said this (p. 11 here):

For a generation we have fretted about whether junior scholars could gain tenure for digital publications. The question rather is how long faculty will be able to get credit for static publications, consisting only of prose and without accompanying machine actionable data.

I am not on the tenure track of course, but I hope to be and would rather not start out behind, which is one of the reasons I signed up for our Digital Humanities class.

In this post, I want to do two things. First, I want to ask what you think about Crane’s statement. He is addressing scholars in the field of Classical studies. Do you think his statement holds in your own field within the humanities? Do you think that there will come a time when publishing machine-actionable data will be required for promotion and tenure, and if so, how soon?

Second, I want to share a bit of my work and how I plan to do some of what Crane suggests: have machine-actionable data to accompany a static publication. Here goes.

As part of a summer institute I attended in 2011, I edited a papyrus letter that a Roman soldier wrote to his family in ancient Egypt, telling them that he would come visit. My edition of the letter is in press, and I just got the proof yesterday. It will be published in the Bulletin of the American Society of Papyrologists (and later available online).

Once I make corrections to the proof, what I need to do next is convert part of my static publication into machine-actionable data, as Crane says.  I will be converting my Greek text and English translation of the letter. (My static publication also features an introduction and line-by-line commentary, most of which will not be converted into machine-actionable data.)

During the summer institute they showed me how to use Papyri.info. It is a digital humanities research tool that allowed me to prepare my edition of the letter. During the summer institute, they also showed me how to contribute content to the site, so that down the road I would be able to add the published Greek text and English translation of my papyrus to the databank of papyri.

In 2011, when they showed me how to contribute content to the site, I entered some short Greek texts into the databank. Here is the record for one of the shortest ones that I entered (o.abu.mina.330). If you go to the editorial history, you will see where I encoded the Greek text in EpiDoc XML, which they taught me during the summer institute. These are the others that I did: o.abu.mina.337, o.abu.mina.340, o.abu.mina.344, o.abu.mina.361, o.abu.mina.377, p.gen.4.191.

That was a while ago, and I need to ask them, namely Josh Sosin at Duke, to refresh my memory. No doubt there are instructions online that he will point me to. I also note that we will be talking about text encoding in class, so that will be good for me.

After I read up on EpiDoc XML, it will still be challenging for me to encode my Greek text and English translation because they are much longer than the ones I entered into the databank in 2011. Plus, the papyrus that I edited is in really bad shape, so I have lots of editorial symbols to encode for. But I can always get help from others, and there is a review team who will make sure that the text and translation are entered correctly.

This is what the record for my papyrus (p.tept.2.583) looks like now. There is already a basic record of it in the databank, even though the letter had not been translated and prepared for publication yet. Now that I have done that, I need to update the electronic record. Currently its usefulness is limited. Once I add my Greek text and English translation, then other researchers will be able to benefit from my work in the same way that I benefited from previous work in the databank. For instance, they will be able to search for the word “furlough,” and the Greek text of my papyrus will come up along with any other papyri that mention military leave.

I could not have edited the letter without using Papyri.info, and I look forward to contributing to it.