Classical Romance and Academia's Greatest Flaw
Recently, I have here written against the grain, arguing that Apuleius cannot be ruled out of consideration as the author of the original Eselroman, or “Ass-Story” as told in his Metamorphoses, or “Golden Ass”. Having published that article, I soon turned my attention to other issues of the Roman Novel as I tied up some loose ends and open questions which remained for my part having spent so much time thinking about Apuleius’ work. In particular, I had briefly touched on Apuleius’ influences, from Varro and others, and on my mind was the fact that Petronius’ Satyricon, as far as antique novels go, was quite early. The only other work which may have been earlier (and in fact I would be perfectly happy considering to have preceded Satyricon1 by maybe 5 or so years).
I have also in the more distant past published a piece with the following conclusionː
This neurosis of classical philology can be found in many different places […]. For every Latin poem, there is a scholar (or two) waiting to tell you what lost Greek poem was its model, and what miserly scrap of Greek papyrus proves this. Granted, there has been a degree of genius invested in these exercises, and where genius is applied, success often follows. But one must recognize that there has been a degree of mass romance (some would say hysteria) in some of these claims, and especially in the thinly-veiled and widely-held belief of Greek supremacy over Latin (take for instance the prejudices of Basil Gildersleeve, founder of the American Philological Society, on ethnonationalist grounds, as described in Habinek 1998), for reasons which are hardly philological or even vaguely scientific.
It was on the issue of Petronius and his influence where I found what must be filed among the absolute worst of the above-described sins. In a paper published in 1977 entitled Petronius, P. Oxy. 3010, and Menippean Satire, Raymond Astbury concludes the following about a certain scrap of Papyrus of fifty partial lines (my emphasis and translation of Knoche’s German):
The arguments presented so far, if accepted, leave us at last with the use of prosimetrum as the only remaining link between Petronius and Menippean satire, and many scholars in the past have reluctantly accepted this matter of external form as the extent of Petronius' debt to Varro and Seneca. Some, it is true, have attempted to break the link by suggesting that other literary influences inspired Petronius' use of prosimetrum, such as the Milesian tale and the mime. However, the evidence for the mixture of prose and verse in these forms is slight, and suggestions based on it have not carried conviction. Ulrich Knoche, with his usual percipience, commented: 'Der Form nach kann man und muss man wahrscheinlich Petrons Roman mit der Menippeischen Satire verbinden, obwohl auch im hellenistischen Roman wahrscheinlich schon hie und da die Erzahlung auch ausserhalb der Reden in Verse iibergehen konnte. Wir sehen das z.B. im Alexander- roman, in der Historia Apollonii, auch ein Vorlaufer Charitons hat es vielleicht so gehalten.' [In terms of form, one can and probably must connect Petronius's novel with the Menippean satire, although even in the Hellenistic novel, the narrative could probably already occasionally transition into verse outside of the speeches. We see this, for example, in the Alexander Romance, in the Historia Apollonii; a forerunner of Chariton may also have done so.] The new papyrus enormously strengthens the case for believing that prosimetrum was not alien to the Greek romance. It is of the second century after Christ, so it is necessary to hypothesize from it the use of prosimetrum in romances earlier than Petronius. This does not appear to be an insuperable obstacle, especially if we recall that. when Richard Heinze first suggested the romance as an influence on Petronius, he was unaware of the evidence for the existence of the romance before Petronius' time: it was only when his article was in the proof stage that he learned of the discovery of the Ninos romance which supported his hypothesis. If we are prepared to accept that prosimetrum was probably found in Greek romances before Petronius, we are then faced with two possible sources for the use of the form by Petronius — the romance and Menippean satire. I suggest that, at the very least. it is more economical to find the source in the romance, which, it is clear, influenced Petronius in other respects, than in Menippean satire, with which he has nothing in common other than prosimetrum.
It may seem unfair for me to respond to this without itemizing Astbury’s arguments which “if accepted” would lead to this appalling conclusion. But I would like to argue that there are no arguments which could ever lead to such a conclusion, because the question itself is ridiculous.
First, let’s address the assumptions he wishes us to make. Most comically, we are to assume, on the basis of this papyrus, which postdates (we are told) Petronius by some hundred years, that prosimetrum (that is, prose interpunctuated by verse), was not alien to Greek Romance at the time when Petronius conceived his Satyricon. Greek Romance, it must be said, was only coming into being in the 1st century AD, while Petronius was executed in 66. Given the apparent extent of Satyricon prior to its demolishment by time, it surely could not have been conceived of any later than 64 or indeed earlier, depending on how much industriousness we are willing to grant Petronius. The Ninus Romance, also only attested by virtue of a group of bemaggotted papyri (if indeed it was a ‘romance’ at all — there is little reason it could not have been a rather florid proto-romantic history), is usually dated to about the turn of the millennium.2 Are we to believe that just as soon as Romance as a form was being breathed life, prosimetrum was already being experimented with thereon?
Astbury then adduces this example wherein Heinze, apparently, prior to the publication of his article, was validated by the discovery of this Ninus romance and its papyrological dating, which, as we will discuss below, can hardly be said to be secure. The difference between Heinze and Astbury is that Heinze only published his article after this papyrological discovery was made public: Astbury asks us to suspend our disbelief in the blind expectation that a prosimetric romance dateable to a period earlier than Petronius, based upon Heinze’s ‘precedent’. This of course is absurd, and, to date, nearly 50 years thence, no such papyrus has been found.
The second assumption we are to make, which is the most tragic, is an assumption which it would seem even Astbury did not conceive of as an assumption, for it is the fundamental problem with the question itself: that Petronius had a singular model, or, that it is ‘most economical’ to assume he had a singular model.
It is true that genre could be quite limiting in antiquity. But if we are already assuming a rapid, unseen development in what was still a “brand-new” genre, that is, prosimetrum, then we can assume just as easily (if not much easier), that what Petronius was doing transcended the rigidity of genre altogether.
We don’t really know much about Varronian satire. We know that it was prosimetric, and we know that they were satirical. We do not know if they had anything to do with Romance. This is another assumption we are being asked to make, that there was nothing in Varro which could have the essential quality of a Romance, that is, two lovers in idyllic bliss, suddenly separated or foiled and in need of reunion. I am preparing many thoughts on Varro presently, but I can say confidently that, as with all fragmentary literature, there are cases to be made in Varro containing everything or nothing.
Returning the the problem with the question itself, we must face a very inconvenient truth: classicists are hideously poor at understanding probabilities. With all the assumptions we are being asked to make, wherewithout this conclusion would disintegrate, it is not in fact at all economical to see Petronius’ as having a singular model in hypothetical antepetronian prosimetric Greek romance. Physicists only have the audacity to propose one kind of dark matter; classicists have the shamelessness to propose a new species of dark matter for every god-forsaken text.
To indict all of classics on account of a single article would seem unfair, were it not the case that this very article was republished as recently as 1999 as an example of the state-of-the-science with respect to Roman Novel in Harrison’s “Oxford Readings in the Roman Novel”. And, as we shall see below in another example of unscience in classics, even the basis upon which this article stands is questionable.
Papyrology, the study of papyrus fragments, in particular of those found fortunately preserved in the aridity of Egypt, and weighs heavily on the dating of Hellenistic romantic material, since we know virtually nothing about the authors of such works (whereas we know much about the authors of Latin Romance). Indeed, most romances, it would appear, have been entirely lost in the paucity of hands available to copy texts.
One fantastic thing about papyrus, when preserved, is that it gives us a sort of insight into the classical world which we typically would only associate with the Classical Near East and its non-biodegradable clay tablets and stone inscriptions. Papyrus fragments have preserved numerous exemplars of various administrative documents from the Roman period, for instance, without which we would be blind to many details of the operation of the Roman state and its various local governments. Papyrus has also preserved numerous literary works.
In turn, one fantastic thing about administrative documents, is that they are very often precisely dateable. The governor or consuls or some other circumstances are recorded along with the document, and can be correlated with other records, for example histories or, perhaps better yet, epigraphy. This can be useful for dating literature, as occasionally we will find literary papyri which were later repurposed for administrative matters. In such a case, it may be relatively safely supposed that the literature preceded the record, giving us a terminus ante quem, a latest possible date, for that text. While it should not at all be deemed impossible that the reverse happened in cases where the flip side of a papyrus was reused (in such cases it is impossible to scientifically surmise which was first), this would seem more or less acceptable to me as admissible dating evidence3, barring any other material complications.
This said, there is one dark art of papyrology which I would suggest to be much further from unimpeachability: paleographical dating.
Paleographical dating is the dating of a particular manuscript of a text on the basis of paleography, that is, the study of handwriting. It has its more secure roots in the study of medieval manuscripts, where the practice of writing was very often only learned by the scribe at the monastery in which he lived. As such, consistent traditions and manners and trends of writing are identifiable. Further, the comparative age of a Medieval manuscript shunted away in some library or monastery shows rather more clearly than the ages of any two scraps of papyri which have been laying in the dirt together for 1-2 millennia.
But as concerns papyri, I would suggest that paleography cannot be so reliable an assistant whatsoever. Personally, my handwriting changes day-to-day. Some days, I write with my “natural” hand, that is, the chickenscratch I picked up (much to the chagrin of my teachers) as a young student; other days, I use a much more affected handwriting which follows a personally adapted ductus of Carolingian manuscripts. Sometimes I mix the two. Sometimes I use a capital script in imitation of my father, hoping to write most clearly. Once I start writing a document or a note with a certain ductus, I rarely switch.
Furthermore, the conditions in which an administrator or a “scribe” (whatever that was: a classical copyist could be a slave, freeman, schoolteacher, student or any combination of the above) learned how to write, or set about stylizing their hand, is more or less completely opaque to us. Individuals may have learned at school, or in some sort of slave-training bootcamp, or indeed even were self-taught by mimicking things they saw, such as epigraphy or documents. Literacy in the classical period, arguably especially in Roman Egypt, was (probably) a much more dizzying spectral expanse than was the case in the Medieval period.
Regardless, paleography has been used, and is being used to this very day, to date papyri not to the span of a century or two (as would probably be the limit of confidence of a medieval paleographer) not even to the span of 50 years. Quite to the contrary, it has become rather common to see papyri dated to a span of just 25 years4 or less.
This is absurd and unscientific. I don’t believe there is any room for equivocation here: it is common quackery. To confidently date a document to such a minute span of time is hardly more scientific than palmistry or tarot. Worse, this is being done through a pseudoscientific slight of hand in which a literary papyrus will be dated by comparison to an administrative exemplar. Is it not inevitable that a person or persons copying a literary text would use quite a different (and probably archaizing) hand from that of an administrator? If a manuscript is being produced at the behest of some highly literate patron, is it not inevitable that such a manuscript be expected to look a little different from some receipt or horrible legal document?5
Lengthy tracts have been composed to try and make some science out of all of this, but there is no real reliable science that can be found here. People are unreliable — especially outside the confines of a monastery. Their hands are even more unreliable, and adminstrative documents are incomparable with literary. Even if we may make some very broad generalizations and periodizations of handwritings, based upon some vague correlations which may be available to our meager sight, I very much doubt that the chaos of a place like Roman Egypt renders these correlations any more than factoidal. The world is far too random for all of this, especially a world we very poorly understand.
Worst of all of this is that, clearly, this unscience is already doing damage. Take for instance this barely coherent line of reasoning for some inscrutable reason, as of today, still on the Wikipedia page6 for Leucippe and Clitophon:
“…if this date is accepted, …”. No. I do not accept this implausibly precise date. Now shut up.
I have examined the original dates provided by Parsons, Cavallo, and Del Corso. Del Corso has already appeared in our footnotes as an example of the very worst of the excesses of papyrology. In all cases, this date is determined purely by paleography, purely by comparison with administrative documents.7 In this case, whoever wrote this along with, apparently, Cavallo and Del Corso, is willing to dismiss out of hand the possibility that an episode in the Romance is historical allusion purely on the authority of papyrologists. The editor here, who is quite obviously familiar with the literature, does not even for a moment question the paleographical dates in light of the possibility of historical allusion, arguably a more secure evidence. Given the author’s ‘erudition’, it would seem, horrifyingly, that surely this is a graduate student of some sort, or worse, a professor, who has rambled on so. Shamefully, they have also violated many stylistic standards of Wikipedia in doing so.
It is astounding that papyrologists feel apparently confident enough to date to within a quarter-century but not even medieval paleographers would have this much confidence despite much better understood circumstances of composition. In Assyriology, one would never find arguments for dating a manuscript on the grounds of the "proportionality of the 2nd and 3rd wedges of the 'a' sign", while adductions of this sort form the basis of classical paleography.
Once again, Classicists are not good at statistics. Nowhere in Cavallo or Del Corso have I yet found an even pretended statistical analysis of a papyrus whereby the relative ‘prominence’ of various features is indexed in relation to other securely dated texts. We are only left to trust that they know their papyri so extremely, divinely well that they almost can tell us when a papyrus was written, by whom, and whether he wiped his ass and/or picked his nose that day. There is hardly a pretense of science in any of this, but where there is, it’s by comparison with some other text. We’re never given any sort of analysis on how likely it is that a 1st century text would look like a 2nd. A true scientist, without knowing a thing about papyrology, would correctly be able to tell you that the probability of such is not zero; a papyrologist won’t even entertain the idea. Certainly, if they would, they don’t write as if it’s a real possibility. Their determinations are proclaimed as fact. As is the case all too often elsewhere in classics, they do not stop even for a second to curb the inevitable gullibility of the student who might encounter them.
Indeed, this problem appears everywhere in classics. Textual criticism in particular has been prone to a pseudoscientism which only in recent years has been at all recognized within academia. Nonetheless, in the blossoming field of papyrology, the last place where there is anything new to be read, this mildew would seem to be growing ever more, completely compromising the literary critical field by corrupting every last date with false evidence and astrological prognostications. As we have seen, with Romance in particular, consequently, we can’t make heads or tails of the dating evidence. Hubris and vanity, supported, funded and reinforced by a blind and misguided hope that we can know what is very likely unknowable, have distorted and corrupted any and all open questions of literary dating, influence and connection.
There can be no science, no reasoning, no truth found in this. If we are to make any suppositions of this sort with any sort of hope of being right, this cancerous quackery must be excised and treated expeditiously with the most aggressive regimens available. Radiation, chemotherapy, amputation: without these, classics stands to to die sickly by a pseudoscientific malignancy of its own making.
Persius, who probably died in 62 (four years before the death of Petronius) according to a biography of his packaged along with Suetonius’ De Poetis, would appear to have referenced the work in his first satire — albeit his satires were published after his death according to the very same biography, so this is not entirely secure.
eg. López Martínez, María Paz. "The Ninus Romance: New Textual and Contextual Studies" Archiv für Papyrusforschung und verwandte Gebiete 65, no. 1 (2019): 20-44. The basis for a date earlier than the 1st c. AD is made entirely on unscientific palaeographical and stylistic grounds, by the same people, Cavallo and the quack Del Corso, discussed here. The material evidence itself only supplies a terminus ante quem for the 1st c. AD. This is all to say, there is no material evidence whatsoever of a romantic tradition in the millennium before Christ.
It must be said that the archaeological contexts in which papyri are found are rarely secure in such a way that would be very helpful for dating. The source of nearly all (if not all?) of our Oxyrrhynchus papyri is a trash heap which contains some 600 years of fragments. Many of these papyri were also collected without their findspot recorded, and only later examined in any meaningful way.
See for instance Del Corso, Lucio. "I Papiri del Romanzo Antico." In Atti del Convegno Internazionale di Studi, Firenze, 11-12 Giugno 2009, edited by Guido Bastianini and Angelo Casanova. Firenze: Istituto Papirologico, 2010.
Granted, some paleographers account for this, eg. Cavallo speaks of a mano libraria, a ‘book hand’. But it doesn’t seem to do him much good. Administrative documents remain the ground truth for dating in his work, with no doubt or very little doubt indeed being cast on a date on account of a ‘book hand’, a concept poorly defined in any scientific sense.
an archived version exists here: https://en.wikipedia.org/w/index.php?title=Leucippe_and_Clitophon&oldid=1273738834#Date
Parsons and Cavallo do not provide much reasoning, just vibes, while Del Corso summons an administrative document which he believes is almost an exact match — but he doesn’t suppose that the same hand wrote both: so what’s the point?