The emperor’s new clothes

Everywhere people are falling over themselves to justify new and exciting uses for AI technology. Many of these are obviously trivial and contrived. Fake video abounds, students employ “content generators” to write their essays for them, and unscrupulous publishers create false scientific publications. Meanwhile, acres of creative writing are silently mined as raw material to train robots to emulate humans. All of this places a strain on existing data centres and boosts carbon emissions at a time when we are supposed to be conserving energy.

Of course, there are undoubtedly good uses of AI technology in the analysis of large scientific data sets, in the detection of patterns and potential correlations, and in the prediction of possible future data trends.

None of this, it seems to me, applies to the Arts and Humanities. Here, research is necessarily driven by human ingenuity.

There are no large scientific data sets that defy human comprehension. No raw statistics whose “crunching” requires machine assistance. Just, it seems, a fair number of researchers anxious to jump on a band wagon and draw upon whatever funding might briefly be available.

Earlier in the year, I became aware of a massively multi-authored paper (published in the journal Nature Vol. 645) entitled “Contextualizing ancient texts with generative neural networks”. I was immediately repelled by its jargon-heavy content, the kind of writing that I have usually found to be designed to exclude all but those with a vested interest. (I commented on this phenomenon back in June.)

Nevertheless, I was intrigued by the charm-offensive mounted on this paper’s behalf by Professor Dame Mary Beard, who pronounced that “it promises to be transformative!” And so, not one to admit defeat easily, I decided to have another go.

Robotic interpretations

The purpose of the paper is to announce a new computerized tool named Aeneas, a “multimodal generative neural network for contextualizing Latin inscriptions”. According to the authors, it is designed to solve the following epigraphic problems:
“letters, words or entire sections of an inscription may be lost over time, with the full extent of the missing text often being unknown. High levels of human mobility, absence of explicit dates and the frequent use of abbreviations, a hallmark of the Latin epigraphic habit, further complicate our interpretation of these inscribed artefacts”.

The writers are not particularly forthcoming about Aeneas’ method, but they cite an altar from Mainz (CIL XIII, 6667) as a successful example of its use. This I find confusing, since the altar (as can be seen here) is entirely undamaged and does not seem to suffer from any of the aforementioned “problems”.

It reads:
Deab(us) Aufan(is) / et Tutelae loci / pro salute et in/col(u)mitate sua / suorumq(ue) om/nium L(ucius) Maiori/us Cogitatus b(ene)f(iciarius) / co(n)s(ularis) vot(um) sol(vit) l(ibens) l(aetus) m(erito) / Idibus Iuli(i)s / Gentiano et / Basso co(n)s(ulibus)

“To the Aufanian goddesses and the guardian of this place, for the health and safety of himself and all his family, Lucius Majorius Cogitatus, beneficiarius of the legate, gladly, willingly, and deservedly paid his vow on the Ides of July when Gentianus and Bassus were consuls” (i.e. 15 July AD 211).

The writers correctly report that the altar is a well-attested type of military inscription, but more curiously they celebrate the fact that “Aeneas’ predictions successfully situate the inscription within this broader epigraphic habit, with a dating estimate (214 CE, within the expected range) and geographical attribution (correctly placing the stone in Germania Superior, with related alternatives in Germania Inferior and Pannonia) that align well with historical expectations”.

The writers divulge none of the actual “predictions” or “historical expectations” to which they obliquely allude. But later in the paper, they again emphasize that “Aeneas’ dating average for this altar is 214 CE, which is well within the 10-year range the model is trained on, and its top-3 geographical attributions are Germania superior (correct), Germania inferior and Pannonia superior”.

Remember, the text of the altar itself explicitly dates it to AD 211. There should be no confusion about this. Am I missing something here? To my mind, Aeneas has done significantly worse than any student who might simply search for this inscription in (for example) the online Heidelberg Epigraphic Database, which gives the correct date and provenance.

The writers praise Aeneas for “the historically enriched depth of the parallels it retrieves”. In the case of the Mainz altar, “even more striking”, claim the writers, “is Aeneas’ top parallel identification for this text: another votive altar, dedicated in 197 CE by beneficiarius Iulius Bellator and found near the same location in Mainz”. (They are alluding to AE 2017, 1072.)

It reads:
Deabus Aufan/is pro salute / et incolumitate / sua suorumq(ue) / omnium Iul(ius) / Bellator be(neficiarius) / co(n)s(ularis) libens posuit / Idibus Iuli(i)s / Laterano et / Rufino co(n)s(ulibus)

“To the Aufanian goddesses, for the health and safety of himself and all his family, Julius Bellator, beneficiarius of the legate, gladly set (this) up on the Ides of July when Lateranus and Rufinus were consuls” (i.e. 15 July AD 197).

Again, Aeneas’ modus operandi is not divulged, but I might guess that, like any moderately skilled epigraphy student, it searched for other dedications to the Deae Aufaniae, for example. Or perhaps it searched on the phrase pro salute et incolumitate. And maybe in conjunction with the donor’s rank of beneficiarius consularis. Certainly, the parallel texts reportedly offered by Aeneas would suggest this method (nine are listed, inconveniently by Heidelberg code, rather than the standard CIL or AE references used by generations of epigraphers). At any rate, it’s hard to know why we need AI technology for this.

Filling in the blanks

One of the features of Aeneas emphasized by the writers is that it allegedly “enables restoration of text sequences of unknown lengths”. Offhand, I can think of half-a-dozen inscriptions with significant gaps or lacunae that researchers have attempted to fill (such as this one). By contrast, neither of the altars cited by the writers has missing text, so they arbitrarily removed the words loci pro from the first one (leaving Tutelae ---/--- salute), in order to “test” Aeneas’ powers of restoration. They excitedly report that five “contextually and linguistically accurate” alternatives were proposed. This seems a rather poor result.

Our moderately skilled epigraphy student would surely predict, as Aeneas did (but only in four of its five attempts), that the word salute should be preceded by pro. (Out of 2,976 instances of the word salute in the Clauss-Slaby Database, it is preceded by pro in 2,920 cases.)

The same student would surely then realize, as the Aeneas program seems to have, that the careful layout of the inscription requires a three- or four-letter word after tutelae. There are not many options. The deity Tutela Augusta is occasionally encountered, which is presumably why Aeneas’ third choice is Aug. (I imagine it realized the four-letter maximum, as its fourth choice is sacr, which in football would be termed a “Hail, Mary” attempt, desperately relying on the stonemason’s incompetence over epigraphic precedence.) Sadly, its first choice, its best guess, is to leave a blank, which makes nonsense of the inscription.

Its second choice is the correct one. I’m not sure if that makes good odds or not. It perhaps falls into the same category as Dr Johnson’s dog walking upon its hind legs: “It is not done well, but you are surprized to find it done at all”.

New horizons?

Unfortunately, more is planned. I’m afraid I don’t agree with the writers when they claim that “Aeneas demonstrates the transformative potential of AI in augmenting historical research”. Nor with Dame Mary when she claims that “Aeneas opens up entirely new horizons”. Based on this paper, it doesn’t seem promising at all.

Incidentally, the Nature paper is not error-free. The plural of beneficiarius consularis is not beneficiarii consulares, since each man is still attached only to one legate, and it is not clear why the writers assume that such men were “usually close to retirement”. Nor is it clear why they think that tutelae pro salute is “the more commonly attested version of the formula”, since the Clauss-Slaby Database has precisely one inscription that matches this pattern (CIL XIII, 11015).

Also, some of the references have come unstuck. No. 71, “Elliott, T. in Epigraphy and Digital Resources (eds Bruun, C. & Edmondson, J.) 78–85 (Oxford Univ. Press, 2014)” — this is the title of Elliott’s paper, not the Bruun–Edmondson volume, which is The Oxford Handbook of Roman Epigraphy. No. 99, “Haensch, R. Steine sind. Benefiziarier, Matronae Aufaniae und die Topographie des Römischen Mainz. Z. Papyr. Epigr. 203, 293–296 (2017)” — the title is actually “Sag mir, wo die Steine sind”. Perhaps we might start with a proof-reading AI?

Caledonia, land of cold and rain (troisième et dernière partie)

It has been a busy twelve months and I find that I have managed to neglect Caledonia. Book One: The Ninth Legion, a wonderfully inventive bande dessinée by Corbeyran and Despujol — une chose impardonnable (see Caledonia, land of cold and rain (deuxième partie) from last December).

When we left the Ninth Legion, centurion Lucius was planning to track down the source of the Caledonians’ power on the Island of Death. “If we manage to unravel its mystery”, he explains, “we’ll undoubtedly gain the upper hand over our adversaries” (p. 32).

After a fierce mêlée (pp. 33–34), a scouting party manages to capture the warrior Cirig (p. 35), and seeing that he and his existing prisoner Leta share a fondness (p. 36), Lucius orders him to be hung on a cross to force Leta’s compliance in brokering a deal with her people (p. 37). Off she runs to the dark broch (p. 38) where she delivers the Roman message, but Cirig’s rival, the dark-haired Fel, shouts angrily, “Our people have no need of an invader to dictate their conduct” (p. 39).

The sacrifice

Despujol’s beautiful graphics depict the dim broch interior, with its drystone walls and intramural galleries and blazing central hearth. It is easy to miss the ominously large skull amongst the collection of antlers on the wall. What kind of creature does it belong to?

The old crone, Isla, reveals that Leta’s young brother, Mil, was sacrificed in order to summon the hell-hounds (p. 40) — the hell-hounds that we saw attacking the Roman fort by night in Part Two. Leta is furious and blames Fel, claiming that, unable to win her heart from Cirig, he has spitefully arranged for her brother’s death.

Worse than that, Mil has died in vain, since Leta was obliged to send the hell-hounds back in order to save her own life. She tells the old chieftain that she must return to the Romans with a deal to save Cirig’s life, but Galam replies, “Such a deal is impossible. You know that” (p. 42).

“But the Romans are here. They’ll never leave,” she says. “You have to negotiate. Or we’ll disappear” (p. 43). Meanwhile, centurion Lucius is apprehensive that his men have started to mistrust him (p. 44), when Leta returns with news that her father wishes to conclude an honorable treaty with the Romans (really?): “You will meet him at the new moon” (p. 45).

Legion on the march

Centurion Lucius arrives at the headquarters camp of the Ninth Legion, where the overweight, fur-clad legate Deodatus Faustus is having another cup of wine. He complains that, every time someone lifts the flap of his tent, they let the cold wind in, “And without fail, it never brings good news” (p. 46).

Lucius proposes that he can negotiate a treaty with Galam, and if Faustus entrusts him with the entire Ninth Legion, he can deter the Caledonians from any aggression (p. 47).

And so, Lucius sets off with his pessimistic optio beside him: “You’re rushing headlong into another trap”, he grumbles, “but this time, you’re dragging everyone along with you” (pp. 48–49). Lucius is undeterred. “This show of force is solely intended to ensure the meeting proceeds smoothly,” he explains (pp. 50–51).

The horror from the woods

The scene is a snowy plain bounded by dense forest. Lucius is convinced that his plan will work. The entire region seems quiet (p. 52).

Spotting Leta on a rock in the midst of the plain, Lucius rides over to her (p. 53). She seems defiant. “My father won’t appreciate you gathering so many soldiers for a simple meeting”, she scowls. “So many thousands of armoured men against a handful of naked warriors. Are you afraid, Roman?”

Meanwhile, the thin line of Caledonians along the forest edge doesn’t move. The crone Isla shakes her magical talisman and the forest starts to crack and groan (p. 54). Lucius’ eyes start from his head. His horse shies. “The shapes in the woods!” he cries. “Yes, Lucius”, laughs Leta. “Your eyes do not deceive you” (p. 55).

Too late, the Romans see the gigantic antlered figures with six-foot blades, lumbering from the forest (hinted at on the book cover — see Caledonia, land of cold and rain (première partie)). “I have betrayed you”, screams Leta. “This plain will soon be littered with corpses. Caledonia will be your grave!” (p. 56). Thus is the Ninth Legion doomed.

Kicking the pricks

I have been reading Blessings in Disguise, the 1985 memoir by actor Alec Guinness, who famously played Obi-Wan Kenobi in the movie Star Wars in return for 2.25% of profits. A shrewd deal when the movie, dismissed at the time by Guinness as “fairy-tale rubbish”, seemed destined to perform poorly.

Guinness later disingenuously claimed that Star Wars enabled him to “live for the rest of my life in the reasonably modest way I am now used to”. He earned an estimated $95 million from the film.

Many of the stories he recounts in the book are from the 1930s, 1940s, and 1950s, about characters who are long forgotten. No one nowadays remembers theatre impresario Tyrone Guthrie (1900–1971), for example, or actress Martita Hunt (1900–1969). Few enough still recognize the names John Gielgud (1904–2000) or Ralph Richardson (1902–1983), both (like Guinness himself) knighted for their services to theatre. But Sir Alec writes in a refreshingly precise and lucid style, so that his stories are charming and readable.

In Chapter 4, concerning his early struggle with religious faith and Roman Catholicism, Guinness refers to “the naive strugglings of an adolescent English schoolboy kicking, so to speak, against the pricks”.

Spurs and pricks

Quite by chance, I had recently come across the same peculiar phrase in the Penguin edition of Cervantes’ Don Quixote (2000).

In Chapter 20, Sancho Pansa hobbles the Don’s horse, Rocinante, with his ass’s halter in order to thwart his master’s desire for continuous (usually disastrous) knightly adventure.

I have no idea what the original medieval Spanish reads, but John Rutherford’s translation (which elsewhere very cleverly brings out Cervantes’ humour by avoiding literality) has Sancho say, “The heavens have ordained that Rocinante can’t move, and if you keep on spurring him again and again, you’ll only annoy fortune and, as they say, kick against the pricks”.

Dictionary definition

Curious to find where this phrase came from, I turned to the Oxford English Dictionary, where I found this definition: “Phrase: to kick against the pricks (spur, goad): to strike the foot against such sharp-pointed or piercing objects; also figurative to be recalcitrant to one’s own hurt”. But where did it come from? The OED gives credit squarely to the fourteenth-century Christian scholar John Wycliffe for coming up with the phrase. But why?

As a student in the 1980s, my go-to dictionary was always Chambers 20th Century Dictionary (“New edition 1983”), beloved of Scrabble-players. It explains this phrase as “to hurt oneself by resisting someone or something, to no avail (Acts ix.5)”.

The biblical reference at the end turns out to be revealing, for these words do not occur in the Greek text of Acts 9:5 (the “road to Damascus” story), but are, in fact, found at Acts 26:14. What has caused the mistaken reference?

Both dictionaries are technically correct, for the phrase occurs in Wycliffe’s translations of Acts (or Deeds of Apostles, as he prefers) at 9:5 — “it is hard to thee to kike agens the pricke”, in Wycliffe’s Middle English. (The phrase occurs on the fourth line from the bottom of the left column of this manuscript page.)

The King James Version of 1611 — along with Webster’s revision of the KJV and one or two other nineteenth-century versions that took their text unacknowledged from the KJV — is the only bible to retain the phrase at Acts 9:5, for in the accepted Greek text, these words only appear at Acts 26:14, when Paul relates the story of his conversion to King Agrippa. Here is the original (translated further below):

Incidentally, the “pricks” in question, better rendered as “goads”, are the sharp wooden implements used by farmers to drive cattle. A goaded ox, for example, once stabbed by the goad, would only hurt itself further if it kicked back against it.

In the words of the King James Version of 1611 (which is easier to decipher than Wycliffe’s Middle English, pictured here), “And when we were all fallen to the earth, I heard a voice speaking unto me, and saying in the Hebrew tongue, Saul, Saul, why persecutest thou me? It is hard for thee to kick against the pricks.”

With all due respect to the Oxford English Dictionary, it seems that the phrase should be credited not to Wycliffe, but to the writer of the Acts of the Apostles, usually taken to be Saint Luke. Or should it?

Doubt only arises because it seems that the same phrase was used half a millennium earlier by the Athenian playwright Aeschylus, in his Agamemnon (line 1624), as we can see here.

Perhaps the OED should be told?

A misunderstanding concerning Hyginus

The sources for the study of the Roman imperial army are many and varied. Snippets of information culled from unlikely places, facts unwittingly preserved by ancient historians, biographers, or even poets. The inscriptional evidence, of course, though it is patchy, increases in variety through the later first century to reach a peak during the time of the Severans and falls off sharply thereafter.

But two written texts stand out as indispensable. And I find it curious that they have attracted so little attention and discussion.

They are Arrian’s Ektaxis (a new edition and translation of which I published in 2022, accompanied by a lengthy commentary) and the Liber de munitionibus castrorum attributed to one Hyginus (a new edition and translation of which I published in 2018).

(As a side note, I am wryly amused to see that both books fall short of an Amazon 5-star rating because one disgruntled purchaser of Hyginus tetchily complained, laconically, that “Left-hand page is in Latin”, while a purchaser of Arrian whined that “the commentary is peppered with Greek, German, French, Italian and Latin, which really disrupted my reading and concentrating, although in almost every case the author gives an English translation” — which only goes to show that, even when we provide pearls, apparently not everyone likes pearls.)

A curt message

On a not entirely separate note, I recently received an unsolicited communication from a stranger, who began, rather abruptly, “Duncan Campbell ...”, as if to catch my attention in a noisy room. Rather than introducing himself or explaining his involvement with Roman studies, both of which certainly used to be standard forms of initiating polite conversation, he brusquely continued, “I have a question about two of your translations concerning Hyginus”. Fair enough.

And straight to the point: “Hyginus 30. Devito has MDC (1,600). Campbell has MD (1,500). Why have you arrived at 1,500 and Devito 1,600. It is because both of you are using different editions?”

Asked and answered, you might think.

Incidentally, the “Devito” to whom my charming interlocutor refers is J.G. DeVoto, a classics professor (I think) at Chicago’s Loyola University in the 1990s, who seems to have been behind the translation of Hyginus that appeared in 1994.

(Doubt only arises because the volume in question includes both Polybius and Hyginus, “Translated and Edited by M.C.J. Miller and J.G. DeVoto”, and the Polybius is explicitly credited to Miller, which only leaves the Hyginus.)

The Liber de munitionibus castrorum of Hyginus (or Pseudo-Hyginus, as some would prefer, to hammer home the point that he is not the Trajanic surveying author of the same name, as if that makes any difference) is a tricky text, known to us via a single manuscript, which inevitably requires careful handling. You can see a portion of the page containing chapter 30 illustrated here. It reads:

Legiones III vexilla|ri MDC. cohors prae|toria IIII. aequites | praetores .CCCC ae|quitatum in singuli | in ped CCCCL. alae mi|liariae .IIII. quingena|riae .V. ma^uri aequites | DC. pannoni beridari | etc.

It is quite clear that the manuscript reads vexillari(i) MDC, which means “1,600 vexillation-soldiers”, but the number is peculiar. Anyone using my edition of Hyginus will see, from the apparatus criticus (at the foot of p. 48), that I have accepted the suggestion of Radboud Schele (in his 1660 publication Hygini Gromatici, et Polybii Megalopolitani, De castris Romanis, Quae exstant, pictured here), who decided to change this to “vexillarii mille quingenti” (i.e. 1,500).

Although Schele gave no reason for his emendation, the first modern editor, Ludwig Lange, Professor of Classical Philology at Prague, made the case (a compelling one, in my opinion) in his edition of 1848. He explained that “we see that, in the army of Hyginus, 500 vexillation-soldiers were added to each legion”, since his model army (quoted above) contained three legions. Three times 500 equals 1,500. (The principle is demonstrated by the army of Vitellius in AD 69, which comprised four legions and vexillations from four others.) As for how the mistake occurred, the extra C could easily have crept into the manuscript on account of the fact that the subsequent word begins with the same letter, a phenomenon known as “dittography”, when the copyist has mistakenly repeated a letter.

Sadly, Lange’s version wasn’t even noted in the “official” Teubner edition of 1879. Nor was it adopted by Alfred von Domaszewski, in his well-known edition of 1887 (pictured here at pp. 48–49, where Domaszewski discusses the vexillarii), though he noted the other emendations required in this short passage: “cohors praetoria” should clearly be cohortes praetoriae, “aequites praetores” should be equites praetoriani, and “aequitatum in singuli in ped” are obviously the equites singulares imperatoris.

Somewhat surprisingly, given Domaszewski’s standing as a pre-eminent Roman scholar, he had misunderstood the significance of the vexillarii as “vexillation-soldiers”. To him, they were non-combatant baggage carriers, so 1,600 seemed a reasonable number and thus didn’t require emendation. It is worth noting that, even when the French scholar Maurice Lenoir produced his edition of the text in 1979, despite recognizing their combat role, he saw no problem with the peculiar figure of 1,600 vexillarii, although legionary vexillations are normally numbered in multiples of 500. In short, I stand by the figure of 1,500.

Be that as it may. It is perhaps worth noting here that DeVoto’s translation, mentioned earlier (and still apparently favoured by some English-speaking readers), is more than a little problematic. Firstly, it is not at all clear where DeVoto obtained his Latin text, though it seems mostly to be Lenoir’s. This may seem an airily esoteric point, but — where Hyginus is concerned — it is critical that the source text be identified. The short passage quoted above gives a flavour of the poor condition in which the text has been transmitted to us, and various editors (there have been at least five) have made their own corrections and rejected others.

And secondly, DeVoto’s translation doesn’t always match the Latin text. As an example, in the short passage mentioned above, where Hyginus enumerates his ideal army — it begins at the bottom of a lefthand column, pictured above, and continues at the top of the righthand column, pictured here —, the list includes, amongst the various military units, cohortes equitatae miliariae II, quingenariae IIII; cohortes peditatae miliariae III, quingenariae III.

Four different types of cohorts, twelve in total.

Here is DeVoto’s translation: “Two 1000-man mounted cohorts, three 500-man ones”. See if you can see where he went wrong.

The Roman army in turmoil

As an independent researcher, I am spared the burden of academic administration that seems to plague the current generation of university staff members. But I am also denied the many benefits that they enjoy, chiefly (to my mind) ready access to a research library. However, this is somewhat mitigated by the existence of the Internet Archive’s indispensable virtual lending library (as I mentioned in the preface to my book Phantom Horsemen).

An overlooked book

I recently came across a book that was uploaded to this library last year but had so far escaped my attention: it is entitled L’armée romaine dans la tourmente, written in 2009 by Yann Le Bohec, Emeritus Professor at the Paris Sorbonne.

No reviews ever appeared in the English-language journals, but maybe they received no review copies. As far as I can see, it seems that the book was noticed only in the German journal Bonner Jahrbücher (in 2010) and in the French-language journal Latomus (in September 2013), on both occasions in broadly congratulatory terms — in the former, we read, “Despite some redundancy, the book is clearly organized, its thesis firmly and confidently presented, in any case worth discussion”, and in the latter, “(Le Bohec) approaches a hackneyed subject with a fresh eye and a readable style”.

Le Bohec begins by promising a detective story (“une véritable enquête policière”) in which the Roman empire is the victim. “How could a body as solidly built as the Roman army”, he asks, “have received such violent blows, been shaken in such terrible difficulties?” Disasters of such magnitude, including emperors killed in combat or taken prisoner, legions routed or wiped out, these and a thousand other misfortunes require an explanation. (Such dramatic hyperbole, incidentally, also requires justification.)

The remainder of the book describes, first of all, the Roman army of the later second century, in which Le Bohec offers an apology for having drawn extensively on his 1989 book L’armée romaine sous le Haut-Empire (apparently in a third edition of 2002, which I have not seen). “To make amends”, he writes, “we shall try to update it and, which is more difficult, to identify the features specific to the end of the second century”. None of which is strictly relevant to the third-century army, but let us proceed.

The rise and fall of the Roman army?

Le Bohec emphasizes (in chapters 1 and 2) that the success of the army of the Antonine emperors (AD 138–192) was founded on high-quality recruiting (“le choix des meilleurs”). He also emphasizes that this excellent second-century army was stationed in a privileged frontier zone, a “golden belt” (“une ceinture dorée”) that “undoubtedly exerted a strong attraction on the barbarians, but also allowed them to be watched and repulsed”.

So far, so good. These points may well be true, but they do not necessarily imply that any disasters that occurred thereafter must have been caused by (a) a deterioration in recruiting standards from the Antonine ideal, and (b) the increased allure of the wealthy frontier zone, that “golden belt”, to the external peoples. But let’s see where Le Bohec is going with this.

In chapter 3, he describes the army of the Severan emperors, from Septimius himself (AD 193–211) through to the last notional “relative” Severus Alexander (AD 222–235). This was the period, Le Bohec claims, when the Roman army “probably reached the peak of its efficiency”. They could easily repulse German and Parthian enemies. And yet it was also the period during which the soldiers, by their inflationary greed, “dug their own graves” (“les soldats creusaient leur propres tombes”).

Le Bohec devotes chapters 4–8 to the main enemies of Rome: Franks, Alamanni, and Goths; the Persians; the northern Britons, the trans-Danubian peoples, and the north African peoples; and finally, the “invisible enemy” of the inflationary economy. This is his main bugbear.

He seeks to blame the soldiers for demanding a pay rise from Caracalla. “The emperors and the soldiers jointly committed this crime against the Empire”, resulting in a financial crisis that he locates squarely in the reign of Gallienus, largely based on the presumption that the loss of the eastern provinces compromised the imperial revenues. He also blames the need to pay “tribute” to victorious enemies, claiming that “the third century provides some examples of this practice”.

Paying off Rome’s enemies?

It is certainly true that, from time to time, the Romans resorted to paying subsidies, but there certainly doesn’t seem to have been a sudden torrent of payments depleting the third-century imperial treasury. Severus Alexander is said to have considered paying off the Germans in AD 234, but probably never did. The Persians claimed that Philippus paid them tribute in AD 244, though no western source records it, while in AD 251 Trebonianus Gallus probably did pay off the ‘Scythians’ who had killed the emperor Decius. The Juthungi certainly requested a subsidy from Aurelian in AD 270, but didn’t get it.

I have a nagging doubt about whether the problem can be diagnosed so simplistically, in any case. Evidence of price inflation, though sparse, seems to begin in the AD 270s, and not earlier. Furthermore, there had already been a debasement of the coinage under Marcus Aurelius, who faced a similar problem to the one that vexed Gallienus — namely, major military operations mounted at a time of rampant plague — and we certainly don’t talk about an “Antonine crisis”.

And yet Le Bohec writes of “a serious military crisis”; of “attacks carried out, more and more often, on two fronts at the same time”; of defeats, “often terrible”, and coups (his preferred term for usurpations); of “a worsening of difficulties in the years 249–258”; and of “a continuation of worsening” up to AD 275. “The inhabitants of the Empire”, he writes, “experienced all possible evils”.

Some of this really needs to be fact-checked.

In support of warfare on two fronts, Le Bohec claims that the Alamanni were threatening northern Italy at precisely the same time that the Goths invaded the Balkans. Unfortunately, the chronology is vague, and the two events may have been consecutive rather than simultaneous. (And they don’t seem to have occurred “more and more often”.) Gallienus certainly managed to visit both theatres. For his terrible defeats, Le Bohec is probably thinking of Valerian’s eastern débâcle (which was not, strictly speaking, a battlefield defeat but a foolish blunder).

I have no idea why he thinks that legions were routed and wiped out (unless he is, again, alluding to the eastern fiasco and the unknown fate of Valerian’s army), but he would like his readers to believe that “this situation encouraged minor peoples who had never troubled Rome to do some looting”. His evidence for this is limited to the case of Gaius Macrinius Decianus (governor of Numidia, probably around AD 254, as he alludes to joint emperors on his dedication, ILS 1194), who on three occasions, defeated a people known as Bavares, once in concert with a people known as Quinquegentanei. But this kind of raiding must have been common in all periods. It cannot, in any case, be used as evidence of a serious military crisis. (Decianus was, after all, successful.) Finally, the allegation that “coups d’état became commonplace” rests heavily on the spurious testimony of the Augustan History and its “Thirty Tyrants”, most of whom are fictitious.

The Roman army in crisis?

Le Bohec’s general conclusion is that three factors conspired to create the grave turmoil of AD 235–284.
1) Stronger barbarians. Really? It is true that, for a few years in the early AD 260s, invaders were able to penetrate much further into the empire before being repulsed, but repulsed they were. Le Bohec thinks that this “proves their strength and, at the same time, the weakness of the Romans”, but only at that precise moment. It is difficult to demonstrate either “a new and violent aggressiveness” or the composition of “leagues that allowed them to help each other” (for which he cites the Quinquegentanei, mentioned above, and the Picts, who first appear in AD 297).
2) Weaker Romans. It is also difficult to demonstrate Le Bohec’s “clear decline” in the Roman army and a “crisis of adaptation”, as the Romans failed to modify their tactics in the face of “barbarians (who) used new weapons and fought better than before”. No wonder “the Romans experienced defeat with the consequence of a weakening of morale and a drop in the quality of recruits”. Unfortunately, Le Bohec doesn’t back any of this up with evidence.
3) The crisis against Rome. Under this heading, Le Bohec cites the (alleged) economic failure, which he considers inevitable, after two centuries of continuous prosperity. Since the Roman army was “less well paid and more poorly fed, they had less and less desire to be killed”. Again, no evidence is adduced.

These are all very interesting ideas. But Le Bohec never comes close to proving any of them, and in the end, we are left wondering if there really was a third-century crisis after all.

Fact Check: Roman cavalry reform

In the school where I do my day job, the assembly hall is dominated by a large painting entitled The Thin Red Line. It depicts Sir Colin Campbell’s 93rd Highlanders receiving the charge of the Russian cavalry at Balaclava in 1854.

Apparently, the highlanders had not been trained to deploy into the textbook square formation, so they stood two-deep and fired at the approaching enemy. The cavalry veered off, rather than engaging them.

It is a commonplace that cavalry will not attack a formed line of infantry. There seem to be two reasons. First, the cavalrymen themselves are unwilling to enter a killing zone, the area in which (in nineteenth-century warfare) they might be peppered with shot or (in ancient times) deluged with arrows and sling bullets. But second, horses are naturally averse to crashing into solid barriers, so that, faced with an unflinching line of infantry, they will normally attempt to go around it. Only the fortuitous breaking of the infantry line would give the cavalry an avenue through the barrier, rather than around it.

Of course, the Roman consul Arrian, governing the armed province of Cappadocia in the mid-AD 130s, was well aware of the characteristics of cavalry warfare. Faced with the threat of invasion by the horsemen of the Alans, Arrian drew up a battle plan whereby their approach would be halted by a solid block of infantry, bristling with spearheads. Just the sort of obstacle that cavalry would normally avoid. All of this is explained in my recent edition and translation of Arrian’s Ektaxis.

A barrage of missiles was planned to dissuade any of the Alans from loitering in the killing zone, while measures were put in place to tackle the horsemen as they inevitably veered off. The main point is that a completely cavalry army could be handled, if not easily, then certainly straightforwardly. Cavalry on its own had very limited utility.

Nevertheless, there is a widespread belief, ingrained from generations of repetition, that — sometime in the late AD 250s or early 260s — the emperor Gallienus developed a solely cavalry army, which the German scholar Robert Grosse dubbed the “battle cavalry”.

Gallienus’ cavalry commanders?

In a recent discussion of my Phantom Horsemen, the book in which I question the very existence of Gallienus’ cavalry army, reviewer Byron Waldron objects on the grounds that “it seems to me that Gallienus did make changes to the cavalry”.

So how might we expect these “changes” to manifest themselves? Surely Dr Waldron has more than a gut feeling to justify planting the seed of doubt in readers’ minds. Surely he won’t simply restate “facts” that I already dealt with in the book. Surely?

Chiefly, Dr Waldron claims that “various cavalry commanders in this period eventually became emperors or usurpers”, by which he means four individuals, none of whom is ever called “commander of the cavalry army” or “commander of the battle cavalry”. But each individual’s link to cavalry, however tenuous, is reinterpreted in a subjective way in order to fit the theory of the (unproven) “battle cavalry”. Here are the four men.

1) Aureolus

When the Byzantine historian Zosimus introduces Aureolus in his description of the events surrounding Gallienus’ death at Milan in AD 268, he describes him as “commander of all the cavalry”. In this regard, we should remember the pertinent observation made by Matthias Springer in his 1988 paper, that the phrase “commander of all the cavalry” is essentially meaningless. (“Only a preconceived opinion”, wrote Springer, “can infer the existence of a special battle cavalry from that phrase”.) After all, it was the emperor himself who commanded all the cavalry, and he did so through the agency of his equestrian officers.

However, the phrase makes perfect sense in the context of a battle, where the cavalry forces might be entrusted to an overall commander, such as the one Septimius Severus explicitly appointed at the Battle of Lugdunum (as I explain in the book). Aureolus, we know, had commanded the cavalry at Mursa against the usurper Ingenuus, because Zonaras tells us so.

Dr Waldron complains that, at the Battle of Mursa (which probably occurred in AD 260), “Zonaras does not speak of ‘all the cavalry’ (Gallienus led the Mauri)”, which is perfectly true. In fact, Zonaras tells us that “commanding the cavalry, Aureolus, by fighting nobly with his horsemen, destroyed many of those who sided with Ingenuus, and turned the remainder to flight”. I think most of Zonaras’ readers would realize that he must have been directing all the cavalry forces present in the battle with the exception of those accompanying the emperor.

Eight years later, Aureolus had moved on with his career, but his claim to fame was having commanded the cavalry and won the day for his emperor at Mursa, the first battle against a usurper. Let us not lose sight of this important fact, which is often forgotten in the headlong rush to try and “prove” the existence of Robert Grosse’s “battle cavalry”.

2) Aurelian

Zonaras records that the future emperor Aurelian “came to (Gallienus) with horsemen” in the struggle against Aureolus in AD 268. (I wondered, in my book, whether this might have been Gallienus’ horse guard, but the phrase is too vague to press the point.) The only other testimony comes from the rather unreliable “Life of the deified Aurelian” in the Augustan History, which claims that “before he took power, Aurelian commanded all the cavalry under Claudius”. (Here is this meaningless phrase “all the cavalry” again, and I wonder whether this, too, might have been a battlefield command.)

3) Claudius II

Dr Waldron reminds us that “Claudius was said to have been a cavalry commander at some point”. This is Zonaras, who quotes a secondhand story referring to him, at some point prior to becoming emperor, as “Claudius the cavalry commander”. It is worth remembering that commanding a unit of cavalry was a standard stage in the career of any aspiring equestrian officer.

Dr Waldron is (naturally) not alone in creating this spurious link between cavalry — any cavalry — and the fictitious “battle cavalry”. For example, Professor Lukas de Blois, who has championed Gallienus’ cavalry army for fifty years, describes Claudius II (in his 2019 book) as “Gallienus’ cavalry commander” and “the commander of Gallienus’ new army”. He wasn’t.

4) Probus

Dr Waldron also highlights the fact that the future emperor Probus “according to the Epitome de Caesaribus was named Equitius”, which might link him to the “battle cavalry”. Like the previous three individuals, Probus was already discussed and dismissed in my book, but perhaps Dr Waldron overlooked that passage.

Equitius, we should note, is not an unknown Roman nomen, so — giving the Epitome the benefit of the doubt — it may well have been Probus’ own name before he took the official nomenclature of Marcus Aurelius Probus.

Incidentally, it is quite unlikely that Aurelius Victor (the ultimate source of the Epitome), writing almost a century after the reign of Probus, was aware of the curious coins (discussed in my book) that each carry one of the letters A, E, Q, V, I, and T; and equally unlikely that he drew the same conclusion as Alfred von Domaszewski, that this was a coded reference to “Equitius” as the emperor’s name.

In any case, it’s a peculiar argument that concludes, from a horse-related knickname found on only one occasion (equus means “horse”, so Equitius ought to mean “pertaining to horses”, so the argument goes), that Probus must have been not only a cavalry commander, but specifically in command of the alleged “battle cavalry”.

5) Diocletian?

Lastly, Dr Waldron throws in a fifth individual: “It may be relevant that Diocletian was a Dalmatian career soldier”. I simply cannot imagine how this is supposed to bolster a case for the existence of a “battle cavalry”, so I’ll leave that one.

Cavalry coins indicating a reform?

But finally on this theme, Dr Waldron supports his belief in Gallienus’ cavalry reform by observing that “numismatic legends and iconography attest to periods in the late third century when cavalry was especially important”. Forgive me my deep scepticism as we explore what he means.

It’s one of these arguments that begins innocently enough. Third-century emperors are depicted on horseback, we are told. Where? On the Arch of Galerius in Thessaloniki. But wait ... that’s not third-century. That’s Tetrarchic.

Here is the reverse of a coin of Galerius, depicting that fourth-century emperor on horseback trampling his enemies. It wasn’t a new motif. Septimius Severus had used it, a century earlier. Severus Alexander, Decius, and Gallienus, too.

But many emperors appear on their coinage riding a horse when the theme is the emperor’s ADVENTVS (“arrival” in Rome). Hadrian used the same motif on his ADVENTVS coins. Likewise, Gordian III, Philippus, Decius, Trebonianus, Claudius, Aurelian, Probus, and Carus (amongst third-century emperors).

When Septimius Severus appeared as a galloping horseman, the theme was obviously the emperor’s VIRTVS (“manly courage”). Nobody would have suggested that it was an indication that the emperor had made some kind of change to the cavalry.

So let’s try and rein in some of these wild unsupported allegations of “changes to the cavalry” under Gallienus.

A troublesome history of the third century

I have been reading Rome in the Third Century. A Troubled Empire by Michael Sage, an emeritus professor of Classics. Reading, but not enjoying. Or agreeing.

I am unsure of Professor Sage’s precise specialism. His “Expert Profile” on the University of Cincinnati website informs us that he is “engaged on a source book on Roman warfare”, but that book appeared seventeen years ago in 2008. (Who knows, maybe there’s another one on the way.)

He holds a PhD, which (I imagine) concerned the third-century St Cyprian, since this was the subject of his first monograph, charmingly produced on a typewriter and published in 1975. Cyprian, bishop of Carthage, was famously martyred in AD 258, during the joint reign of the emperors Valerian and Gallienus. This, I assume, is the basis of the professor’s interest in the third century.

At any rate, Rome in the Third Century seems to have managed to avoid any editing process. Misprints may be due to inaccuracies in my eBook version. But there are mistaken references and much repetition of irrelevant material (irrelevant to the third century, that is), while the narrative is often confusing.

I’ve drawn some examples from the years AD 253–268, having just published my own book on this period.

Valerian and Gallienus (AD 253–268)

Sage employs a to-and-fro timescale that I found rather confusing. In a single paragraph, we read that “Valerian was in Germany during 256–257” and “the Goths, along with the Borani, (...) in 254 or 255” raided Asia Minor, while “in 266 the Goths and Borani launched their most devastating raid (and) in 267 they even penetrated into Greece”, whereas “unlike his father, Gallienus remained in the west dealing with the German tribes”.

We then jump back to AD 253 for Shapur’s invasion of Syria, but “by the next year, the Romans had re-established themselves in Antioch” under Valerian (pictured here), while Gallienus “was active on the Danubian front”, but “the quiet was broken by the emergence of a new threat along the Rhine frontier” — this was the emergence of the Franks “in 233 or a little later” and the Alamanni “in 213”. Remember these dates, because we then read that “Gallienus probably moved to the Rhine frontier in response to these developments in late 255 or 256” — so, his response came at least twenty years after the fact? That’s confusing.

Then, “Valerian was with Gallienus at Cologne in 257” (fair enough, we read earlier — see above — that he was in Germany in 256–257) but “he was soon drawn away to the east” (if we hang on for a few pages, we’ll discover that this was spring AD 258), so “in 257 or 258 he installed his elder son Valerian at Viminacium” on the Danube.

Let’s read that again. The emperor Valerian installed his elder son Valerian at Viminacium? No, Valerian Caesar (pictured here) was Gallienus’ elder son (thus, Valerian’s grandson) about whom nothing is known beyond his name. (That name appears briefly on the coinage during AD 257 and disappears in AD 258.)

Meanwhile, Valerian went to Rome in AD 257 and “(Gallienus’) campaign in Gaul was interrupted by the proclamation of Ingenuus by his troops as emperor” on the Danube, and “it seems likely that Valerian the younger perished in the course of the revolt” (actually, that’s not at all likely, given that we know nothing about Valerian junior), but “Gallienus’ commander Aureolus fought a decisive battle with Ingenuus at Mursa” (it was actually Gallienus himself — pictured here — who fought and won the battle; Aureolus only commanded the cavalry).

Nevertheless, “the failed rebellion was followed by another in 259” by Regalianus “governor of Illyricum” (note that there was no Roman province of Illyricum) who died under mysterious circumstances: “some claim that he was killed in battle against the Roxolani, while others state that he was assassinated by his own troops” (in fact, the Augustan History — the only source to mention it — claims both when it records that “he was killed at the instigation of the Roxolani and with the consent of the soldiers and the provincials”).

In the meantime, Gallienus’ absence from the Rhine (we’re in AD 259, now) prompted “multiple attacks along the frontier”, which were only finally “stopped by a governor of Raetia” (for this fascinating episode, passed over in half-a-dozen words here, see pages 56–59 of my book, where it is dated to AD 260 or 261), but in the process, Gallienus’ younger son Saloninus (pictured here) was murdered by Postumus, the governor of Lower Germany, who promptly set up his breakaway Gallic Empire. “Finally, in early summer 260, Gallienus was able to extricate himself from the Danube frontier”, in order to defeat another band of Alamanni at Milan (actually probably the same Alamanni as were mentioned earlier).

East and West

“Meanwhile, during the spring of 258 Valerian returned to the east by way of Illyricum where the situation had become close to disastrous”. Professor Sage’s narrative has become so convoluted that we must take a moment (a) to remember where on earth Valerian had been (both emperors had perhaps spent AD 257 on the Danube front), and (b) to consider why Illyricum was a disaster zone (this must be harking back to Ingenuus’ revolt, which more probably occurred in AD 260). When Valerian arrived in the east, “the campaign was abandoned before it could begin” because “the army was infected by the plague”, causing the “further weakening of an army that had already seen combat in Illyricum”. But wait. If this is just AD 258, whom had they been fighting in Illyricum?

Let’s be charitable and call Professor Sage’s chronology extremely speculative. I, for one, do not agree with his version of events involving Ingenuus, Regalianus, and Postumus. At any rate, he proceeds to recount Valerian’s defeat and capture in 260, followed by the revolt of Macrianus and its defeat by Aureolus (“the general who had earlier defeated Ingenuus”), and the transfer of control in the east to “Odenathus (who) now launched two attacks on the Persians, the first in 262–3 and the second in 267”.

We are then swept back in time again: “there were invasions by the Goths in 262, 266 and the most serious in 267/8”. Then “Gallienus launched an attack against Postumus in 261, but it failed”. His tenth anniversary celebration in AD 262 “must have been a happy occasion as he had finally established general peace in the provinces that he controlled”. But wait. Didn’t we just read that the Goths were invading?

By this stage in the narrative, it only remains to document the revolt of “Marcus Aelius Aureolus”, a novel concoction of names for the man whom history knows only as “Aureolus” (as I explain in my book, he is often mistakenly named Manius Acilius Aureolus, for historic reasons dating back to the seventeenth century, but never Marcus Aelius), and the murder of Gallienus, who — we are told — “had remained on the throne for eighteen years”. But there were only fifteen years from his father’s accession in September AD 253 until his own death in autumn AD 268.

Military reforms?

Inevitably, we read that “Gallienus introduced or developed a number of reforms that were vital to the empire’s survival”, and that “his reforms were primarily directed towards creating a more powerful military and a more effective administration to support it”. Unfortunately, this common belief has no basis in fact (as I argue in my book).

“The most striking change was the growing importance of cavalry”. This is a curious claim. How does Professor Sage define “importance”? I have shown in my book that there was no increase in cavalry numbers, and there is no sign of cavalry-only task forces being employed.

He gives credence to a bizarre statement made by Professor Brian Campbell in 2005, who wrote, in the context of the Severan army, that “A further development was the creation of milliary units, which consisted of between 800 and 1,000 men”. But this development occurred in the later first century, long before Septimius Severus and fully a hundred and fifty years before Gallienus. As far as cavalry was concerned, when the ala II Septimia Syrorum miliaria appeared under Severus in north Africa, that took the total of “double-strength” squadrons to nine, distributed all around the frontiers. Yes, nine. Hardly an overwhelming trend.

Professor Sage adds that “there was also some innovation in cavalry equipment beginning under the Emperor Hadrian who introduced mailed cavalry”. (But what were they wearing before this time, if not mail? In fact, he seems to have misunderstood the significance of the term catafracta — “fully armoured” — in the name of the ala Gallorum et Pannoniorum catafracta, a squadron whose earliest appearance is during the reign of Hadrian.)

“These changes”, he writes — the gradual creation of a handful of double-strength cavalry squadrons over the course of the second century and the existence of a single fully-armoured squadron — “culminated in the third century (when) the key development was of a large, independent cavalry army in the reign of Gallienus”. So this is the basis of the “number of vital reforms” that he alleges.

His only doubt is whether Gallienus deliberately created the cavalry army, or whether “it simply emerged from existing trends” (what existing trends?). He decides that “it was probably the result of both factors”. So it was deliberate ... and also not deliberate. And if this level of argumentation is not astounding enough, the professor goes on to claim that “it was the beginning of a major change in the way Rome used its armies. It led to the separation of the army into two types of forces, a field army and a border or frontier army”. Oh, dear.

A troubled narrative

Professor Sage doesn’t provide an epilogue, a summing up, or any kind of final thoughts. The book simply seems to peter out. Along the way, there is quite a bit of padding, as he takes each subject back to the days of the Roman Republic. For example, “the Roman army had developed from a militia that was raised for individual campaigns and commanded by senators who were not military specialists”. This extreme level of scene-setting is unnecessary in a book purporting to be about the third century.

The frequency of vague claims also generates an uncomfortable feeling that the reader is not in safe hands. “The army has been criticized by some scholars for lacking in innovation before the third century” (unfortunately, none of these scholars is named) “but there were some adjustments in tactics and weaponry”. He cites Arrian’s Ektaxis (sending the reader in search of a 1979 paper on “The legion as phalanx”, which is not the surest guide to this work) as proof that “the auxilia with its diverse arm and equipment was a weapon of increasing importance”. I find this conclusion baffling. He also likens the Roman army to a wall, “but in the course of the third century this wall had to be rebuilt”.

Well, if we’re relying on the existence of a fully-armoured Hadrianic cavalry squadron and Gallienus’ imaginary mobile cavalry army, I’m afraid it will need to remain unbuilt. Verdict: not recommended.

... and he builds a privy

The eighteenth-century merchant Jonas Hanway, who achieved notoriety as an anti-tea-drinker, briefly travelled in Afsharid Persia and published his observations as An Historical Account of the British Trade over the Caspian Sea in 1753.

In Volume 2, subtitled The History of the Celebrated Usurper Nadir Kouli, he described Nadir Shah’s recapture of Isfahan from its Afghan occupiers in 1729, during which a most peculiar event occurred.

The inhabitants, keen to expunge all traces of the Afghans from their town, directed their wrath towards the mausoleum of Mahmoud al-Afghani.

“Having obtained the permission of the Shah for this purpose, they levell’d this edifice to the ground, with the fury of an enraged mob; they would not even suffer the repose of his bones; and, to add yet a greater mark of contempt and abhorrence, in the very place on which this mausoleum had been erected, they built a public jakes” (i.e. a latrine).

History repeats itself

Perhaps the Isfahanians knew their Old Testament. Certainly, on reading this curious story, I was reminded of King Jehu’s treatment of Ahab’s family and the prophets of Baal (2 Kings 10:1–28). At verse 27, καὶ κατέσπασαν τὰς στήλας τοῦ Βάαλ καὶ ἔταξαν αὐτὸν εἰς λυτρῶνα ἕως τῆς ἡμέρας ταύτης, or in the sonorous King James Version, “they brake down the house of Baal, and made it a draught house unto this day” (draught-house being a euphemism for latrine).

It is interesting that the tenth-century Byzantine chronicler Symeon Logothetes, in his abbreviated account of biblical history, chose to highlight this particularly colourful event: “Then, gathering all the prophets of Baal and the priests and the people, by pretending to order a public sacrifice, (Jehu) slays them all at once and he destroys the sanctuary of the idol and he makes it a latrine” (Chronicon 44:10).

Confusion and battle cavalry

I was reminded of all these interconnections this week, on reading a remark made by Australian classicist Byron Waldron, a post-doctoral researcher at the University of Cyprus, in his review of my book Phantom Horsemen (in which I question the existence of Gallienus’ battle cavalry).

Briefly stated, the evidence — such as it is — for the emperor Gallienus’ battle cavalry, beloved of generations of writers on third-century history, is a sentence written by the above-mentioned Byzantine chronicler Symeon Logothetes in his Chronicon.

A century and a half later, George Cedrenus then lifted it for his own Compendium of Histories, which is where twentieth-century scholars found it. I have told the whole story in Chapter 2 of my book.

Cedrenus, it should be explained, followed the example of Symeon (and his other tenth-century source, George Monachus) in presenting — when his world chronicle got to the period of the Roman empire — one or two salient facts about each emperor’s reign, especially if they involved church-related events.

For example, in the paragraph on the emperor Severus Alexander, he records that Alexander’s mother Mamaea “happened to be a Christian, and she summoned Origen, who resided in Antioch, as she wished to be taught the mystery of Christ”. Cedrenus lifted this almost verbatim from Symeon’s Chronicon.

For the emperor Gallienus, the chosen “fact” — tellingly characterized as “Cedrenus’ confused remark” by the earliest modern writer to cite it — was that “he first appointed units of cavalry, for in general the soldiers of the Romans were infantry”. This is patent nonsense. As I explain in my book, neither Symeon nor Cedrenus is a particularly trustworthy source for Roman imperial history. They periodically make some very odd allegations. (I gave some examples in my book, to which we might add the “bread falling from heaven” during the reign of Probus.) My own instinct is to reject “Cedrenus’ confused remark” as either corrupt or mistaken, or both.

Choosing the right verb

However, I threw out the possibility — an intriguing one, I thought — that Symeon (or his unknown source) may have misconstrued a remark that originally referred not to “units of cavalry” (ἱππικά τάγματα), but to Rome’s “equestrian order” (ἱππικόν τάγμα), that wealthy class of citizens who stood second only to the senators in terms of prestige. I have explained my reasoning in the book. In short, there was a belief, current in the centuries after Gallienus’ death, that he had been responsible for removing senators from legionary commands and appointing equestrians in their place.

I deliberately translated the Greek verb καθίστημι (kathistêmi) in the above-quoted remark using the rather awkward “appointed” (as I explained in my book) because I found it an odd choice of word to convey the meaning that modern researchers wish to give it — namely, that Gallienus “created” a new style of military unit.

The verb, in my opinion, usually implies “appointing” someone to a role or “posting” someone somewhere. To be sure, it can also (when suitably qualified) indicate “imposing” or “rendering” or “bringing about” a particular state of affairs.

Choosing the right translation

When we consider other ancient Greek writers, Josephus consistently uses this verb for appointing people to particular positions or tasks. For example, in his description of King Solomon’s army, he writes that David “appointed as commander of each division a man whom he knew to be brave and just” (Jewish Antiquities 7.369; many other examples at 4.165, 7.232, 8.318, 9.4, 9.188, 10.97, 10.182, 10.249, 12.254, 13.146, 15.322, 18.95, 18.237, 20.247 and 248). Xenophon uses it in this way, too (e.g. Hellenica 3.4.29, 7.1.45).

As for the meaning of posting existing troops, Xenophon uses the verb in this way, for example, when he writes that the Spartan Pharnabazos “armed the sailors and posted them as guardians of his own coastal territory” (Hellenica 1.1.24; other examples at 6.2.33 and Cyropaedia 5.3.1, 7.5.69, 8.6.17). Josephus often uses the verb in this sense, as well (e.g. Jewish Antiquities 8.181, 9.287, 13.157, 13.227, 15.184 and 185).

And it can imply the arranging of existing troops, as when Xenophon describes how the Spartan Clearchus “arranged the army so that it presented a fine appearance from every side as a compact phalanx” (Anabasis 2.3.3). But I cannot find an instance where it means creating troops or raising troops, in the manner required by “Cedrenus’ confused remark”.

Rendering a privy

Of course, it is always important to consider how a particular author uses a particular word. Each may have his own idiosyncrasy. I found that Symeon — again as I wrote in my book — often uses this verb to mean appointing people. I gave four examples (48.2, 53.6, 54.4, 88.2). I could have added more (34.6, 42.16, 42.19, 45.2, 45.4, 48.1). And he uses the related verb ἐφίστημι (ephistêmi) with this meaning, also (83.1). But I could not find a single example where it is supposed to mean creating an institution.

Returning to Waldron’s review, he is lukewarm about my tentative proposal to reinterpret the offending sentence (to mean Rome’s “equestrian order” rather than “units of cavalry”), although he concedes that “this is an intriguing idea that should be carefully considered”. Far from considering it, though, he immediately claims, by way of rebuttal, that “Symeon on three other occasions uses καθίστημι to mean ‘make’ or ‘establish as’ (42.17, 44.10, 49.2)”. I’m not convinced that “Gallienus was first to establish as cavalry units” is much of an improvement. The grammar still strikes me as odd.

But what about those three occasions cited by Waldron on which Symeon allegedly used the verb in a sense that would really help us here? Well, prepare for disappointment.

In the first one, Symeon describes how Nebuchadnezzar, the Babylonian king, invaded the Levant and “made all of Judaea subject to taxation”. Here he uses καθίστημι in the sense of “imposing” or “rendering” a particular state of affairs, qualified by the adjective ὑπόφορος (hypoforos, “liable to pay taxation”). In our sentence, Gallienus did not render a particular state of affairs on the cavalry units; he just “rendered” them — there’s no qualifying adjective!

Jumping to the third one, here Symeon uses the same construction to describe how Julius Caesar conquered parts of Germany, Gaul, and Britain, and “made 500 inhabited towns subject to Roman taxation”. Again, it is καθίστημι with the adjective ὑπόφορος and the same objection applies.

Waldron’s second reference is more interesting. Certainly, it is as irrelevant as the other two, but it brings us back to Jehu’s desecration of the precinct of Baal, quoted at the start. Here, perhaps uniquely in the Chronicon, Symeon uses καθίστημι not with an adjective but with a noun. The verb is, again, used in the sense of “rendering” or “bringing about” a particular state of affairs, but in this case, the subject of the sentence is the destroyed sanctuary of Baal, and it is being “rendered” as a λυτρών (lutrôn, “latrine”). Or as Hanway would have said, a jakes.

Final thoughts

Perhaps I was too hasty in excluding the meaning of setting up an institution. To be sure, Symeon doesn’t ever use the verb with this meaning. (So his statement about cavalry units is still suspect.) But I have found an instance of the verb’s use by Josephus referring to the Roman general Gabinius “setting up five councils” in Judaea (Jewish Antiquities 14.91). So if Josephus had ever described the creation of cavalry units, he might have used this verb.

Just to clarify, this is a grammatical argument only. I am still convinced that “Cedrenus’ confused remark” is either corrupt or mistaken, or both!

Fours and nines and Roman numerals

At primary school, I took the leading part in a play written by one of the teachers. It was a pastiche of the then-current and popular TV series Up Pompeii, of which I (being around 10 years of age) was entirely ignorant (it being rather risqué) — but the parents would definitely have appreciated the homage.

The show was titled Up Wumpeii, after the nearby housing estate built by developer George Wimpey Limited.

A few lines stick in my mind, across the gulf of fifty years. I opened the show, striding confidently down the aisle in the assembly hall, and as I headed for the stage, I intoned the immortal phrase, “Friends, Romans, countrymen, lend me your ears”, to which the chorus responded with “’ere, ’ere!”

At one point, a squad of ten Roman soldiers marched on and performed a gag about a roll call, which prompted a litany of baked products, including “baps, softies, MacKechnies” — a reference to the nearby bakery run by the firm of MacKechnie’s, where some of the parents may have been employed. The legionaries then proceeded to number off, Roman-style, to the amusement of the audience: “Aye, Aye-Aye, Aye-Aye-Aye, Aye-Vee, ...”, and so on, finishing with “Aye-Ecks, Ecks”.

How Roman numerals work

As school children the world over will tell you, the Romans did indeed form numbers by repeating symbols — thus, III (“Aye-Aye-Aye”) = 1 + 1 + 1 = 3 — though, in practice, only the symbols I (one), X (ten), C (one hundred), and M (one thousand) could be repeated — not V (five), L (fifty), or D (five hundred).

Some numerals can be seen on this graffito from Pompeii (CIL IV, 5380), which is probably someone’s expenses, scratched on the wall.

Lines 1, 5, and 12 are consecutive dates in an unknown month: VIII Idus (“the eighth day before the Ides”) — the S has a flamboyant tail that trails diagonally down the wall to the next day’s entry: VII Idus. And there is a fine selection of numerals on display (presumably the cost of each item in asses, the smallest commonly available Roman coin).

Fours and nines

It is widely believed that, when writing numbers, rather than repeating a given symbol thrice (e.g. IIII), an alternative method was employed, whereby the symbol was placed to the left of the next-greater symbol. Since, in the case of I, the next greater symbol is V, rather than writing IIII, the alternative form IV was preferred. We can certainly see this on the graffito. In line 13 (third from bottom in the picture), the entry reads puero pane IV (“bread for the boy, 4”, presumably referring to a slave). (Notice that the character E is frequently represented in Roman handwriting by two vertical lines, so pane — also on lines 2, 6, and 12 — looks like PAN||.)

As with IIII and IV, rather than writing VIIII, the alternative “subtractive” form IX was preferred. Hence, the ninth Roman soldier in Up Wumpeii chanted “Aye-Ecks” (IX, geddit?). Unfortunately, the graffiti artist didn’t purchase anything that cost 9 asses, so we cannot say for sure that he would have used this form.

Convention or preference?

I was reminded of all this by a remark made in one of the reviews of my book The Fate of the Ninth (I don’t recall which one). The author had suggested that the Romans preferred to write the number 9 as VIIII, and not IX. This is definitely something crying out to be fact-checked.

It is easy to find monumental inscriptions with numerals. Every Roman emperor advertised how many times he had held the consulship, or had been acclaimed as imperator (“conquering general”), or — more importantly — had held the tribunicia potestas (“tribunician power”, which made the holder sacrosanct, and was thus fastidiously renewed every year).

On the left is Septimius Severus, who held the tribunician power for the ninth time in AD 201 (CIL VIII, 10992). He was reigning jointly with his elder son, Caracalla, who appears here as TR P IIII. (The numerals are highlighted in red.)

It is interesting that the “subtractive” method is used for 9, but the “additive” method for 4. This may have been down to the stonemason’s personal preference. It’s true that there wouldn’t really have been space to write TR P VIIII while keeping Severus’s titulature within the first two lines. But equally, the stonemason could have saved a little space on line 4 by writing TR P IV there.

Here is Septimius Severus again, in the same year, with the same co-regent. It’s even from the same site — Bu-Njem in Tripolitania! Each inscription sat above one of the fort gateways.

But this time, the stonemason has used the “additive” method for Severus’ tribunician power: TR POT VIIII. Was it simply down to choice?

I have a nagging suspicion that, where monumental inscriptions are concerned, the stonemason could use his own discretion to lay out a nicely balanced panel. Sometimes VIIII would work, and sometimes it had to be IX.

Official Roman documentation

But there is one genre that, we know, was governed by the imperial chancellery at Rome: namely, military diplomas. These were strictly legal documents, produced under the emperor’s auspices in Rome and conferring certain legal privileges upon the bearer. Their wording would have been carefully controlled. Can we see a pattern governing the representation of the number 9 here?

This is a diploma (CIL XVI, 95) issued by the emperor Antoninus Pius on 29 February AD 148 to the men “who served in my praetorium in the ten cohorts” — in other words, the Praetorian Guard.

In the list of ten numerals denoting the ten cohorts (the long list of numbers highlighted here), it can clearly be seen that the forms IV (for four) and IX (for nine) have been used.

Interestingly, on line 5, at the end of Antoninus Pius’ titulature, he is COS IIII (“having been consul four times”)! The scribe has used the “additive” method, although the “subtractive” method was used in the numbering of the Praetorian cohorts.

For comparison, here (below) is a section of a badly bent and twisted diploma, issued (again, to men of the ten Praetorian cohorts) by the emperors Valerian and Gallienus on 7 January AD 254 (CIL XVI, 155). It is, incidentally, one of the last known diplomas.

The ten cohorts are numbered on the third line, and it can clearly be seen that, by contrast with the Antonine scribe above, Valerian and Gallienus’ scribe has employed the “additive” method for both IIII and VIIII.

(It even seems that the scribe preparing the outer text of the diploma may often have had different ideas from the scribe responsible for the inner text, as RMD III, 185 shows.)

Method in the madness?

Perhaps a more detailed critique would uncover a hidden method, but it certainly looks as if VIIII and IX were used interchangeably throughout Roman imperial history. Neither one was preferred over the other.

The whole question arose, I suppose, in connection with my book The Fate of the Ninth, and the original comment was presumably based on the supposition that the Ninth Legion called itself legio VIIII Hispana, by preference. So, is this true? Can this be demonstrated?

Probably the most famous inscription of the Ninth Legion is the building slab that presumably graced one of the fortress gateways at York (RIB 665). (There’s a better image on p. 21 of my book.)

Here, on the bottom line, the legion calls itself LEG(io) VIIII HI[SP(ana).

It appears with the same style of numbering (but without Hispana) on the tombstone of Lucius Duccius the signifer (RIB 673, on p. 18 of my book), and also on the career inscription of Titus Aninius Sextius Florentinus (CIL III, 14148¹⁰), the memorial to Lucius Aemilius Karus (CIL VI, 1333), and the dedication to Lucius Novius Crispinus Martialis Saturninus at Lambaesis (CIL VIII, 2747). A couple of early tombstones also use this form.

By contrast, it appears as LEG(io) IX Hisp(ana) on the career inscription of Lucius Burbuleius Optatus Ligarianus (CIL X, 6006), the career inscription of Lucius Roscius Celer (CIL XIV, 3612), and one or two early tombstones.

Perhaps (it might be objected) those men chose their own way of representing the legion’s title. Surely the legion’s own gateway inscription should decide the issue?

However, we should be wary of such an argument, remembering the gateway inscriptions from Bu-Njem (above). I wonder — if we found the inscription from a different gateway of the York fortress, would the numeral necessarily take the same form?

Finally, we may consider the legion’s own tile-stamps, produced at their own tile-works. Shouldn’t they carry the official form of the legion’s numeral? The tile from Nijmegen (illustrated on p. 122 of my book) certainly reads LEG VIIII. But the tiles from northern England (two of them are illustrated here) use the “subtractive” method.

In any case, the Roman soldiers’ gag doesn’t seem quite so funny with “Vee-Aye-Aye-Aye, Vee-Aye-Aye-Aye-Aye, Ecks”.