An Early Islamic papyrus fragment from Faynan

I promised previously that I would discuss Hannah Friedman and colleagues' paper, "Fragments of an early Islamic Arabic papyrus from Khirbet Hamrā Ifdān" — published in the same issue of Arabian Archaeology and Epigraphy as our paper on Khirbat al-Manā'iyya — in more detail. Since then I've been distracted by several projects (more on which soon eventually), but wanted to return to this paper, as their conclusions are relevant to some of the points I make about Khirbat Hamra Ifdan (KHI) in my dissertation, and the find itself is quite fascinating.

The key find discussed in the paper is, as the title suggests, a poorly preserved four line papyrus fragment with an Arabic inscription. There isn't much to say about the content of the inscription, because it's quite fragmentary, with only one complete word preserved: "Allah." (Given the content, it's worth noting that I don't find their suggestion that "it is highly probable" the author of this text was a Muslim [Friedman, et al. 2017: 291] compelling, for reasons I'll discuss below.) On paleographic grounds, they date the inscription to somewhere between the late 7th and mid-8th centuries AD. I am not, by any stretch of the imagination, a papyrologist, so I can't comment on their paleographic arguments, but I can comment on the archaeological context. They were working in the northern portion of KHI, to the north of the large Early Bronze Age settlement that makes up the most famous component of the site. By contrast, my work with ELRAP has concentrated on the southern portion of the site. The primary feature relevant to the discussion here is Area L, a large square structure excavated in 2000 and, unfortunately, still essentially unpublished (I'm working on it, though). It was identified at the time as a caravanserai, but I think it's more likely to be a farmhouse reusing portions of an earlier Roman tower, a point I discuss in my reevaluation of this material in my dissertation (the papyrus fragment doesn't necessarily point to one identification over the other). In Figure 1 (the image quality for this portion of Wadi Fidan in Google Earth is unfortunately rather bad), the black patch south of the red pin is Area E, a copper slag mound dating primarily to the Iron Age II, and the square structure barely visible to the north of the pin is Area L.

The Barqa Landscape Project team, in addition to the papyrus fragment, claims to have found "Roman, Byzantine, and Islamic pottery" in the northern, thinner portion of the site, although the ceramics aren't yet published and the exact location of their excavations isn't clear from the published report. The basic claim isn't surprising, though, and is consistent with what we know of the site. In our 2014 chapter we refer to a late 8th-late 9th century AD radiocarbon date from Area L — which has since been properly published in a chapter in Mining for Ancient Copper: Essays in Memory of Beno Rothenberg I'm mistaken and must apologize; we took this out of the submitted version — to which Friedman, et al. (2017: 292) refer. In addition to this, the 2000 excavations in Area L produced pottery dating to the late 8th or 9th century, and a surface collected coin likewise dates to the 8th century. This material is unpublished (but written up in my dissertation and in my publication queue), and I point it out not to critique Friedman and colleagues for not being aware of something I didn't tell them about, but rather to point out that their interpretation fits what we know from the southern portion of the site. Certainly I'm surprised to read that they found a papyrus fragment since, as they point out, papyri from this period in southern Jordan are exceedingly rare, but I'm not surprised that they found Early Islamic period material at KHI. I'm curious to read more about the context they excavated, and in particular the depth of the later occupation, however. Many of the primarily Early Bronze Age areas in the central portion of the site had fairly shallow Byzantine and Early Islamic reoccupations above, but only Area L had relatively deep Early Islamic period loci. It will be interesting to see how the northern building compares to these. (MacDonald [1992: 252] also noted slag in the northern part of the site, and while I'd love to be proven wrong on my assertion that there isn't any Early Islamic period copper smelting in Faynan, this is probably something they would have mentioned if that were the case.)

Comparison to Khirbat Faynan

Later in the paper, Friedman, et al. discuss the broader Faynan region, and particularly Khirbat Faynan, a large site in the eastern portion of the Wadi Fidan/Faynan system. Khirbat Faynan was the Roman and Byzantine town of Phaino, site of an imperial metallum, and was certainly occupied into the Early Islamic period. In general I think this discussion is sensible, but one part of it took me by surprise. On pp. 291-292, Friedman, et al. state that "Khirbat Faynān ... has never been excavated." Part of why this surprised me so much is that the citation following this statement includes "Jones et al., 2014: 184," where I make reference to the ELRAP excavations at Khirbat Faynan. In fact, as shown in Figure 2, ELRAP conducted excavations in 2011 and 2012 in three areas of the site. You can even make out the step trench we excavated in 2011 in Area 16 on the satellite imagery (the portion to the left of the "Area 16" pin is particularly easy to make out).

In fairness to Friedman and colleagues, these excavations are not as well published as they should be. I do, however, refer to them in our 2014 chapter, and the 2011 excavations are discussed in a paper in the Annual of the Department of Antiquities of Jordan, a scan of which is available here. The ADAJ paper only discusses earlier occupations at Khirbat Faynan, but even without the Early Islamic area we excavated in 2012 — Area 18 — it would be incorrect to say the site had never been excavated. Nonetheless, Friedman and colleagues correctly note that the site continued to be occupied into at least the later 8th century. This is basically consistent with the dating of the Early Islamic phases in Area 18, although the latest of these contained several forms that continue into the 9th century, as well.

The Islamization of Faynan: An Early Islamic Mosque in Wadi Fidan?

I should now return to the point I brought up above concerning the religious identity of the author of the inscription, which isn't the critical point of the paper, but is the major point I disagree with. Friedman, et al. (2017: 291) point out that the content of the inscription does not actually point to an identification of the author as a Muslim. As the Arabic word for "God," "Allah" appears in Arabic Christian texts of this period (and later, of course), as well. Friedman, et al. provide some examples, to which more could certainly be added. Nonetheless, they also think it is "highly probable" that the author was a Muslim, apparently on the basis of the archaeology of the site.

What about the archaeology makes them say this? On pp. 285-286, Friedman and colleagues note that one of the authors, Russell Adams, visited the site in the mid-1980s as part of Burton MacDonald's Southern Ghors and Northeast 'Arabah Archaeological Survey (SGNAS) and observed a structure that looked like an open-air mosque, but with its mihrab facing Jerusalem, rather than Mecca. They go on to say that when he and my advisor, Tom Levy, began excavating the site in 1999, the structure was gone. It's unclear what this structure is, as it was apparently never mapped. I would note, however, that this sounds rather like a structure visible on a map of the excavations published by Levy, et al. (2002: 434, Fig. 5A), which was assigned to site Stratum II, dating to the Early Bronze Age IV. This isn't an exact match (the "mihrab" faces slightly northeast, more or less toward Damascus, rather than Jerusalem), but it's pretty darn close, considering the reconstruction is based on memory. Friedman and colleagues, rightly, don't seem to regard this as evidence that there was a mosque on the site. It is perhaps also worth noting WFD 105, published by Levy, et al. (2001: 176, Table 2) as an Islamic period cultic site. If that identification is correct, it would also be quite odd, as that structure oriented to the west, rather than the south. (To this discussion we could also add FJHP Site 136, a structure near Petra with a niche in its eastern wall [Kouki 2013].) The discussion of whether these structures, particularly FJHP Site 136, could be mosques revolves to a large extent around the mosque or musalla built of copper slag at Be'er Ora, which seems to have niches in its eastern and southern walls, although the eastern one is less clear. I don't really want to get into this in much detail. The key facts are that this mosque was suggested to have an east-facing mihrab later converted to a south-facing one, and that the excavations did not actually produce evidence that would support (or, to be fair, rule out) this claim (Sharon, et al. 1996). In short, there isn't a lot to go on here. It's not possible to rule out the possibility that WFD 105, FJHP 136, or the now-missing structure at KHI could have been mosques, but that isn't the same as being able to say they are.

Back to the point. Why do Friedman, et al. think the author of the papyrus was likely Muslim? The beginning of their discussion is not promising: "Archaeological suggestions of an early Islamic mosque at KHI dating to the seventh or eighth century are supported by the find of the papyrus" (Friedman, et al. 2017: 292). This is, of course, circular reasoning. The papyrus is likely to have been written by a Muslim if there is a mosque at the site, which in turn is likely if there's a papyrus written by a Muslim at the site. I'll return to this briefly, but first let me lay out the rest of their argument. They first suggest the "Negev desert mosques" as a model, and place the hypothetical mosque near 'Ayn Fidan, the spring to the south of the site. In the next paragraph, they suggest that "structures surrounding the missing mosque functioned as a venue for economic or social purposes," citing the farmstead at 'En 'Avrona (in the southern Wadi 'Araba, near 'Aqaba) as a parallel (Friedman, et al. 2017: 292). I don't think both of these models should be applied to the same site, and I would suggest that 'En 'Avrona, Nahal La'ana (Nahlieli, et al. 1996] and a number of other sites — which have rooms that served as mosques inside structures that look rather like KHI Area L — are probably a closer parallel to KHI than the hilltop Negev mosques. In other words, further excavation in Area L could probably tell us whether there is a mosque at the site or not. Which brings us back to their initial reasoning. The idea that there should be a mosque at the site assumes more about the Islamization of Faynan than we actually know. What we can presently say is that the region continued to be occupied well into the Early Islamic period. That doesn't, however, tell us much about the religious identity of the people living there. One can assume that these people were either early converts or Muslim newcomers to the region, but at our present state of knowledge, this is an assumption. Looking to Petra rather than the Negev may caution against this, as excavations at Khirbat al-Nawafla in Wadi Musa demonstrated that the population was either primarily Christian or confessionally-mixed into at least the late 8th century, and probably rather later ('Amr, et al. 2000). It's also possible that KHI and Khirbat Faynan don't follow the same pattern. Further excavation could very well find a mosque at KHI and continued use of churches at Khirbat Faynan. Presently, we simply don't know.

That said, however, I should reiterate that this is a really incredible find, and I'm definitely looking forward to hearing about the rest of the material they've found.

---Friedman, Hannah, Tasha Vorderstrasse, Rachel Mairs, & Russel Adams (2017). Fragments of an early Islamic Arabic papyrus from Khirbet Hamrā Ifdān Arabian Archaeology and Epigraphy, 28, 285-296 : 10.1111/aae.12099 

Other Works Cited

'Amr, Khairieh, Ahmed al-Momani, Naif al-Nawafleh, and Sami al-Nawafleh. 2000. Summary Results of the Archaeological Project at Khirbat an-Nawāfla/Wādī Mūsā. Annual of the Department of Antiquities of Jordan 44:231-255.

Jones, Ian W. N., Mohammad Najjar, and Thomas E. Levy. 2014. “Not Found in the Order of History”: Toward a “Medieval” Archaeology of Southern Jordan. In From West to East: Current Approaches to Medieval Archaeology. S.D. Stull, ed. Pp. 171-206. Newcastle-upon-Tyne: Cambridge Scholars.

Jones, Ian W. N., Mohammad Najjar, and Thomas E. Levy. 2018. The Arabah Copper Industry in the Islamic Period: Views from Faynan and Timna. In Mining for Ancient Copper: Essays in Memory of Beno Rothenberg. E. Ben-Yosef, ed. Pp. 332-342. Tel Aviv University Sonia and Marco Nadler Institute of Archaeology Monograph Series, Vol. 37. Winona Lake, IN and Tel Aviv: Eisenbrauns and Emery and Claire Yass Publications in Archaeology.

Kouki, Paula. 2013. Site 136, an Open Mosque? In Petra — The Mountain of Aaron: The Finnish Archaeological Project in Jordan, Volume III: The Archaeological Survey. P. Kouki and M. Lavento, eds. Pp. 317-321. Helsinki: Societas Scientiarum Fennica.

Levy, Thomas E., Russell B. Adams, Andreas Hauptmann, Michael Prange, Sigrid Schmitt-Strecker, and Mohammad Najjar. 2002. Early Bronze Age metallurgy: a newly discovered copper manufactory in southern Jordan. Antiquity 76:425-437.

Levy, Thomas E., Russell B. Adams, Alan J. Witten, James Anderson, Yoav Arbel, Solomon Kuah, John Moreno, Angela Lo, and Mark Wagonner. 2001. Early Metallurgy, Interaction, and Social Change: The Jabal Ḥamrat Fīdān (Jordan) Research Design and 1998 Archaeological Survey: Preliminary Report. Annual of the Department of Antiquities of Jordan 45:159-187.

Levy, Thomas E., Mohammad Najjar, Aaron D. Gidding, Ian W. N. Jones, Kyle A. Knabb, Kathleen Bennallack, Matthew Vincent, Alex Novo Lamosco, Ashley M. Richter, Craig Smitheram, Lauren D. Hahn, and Sowparnika Balaswaminathan. 2012. The 2011 Edom Lowlands Regional Archaeology Project (ELRAP): Excavations and Surveys in the Faynān Copper Ore District, Jordan. Annual of the Department of Antiquities of Jordan 56:423-445.

MacDonald, Burton. 1992. The Southern Ghors and Northeast 'Arabah Archaeological Survey. Dorchester: The Dorset Press.

Nahlieli, Dov, Yigal Israel, and Yehudit Ben-Michael. 1996. The Nahal La'ana Site: An Early Islamic Farm in the Negev. 'Atiqot 30:67-78, 130.

Sharon, Moshe, Uzi Avner, and Dov Nahlieli. 1996. An Early Islamic Mosque near Be'er Ora in the Southern Negev: Possible Evidence for an Early Eastern Qiblah? 'Atiqot 30:107-114.

Some shameless self-promotion

Posts have been a little light here for a while as I finish my dissertation, but enough things have come out in the last month or so that I should really mention them here. First, and certainly most excitingly, I was profiled in the Jordan Times on December 9 in an article by the amazing Saeb Rawashdeh. Saeb did a great job of presenting the key arguments of my dissertation research and the significance of that research for Jordanian archaeology. I'm somewhat biased, but I think you should check it out!

In the realm of peer-reviewed, unfortunately closed-access work, my co-authors and I published a paper in the most recent issue of Arabian Archaeology and Epigraphy (if you don't have access to AAE and would like an offprint, please feel free to contact me). In it, we report on ELRAP excavations at a site in southern Jordan called Khirbat al-Manā'iyya in 2012. As I point out in the abstract, Khirbat al-Manā'iyya is exciting because (among other things) it's the first Early Islamic period copper smelting site known in the southeastern Wadi 'Araba (actually, in the entire eastern Wadi 'Araba). In addition to reporting the site, we also discuss how Khirbat al-Manā'iyya fits into the system of industrial settlements, including other copper smelting sites, in the southwestern 'Araba, and how this system articulates with Early Islamic mining in northern Arabia, expanding on arguments we first made in our "Not Found in the Order of History" chapter. I should also note that Brita Lorentzen," who I've mentioned previously on this blog, analyzed the charcoal assemblage from the site and found evidence for the use of deadwood, which tells us some interesting things about how the sparse wood resources of the southern Wadi 'Araba were managed during this period. I was quite excited to be able to work on this site, and I'm even more excited that the publication is out. Note also that it came out in the same issue as David Kennedy's paper on the "gate" features in Saudi Arabia, which got a bit of press, and a very interesting paper by Hannah Friedman and colleagues about an Early Islamic papyrus from the Faynan region, which I hope to discuss in slightly more detail in an upcoming post.

Lastly, I've been co-editing a book called Cyber-Archaeology and Grand Narratives: Digital Technology and Deep-Time Perspectives on Culture Change in the Middle East with my advisor, Tom Levy, and it has also just come out in the Springer One World Archaeology series. It's a cool volume, based on a session at the 7th World Archaeological Congress and a workshop at UC San Diego (the chapters have been updated since then, of course). The idea was that contributors would consider how digital archaeology can contribute to investigations of archaeological "grand narratives," and the contributions both explore the potential of new methods and provide insightful critiques of existing methods (you can check out the table of contents here). In addition to being one of the editors, I'm also first author in the intro chapter, which I think is worth a read (as is the entire volume!).

The southernmost Levant in the news!

Several interesting news stories have popped up in the last few days about archaeological research in southern Jordan and Israel. First, I was excited to see that my good friend and colleague Erez Ben-Yosef's work in the Timna Valley was featured in National Geographic. The gist of the article is that he and his team analyzed donkey dung (kind of a shitty project, to steal a joke from Ben Saidel), and were able not only to radiocarbon date the dung to the 10th century BC, but also to narrow the source of the donkeys' food to regions with Mediterranean climates, hundreds of kilometers to the north. That research has also been published in Journal of Archaeological Science: Reports. It's behind a paywall, but you can at least read the abstract for free. I would do a proper Research Blogging summary, but I owe Erez revisions on a book chapter, so I should probably do that instead. . . But before that:

Second, John Oleson's work at al-Humayma has been highlighted in a Jordan Times article (I've seen it reposted in a number of other places, too) focusing on the earliest Nabataean settlement at the site. The article then goes into a brief history of the site itself — including its role in the 'Abbasid Revolution — and of the extensive research that Oleson has conducted there.

I'm a bit biased, as I'm always glad to see my general research area in the news, but these are both great projects, and the stories are definitely worth a read.

That new coin hoard from the Sasanid invasion

Hello, and welcome to the latest installment of "Ian rambles about the latest archaeology news." I've seen this story shared in various forms on the various Near Eastern archaeology social media thingummies, and I've read it with interest each time. This is, in part, because much of the reporting on it has been . . . rather strange. The gist of the story is that archaeologists from the Israel Antiquities Authority found a hoard of nine 6th and early 7th century AD copper folles in salvage excavations at 'En Hemed, and based on the dating of the latest coins have suggested that they were hidden prior to the Sasanid invasion of 614.

It's been reported in the Jerusalem Post (use an ad blocker!), which seems mostly to rely on the IAA press release, and a much stranger article has also appeared in the Washington Post. I'm sure there are other versions out there, too, though I can't imagine any are quite as strange as that one.

First, the less strange one. As I mentioned, the Jerusalem Post story mostly seems to repeat the IAA press release, but with nicer photos. The major contribution seems to be that in the JPost story the hoard is called "rare," which is probably not the word I would use, as quite a few hoards can be attributed to the Sasanid invasion. In fact, to quote Alan Walmsley (2007: 324), "Quite a few hoards can be attributed to the Sasanid invasion." Beyond that, it's a bit weird that both the JPost article and IAA release give dates of birth and death for each emperor, which, when discussing coins, is somewhat less useful than regnal dates. (It would, indeed, be quite rare to find a coin of Phocas dating to 547 AD, as that's 55 years before he became emperor. . .) This isn't a huge issue, though, and the rest of the release seems pretty straightforward.

The dating issue is corrected, at least for the reign of Phocas, in the Washington Post version. Sort of. They give the dates as 604-609, which is not quite right for the reign of Phocas, but I suspect the archaeologist, Annette Landes-Nagar, has narrowed this down on numismatic grounds. I'm not a numismatist (and the coins are, as would be expected in a news story like this, not presented in much detail), so I'm not totally sure. I am, however, more sure that it is not accurate to say that Landes-Nagar "estimated that the coins were minted sometime between 604 and 609 because they bear the faces of Byzantine emperors of the time," considering that the hoard also contained issues of Justinian and Maurice, neither of whom was Byzantine emperor at that time.

I imagine that at this point you're saying to yourself, "I don't know. That doesn't seem that weird." Agreed. That is just me being nitpicky. What's weird is that, of the 18 paragraphs that make up the article, only five are about the 'En Hemed excavations. The rest seems to be about the archaeological evidence for early Christianity in Israel and, uh, the existence of Jesus, for some reason. The first, we're told, is "a potent point, offering proof of the Christian connection to the Holy Land and the Middle East, alongside that of Judaism and Islam." This seems fairly obvious, but on the other hand, I don't think anyone who follows the archaeology of Israel will be surprised to see the phrase "proof of the X connection to the Holy Land" in an article. So, fair enough?

Stranger is the point that archaeologists haven't found "physical evidence of [Jesus'] existence." True, but . . . what does that have to do with a hoard of 6th and 7th century coins? I think the idea here is something to do with the development of the Christian community, but it seems like a strange way of introducing this coin hoard, especially since that connection is never made very clear.

None of that is, of course, a critique of the excavations, and I'll be curious to see the publication when it comes out. Considering that the hoard was found in collapse, I'll be interested, in particular, to see if they consider the possibility that it might actually have been dropped during the earthquake of 633, as Russell (1985: 46) suggested for three houses at Bet She'an (where the latest coins, it turns out, were also issues of Phocas). That does seem less likely in this case, but still possible.

Works Cited

Russell, Kenneth W. 1985. "The Earthquake Chronology of Palestine and Northwest Arabia from the 2nd through the mid-8th Century A.D." Bulletin of the American Schools of Oriental Research 260:37-59.

Walmsley, Alan. 2007. "Economic Developments and the Nature of Settlement in the Towns and Countryside of Syria-Palestine, ca. 565-800." Dumbarton Oaks Papers 61:319-352.

Filling gaps in Middle Islamic settlement

My colleague and good friend Kyle Knabb just posted an abstract for a paper he's giving at the SAAs in Honolulu this year, and I thought, "Oh, I'm also presenting half of that paper. I should probably mention it."So, as Kyle said, we're working together right now to analyze some of the pottery from an intensive survey he led in Wadi al-Faydh, near Petra, in 2009. I was a member of his survey team, so it's exciting to get to analyze a lot of this material finally. As Kyle also mentioned, the majority of the assemblage is made up of rather coarse hand-made pottery (how coarse, you ask? Here's an example collected in Petra by the International Wadi Farasa Project, which gives you an idea of what we're talking about). In addition to not being the most attractive pottery (though I would argue that it has its charms), much of it is also rather difficult to date, especially when it comes from surveys, rather than excavations. One of our arguments, however, is that recent (and in some cases not-so-recent) excavations have produced evidence that enables us to date some of the ceramics Kyle found – especially some distinctive decorated forms – to the 11th and early 12th centuries AD.Our abstract begins with a related problem, which is that evidence of settlement during this period has been somewhat tricky to actually find. To get an idea of this, we just have to check out the DAAHL's (that's the Digital Archaeological Atlas of the Holy Land, for those not yet in the know) Archaeological Periods page. If you zoom to the study area and select Fatimid or Crusader, you see relatively little settlement. If you click Ayyubid, Ayyubid/Mamluk, or Mamluk, the picture is different (and this is also true if you select 'Abbasid/Fatimid, actually). If, as we're going to argue, many ceramics of the Fatimid and Crusader periods have been misclassified as Ayyubid/Mamluk – especially in the south – we have to wonder how good our understanding of settlement patterns in the Middle Islamic I (1000-1200 AD) actually is.These dating concerns have implications beyond simply establishing the chronology of settlement in the region. The biggest issue for me is that sites which weren't occupied during the same period obviously can't be directly connected to one another. This is important for reconstructing local patterns of trade, and one of the things I'm concerned with in terms of my work in Faynan. As the dating of sites becomes more precise, connections that seemed obvious when all the ceramics were lumped together as "Ayyubid/Mamluk" suddenly disappear. But that's a different story for another day. . .