The famous papyrus edited in 2008 by Claudio Gallazzi, Bärbel Kramer and Salvatore Settis is called the Artemidorus papyrus because the first fourteen lines of its summarizing description of Spain offer the slightly amplified version of a fragment of this late Hellenistic geographer². Besides this, the papyrus contains a comparision between geography and philosophy³; it also contains a hitherto unidentified map, and drawings of human heads and limbs. The entire verso is filled with a series of drawings of animals (some real, others imaginary). No similar papyrus is known, and, starting with Luciano Canfora, its authenticity has been challenged. It has been regarded as a XIX^th-century fake made by the notorious Konstantinos Simonidis⁴.

My contribution summarizes the most frequent arguments for and against the authenticity of the papyrus. Owing to limited space, only one very decisive philological argument will be treated in greater detail. As a result, I cannot include a discussion about the important consequences of Giambattista d’Alessio’s convincing recomposition of the papyrus fragments, which inverts the sequence of the two geographical texts and reopens the question of the authorship for the comparison between geography and philosophy in columns i-iii⁵.

First of all, I will briefly mention three supposed proofs against authenticity which in one way or the other go beyond the competence of philology. The most striking and immediate proof of Simonidis’ forgery, at least according to Canfora, consists in the textual parallels which the art historian Maurizio Calvesi sees between the beginning of the papyrus text and the introduction of Karl Ritter’s Erdkunde in the French translation of 1835⁶. Canfora gave the following synopsis of the relevant parts of the two texts⁷:

Does this comparision prove that the writer of the papyrus text was inspired by Ritter? Let us look at the first lines of Ritter’s plain text⁸. «Dans l’introduction à un ouvrage qui a pour but de réunir en un corps intimement uni dans ses parties et plus scientifique les notions diverses sur la terre, il est indispensable, avant d’exposer le plan, la méthode et les sources, d’indiquer ce qui, dans cette science, a directement rapport à l’homme, car c’est ce point de vue qui donne à cette étude son importance et sa valeur. Ce point de vue, qui touche les rapports de la nature à l’histoire, de la partie au peuple, et surtout, de l’individu à l’univers, ne sera ici que légèrement indiqué, seulement pour attirer l’attention sur le but dernier de nos recherches. L’homme moral, pour accomplir sa fin, l’homme qui veut agir d’une manière efficace, doit avoir la conscience intime de ses forces, connaître ce qu’il reçoit du dehors, ce qui l’entoure, et les rapports qui l’unissent avec ce qui n’est pas lui: toute association d’hommes, tout peuple, pour ne pas manquer son but, doit connaître aussi ses forces intérieures et extérieures, celles de ses voisins, et la place qu’il occupe au milieu des rapports qui agissent sur lui du dehors. Des efforts aveugles, une volonté instinctive, ne peuvent donner à l’homme, malgré toute la tension et l’énergie possibles, cette puissance qui fait être, qui fait agir; ce sont des efforts plus réfléchis, plus sentis, c’est l’accord de la volonté avec la force, qui là où la clarté s’unit à la vérité, se manifestent en actes éclatants, sublimes et éternels.»

Now let us compare this with the first eight lines of the plain papyrus text as constituted by Canfora⁹: τὸν ἐπιβαλλόμενον γεωγραφ[ίᾳ] | τῆc ὅληс ἐπιсτήμηс ἐπίδε[ιξιν] | ποιεῖсθαι ἑαυτοῦ δεῖ πρὸ [ταλ]αν|τεύсαντα τὴν ψυχὴ[ν εἰc ταύ]|την τὴν πραγματείαν [νι]|κητικωτέρᾳ τῇ θεληсε[ι κατὰ | τ’ ἐπα]γγελίαν ταύτην κατ[ά τε τὴν] | τῆc ἀρετῆc δύναμιν [ποιεῖν]. I do not see the immediate evidential value which Canfora attributes to the linguistic analysis given by the art historian¹⁰.

The second proof proceeds the other way round. Richard Janko believed that a head of Matthew incised in the frontispice of a book published by Simonidis in 1861 «is remarkably similar to a head of the Artemidorus papyrus (…) except for the latter’s heavy browridges»¹¹. These heavy brow-ridges are explained by Janko as an idealized self-portrait of Simonidis, quoting the description which Simonidis gave of himself in a pseudonymous letter: «[i]mmense black whiskers, moustache and imperial; huge black eyebrows; an enormous mass of jet-black and glossy hair (…); deep-sunk but fiery and piercing eyes; dark swarthy visage; massive lips and strongly marked mouth make up a face not easily forgotten.»¹² In a second drawing, Janko believed that he could recognize Simonidis’ profile.

Richard Janko is a philologist. I gave these images to Uwe Westfehling, an art historian and specialist of XIX^th century drawing, who became famous for discovering fakes in German museums. He wrote to me that there are great differences, and that the common features are so conventional that the incision of Matthew cannot prove that Simonides drew the papyrus heads.

In 2009 a worrying accusation was made. Silio Bozzi, an Italian police officer, analysed the published photo which shows pieces of the Artemidorus papyrus still joined with other documentary papyri of the late first century AD. Bozzi claimed that this photo had been manipulated on the computer and that the writing of the Artemidorus papyrus had been later added to an existing convolute, clearly in order to defend the authenticity of the papyrus by criminal means¹³. First of all, I wonder why this police investigation, concerning a papyrus bought by an Italian institution for 2 750 000 euros, has not led to any legal proceeding¹⁴.

Moreover, I am not convinced by the logic of some of the conclusions contained in Bozzi’s publication. I contacted specialists at the German Bundeskriminalamt in Wiesbaden, who recommended Dr. Hans Baumann as the best specialist in this field¹⁵. Baumann’s analysis definitely rules out that the writing on this photo was added later or that there was any criminal manipulation. He also rejects the more recent accusations by the Neapolitan police officer Salvatore Granata¹⁶. Besides, Granata’s conclusions are incompatible with Bozzi’s theory. Baumann presented his results in a workshop at Cologne University on 20 September 2010. An English version is being prepared and will be published, together with a short introduction written by myself, in 2012¹⁷.

We should now turn to the very few facts which are agreed on by those who believe that the papyrus is a fake and those who consider it as genuine. Carbon-14 analysis has dated the papyrus material to the late first century BC or early first century AD¹⁸. Two further facts agreed on lead to opposing conclusions. On the one hand, the spectral analysis of ink revealed that its composition matches that of ancient non-metallic ink¹⁹. But admittedly, similar ink can be produced in modern times. On the other hand, Daniel Delattre has shown that the handwriting of the papyrus matches modern disegni of Herculaneum papyri, published from 1862 onwards. The forger Simonidis probably was aware of them. However, as Delattre remarks himself, the disegni reproduce the handwriting of real ancient papyri of the first century BC, not dissimilar to that of the Artemidorus papyrus²⁰.

Three general considerations make a modern fake very unlikely. First of all, it would have been very difficult to find such a large unwritten piece of papyrus²¹. Furthermore, the ink of the recto has left offset images on the verso, and there are also some traces of the verso on the recto. The regular distances between the original writing and drawings on one side and the offset images on the other show that this incident happened when the papyrus was rolled²². Would it have been possible for a modem forger to roll such fragile ancient papyrus material so tightly? Finally, is it just coincidence that not only the form of writing and the other documentary papyrus pieces contained in the convolute, but also the carbon-14 analysis point unanimously to the same period²³? No forger, not even Simonidis, could foresee this.

Canfora published several pictures of Konstantinos Simonidis’ uncontested fakes, which are kept in Liverpool, so that it is now easier to get an idea of this kind of forgery²⁴. As far as we know, Konstantinos Simonidis was born between 1820 and 1824. After the most astonishing vicissitudes, he must even have forged the announcement of his death in 1867, since we know about his later activities and an obituary was published in 1890. At that time Simonides would have been between 66 and 70 years old.

At any rate, he did not live up to the time when, starting in 1896, the French epigraphist Bernard Haussoullier discovered the inscriptions in the Didymaion near Miletus which revealed a rare method of indicating the number 1000 and its multiples. This hitherto unknown method employs the so-called sampi, which in the normal Ionian numeric system is used as a sign for 900, while 1000 and its multiples are normally rendered by the letters Α-Θ, preceded by some additional mark of varying shape. Now, in some of the new Didymaion inscriptions, the same sign was used with a further alphabetic modifier on top of it to indicate 1000 and its multiples. We know that Haussoullier spoke in 1897 about this discovery at a conference in Paris²⁵. But the first published explanation of these rare numerals was given by Bruno Keil only in 1907 in his appendix to Otto Rubensohn’s edition of the Elephantine papyri, who made use of a private communication by Haussoullier.

It is a surprising fact that the same method occurs in the passage of the Artemidorus papyrus which enumerates the distances between places on the Spanish coast and which had not been taken into consideration by Canfora before the editio princeps appeared²⁶. The editors of the papyrus pointed out that this use of the sign was unknown in Simonidis’ lifetime, so that he could not have forged the papyrus²⁷. Canfora and his pupil Giuseppe Carlucci disputed this²⁸; recently they announced that, on the basis of Haussoullier’s unpublished papers, they could refute some of my further remarks where I assert that Simonides did not live long enough to understand and to falsify this numeric system²⁹. I decline here to respond to their most recent criticism and prefer to await the publication of their announced investigations. These will in any case improve our knowledge of the course of the epigraphic discoveries at Didyma³⁰.

I prefer now to deal with the most important and fundamental controversy. Are the differences between the new papyrus text and Artemidorus fr. 21 Stiehle, indirectly transmitted by Constantine Porphyrogennetos, as Canfora believes, due to the work of a modern forger who reveals himself by using inappropriate modern conjectures and giving anachronistic geographical information³¹? Or does the new text rather offer a splendid confirmation of the results of modern philological work, and fill, in one case, the lacuna of a corrupt passage which no philologist had been able to heal, and certainly no forger could have emended?

Canfora tried to restore the medievally transmitted Artemidorus quotation without adopting three modern philological emendations and operations³². I have argued that in all three cases the medieval text is corrupt³³. Canfora replied in the same volume, but his response does not adress the most important problems³⁴:

1. the awkward word order of ἕωс τῶν κατὰ Γάδειρα τόπων ἐνδοτέρω in the medieval tradition of Artemidorus fr. 21³⁵;

2. the difficulty of supplying a suitable subject to καὶ (understood as «also») сυνωνύμωс Ἱβηρία τε καὶ Ἱсπανία καλεῖται³⁶. Canfora proposed «Iberia». But this would mean that «Iberia is also called synonymously Iberia and Spain»;

3. the impossible break in the syntax after διατείνουсα ἀπὸ τῶν πυρηναίων ὀρῶν ἅπαсα, which leaves διατείνουсα ἀπό without the necessary μέχρι³⁷.

For the moment I leave this to the judgement of the readers³⁸; let us consider the expression which for Canfora forms an insurmountable obstacle to attributing the papyrus text to Artemidorus more closely. According to P.Artemid. iv 11-14, the second of the two Roman provinces contains the lands up to Gadeira and τὰ κατὰ τὴν Λυсιτανίαν πάντα. The medieval quotation of Artemidorus displays instead μέχρι Γαδείρων καὶ Λουсιτανίαс.

Canfora has pointed to a historical inconsistency in the papyrus³⁹. In Artemidorus’ time, the late second century BC, the geographical notion of Lusitania went well beyond the power of the second Roman province. If col. iv 13-14 is translated with «and all the land in Lusitania», meaning «the whole of Lusitania», this would for Canfora be an anachronism. On the other hand, he points out that it would even be wrong to exclude Lusitania from the second province completely.

The territories south of the Tagus were at that time certainly under Roman control⁴⁰. In the Rovereto proceedings, Canfora gave a sarcastic review of the different attempts at interpretation of this expression⁴¹. At the same time he emphasized the historical accuracy of the medieval tradition of Artemidorus which just reads «up to Gadeira and Lusitania»⁴². But is the medieval Artemidorus excerpt as it is preserved in Constantine Porphyrogennetos historically exact? What does «up to Gadeira and Lusitania» mean? Is Lusitania included by μέχρι, like Gadeira? Or does Lusitania serve as an indication for the frontier and is excluded? And what about the regions south of the Tagus which – as Canfora stresses – were under Roman control? The medieval tradition offers the same historical problems as the papyrus⁴³; the only difference is that, in the case of the papyrus text, a solution can be found.

Let us first consider the meaning of κατά in P.Artemid. iv 13. Some lines later there is a mention of the «sea belonging to the Ocean» (P.Artemid. v 2: [τῷ] κατὰ τὸν Ὠκεανὸν πελάγει) which touches the North coast of Spain. Κατά obviously means that the πέλαγοс belongs to the Ocean. It is not as large as the Ocean but is part of it. Likewise, τὰ κατὰ τὴν Λυсιτανίαν can easily indicate territories which belong to Lusitania without having the same extension⁴⁴. They can be identified with the territories south of the Tagus which Canfora misses in the description offered by the papyrus.

I agree that this interpretation leaves two problems open which Canfora was right to point out. The first is the meaning of πάντα in this context. Why should the author stress that all the territory in Lusitania belongs to the second province? The second problem is that the province is described as encompassing the coast up to Gadeira and the territory belonging to Lusitania south of the Tagus, but its most important part, Baetica, is omitted.

So we should adopt Benedetto Bravo’s proposal to read τῶν κατὰ τὴν Λυειτανίαν instead of τὰ κατὰ τὴν Λυсιτανίαν⁴⁵. This easy emendation solves both the problem of πάντα and of the missing mention of Baetica. Πάντα would now refer to the remaining τά in P.Artemid. iv 12, meaning that «all the land up to Gadeira and up to the territories in Lusitania» belongs to the second region. The emended expression τῶν κατά τὴν Λυсιτανίαν depends on μέχρι and indicates the outermost extension of the province to the North: there is no need to mention Baetica in this context.

By these words, the author, whom we can confidently call Artemidorus, reflects the gradual seizure of Lusitanian territory by the Romans in his time. When the Artemidorus excerpt of the medieval tradition was made, this fine differentiation was no more understood, so that it was abridged into an anachronistic μέχρι Γαδείρων καὶ Λουсιτανίαс⁴⁶.

I hope to have given some insight into the Artemidorus question and to have shown that, in spite of the sometimes harsh and disrespectful tone of the debate, the ongoing discussion has led to some definite progress. I thank all those who have contributed to this, regardless of their opinion about the papyrus, with their various observations and proposals.