In volume 220 of Zeitschrift für Papyrologie und Epigraphik (2021), there is a short article by Cornelia Römer: “Die Datierung des Kölner Mani-Kodex” (pp. 94-96).
The article reports the results of AMS radiocarbon analysis of the Cologne Mani codex. For those who don’t know this remarkable little book (TM 64574): It is a miniature parchment codex whose leaves are about 3.8 cm wide and 4.5 cm high. It contains a biography of the prophet Mani (c. 216-276 CE) in Greek. It is copied in finely executed letters that are just 1 millimeter high.
The codex was reportedly bought in Egypt in the late 1960s, but the details are very sketchy. According to Albert Henrichs (in 1979) “next to nothing is known about the fate of the Mani Codex before it reached Cologne.” Rumors that the codex had been subjected to radiocarbon analysis have been circulating since the 1990s, but no publication had appeared. Until now.
To briefly recap the ranges of dates that have been proposed for the codex: When the existence of the codex was first reported in 1970, Albert Henrichs and Ludwig Koenen believed on the basis of its script that the codex was likely produced in the fifth century CE (“Aus paläographischen Gründen ist er wahrscheinlich dem 5. Jh. zuzuweisen”). In 1977, Eric Turner opted for fourth century or fourth-fifth century. In 1990, B.L. Fonkič and F.B. Poljakov assigned the copying of the codex to the eighth century CE. More recently, Pasquale Orsini has argued that the codex preserves the work of two copyists working in two different periods. One hand that “could be dated to the sixth century” was the initial copyist, and then “some time between the second half of the eighth century and the beginning of the ninth century” a second copyist, probably “intervened later to ‘restore’ parts at the beginning and end of the codex which had been damaged.”
Now to the radiocarbon analysis: The process was undertaken at the initiative of Ludwig Koenen in 1995. What was sent off for analysis was a 7.25 mg sample taken from the binding thread of the codex. To supplement the report provided in ZPE: It is very important to recall that the codex was in a delicate state when it was purchased, and it underwent an extensive process of conservation in the hands of Anton Fackelmann. I am not aware if there is a detailed and reliable account of the exact procedures used for conservation. In 1979, Albert Henrichs reported that the codex arrived in Vienna as a collection of four clumps of leaves gummed together. Then, “with the help of a chemical solvent manufactured in the United States, Dr. Fackelmann managed to soften the brittle material. …[T]he pages came off much faster than I could transcribe them.” It would be very good to know what exactly this chemical solvent was–and whether it came into contact with the binding thread of the codex. If the solvent contained carbon, the possibility of contamination arises.
The results of the analysis were reported as follows in a letter from the AMS facility at the University of Arizona dated 20 January 1996:
“Radiocarbon Age: 1,652 ± 60 yr BP”
“Calibrated Age: 267 – 434 AD (1 sigma, 68% confidence)
240 – 540 AD (2 sigma, 95% confidence)”
Römer supplements these results with a more recent IntCal13 calibration expressed in the following way:
“Für die 2 Sigma Fehlergrenzen (95%) erhält man folgendes detaillierte Resultat:
Wahrscheinlichkeit
263 – 274 AD (0,048)
330 – 432 AD (0,745)
490 – 533 AD (0,207)”
We can now adjust again with the still more recent IntCal20 data using the OxCal v4.4.4 program:
The estimated date ranges at 95.4% probability are:
251 – 294 CE (09.7%)
314 – 555 CE (85.7%)
The results across all three sets of calibration data are relatively uniform. The latest calibration offers a slightly wider range of possible dates–from the middle of the third century to the middle of the sixth century. The third century dates can probably be safely excluded, as the biography of Mani contained in the codex is generally thought to have been composed in the fourth century.
The original date assigned by the editors (fifth century) along with Turner’s wider range (fourth-fifth century) are both consistent with the radiocarbon data. That the codex was originally produced in the eighth century, as Fonkič and Poljakov had argued, now seems most unlikely. These radiocarbon results do not, however, rule out Orsini’s hypothesis (that the book was originally copied in the sixth century and then damaged with portions later recopied in the eighth or ninth century). If, however, the codex (or a part of it) was disassembled and rebound, it would be somewhat surprising that the same (older) thread would be used again. But the report does not inform us of the exact place from which the sample of thread was taken (that is: From which of the several quires that have thread preserved was this sample taken?). This would be very helpful information to know, along with the identity of the aforementioned chemical used to separate the leaves. It would also be interesting to test further samples from the quires assigned to different copyists.
Finally, what is the best way to describe the date of the codex in light of all the data (palaeographic and radiocarbon)? We can now say with a high degree of confidence: “fourth century to mid-sixth century.”
Despite the remaining questions, it is good to have this new data more widely available.
Sources mentioned:
Fonkič, B.L. and F.B. Poljakov. “Paläographische Grundlagen der Datierung des Kölner Mani-Kodex.” Byzantinische Zeitschrift 83 (1990), 22-30.
Henrichs, Albert. “The Cologne Mani Codex Reconsidered.” Harvard Studies in Classical Philology 83 (1979), 339-367.
Henrich, Albert and Ludwig Koenen. “Ein griechischer Mani-Codex (P. Colon. inv. nr. 4780).” Zeitschrift für Papyrologie und Epigraphik 5 (1970), 97-216.
Orsini, Pasquale. Studies on Greek and Coptic Majuscule Scripts and Books, trans. Stephen Parkin and Laura Nuvoloni. Berlin: de Gruyter, 2019.
Römer, Cornelia. “Die Datierung des Kölner Mani-Kodex.” Zeitschrift für Papyrologie und Epigraphik 220 (2021), 94-96.
Turner, Eric G. The Typology of the Early Codex. Philadelphia: University of Pennsylvania Press, 1977.
Variant Readings 
This is excellent. Though you have misgivings on the enterprise, seems that manuscripts getting the radiocarbon treatment are more or less matching the (at least) century date(s) which palaeographers are assigning to the manuscripts. Has there been any manuscript which, when being radiocarbon dated, shows the palaeographical date is widely off the mark? IIRC, the Washington MS of the Gospels (Codex W) when radiocarbon dated, was better placed in the 6th century, as opposed to the usually assigned 4th-5th century. But even then this isn’t too far off the palaeographical dating.
Yes, it is often the case that the results of radiocarbon analysis are not inconsistent with palaeographic assessments. In this case, the original editors (and Turner) both proposed dates that turned out to be within this rather wide radiocarbon date range, while the date proposed by Fonkič and Poljakov falls well outside that range. I think in cases like these (when we have widely divergent palaeographic assessments), AMS analysis can be helpful in at least ruling out some possibilities. Do you have a reference for the Freer codex? I was not aware that it had been subjected to radiocarbon analysis.
Seems I’m sort of recollecting a passing comment by Ulrich Schmid, “Reassessing the Palaeography and Codicology of the Freer Gospel manuscript” in “The Free Biblical manuscripts: Fresh Studies of an American Treasure Trove” (pp. 227 – 249) who argues (on palaeographical grounds) for a 6th century date for the codex, and finishes by saying “the last item on my wish list is a state-of-the-art radiocarbon dating of the two parts of the Freer Gospels.” So apologies; I wasn’t remembering correctly on this one. Interestingly though, he does mention the Cologne Mani Codex too. 🙂
Regarding the use of AMS analysis: absolutely! There’s more than enough manuscripts with a large amount of “extra” blank portions of a folio which could be radiocarbon dated, and give us more than a few 10s of dateable manuscripts to compare the non-dateable manuscripts to (Grant Edwards Database of dateable Greek book-hands only has 183 examples https://tinyurl.com/airtabledgbh – there’s a lot more papyri than that!).
Pingback: Radiocarbon Analysis of Papyrus and Parchment Manuscripts: A List | Variant Readings