The Jerusalem Papyrus and the forged words on it
In October 2016, a papyrus mentioning the city of “Jerusalem,” “the king,” and “jars of wine” was published, amid much pomp and circumstance.1 The script and language were Hebrew, the words penned in ink on the papyrus. It seemed to some to be a truly sensational find. Based on the readings and restorations in the original publication, the inscription was interpreted to say: “…[Maidserv]ant of the King, from Na‘arat, two jars of wine to Jerusalem.” This content is quite sensational, as it mentions both the city of “Jerusalem” and “the King” (i.e., a king of Judah) in the same inscription, a true rarity to have both in the same inscription.
Significantly, this papyrus inscription had not been discovered on a scientific archaeological expedition, but rather had “appeared” on the antiquities market. Indeed, more than two years prior (in late 2013 or early 2014), an antiquities dealer had contacted me in Jerusalem and invited me to look at some high resolution images of it. Thus, I had seen the papyrus before. And the antiquities dealer had told me that it was for sale. Naturally, I was not interested. And even at that time, I had strong suspicions that it was a modern forgery, based on some discernible anomalies in the script and language.
Nevertheless, in the editio princeps (first publication) of this papyrus, the authors stated quite confidently that it was an authentic inscription, not a modern forgery. They dated it to the seventh century B.C.E. After all, they contended, the papyrus itself has been carbon dated, and it was shown to be ancient. Moreover, they stated that neither the script nor the language of the inscription contained serious anomalies. In short, they said it was indeed an ancient Hebrew inscription. Quite a number of scholars agreed. But the fact of the matter is that there were a lot of assumptions that were being made, and that is usually not the best manner of attempting to get at the facts. I rapidly posted some of my serious concerns about the authenticity of this inscription.2
First and foremost, assumptions were being made about the certitude of the authenticity of the inscription, based on the carbon dating of the papyrus. This is a serious problem. Obviously, carbon testing is among the most important tools in the toolbox, but the antiquity of the medium does not ensure the antiquity of an inscription. Indeed, I have stated for many years now that it is not all that difficult for someone to acquire “ancient potsherds, ancient metals, stones of Levantine quarry, small pieces of ancient papyrus, or vellum.”3 Therefore, the antiquity of the medium (e.g., papyrus, vellum, potsherd, or metal) is certainly no guarantee of the dating of the writing on that medium. To put it differently, only the dullest of forgers would forge an inscription on modern papyrus, modern vellum, modern potsherds, or modern metals. After all, most forgers are quite sharp and they know that laboratory tests are routinely performed, and so the forgers know that it is important for them to use ancient materials from the correct period as their medium (e.g., using a piece of Iron Age pottery and then write on it using a correct script from the Iron Age, or using a piece of papyrus that is putatively from the Iron Age for an Iron Age inscription). And forgers have produced a fair number of forgeries in the last 40 or 50 years, and this is the way they do it. They know the drill.
There are also additional aspects of the carbon-14 test that deserve scrutiny. Namely, quite a number of people said to me that the papyrus was carbon dated to the seventh century B.C.E., and the script is also dated to the seventh century B.C.E.; therefore, they said, that sort of correspondence is very good evidence for the antiquity of the writing. After all, it might be difficult to find a piece of papyrus that was from the seventh century. Alas, that too is an interesting (and problematic) line of reasoning for the Jerusalem Papyrus. After all, for carbon dating carbon materials from antiquity, there is normally a fairly substantial plus or minus range. Thus, I found it hard to believe (in spite of the initial press reports) that the carbon date for this papyrus would fall in, and only in, the seventh century B.C.E. And sure enough, the devil was in the details here, as well—even more seriously than I had initially thought. Basically, the carbon dates for the Jerusalem Papyrus fell into the Hallstatt Plateau, and so all that can actually be said is that this papyrus dates to sometime between 800 B.C.E. and 400 B.C.E. (I’m grateful to Israel Finkelstein for mentioning the problems with the Hallstatt Plateau to me). That’s a fairly broad range. Obviously, therefore, it cannot be said that the papyrus itself definitely dates to the seventh century B.C.E. It might come from the early eighth century. Or it might come from the late fifth century. And it might come from anywhere in between. In other words, there is not some sort of dramatic convergence of the carbon date and the putative date of the script.
At the time of the publication of the Jerusalem Papyrus, no testing of the chemical composition of the ink had been done (e.g., with something such as a scanning electron microscope equipped with an energy dispersive spectrometer). Fortunately, this testing has reportedly been done just recently, although the results are not yet published. But even this is not the sort of thing that can prove that the writing on the Jerusalem Papyrus is ancient. After all, it has been known for many decades now that the core element of carbon-based inks is (of course!) ancient carbonized remains. Such remains are readily available (e.g., on excavations or from the antiquities market) in the form of burned wood, charred beams, or (as Yuval Goren mentioned to me recently) even by simply scraping of the carbon off from a cooking pot that had been used in antiquity. Once some carbonized remains are in hand, a savvy forger can readily make a nice carbon-based ink, and one that would even yield an ancient carbon-14 date. In short, ink too can be faked.
But there is more. Within the Semitic languages, including Hebrew and Aramaic, is a linguistic construction called a “construct.” In its most basic form, a construct chain consists of the juxtaposition of two nouns (or nominals). So, for example, the phrase “Law of Moses” is a construct chain, and the phrase “Song of Songs” is as well. One of the most important features of a construct chain is the form of the first noun in the construct chain. That noun is said to be in the “construct state” (and the noun that follows it is said to be in the absolute state). And when that first noun is a masculine plural or dual plural, its construct form is quite different from its absolute form.
In the Jerusalem Papyrus inscription, there is paradigmatic construct chain, namely, “jars of wine.” We have that very same phrase, with the very same words, in the Hebrew Bible: “jars of wine” (1 Samuel 25:18; cf. also Job 38:37; Lamentations 4:2). But there is a subtle difference. In the Hebrew Bible, it is spelled nbly yyn. That’s the correct spelling. However, in the Jerusalem Papyrus, it is spelled “nblym yyn.” The problem with the spelling in the Jerusalem Papyrus is that the m is not supposed to be there. It’s not the sort of mistake that a native speaker of ancient Hebrew would make (and certainly not a scribe!). Significantly, within modern Hebrew, a circumlocution is often used to avoid construct forms (namely, the word šĕ), but in ancient Hebrew (in speech and in writing), the construct form was the way to do this. And, of course, the fact that we have the construct form of “jars of” (i.e., nbly) used multiple times in the Bible, including the very phrase “jars of wine,” demonstrates that this was certainly the way it should have been done in the Jerusalem Papyrus. But it wasn’t (and I find the logic of the authors of the editio princeps to account for this problem to be strained, special pleading). This is really quite a rookie mistake for the forger, and my strong suspicion is that the forger of this text is reading up right now on the proper construct forms in ancient Hebrew. I doubt that he will make that mistake again. There are also some problems with the script, some very fine anomalies. I may discuss those in a future publication…or I may not do so, in order to avoid educating the forgers. After all, for the past century and a half, forgers have been reading the things scholars write and learning more and more about how to avoid blunders in their forgeries.
Ultimately, the case against the Jerusalem Papyrus is pretty strong. To be sure, there are, and will continue to be, people who believe that it’s ancient. But for my money, I think that it’s of recent vintage. And the modern forger is pretty good at his craft, but not perfect. And, as I mentioned, I suspect that the forger of this inscription is studying up on construct forms right now.
Christopher Rollston is Associate Professor of Northwest Semitic Languages and Literatures at The George Washington University.
1. S. Ahituv, E. Klein, and A. Ganor, “To Jerusalem: A Seventh-Century BCE Shipping Certificate,” New Studies in the Archaeology of Jerusalem and Its Region 10 (2016), pp. 239–251 [Hebrew]. For my full treatment of this inscription, see C. A. Rollston, “The Putative Authenticity of the New ‘Jerusalem’ Papyrus Inscription: Methodological Caution as a Desideratum,” in Oded Lipschits, ed., Rethinking Israel: Studies in the History and Archaeology of Ancient Israel in Honor of Israel Finkelstein (Winona Lake: Eisenbrauns, 2017), pp. 321-330. This inscription will also be discussed in my forthcoming monograph entitled Forging History in the Ancient World of the Bible & the Modern World of Biblical Studies (Grand Rapids, Eerdmans).
2. C. A. Rollston, “The New ‘Jerusalem’ Papyrus: Not so Fast…” (posted October 26, 2016) and “The Jerusalem Papyrus: Complementary Notations” (posted November 1, 2016), Rollston Epigraphy (blog).
3. C. A. Rollston, “Non-Provenanced Epigraphs I: Pillaged Antiquities, Northwest Semitic Forgeries, and Protocols for Laboratory Tests,” Maarav 10 (2003), pp. 135–193, quotation from page 139.
Biblical History at What Cost? by Roberta Mazza
Precursor to Paleo-Hebrew Script Discovered in Jerusalem
2,000-Year-Old Jerusalem Inscription Bears City’s Name
Archaeological Looting and the Destruction of Cultural Heritage
Is the Gospel of Jesus’ Wife a Fake?
“Down the Rabbit Hole”: Owner of the Gospel of Jesus’ Wife Papyrus Unmasked
Does Radiocarbon Dating Accuracy Help Us Determine Bible Chronology?
Sign up to receive our email newsletter and never miss an update.
Dig into the illuminating world of the Bible with a BAS All-Access membership. Combine a one-year tablet and print subscription to BAR with membership in the BAS Library to start your journey into the ancient past today!Subscribe Today
Probably, the forger has simply consulted §131d in Gesenius’ Hebrew Grammar, and decided to use quite a regular Ancient Hebrew structure of the apposition, compare also מצלתים נחושת, ככרים כסף etc.
One would expect a modern forger ignorant of Biblical Hebrew to use a modern form like nblym shl yyn and not an unknown form like nblym yyn which is neither the Biblical form nor the modern form. It looks more like an ancient form that is a variant of what is used in the Bible.
A forger who doesn’t have a grip on what you are supposed to learn a few weeks into first semester Biblical Hebrew? Maybe so, or maybe he just misspelled – like we all do sometimes.
Rollston makes a strong argument but there are other considerations. Nowadays, people who write shipping certificates and waybills are usually not well educated nor punctilious about their grammar or spelling. Maybe the same was true in First Temple times although reading and writing seem to have been common abilities in Israel in those days, especially compared to other lands.
Further, even in the Hebrew Bible, the MT, there are divergent spellings for the name of the same person and for various other words plus grammatical irregularities [Consider the name spellings Yukal and Yehukal for the same person]. So does the irregular smikhut [construct state] in the word for jars on the papyrus prove that this inscription is forged? Or maybe we cannot tell given that fraudsters may be clever and the argument can go both ways.
Now in Second Temple times for which remains of writings preserved are more common, we can see that on ossuaries, potsherds, mosaics, etc irregular spellings and grammatical forms appear. Is this due again to workmen who were not well educated or to the influence of Aramaic and Greek on Jews in Judea at the time? Or both? Maybe we cannot at this time reach a definitive conclusion about this particular papyrus. But Rollston’s method is intriguing in that it seems to use a linguistic analysis which is what Yehezqel Kauffmann did in his argument against the documentary hypothesis. Can we look to Kauffmann for enlightenment?
While modern error does seem the more likely, we do know of an Egyptian? scribe of notable incompetence. His preserved errors tell us he was able to make an error this obviously wrong. Since he was surely not the only scribe chosen by political pull rather than knowledge or talent, we can’t rule out that the error was ancient. Since the scribe also had a long list of complaints against him, we can assume he was way below average, and the chance our papyrus was ancient is low, but we have to admit there is a chance.
Ancient inscriptions commonly contain unexpected “anomalies” in their language. Languages change over time and have regional differences that are poorly documented because of the lack of material.
Contrary to the argument presented in this article, the fact that this text uses a different spelling to that preserved in the Hebrew Scriptures probably suggests it is authentic.
Most forgeries would copy the phrase exactly from the extant Biblical text… but then again a truly smart fraudster would deliberately introduce something unique.
If the text exactly matched the construct preserved in scripture the ‘experts’ would argue ‘it must be a forgery because the fraudster just copied the text right out of the Bible’.
So as long as the text does not have serious grammatical flaws or obviously reflect modern Hebrew an ‘expert’ can argue either way based on “anomalies” in the language.
The use of the enclitic m is thought to have declined after the exile, but it’s assumed to be from the first Temple anyway, based on the calligraphy, as the article mentions. In any case, I checked the original publication (here in Hebrew, p. 245), and it reads nblym as the dual form נִבְלַיִם, which is also grammatical in construct form (as in וְסָאתַיִם שְׂעֹרִים, 2 Kings 7:1), in which case it isn’t a certain enclitic m. Especially considering this reading of the word, I’m not sure what reason there is to believe the inscription to be ungrammatical.
aew – Again, very interesting. In that case, would the use of the enclitic m narrow the date range given by the carbon dating, i.e. would its use make it more likely the papyrus (if authentic) is from an earlier date in the 400 to 800 BCE range? In any case it will be interesting to see the dating results when they test the ink.
The m at the end of nblym yyn can be interpreted as an enclitic m, which is known to exist in Ugaritic, and seems to also exist in older Hebrew. Nblym yyn should be absolutely expected in a text from that era. On the contrary, the fact that it doesn’t copy the Biblical phrase nbly yyn proves either that the papyrus is authentic, or that the forger was extremely well-versed in the details of Biblical Hebrew.
Very interesting. Is it odd that a forger who went to enough trouble to get an old papyrus and (depending on the dating of the ink) make his/her own ancient ink would get the spelling of a word wrong? Especially if the phrase is one that actually shows up in the Old Testament in its proper form, and therefore would readily accessible.