One of the more exciting textual finds of the last decade is undoubtedly the identification by Gunther Martin and Jana Grusková of a series of fragments written in classicizing Greek describing the barbarian invasions of the mid-third century a.d.Footnote 1 The fragments were preserved in what appears to be an eleventh-century palimpsest bound into Codex Vindobonensis Hist. gr. 73, which resides in the collection of the Austrian National Library in Vienna.Footnote 2 With the aid of digital technology, Martin and Grusková have been able to read and transcribe the longest of these fragments, a continuous passage preserved on two leaves, which was published in 2014.Footnote 3 The fragment under discussion in this paper sheds new light on historical events during the sole reign of the emperor Gallienus (a.d. 260–8), about which much still remains uncertain owing to the generally unsatisfactory nature of the extant literary sources.Footnote 4 In particular, it contributes significantly to the debate on the chronology and extent of the Gothic invasions of Greece in the 260s.Footnote 5 The fragment describes an attack on the Danubian and Greek provinces by a host of ‘Scythians’ and the attempts of the inhabitants of Thessalonica and the province of Achaia to repel the incursion. It is particularly notable for the detail it provides of the Greek resistance, including the fortification of the pass at Thermopylae against the barbarians, the election of three generals to supervise the Greek war effort, and the beginnings of a rousing speech in oratio recta delivered by a leader named Marianus.
Martin and Grusková have identified the author of this text as the Athenian historian P. Herennius Dexippus, who is best known for writing the Scythica, a now-fragmentary account of Rome's Gothic wars in the middle decades of the third century.Footnote 6 The style and content of the Vienna fragment make the identification a compelling one. However, some of Martin and Grusková's conclusions on prosopographical and contextual matters are less secure, including their identification of the general Marianus with Gallienus’ dux Aurelius Marcianus. Most importantly, Martin and Grusková argue that the fragment describes the events of the famous ‘Herulian invasion’ of 267/268, late in the reign of the emperor Gallienus. However, this dating is called into question when the Vienna manuscript is compared with parallel accounts of the 250s and 260s in the Historia Augusta, Zosimus, and Zonaras.
The aim of this article is to examine the text and significance of the new Vienna manuscript, concentrating primarily on historical matters. In Section I, we present a Greek text and English translation, based entirely on Martin and Grusková's published transcription of the palimpsest.Footnote 7 This is followed in Section II by a discussion of the authorship of the text, in which we endorse Martin and Grusková's identification of the author with Dexippus, while also considering other possible authorial contenders. Section III examines the three individuals named in the fragment as generals elected to supervise the defence of Greece: Marianus, Philostratus, and Dexippus. We shall argue that the identity of Marianus is crucial to the dating of the Scythian invasion described in the text, which is the subject of Section IV. We propose that the attack described in the fragment actually took place in the early 260s, several years before the ‘Herulian invasion’. When placed in this context, the Vienna manuscript makes a notable contribution to our understanding of historical events in the Balkans and Greece in the reign of Gallienus, as well as the manner in which they were portrayed by Dexippus.
I TEXT AND TRANSLATION
Text
(NB vertical lines denote line breaks in the MS.)
(1) … <Θρᾴ>|κηνFootnote 8 καὶ Μακεδονίαν, καὶ ἐληίζοντο τὴν αὐτό|θι γῆν σύμπασαν. καὶ ἐπὶ τούτῳ τῇ Θεσσαλο|νικέων πόλει προσβαλόντες, ἀθρόοι ἐπεί|ραζον αὐτῆς τὴν ἅλωσιν· ὡς δὲ οἵ τε ἀπὸ| τοῦ τείχους εὐρώστως ἠμύνοντο, πολυχειρίᾳ| τὰς τάξεις ἀμύνοντες, καὶ προυχώρει οὐδὲν ἐς| ἐλπίδας, λύουσι τὴν πολιορκίαν. (2) καὶ γνώ|μη <ἡ> πλείστη <ἦν>Footnote 9 ἐπί τε Ἀθήνας καὶ Ἀχαιί|αν ὁρμηθῆναι τῷ στρατῷ, δόξῃ τῶν ἐν τοῖς| Ἑλληνικοῖς ἱεροῖς χρυσῶν καὶ ἀργυρῶν ἀνα|θημάτων καὶ ὅσα πομπεῖα· πλουσιώτατον |γὰρ ταύτῃ τὸ χωρίον εἶναι ἐξεπυνθάνοντο.| (3) ἐπεὶ δὲ ἐς τοὺς Ἕλληνας ἐξηγγέλθη ἡ ἔφοδος τῶν| Σκυθῶν συνήεσαν ἐς Πύλας, καὶ κατὰ ταὐ|τοθι στενὰ τῶν παρόδων ἐξείργειν αὐτοὺς ὥρ|μηντο. ἔφερον δὲ οἱ μὲν δοράτια, οἱ δὲ πελέ|κεις,Footnote 10 οἱ δὲ ξύλα κατακεχαλκωμένα καὶ σεσι|δηρωμένα ἄκρα, καὶ ὅπως ἑκάστῳ ὁπλίσασθαι| δυνατὰ ἦν. καὶ ἐπειδὴ ἠθροίσθησαν τό τε διατείχισμα ἐξετείχισαν καὶ τῇ φυλακῇ προ|σεῖχον σπουδῇ. (4) ἐδόκει δὲ τὸ χωρίον καὶ ἄλλως| ἀσφαλέστατον εἶναι, οἷα δὴ τῆς ὁδοῦ διὰ δυσχωρί|αν στενῆς οὔσης καὶ ἀπόρου, ἣ φέρει ἐπὶ τὴν| εἴσω Πυλῶν Ἑλλάδα· παρατείνουσα γὰρ ἐπὶ μή|κιστον ἡ ἐπ’ Εὐβοιας θάλασσα τά τε ἀγχοῦ τῶν| ὀρῶν <δάπεδα>Footnote 11 δυσεμβολώτατα διὰ πηλὸν| ἐργάζεται, καὶ ἐπιλαμβάνουσαFootnote 12 <ἐπὶ> τούτοις ἡ Οἴτη| τὸ ὄρος ********α πεζῇ τε καὶ ἱππικῇ| διὰ τῆς ἐγγύτητος τῶν πετρῶν ἀπορώτατον ἐρ|γάζεται τὸ χωρίον. (5) στρατηγοὶ δὲ τοῦ παντὸς πολέ|μου αἱρετοὶ ἀπεφάνθησαν ὑπὸ τῶν Ἑλλήνων·| Μαριανός τε ὃς δὴFootnote 13 προαιρεθεὶς ἦν ἄρχειν τῆς| Ἑλλάδος ἐκ βασιλέως τῆς ἐντὸς Πυλῶν,| καὶ ἐπὶ τῷδε Φιλόστρατος τε Ἀθηναῖος ἀνὴρ λόγους| καὶ γνώμην κρατίστος, καὶ Δέξιππος ὃς δὴ πέμ|πτον εἶχε τῆν ἐν Βοιωτοῖς ἀρχήν. (6) ἐδόκει τὸ| συμφορώτατον εἶναι καὶ θαρσύναι αὐτοὺς λόγῳ| καὶ ἐς ἀνάμνησιν τῆς τῶν προγόνων ἀρετῆς ἀγαγ|εῖν ὡς ἂν προθυμότερον τοῦ παντὸς πολέμου| ἅπτοιντο καὶ μὴ ἀπαγορεύοιεν πρὸς τὴν φυλα|κὴν <τήν τε ἀθροωτάτην>Footnote 14, ἤν τε καὶ χρόνιος τοῖς| ἐναντίοις ἡ πείρασις τοῦ τειχίσματος γίγνηται. καὶ συνελθόντων Μαριανός, ᾧ κατ’ ἀξίωσιν εἰπεῖν| ἀπεδόθη, ἔλεξε τοιάδε· (7) ὦ Ἕλληνες, ἥ τε πρό|φασις τῆς σωτηρίας ἡμῶν καθ’ ἣν ἠθροίσθητε καὶ| τὸ χωρίον ἐν ᾧ παρατάσσεσθε, ἱκανώτατα ἄμ|φω μνήμην ἀρετῆς παρασκευασαι· οὐδὲ γὰρ οἱ πρό|γονοι ὑμῶν ἐν τῷδε μαχόμενοι ἔσφηλάν ποτε τῆς| ἐλευθερίας τὴν Ἑλλάδα· ἔν τε γὰρ τοῖς Μηδικοῖς| ἄριστα ἠγωνίσαντο καὶ ἐν πολέμῳ τῷ Λαμϊακῷ| κληθέντι καὶ αὖ ὅτε Ἀντίοχον τὸν ἐξ Ἀσίας δυνά|στην ἐτρέψαντο ‛Ρωμαίοις ἄρχουσιν ἤδη συνιστάμενοι· (8)| ὥς που συνκληρωθὲν Ἕλλησι κατὰ τὸ δαιμόνιον| τοῖς κατὰ τῶν βαρβάρων ἀγῶσιν ἐν τῷδε τῷ χώρῳ| ἐνευτυχήται, ὑμῖν δὲ οἰκεῖαι τῶν πολέμων ὑ|ποθέσεις πάλαι ἐξήκουσιν. ἀλλὰ θαρρεῖν ἐστι τῇ τε| ὑμετέρᾳ αὑτῶν παρασκευῇ καὶ τοῦ χωρίου τῇ ἰσχύϊ,| καθ’ ἣν κἂν ταῖς προτέραις ἐφόδοις φοβεροὶ τοῖς| πολεμίοις ἐφάνητε, ἀφ’ ὧν καὶ τὰ μέλλοντά μοι οὐ|κ ἀνέλπιστα εὖ γιγνώσκοντι, ὡς εἰς ἄμεινον α*| * * *
Translation
(1)[(they invaded?) … Thra]ce and Macedonia, and plundered the entire countryside therein. And then, making an assault upon the city of the Thessalonians, they tried to capture it as a close-packed band. But since those on the walls defended themselves valiantly, warding off the battle columns with the assistance of many hands, and as none of the Scythians’ hopes came to pass, they abandoned the siege. (2) The prevailing opinion of the host was to make for Athens and Achaia, envisioning the gold and silver votive offerings and the many processional goods in the Greek sanctuaries: for they learned that the region was exceedingly wealthy in this respect. (3) When the approach of the Scythians was reported to the Greeks, they gathered at Thermopylae, and set about <the task of> blocking them from the narrow passes there. Some carried small spears, others axes, others wooden pikes overlaid with bronze and with iron tips, or whatever each man could arm himself with. And when they came together, they completely fortified the perimeter wall and devoted themselves to its protection with haste. (4) And it seemed that the area was otherwise very secure, since the road which led to Greece beyond the gates was narrow and impracticable on account of the harsh terrain. For the Euboean Sea, at its greatest extent, stretches up to the flat lands near the mountains and makes them most difficult to access on account of the mud, and adjacent to these extends Mt Oeta [, which …] on account of the closeness of the rocks, makes the place almost impassable for both infantry and cavalry. (5) The generals elected for the entire war were proclaimed by the Greeks: first, Marianus, who had been chosen previously by the emperor to govern Greece inside the Gates; in addition to him, Philostratus the Athenian, a man mighty in speech and thought; and also Dexippus, who was holding the chief office among the Boeotians for the fifth time. (6) It seemed that the most prudent course was to encourage the men with a speech, and to recall the memory of their ancestors’ valour, so that they would undertake the entire war with greater heart and not give up either during an extended period of watch, or during an attempt on the wall, if such an attempt were to take place at some point in time. When the men had gathered together, Marianus, who had been given the responsibility of addressing them on account of his status, spoke as follows: (7) ‘O Greeks, the occasion of our preservation for which you are assembled and the land in which you have been deployed are both truly fitting to evoke the memory of virtuous deeds. For your ancestors, fighting in this place in former times, did not let Greece down and deprive it of its free state, for they fought bravely in the Persian wars and in the conflict called the Lamian war, and when they put to flight Antiochos, the despot from Asia, at which time they were already working in partnership with the Romans who were then in command. (8) So perhaps it may be good fortune, in accordance with the daimonion, that it has been allotted to the Greeks to do battle against the barbarians in this region (indeed your own principles of fighting the wars have turned out to be valid in the past). But you may take confidence in both your preparation for these events and the strength of the region — as a result of which, in previous attacks you seemed terrifying to the enemies. On account of these things future events do not appear to me not without hope, as to better …’
II AUTHORSHIP
The style and content of the Vienna fragment suggest that the author of this text dealt with the Gothic invasions of the mid-third century in considerable detail. Martin and Grusková have adduced significant and convincing positive arguments in favour of the author being the Athenian historian P. Herennius Dexippus. These are the content of the fragment, the use of a speech in oratio recta, and the style, which they regard as particularly reminiscent of Thucydides.Footnote 15 We agree wholeheartedly with their identification, and will add further reasons of our own in favour of Dexippan authorship.
On the matter of the content, there can be no doubt that the new text reads like a narrative written by a near contemporary invested in the situation he describes. Dexippus wrote two works that dealt with events of the mid-third century: the Chronicle and the Scythica. According to the waspish remarks of Eunapius of Sardis, the former was little more than a jejune annalistic narrative.Footnote 16 The Scythica, on the other hand, was a detailed history of the Gothic invasions that beset the Balkan and Anatolian provinces of the Roman Empire in the mid-third century, and probably culminated with the victories of Aurelian.Footnote 17 The work was noteworthy for its elaborate siege narratives and long speeches, which attracted the attention of the tenth-century compilers of the Excerpta Constantiniana. Footnote 18 As with the surviving excerpts of the Scythica, the new fragment is clearly part of a much larger narrative history of considerable scope. In recounting the fortification and defence of Thermopylae, the author devoted time to describing the geographical situation, the weaponry of the defenders, and the different generals, before launching into an extended and impassioned pre-battle exhortation. We do not know of any other history of the mid-third century written in Greek that would have the scope to describe events of Gallienus’ reign in such detail, apart from Dexippus’ Scythica.Footnote 19
The concordances between the author of the Vienna fragment and Dexippus run deep. Both are cautious in avoiding Roman terms or Latinisms.Footnote 20 When the former speaks of Achaia, he is referring clearly to the region of Achaia in the Peloponnese, not to the Roman province. The Roman province of Achaia is, instead, referred to as ‘Greece inside the Gates’ (τῆς Ἑλλάδος … τῆς ἐντὸς Πυλῶν) and the emperor's representative, Marianus, is given no official title. There is also a clear similarity in the way in which the Vienna fragment and Dexippus conceive of the relationships between the Greeks and the Roman imperial authorities. Neither author ignores Roman emperors or administrators, but much of their attention is drawn to regional or civic groupings, which fight in partnership with the Roman authorities.Footnote 21 In our text, the Roman emperor's representative Marianus is portrayed as working in tandem with the Greek generals, Dexippus the Boeotian and Philostratus the Athenian.
The content and themes of Marianus’ speech are firmly focused on the valour and achievements of the Greeks themselves.Footnote 22 His emphasis on Greek freedom recalls the Scythica's famous ‘Speech of Dexippus’, which was delivered to rouse the Athenians to arms after the sack of their city by the Heruli.Footnote 23 In the Vienna fragment, Marianus recalls a series of notable occasions on which the Greeks had displayed their fighting spirit in the past, during the Persian wars, the Lamian war, and the conflict with Antiochus.Footnote 24 These sentiments have a greater impact precisely because they are delivered by a Roman representative admiring the past achievements of the Greeks. Marianus also alludes to the necessity of co-operating with the Roman authorities in his reference to the war against Antiochus, which was conducted under Roman leadership (‛Ρωμαίοις ἄρχουσιν ἤδη συνιστάμενοι).Footnote 25 Even in the ‘Speech of Dexippus’, the Athenians are advised to join together with the imperial fleet to attack the Heruli, for it is only then that they will be the most effective fighting force.Footnote 26 The conception that the Greeks and Romans are partners in the battle against the invading Scythians thus represents a striking consonance between our text and the Scythica.
In addition to the content of the fragment, there are several stylistic clues that can assist in identifying the author of the fragment. Martin and Grusková have drawn attention to his tendency to use stock commonplaces or banal phraseology to describe people and events, which is consonant with what we know of Dexippus’ prose.Footnote 27 There may also be several Thucydidean echoes in the fragment.Footnote 28 Dexippus’ debt to Thucydides is well noted in scholarship: in an oft-cited passage Photius remarked on Dexippus’ Thucydidean style, which was particularly apparent in the Scythica.Footnote 29 In this vein, the author's use of the word πολυχειρία (‘many hands’) may owe something to a passage from Thucydides’ narrative of the siege of Plataea.Footnote 30
Other verbal parallels point even more clearly to Dexippus as the author. At the conclusion of his account of the siege of Thessalonica, the Vienna text uses the expression καὶ προυχώρει οὐδὲν ἐς ἐλπίδας (‘and none of their hopes came to pass’), which finds a parallel in Book 6 of Thucydides.Footnote 31 Yet even more significant is the fact that the near-exact phraseology resurfaces in Dexippus’ narrative of another failed siege, that of Pamphylian Side.Footnote 32 Furthermore, a variation of this formulaic phrase may also be seen at the end of Dexippus’ narrative concerning the embassy of the Juthungi to the emperor Aurelian.Footnote 33 It is possible to identify one final stylistic connection between the new text and the Scythica. The end of the surviving portion of Marianus’ speech includes a mention of the daimonion, or ‘the heavenly power’ (τὸ δαιμόνιον). Although other imperial Greek historians make reference to the daimonion in their works, it was by no means a universally applied commonplace. It is, however, a noteworthy feature of Dexippus’ prose — especially in his speeches.Footnote 34
Such positive arguments are difficult to make about other possible authorial candidates. Although we know the names of several historians of the mid-third century, precious little is actually known about their works.Footnote 35 Indeed, it seems likely that the histories of such authors as Philostratus the Athenian, Eusebius of Nantes, and Nicostratus of Trapezus perished before the advent of the tenth and eleventh centuries — a period which was so crucial for the preservation of antique authors and to which the Vienna manuscript belongs.Footnote 36 Of later authorities, the sixth-century historian Peter the Patrician is known to have covered the period of the third century in some detail.Footnote 37 However, he must be dismissed on stylistic grounds since, unlike the author of the Vienna fragment, his works contained many Latinisms. For example, in a famous fragment dealing with events in the reign of Gordian III, the governor of Moesia Inferior, Tullius Menophilus, is described as δοὺξ … Μυσίας and the emperor himself is referred to as αὐτοκράτωρ, whereas the Vienna author uses βασιλεύς for the unnamed emperor in the fragment.Footnote 38
Therefore, we may be confident in accepting Martin and Grusková's identification of the author of the fragment as P. Herennius Dexippus. With the question of authorship largely beyond dispute, we may now turn to the interpretation of the fragment's content, in particular to the key elements of prosopography and dating.
III PROSOPOGRAPHY
The Vienna manuscript names three individuals selected by the Greeks to serve as their generals: Marianus, Philostratus, and Dexippus. Martin and Grusková proposed that Marianus, the emperor's representative in Greece, should be identified with the dux Mar<c>ianus of HA Gallieni duo 6.1. They suggested that Philostratus was a prominent sophist, perhaps even the historian of the same name. Finally, they argued that Dexippus could only be the Athenian archon and historian P. Herennius Dexippus.Footnote 39 All three individuals warrant further discussion and consideration, especially since they have a significant bearing on the dating and interpretation of the new fragment.
Marianus
Martin and Grusková have identified this man as Marcianus, one of Gallienus’ generals, who was known to have fought against the Goths in the late 260s.Footnote 40 They have also suggested that the MS reading Μαριανός should be emended to Μαρκιανός accordingly.Footnote 41 Their argument rests on two points. Firstly, that the Marianus found in the Vienna manuscript is identical with the dux Marcianus attested in Gallieni duo 6.1. Secondly, they have suggested that the copyist of the Vienna MS substituted a familiar name (Marianus) for an unfamiliar one (Marcianus).Footnote 42 In contrast, we will propose that the text should not be emended, and that Marianus is a previously unknown Roman official, probably a governor of the province of Achaia.
The dux Marcianus initially appears to feature in the Historia Augusta in three key events: (i) he defeats a Gothic attempt to invade Achaia, which occurs early in Gallienus’ sole reign in the HA's account (HA Gall. 6.1); (ii) he leads the Roman forces against the Scythians, after they have been repelled from Athens by the historian Dexippus (the so-called ‘Herulian’ invasion of 267/8) (HA Gall. 13.10; Claud. 6.1, 18.1); and (iii) he participates in a plot to murder Gallienus in 268 (HA Gall. 14.1, 14.7, 15.2). This Marcianus is usually identified with Aurelius Marcianus, one of Gallienus’ generals.Footnote 43
Aurelius Marcianus was honoured with a statue by the city of Philippopolis, which acclaimed him as its ‘saviour’ in the inscription on its base. The text of this inscription reveals that he was a vir perfectissimus, protector and dux of Gallienus.Footnote 44 The identification of this man with the Marianus of the Vienna manuscript would seem secure, were it not the fact that the first appearance of Marcianus in Gallieni duo 6.1, is actually itself the product of an emendation.
It is worth citing the relevant passage of the Historia Augusta in context, as it appears in Hohl's Teubner edition (HA Gall. 5.6–6.1):Footnote 45
… occupatis T<h>raciis Macedoniam vastaverunt, Thessalonicam obsederunt, neque usquam quies mediocriter sal[u]tem ostenta est. quae omnia contemptu, ut saepius diximus, Gallieni fiebant, hominis luxuriosissimi et, si esset securus, ad omne dedecus paratissimi. pugnatum est in Ach<a>ia Mar<c>iano duce contra eosdem Gothos, unde victi per Ach<a>eos recesserunt.
… after Thrace had been seized, they [sc. the Goths] laid waste to Macedonia, and besieged Thessalonica; nowhere was there the slightest respite offering up salvation. All these things happened as a result of (as we have mentioned frequently) Gallienus’ contempt for the situation, since he was a totally dissolute fellow, always prepared to undertake any shameful act, as long as he was safe. The battle in Achaia was fought by the general Mar<c>ianus against these very same Goths, and after their defeat, they retreated through the lands of the Achaians.
The correspondence between this passage in the Historia Augusta and the events recounted in the Vienna manuscript is striking. Both works describe the Gothic/Scythian sack of the provinces of Thrace and Macedonia, the attempted siege of Thessalonica, and the incursion into Achaia. The decision to emend the name of the dux from ‘Mariano’ to ‘Mar<c>iano’ in Gallieni duo 6.1, based on the appearance of Marcianus in later events in the Gallieni duo and Divus Claudius, has become entrenched editorial practice, originating in the editio princeps of Salmasius in 1620. But it is important to note that ‘Mariano’ is the reading of the authoritative Codex Palatinus 899 (Ρ), and its derivatives, as well as all the so-called Σ-group MSS.Footnote 46 We would argue that the emendation is now unnecessary, since both the Historia Augusta and the Vienna fragment describe the Roman leader as Marianus/Μαριανός.Footnote 47 He must, therefore, be disassociated from the Aurelius Marcianus, protector, tribunus and dux, who fought against the Scythians in 267/8, before he conspired to murder Gallienus in 268. This has significant ramifications for the dating of the events described in the manuscript, which we will discuss in detail in the next section.
For Martin and Grusková, however, it would seem to be a coincidence that both the Historia Augusta MSS and the Vienna fragment should name the Roman commander Marianus. This has led them to justify the emendation on the grounds of nomenclature. Since Marianus is a common name in the Byzantine Middle Ages, they argue that this resulted in the eleventh-century scribe committing a simple error, substituting a familiar name (Marianus) for an unfamiliar one (Marcianus).Footnote 48 It is true that Marianus was a common name during the period when the Vienna manuscript is believed to have been copied. However, it should be added that the name Marcianus was also not unknown during the tenth or eleventh centuries, so such a solution is inconclusive.Footnote 49 Perhaps more importantly, there is nothing implausible about Marianus being the name of the Roman commander. Marianus was a cognomen used by members of the senatorial and equestrian orders in the imperial age.Footnote 50 These included the third-century governor of Pontus, [S]ept(imius) Maria[nus], and P. Vibius Marianus, procurator and praeses of Sardinia.Footnote 51 There is therefore no justification for emending the text on the grounds of nomenclature.
Indeed, it is probable that Marianus was actually a provincial governor. The text states that ‘he had been chosen previously by the emperor to govern Greece inside the Gates’ (προαιρεθεὶς ἦν ἄρχειν τῆς Ἑλλάδος ἐκ βασιλέως τῆς ἐντὸς Πυλῶν). This expression accurately describes the boundary of the Roman province of Achaia in the mid-third century.Footnote 52 The northern boundary of the province lay directly north of Thermopylae, probably at the Gorgopotamos gorge on the eastern face of Mt Oeta.Footnote 53 The road station of Thapedon, which served as a suitable border crossing between the Roman provinces of Achaia and Macedonia, has been plausibly located at the base of the mountain.Footnote 54 We propose, therefore, that Marianus was a senator of praetorian rank, who had obtained the proconsulship of Achaia for the year.Footnote 55 The proconsul traditionally had limited auxiliary forces at his disposal, which accounts for the reliance on local men to defend the pass at Thermopylae in the Vienna text.Footnote 56 The fact that the emperor chose Marianus as his representative does not preclude him from being a proconsular governor, although the proconsular position was traditionally selected by sortitio. Gubernatorial appointments to ‘public’ provinces made extra sortem are well attested throughout the imperial period, and such practices seem to have become particularly prevalent in the mid-third century.Footnote 57 Furthermore, the fact that the Historia Augusta gives Marianus the title of dux is not a serious impediment to this identification, since its author likewise routinely applies incorrect titles to senatorial and equestrian officers and administrators.Footnote 58 Given the geographical description of Marianus’ sphere of responsibility, the proconsulship of Achaia remains his most likely official post.
Philostratus
Philostratus is described as ‘mighty in speech and thought’ (λόγους καὶ γνώμην κρατίστος), implying that he was famed for his particular philosophical, literary, or oratorical prowess. This suggests that he could be identified with a little-known historian of the late third century, Philostratus the Athenian, mentioned above in Section II.Footnote 59 A second promising candidate is L. Flavius Philostratus, of the deme Steiria, who is attested as archon in Athens in the mid-third century, probably in 255/6.Footnote 60 He may well be identical with the historian, but this is far from certain, given the ubiquity of the name Philostratus in Greece, and most scholars have exercised caution.Footnote 61
The Flavii Philostrati of Steiria were a politically distinguished Athenian family in the third century. Their most notable member was L. Flavius Philostratus, who was a hoplite general c. 205, and possibly a prytanis as well.Footnote 62 He is best known, however, as the Philostratus who wrote the Lives of the Sophists, the Life of Apollonius of Tyana, and Eikones. Footnote 63 This Philostratus was well connected: he was an associate of Julia Domna, and of the consular Antonius Gordianus, to whom he dedicated the Lives of the Sophists.Footnote 64 His presumed son, L. Flavius Philostratus the younger, is attested as a hoplite general c. 225.Footnote 65 This means that the Philostratus who was archon in 255/6 was probably his grandson.Footnote 66 The family tree of the author of the Lives of the Sophists, and especially the assignment of particular literary works to his various descendants, is a well-known prosopographical conundrum, which lies beyond the scope of this paper.Footnote 67 We do know that one of Philostratus’ grandsons wrote a second Eikones, and he could possibly be the archon of 255/6.Footnote 68 It was not unheard of for sophists and rhetoricians to turn their hands to history, as in the case of Dio of Prusa's Getica, though the fact that a historical work is not ascribed to him in the Suda's account of the Philostrati might count against the identification of Philostratus the historian with the author of the second Eikones.Footnote 69
On current evidence, the Philostratus of the Vienna fragment could conceivably be identified with: (a) Philostratus the Athenian historian; (b) L. Flavius Philostratus, archon of 255/6; or (c) Philostratus, author of the second Eikones. They could all be the same man, or they could all be different men. What we can say is that our Philostratus numbered among a class of Greek notables, famed for literary and rhetorical talents, who held prominent local office and who — like Herennius Dexippus himself — took a leading rôle in mounting the defence of Greece in the face of barbarian invasions.
Dexippus
The final figure is Dexippus, who ‘was holding the chief office among the Boeotians for the fifth time’ (ὅς δὴ πέμπτον εἶχε τῆν ἐν Βοιωτοῖς ἀρχήν). Martin and Grusková's suggestion that he should be identified with the historian Dexippus, who was an Athenian, is unlikely.Footnote 70 Instead, he was probably a member of an eminent Boeotian family, either identical with, or related to, a certain Cn. Curtius Dexippus from Chaeronea.Footnote 71 Curtius Dexippus is known to have erected a commemorative statue of his mother, Flavia Lanica, a priestess in several local cults.Footnote 72 Although this monument was initially dated to the early third century, Fossey has assigned it to the mid- to late third century on the basis of letter forms in the inscription.Footnote 73 This inscription records that when the monument was erected, Curtius Dexippus was Boeotarch for the third time, high priest of the emperors for life, and logistes (curator rei publicae) of the city of Chaeronea.Footnote 74 The curatores rei publicae could either be senators or equestrians from outside the community, or members of the local aristocracy, as was the case with Curtius Dexippus.Footnote 75 The revival of the office of Boeotarch in the Roman imperial period was possibly inspired by other Greek koina who had similar offices (such as Macedoniarch or Asiarch), and it has been argued that it was largely a religious and ceremonial post at this time.Footnote 76 Having held this position five times, the Dexippus of the fragment was undoubtedly one of the most prominent Boeotians of his era, and perhaps the leading citizen of Chaeronea.
The ‘Greeks’
All three men named in the fragment — Marianus, Philostratus, and Dexippus — are said to have been elected generals for the war against the Scythians by ‘the Greeks’. The only Greek political institution in existence in the third century that encompassed both Attica and Boeotia was the Panhellenion. This league of Greek city-states and federations was established in the reign of Hadrian, but is attested epigraphically into the 250s.Footnote 77 Each of the elected generals represented a different region of mainland Greece. The Boeotians, who were represented by Dexippus, were one of the federal members of the Panhellenion, alongside the Achaians, Phocians, and Thessalians.Footnote 78 The fact that Boeotia (and specifically the town of Chaeronea) was in the immediate path of the invading Goths may have ensured that one of their members was elected as a general. Athens, represented by Philostratus, was the capital of the Panhellenion, as well as being the intellectual and cultural centre of Greece. Marianus, as proconsul of Achaia, was based in the provincial capital of Corinth, and thus could be said to represent the Peloponnese.Footnote 79 Hitherto, the Panhellenion is not known to have elected ‘generals’ (strategoi) among their regular roster of officials, but the Scythian invasion was surely an extraordinary circumstance, which required military leadership.Footnote 80
IV DATE AND HISTORICAL CONTEXT
The familiar story of barbarian invasion and local resistance described by the Vienna manuscript not only fits with what we know of Dexippus’ Scythica, but also with our conception of the barbarian invasions during the ‘third-century crisis’ and the reign of Gallienus in particular.Footnote 81 But determining precisely when these events took place in Gallienus’ reign is a more challenging proposition. The view of Martin and Grusková is that the fragment belongs to the period of the Herulian invasions of 267/8.Footnote 82 There are, however, several problems with this dating. The Herulian invasion, which originated on the coasts of the Black Sea, is described in our sources as a predominately seaborne invasion.Footnote 83 In the new fragment, there is no mention of the Scythians being supported by an armada of any description. Furthermore, according to Zosimus, the sack of Athens by the Heruli actually preceded the siege of Thessalonica, which was lifted only by the timely appearance of imperial reinforcements.Footnote 84 In the Vienna fragment, however, the inhabitants of Thessalonica secure the safety of the city by their own force of arms, suggesting it should be identified with an earlier siege of the city. Finally, and perhaps most significantly, nowhere in the tradition concerning the Herulian invasion do we hear about the fortification of Thermopylae. Such discrepancies challenge us to consider alternatives. As we shall argue, the evidence leads us to suggest that the events described in the Vienna manuscript refer to an earlier invasion of Greece, which occurred c. 262.
There are grounds for suggesting that Marianus, our Roman governor, was in office in Achaia in 262. It is a rare thing for the later biographies in the Historia Augusta to offer illumination rather than obfuscation, but this most difficult of works may shed light on the dating of the events described in the Vienna fragment. Chs 5.2–6.1 of Gallieni duo describe a series of events which leads up to Marianus’ defence of Greece against the Goths. At the start of this section of the narrative, the author provides us with a verifiable date in the form of the consular pair of the year 262:Footnote 85
Gallieno et Faustiano conss. inter tot bellicas clades etiam terrae motus gravissimus fuit et tenebrae per multos <dies> …
In the consulship of Gallienus and Faustianus, among countless military disasters, there was even a very serious earthquake and darkness for many days…
Just as the MSS of the Historia Augusta are unanimous in their attestation of Marianus as the Roman commander at Gall. 6.1, the consular pair ‘Gallieno and Faustiano’ has the corroboration of P and at least one text in the Σ-group MSS.Footnote 86 In the past, the name of Gallienus’ colleague has often been emended to ‘Fausiano’ on the basis of an inscription from Ostia and the Chronographer of 354, which give his name as Nummius Fausianus.Footnote 87 Yet the publication of an inscription from Thugga in Africa Proconsularis in the 1990s has subsequently made it clear that the consul's name is L. Mummius Faustianus,Footnote 88 thus vindicating the transmission of the name in the Historia Augusta MSS.Footnote 89
The quality of the annalistic data provided by the pseudonymous scriptor of the Historia Augusta is noteworthy. It has long been recognized that the consular dates in the Gallieni duo (1.2, 5.2, 10.1, 12.1), as well as those in the Maximus et Balbinus, are authentic and probably derived from Dexippus’ Chronicle.Footnote 90 But of recent scholars, only Armstrong has accepted the possibility that there was an actual Scythian invasion of Achaia in 262.Footnote 91 The majority of scholars have tended to disassociate the vague catalogue of natural disasters,Footnote 92 plagues, and revolts of Gallieni duo 5.2–5, from the attack on Thessalonica and the invasion of Greece described in 5.6–6.1.Footnote 93 According to Barnes, the author of the Gallieni duo ‘has wantonly transferred the episode to a false date’.Footnote 94 However, this view is built on the assumption that the Marianus named in Gallieni duo 6.1 is in fact the dux Marcianus named at Gallieni duo 13.10 and elsewhere, and that the siege of Thessalonica described at this point occurred c. 268, rather than earlier. The appearance of the Vienna fragment has undermined these assumptions, and such a conclusion is no longer inevitable.
How easily can we reconcile the dating supplied by the Gallieni duo 5.2 with our other evidence? It is worth noting that military activity in the Greek provinces is attested in the early 260s in the Latin chronographical tradition. Jerome notes that in 263, ‘Greece, Macedonia, Pontus and Asia were pillaged by the Goths’ (‘Graecia Macedonia Pontus Asia depopulata per Gothos’).Footnote 95 The so-called Consularia Constantinopolitana, whose material seems to be independent of the Eusebius-Jerome tradition, notes a barbarian incursion that began in 261: ‘many barbarians invaded Roman lands’ (‘hostes multi inruerunt in Romania’).Footnote 96 More tellingly, there are important correspondences with the later Greek historiographical traditions.
Georgius Syncellus, in his Selection of Chronography (Ecloga Chronographica), records the following events:Footnote 97
Ἐπὶ Οὐαλεριανοῦ δὲ καὶ Γαλιηνοῦ πάλιν οἱ Σκύθαι διαβάντες τὸν Ἴστρον ποταμὸν τήν τε Θρᾴκην ἐδῄωσαν καὶ Θεσσαλονίκην ἐπολιόρκησαν τὴν Ἰλλυρίδα πόλιν, οὐδὲν ἄριστον ἐπ’ αὐτῇ δράσαντες τῇ τῶν ϕυλάκων ἀνδρείᾳ. διὰ τοῦτο ταραχθέντες Ἕλληνες τὰς Θερμοπύλας ἐϕρούρησαν τό τε τεῖχος Ἀθηναῖοι ἀνῳκοδόμησαν καθαιρεθὲν ἀπὸ τῶν Σύλλου χρόνων, Πελοποννήσιοι δὲ ἀπὸ θαλάσσης εἰς θάλασσαν τὸν Ἰσθμὸν διετείχισαν, οἱ δὲ Σκύθαι μετὰ πολλῶν λαϕύρων εἰς τὰ ἴδια ἦλθον.
During the reign of Valerian and Gallienus, the Scythians again crossed the Danube and ravaged Thrace and besieged the Illyrian city of Thessalonica. But because of the courage of those guarding the city, they achieved nothing great against it. Thrown into panic because of this, the Greeks stationed guards at Thermopylae, and the Athenians rebuilt the wall that had been taken down from the time of Sulla. And the Peloponnesians fortified the Isthmus with a wall extending from sea to sea. And the Scythians returned home with much booty. (trans. Adler and Tuffin, slightly adapted).
The events of this passage closely follow those described by the Vienna fragment and the Historia Augusta.Footnote 98 The Scythians follow the same route through Thrace to besiege the Macedonian city of Thessalonica, where the inhabitants successfully repel them. In response to this attack, the Greeks decide to strengthen their defences at Thermopylae in order to prevent the Scythians’ progress further into Greece. Syncellus then proceeds to give further details, which lie beyond the chronological scope of the surviving narrative of our new text, namely that the Athenians refortified their city, and the Peloponnesians built a wall across the Isthmus of Corinth. These last details can be found in Zosimus and Zonaras, both of whom describe the refortification of Athens after the previous wall had been dismantled by Sulla, as well as the construction of a wall across the Isthmus of Corinth.Footnote 99 The rebuilding of the Athenian walls prior to the Herulian invasion of 267/8 is archaeologically verifiable. The fortifications largely followed the course of the Themistoclean wall, but expanded in the eastern sectors in order to account for Athens’ growth in the second century.Footnote 100 There is no extant archaeological evidence for the fortification of the Isthmus of Corinth, and it has been called into question whether it was actually built.Footnote 101 The crucial fact is that in all three writers, the building of the new walls is a reaction to a Scythian attack on Thessalonica.Footnote 102
Here we encounter a potential hurdle. Scholars have traditionally identified two major sieges of Thessalonica in the literary sources. The second, it is clear, occurred during the Herulian invasion. The first attack appears early, seemingly in the joint reign of Valerian and Gallienus, before Valerian's departure to the East. This is at least how it is presented in the accounts of Zosimus, Syncellus, and Zonaras.Footnote 103 Thus, the contention that the siege occurred in either 253 or 254 has become canonical.Footnote 104 This date is not unproblematic, for the evidence from which it has been derived is slender. Zosimus’ unreliability in terms of chronology is well known.Footnote 105 Therefore, much rests on the interpretation of Syncellus’ testimony. It seems doubtful whether we should regard Syncellus as referring precisely to the years 253–260 when he places events in the reign of ‘Valerian and Gallienus’.Footnote 106 Unlike certain ancient biographers (or modern text books), Syncellus thought of the period from the proclamation of Valerian to the demise of Gallienus as being a joint reign totalling fifteen years.Footnote 107 As such, it is unwise to assume that the passage above fell into the early part of their reign, just because it is introduced that way in the text.Footnote 108 More worryingly, we cannot even assume that Syncellus presented his material in the correct chronological order. Indeed the slipshod nature of Syncellus’ method of composition tells against such assumptions.Footnote 109 It is clear that Syncellus used several sources (or a tradition that drew upon several sources) for his material for the reign of Valerian and Gallienus, following, excerpting, and abbreviating each in turn as it took his fancy.Footnote 110 For Syncellus it was sufficient to place the historical events in the reign of the correct emperor and nothing more.Footnote 111 Therefore, we need not (and perhaps should not) take his narrative of the Persian wars, which follow his account of the Scythian invasion, as the basis for reconstructing the chronology of these events.Footnote 112 One point is clear: in all these accounts of the siege of Thessalonica, including that of the new Vienna text, the barbarians are repelled by the inhabitants of the city. This firmly dissociates it from the later siege in the Herulian invasion, when the enemies leave only after hearing of the approach of the imperial fleet.
The correspondence between the narrative described in the Gallieni duo and the events of the Vienna manuscript, Zosimus, and Syncellus is striking and suggestive, so as to indicate they are describing the same events.Footnote 113 The discussion, so far, may be summarized by the following table:
The Gallienic date for the first Scythian invasion of Greece finds circumstantial support in archaeological, epigraphic, and numismatic evidence. We know, first of all, from the literary sources, that Marianus and his forces were successful in their endeavours, even though the marauding Goths did seize some booty. One would expect such a victory to be commemorated with an imperatorial acclamation for the emperor Gallienus, even though he himself was in Byzantium in 262.Footnote 114 As is well known, there are numerous problems with dating the imperatorial acclamations of Gallienus in the early 260s.Footnote 115 Nevertheless, numismatic evidence indicates that he was named imperator for the sixth time either in 262 or 263, as shown by a gold medallion from Rome that bears the reverse legend PM TR P IMP VI COS V P P.Footnote 116 Since Gallienus’ fifth consulship fell in 262 and his sixth in 264, we can safely date the sixth acclamation to either 262 or 263.Footnote 117 The number of acclamations rose steadily in this period, since the emperor was IMP XII by 265 and IMP XV by 268.Footnote 118 It could be the case that Gallienus’ imperatorial titles increased on an annual basis and are not connected with specific victories.Footnote 119 If, however, they were assumed as a result of the achievements of his generals, then the defeat of the Goths by Marianus and the Greeks provides a plausible reason for one of the imperatorial acclamations in the early 260s.
If Marianus was, as we have argued, a proconsul of Achaia, it should be noted that there is sufficient space for him in the provincial fasti for the years in question.Footnote 120 A certain Valens is attested as proconsul by the Historia Augusta around the year 261, when he is said to have staged a revolt against Gallienus.Footnote 121 One might doubt this governor's historicity, were it not for the fact that he appears in both the Epitome de Caesaribus and Ammianus Marcellinus.Footnote 122 Valens’ revolt was allegedly suppressed by a certain Piso, who was then himself proclaimed emperor.Footnote 123 Ammianus gives Valens the cognomen ‘Thessalonicus’, but does not specify how he earned this name, whereas the HA says Piso earned the name ‘Thessalicus’ after being acclaimed emperor in Thessaly.Footnote 124 The precise chronology of these events is beyond redemption, given the garbled narrative presented by our disparate sources. However, there is no proconsul attested later in Gallienus’ reign, now that Panathenius’ proconsulship, initially dated to 267, has been relocated to the fourth century.Footnote 125 Gallienus might have appointed Marianus proconsul of Achaia in the wake of the mutinies of Valens and Piso, though this remains speculative.
Finally, we may note that a date in the early 260s dovetails neatly with recent archaeological conjecture. The construction of the third-century walls of Athens (the so-called ‘Valerian’ wall) has long been regarded as a response to the siege of Thessalonica and invasion of Achaia in the accounts of Syncellus, Zosimus, and Zonaras. These events, as we have seen, have traditionally been dated to 253/4. Very recently, however, it has been proposed that the ‘Valerian’ walls were actually built in the sole reign of Gallienus.Footnote 126 This conclusion is suggested by the vast outlay of new coinage minted in Gallienus’ reign, which was necessary to pay for the construction of the walls.Footnote 127 The threat to Greece, and to Athens in particular, probably explains Gallienus’ journey to the city in September/October of 264, which was clearly an expedition of importance, since it was the emperor's first foray outside Rome in several years.Footnote 128
The evidence suggests that the events described by the Vienna fragment took place early in the sole reign of Gallienus. The new text certainly describes the same events as Syncellus, Zosimus, and Zonaras. But rather than dating the siege of Thessalonica and the invasion of Greece to 253/4, as has become the orthodox position, c. 262 seems a much more plausible date. This is suggested by the accurate consular date of 262 in Gallieni duo 5.2, which probably refers to the Athenian archon year of 261/2.Footnote 129 The HA cites this date immediately preceding the reference to the siege of Thessalonica and the invasion of Greece. The name of the Roman commander Marianus, accurately recorded in the Vienna fragment, connects the events described there to Gallieni duo 6.1 (and by extension the Byzantine Greek tradition of Zosimus, Zonaras, and Syncellus). The results of this discussion may be best summarized with the following table:
V CONCLUSION
The relationship between our sources for the late third century is not a topic to be dealt with lightly. The place of Dexippus in the succeeding Greek and Latin historiographical traditions is especially uncertain. The foregoing discussion has identified several important parallels between Syncellus, the Historia Augusta, Zosimus, and the new Vienna fragment, the author of which we have identified as Dexippus, building on the arguments of Martin and Grusková. This leads us to the conclusion that Dexippus was probably the original source for the events described by these later writers, even if his account was mediated through, and perhaps even supplemented by, the accounts of other historians. But did these authors or their source follow the Scythica or the Chronicle? It seems likely that the Vienna fragment is from the Scythica for the reasons outlined in the previous section. Yet Dexippus presumably covered the same events in less detail in the Chronicle as well.Footnote 130 Given what we know of the Chronicle, with its abbreviated style but practice of naming the consuls, it seems likely that this was the source followed by the HA.Footnote 131 The similarities in the accounts of Syncellus, Zosimus, and Zonaras suggest that all three relied on the same source, either directly or indirectly, of which Dexippus’ Scythica was perhaps the ultimate ancestor.Footnote 132 The Vienna manuscript will no doubt play an important rôle in ongoing scholarly investigations of the relationship between the different Greek accounts of the third century, a discussion that lies beyond the scope of this paper.
Martin and Grusková's discovery of the new fragment of Dexippus’ Scythica is especially important for its potential to shed new light on historical events in Greece and the Balkans early in the sole reign of Gallienus. The foregoing investigation of the historical context of the Gothic invasion leads us to propose a new reconstruction of these years.
Some time in late 261 or early 262, a host of Goths invaded and laid waste to the provinces of Thrace and Macedonia. Before long, the barbarians turned their attention to Thessalonica, capital of the Roman province of Macedonia. The valiant actions of the inhabitants of the city meant that their siege was unsuccessful and the Goths set their sights on the province of Achaia. Upon hearing the news of the attack on Thessalonica, the Greeks — probably through the Panhellenion — set about organizing the defence of the ungarrisoned province under the supervision of Marianus, the Roman proconsul of Achaia. Together with Marianus, the Panhellenion placed Philostratus the Athenian and the Boeotarch Dexippus in command of a hastily organized militia.
The Athenians, probably on Marianus’ orders, began rebuilding the walls at Athens, which had been destroyed by Sulla. To pay for these building works, a large quantity of bronze coinage was struck at Athens. By late 262 or early 263, the Goths had invaded the province proper but were turned back by Marianus and his Greek militia. Nevertheless, the Goths left Achaia with much plunder, taken from the Hellenic shrines, including perhaps from the great sanctuary at Eleusis. Some of the Goths, who may have broken away from the army immediately following the failed siege of Thessalonica, or after they ravaged Achaia, then attacked the province of Asia. Yet in Achaia at least, the Gothic threat had passed by the end of 263 and thanks to the efforts of Marianus and the provincials, Gallienus was acclaimed imperator for the sixth time. As a mark of respect to the provincials, Gallienus toured Greece in 264, and held the archonship at Athens.
The author of the HA presents the emperor's visit as a largely frivolous exercise motivated by personal vanity, since Gallienus wished to become an Athenian citizen, archon, and a member of the Areopagus council, and asked to be ‘initiated into all sacred rites’ (‘sacris omnibus interesse’).Footnote 133 But there may well have been legitimate strategic reasons for the journey, which were neglected by the author of the HA in the pursuit of his desire to portray Gallienus as an indolent ruler, neglectful of the Empire's security.Footnote 134 In December of 265, a year after his initial visit, Gallienus sent a letter to the people of Athens concerning the fortification of Eleusis.Footnote 135 Armstrong suggested that Eleusis was fortified because it had been attacked by the Costobocoi in 170.Footnote 136 While the memory of this event undoubtedly lived on in the third century, the new fragment of Dexippus’ Scythica suggests that there were more pressing reasons for such fortifications, namely the barbarians’ desire to make away with the votive offerings and processional goods in the Greek sanctuaries. Gallienus’ visit to Athens in 264 was therefore no whimsical tour by a dissolute emperor, but was motivated by defensive exigencies: he wished to ensure the security of the city following the earlier invasion of c. 262, which resulted in the construction of the ‘Valerian’ wall. If this reconstruction of events is accepted, it suggests that Gallienus may have been a more engaged and proactive emperor than traditionally supposed.Footnote 137 Several decades ago, Armstrong argued that the portrayal of Gallienus in Dexippus’ Scythica was largely a positive one, as the result of personal contact with the emperor during his visit to Athens.Footnote 138 The speech of Marianus in the Vienna fragment certainly does not give the impression that Dexippus viewed the Roman Empire and its representatives as ineffective and indolent. Instead, in his vision, it was co-operation between Greeks and Romans that would ultimately lead to them surviving the dark days of the mid-third century.
