The Greek version of the Myth of the Sun s Eye, P.Lond. Lit. 192 (MP³ 2618; LDAB 5054), has often been the topic of discussion by scholars in both the fields of classical and demotic studies¹. However, perhaps because of its idiosyncratic nature, the text has hardly undergone a full systematic treatment, and much is still to be achieved: the current paper intends to add some new, unpublished information about the text, and discuss select issues that have emerged over the preliminary phases of the preparation of a new edition².

The papyrus itself, although known for more than a century, has never received a complete description as an artifact. It will be useful to trace quickly its known modern history. It was first mentioned in 1894 in the Catalogue of Additions to the Manuscripts in the British Museum, where it received the inventory number 274, being listed among the purchases of the year 1893 (together with papyri 229 to 458 inclusive) and described as an unidentified romance³. It is mentioned again in almost the same terms by Frederic Kenyon in 1898, in P.Lond. II, with its number 274. The first announcement of its identification as a Greek version of the demotic Myth of the Sun’s Eye by Richard Reitzenstein precedes by some years its editio princeps, which appeared in 1923⁴. The editor, however, did not work on the original papyrus, but on a transcript prepared by Wilhelm Crönert, and thus devoted little space to the description of the manuscript. In 1927, the papyrus was then included by Herbert Milne in P.Lond. Lit. under the number 192, with some improved readings (but again no full description) and a photograph of its main fragment⁵.

Only in 1969, with Stephanie West’s re-edition, did the text receive a satisfactory treatment, following a re-examination of the papyrus which also allowed some new joins to be made, although West’s focus is on the text itself, and not on the manuscript⁶. Several other papers have appeared about P.Lond. Lit. 192, especially over the last decade, but these are mainly speculative and disregard the importance of the direct analysis of the papyrological evidence. A useful addition that appeared in 2005 is an article that publishes for the first time photographs of all the fragments of the papyrus⁷.

I present here the results of a new inspection of P.Lond. Lit. 192, focusing specifically on the data not provided in earlier studies. The manuscript is now preserved in the British Library, as are all Greek papyri formerly deposited in the British Museum⁸.

The maximum measures of the seven fragments are as follows (w. x h. in cm): fr. A 12,3 x 19; fr. B 4,5 x 14,6; fr. C 6,2 x 11; fr. D 16,1 x 34; fr. E 14 x 9,3; fr. F 11,8 x 9,5; fr. G 6,5 x 5,5. The height of an entire column is observed in fr. D: col. II, 27,7 cm high, has 69 lines, col. I probably had between 71 and 69. Line length is 6,5 to 7,5 cm in average. The intercolumnium measures 1 to 1,5 cm. The large surviving top (at least 2,7 cm) and bottom (3,5 cm) margins are one of the signs of the care with which a professional scribe produced this quality roll. The text is written along the fibres. No traces of kolleseis could be found. The verso is blank. However, the main fragment, D, has four very small fragments bearing exiguous remains of writing attached to its back (three around the edge of the lower part of col. I, one on the edge of the upper part of the same column). These unrelated fragments must have remained stuck to P.Lond. Lit. 192 since the time it was unearthed: two of them are in Greek cursive writing, two in demotic⁹.

The dating of P.Lond. Lit. 192 seems to be not as unambiguous as most modern commentators have believed¹⁰. When Milne included the text in P.Lond. Lit., he dated it in absolute terms to the third century AD, a date that has since been repeated in almost all of the relevant studies¹¹. Nevertheless, Kenyon, the first to analyse the papyrus, always dated it to the second century, a date which, in spite of Milne’s statement, was still accepted by a few later commentators¹².

Despite the difficulty of dating a Greek literary hand to a precise century, the very small, neat, informal uncial hand of the manuscript shows relevant similarities with others dated to the second century, such as e.g. P.Oxy. VI 853, P.Oxy. II 221, P.Oxy. XXXI 2536 (= Turner/Parsons, GMAW², no. 61; first hand). I am therefore inclined to suggest a dating to the second half of the second century, or even, to be cautious, from the second to the mid-third century¹³.

The provenance of the papyrus is utterly unknown according to the published records. An inspection of archival records in the British Library has however revealed some interesting information. P.Lond. Lit. 192 was purchased on 11^th April, 1893, by the Department of Manuscripts from Theodor Graf¹⁴. It came to London along with other papyri, mainly documentary texts, that Graf sold to the British Museum in two lots, respectively P.Brit. Lib. inv. 256-271 (purchased on 11^th March, 1893) and 272-400 (11^th April), all of which can be found in P.Lond. II (with the same numeration)¹⁵.

The sale of part of one of Graf’s collections to the British Museum is an interesting episode from the time of the «gold rush» of papyrology. Due to recent disputes with Josef Karabacek and the unsatisfying sum that he had offered him for his last collection, Graf started negotiations with the British Museum, in late 1891 or early 1892¹⁶. As emerges from unpublished correspondence in the British Library archive, Kenyon, then a young assistant in the Department of Manuscripts, was sent to Vienna, where he stayed from 12^th-23^rd September, 1892, surveying Graf’s collection¹⁷. Graf’s intention was to sell his lot en masse, but his attempts to put pressure on the British Museum authorities failed, and he finally agreed to sell them only selected parts of the collection. Later, the remainder ended up in Karabacek’s hands (as had happened with the previous collections assembled by Graf), and is now in Vienna.

Among the items selected for the British Museum were the fragments of P.Lond. Lit. 192. A letter containing an overview of Graf’s collection was sent by Kenyon to Edward Scott, Keeper of Manuscripts, at the end of his first day at Graf’s, and it includes what I believe must be the first mention of our text: «Several fairly large fragments of a prose literary work, apparently consisting of fables; in a small square plain uncial hand.»¹⁸

Two main observations can be drawn from knowing that the papyrus came from Graf’s collection. First, it is possible to make a conjecture about the findspot of P.Lond. Lit. 192. The collection split between Vienna and London is known as that of the «III. Papyrusfund», a collection that Graf had put together in Cairo and shipped to Vienna in 1891¹⁹. As was the case with Graf’s previous collections, the papyri had been bought from antiquities dealers, and had been excavated by sebakhîn, mostly in the Fayum. The Vienna/London lot in particular has often been said to have come from Soknopaiou Nesos. However, although several papyri of this group undoubtedly come from this town, it has been correctly argued that parts of this lot do not originate from Soknopaiou Nesos, but from other sites of the Fayum²⁰. Kenyon himself, the year after the purchase of the papyri from Graf, was cautious enough to indicate the whole Fayum as a reported provenance for one of the literary items²¹.

It is noteworthy that the documentary texts that came to London along with P.Lond. Lit. 192 originate from the Fayum, when they can be certainly provenanced on the basis of internal elements: many of them do come from Soknopaiou Nesos, but several others are from different centres in the Fayum²². We cannot have any certainty about the provenance of P.Lond. Lit. 192, as any papyrus from virtually anywhere in Egypt could have been added to the lot assembled on the antiquities market by Graf. Nevertheless, on the basis of the above information, I think it is possible to consider the Fayum as a likely area where our text could have been unearthed in the very late 1880s or very early 1890s.

It is also worthwhile mentioning at this point that the Fayum was clearly one of the main centres, if not the main one, for the preservation of Egyptian culture in the indigenous demotic language in Roman times, as the papyrological evidence from the second and third centuries AD in particular shows. Though we must allow for the distorting effect of the chance of papyrological survival, the importance of the temple libraries of the Fayum is undeniable. Fragments of a relatively high number of demotic copies of the Myth of the Sun’s Eye survive, proving the popularity that this text enjoyed in the Roman age; and most of them come from the Fayum (see below).

A second observation resulting from the acquisition history of our text is based on a comparison with the other literary papyri that arrived in London with it, and concerns the possibility that additional fragments of our text might still be found in Vienna.

As already said, the majority of the texts in the British Museum’s lot is documentary. Out of the whole purchase (P.Brit. Lib. inv. 256-400; 450-464; 2037a-f; Add. MS. 34473 art. 1-7), only twelve items, i.e. less than 8 %, are literary. These are: P.Brit. Lib. inv. 256; 271; 272; 273 (and 273b); 274; 275; 455; 459; 462; 464; 2037a-f; Add. MS. 34473 art. 1-7.

To further reduce the number of manuscripts to be dealt with, only the literary texts of non-Christian content will be taken here into consideration, i.e.: P.Brit. Lib. inv. 256; 271; 272; 273 (and 273b); 274; 275; 2037c; Add. MS. 34473 art. 1-6. Of these few items, a significant number has matching fragments in the Rainer collection in Vienna: MP³ 1039, 0115, 2531, and 0293²³. It appears that, in several cases, fragments from the same manuscripts were divided into different lots, and some were shipped to London, whilst others remained with Graf, and eventually entered the Rainer collection.

It would not be naive to hope that, especially in the case of such a fragmentary text as P.Lond. Lit. 192, further fragments could be identified in the collection of the Österreichische Nationalbibliothek. Searching for one or a few fragments of a specific text amongst all unpublished items of this immense collection would be utopia. But a look at the inventory numbers of the matching fragments listed in n. 23 clearly shows that in three instances the Vienna items are in spans between P.Vindob. inv. G 29292 and 29834, and in one between 26746 and 26760. Within and about these intervals, a survey is currently under way in the search for possible additional fragments to P.Lond. Lit. 192²⁴.

As mentioned at the beginning of this paper, several contributions have appeared in recent years concerning the Greek version of the Myth of the Sun’s Eye, but most show a tendency toward a certain redundancy, in the repetition of a hermeneutic vulgate that seems to have prevailed about the topic. This is specifically the case where scholars with no working knowledge of the Egyptian language draw conclusions on the basis of an ill-founded comparison of the demotic and Greek texts²⁵. Focusing on this issue, a couple of points will be discussed briefly in the remainder of this paper.

In comparing the demotic text and the Greek version where possible, it has generally been suggested that, when the two texts diverge significantly, this must be due to a choice of the translator, who adapted the text to an alleged Greek literary – if not religious – sensibility, further assuming that the demotic copy he was working on was almost the same as that we have in the Leiden papyrus²⁶. There are two main faults with this view.

The first concerns the nature of ancient Egyptian textual tradition. Until now, those who dealt with the topic of our paper had to rely almost completely on the demotic text of P.Leid. Dem. inv. I 384, as the published fragments of other demotic manuscripts bear little (P.Tebt. Tait 8) or no material (P. LilleDem. inv. 31) for comparison with the Greek version. Yet this should not lead one to forget that the Myth of the Sun’s Eye was a very popular text, preserved in numerous copies²⁷. And different copies of the same demotic text can bear major differences: this is a distinctive feature of the textual transmission that can characterise ancient Egyptian – and not only demotic – literature, which is often an open transmission, where textual contamination plays a major role, much more than in its Greek and Latin counterparts²⁸. The Leiden text itself incorporates several alternative readings, which is again customary for demotic literature: these are clearly introduced by the expression ky d‘m «another manuscript» (i.e. «textual variant»), a relevant instance of which can be seen in P.Leid. Dem. inv. I 384 col. XXI 2: here, this suggests that the whole section mentioning the town of Nekheb (and the goddess’ connected hypostasis as a vulture) was probably not found in all manuscripts.

The as of now identified (fragmentary) copies of the demotic text will shed much light on the problem of the textual transmission of this text. An international project aiming at a full edition of all of them (and a re-edition of those already published) started in early 2010, based at the Papyrus Carlsberg Collection in Copenhagen. A total of at least eight demotic manuscripts will undergo a full study in the context of this major project²⁹.

The second weak point of the common view on the nature of the translation concerns the alleged reshaping of some passages by the translator. I am not suggesting that the translator did not at all influence the Greek text he was composing, which would be impossible, especially in the context of Hellenistic and Roman antiquity, where, in most cases, it is more appropriate to talk of version than of translation for literary texts. Nevertheless, it seems excessive to talk of a deliberate intention to remove all the most local elements of Egyptian religion and culture from the Greek version, if not even of censorship. By way of example, it has been often stated that in the Greek version Tefnut is presented simply as ἡ θεόc (e.g. fr. F col. II 62) since she was not a goddess included in the standard repertoire of the interpretatio Graeca. But how could one then explain the presence of such an uncommon name as that of the god Ἀρε[cνούφιοc] (genitive; fr. A col. I 58-59) to render the Egyptian god Shu, Tefnut’s brother and spouse? The reason can be found in a simple observation: in all the parallels preserved between the (Leiden) demotic and Greek versions, Tefnut is never mentioned by name, but in the demotic text too she is called t3 ntr. t «the goddess» (e.g. col. XX 3); only once in the Leiden papyrus is she Tfny.t «Tefnut» (col. XXII 2), at a point for which no counterpart survives in P.Lond. Lit. 192. Similarly, why should we suspect the translator of having suppressed the mention of the goddess’ feline hypostasis (P.Leid. Dem. inv. I 384 col. XV 24; P.Lond. Lit. 192 fr. D col. II 16-17), when her metamorphosis into a gazelle seems to be no problem for him, and is kept in the Greek version (P.Leid. Dem. inv. I 384 col. XXI 9 and 22; P.Lond. Lit. 192 fr. F col. II 58-59 and 64-65)?

The Myth of the Sun’s Eye was translated into Greek because it was considered a very important text, as clearly shown by the multiple surviving demotic copies. As such, and because of its primary nature as a religious text, it was needed as it was – and certainly not in any «censored» version – by some specific social group, not implausibly by the Egyptian priesthood itself³⁰. Therefore, there would have been no point in obscuring its most specific Egyptian features, something that is easily forgotten when one thinks of P.Lond. Lit. 192 as a version composed simply for literary entertainment, if not just as a novel³¹.

A last issue to be mentioned is that concerning the nature of the translation process that led to the text of P.Lond. Lit. 192. The idea of a possible twofold process, with a first rough translation by an Egyptian scribe and a refined literary version reworked by a learned individual educated in the Greek paideia, still appears in the literature on the subject³². It should be sufficient to note, in order to reject this theory, how many passages of the Greek closely reproduce the demotic.

But there is now a further, external argument for refuting the abovementioned view, a view that is substantially based on the prejudice that an indigenous Egyptian priest, the most likely repository of demotic literary tradition in Roman times, could not have mastered Greek writing at a sufficient level to produce a composition in an agreeable literary koine such as that of P.Lond. Lit. 192. The evidence from the demotic and Greek ostraca of Narmouthis, found in an Egyptian temple in the Fayum, and from the second/third centuries AD, proves in fact that priests in traditional Egyptian temples not only could write Greek, but were taught composition in that language, and could create complex texts and master unusual, selected vocabulary³³.

The Greek version of P.Lond. Lit. 192 is the product of such a mixed cultural milieu. One of the unpublished manuscripts of the demotic text, P.Carlsberg 600, nicely exemplifies this. As a peculiar material feature, it shows a dual pagination in both demotic and Greek: that is, its columns are numbered with demotic numerals in the upper margin, and with Greek letters in the lower, both writings used together to assist the scribe³⁴.