Emperor leo iii the isau.., p.5
Emperor Leo III the Isaurian, page 5
Unfortunately, the development of this thematic reorganisation received little focus from the sources. And even material we do have from the likes of Theophanes may be tarnished by him filling gaps in the record with the provincial layout of his period a century later. What can be gleaned from the surviving material suggests that the transformation from the previous Roman provincial set up into the ‘theme system’ was a gradual one, rather than the result of the choices of a single emperor such as Heraclius or Constans II.13 And even this ‘gradual transformation’ containing notions of planning seems out of place. The ‘evolution’ of this new provincial layout was uneven and haphazard, to the point that the very name ‘theme system’ is a misnomer. There was nothing ‘systematic’ about the ‘theme system’.
The basis for these new provincial units was the retreating imperial field armies. This is reflected in the usage of the word ‘theme’ (θέμα/thema, pl. Θέματα/themata) to describe these new entities. Although its exact etymology is uncertain, ‘theme’ had been used to mean either a military unit or army, including these retreating field armies. And as these military occupations became a lot less temporary, the geographic regions made up of several former provinces that had been attached to these armies in order to support them took on the name of the army that was stationed there14 – the Anatolic theme (Θέμα Άνατολικῶν/Thema Anatolikōn) was born out of the eastern field army of the magister militum per Orientem; the Armeniac theme (Θέμα Άρμενιάκων/Thema Armeniakōn) out of that of the magister militum per Armeniae; and the Thrakesian theme (Θέμα Θρᾳκησίων/Thema Thrakēsiōn) out of that of the magister militum per Thracias. The four earliest themes were rounded out by the Opsikon theme (Θέμα Ὀψικίου/Thema Opsikiou), which was formed from the Obsequium, the imperial retinue of the emperor on campaign.
The commanders of these new thematic armies were the strategoi – ‘generals’ – who held similar civil and military authority to the exarchs.15 The Opsikon theme was slightly different in having a comes as its commander. The soldiers of these ‘thematic’ armies were likely paid initially in cash, only to then be allotted land to live on and to farm to support their families and the army. Such allotments likely allowed the empire to reduce a soldier’s pay while also giving him a vested interest in defending his new home. Service in the thematic armies was to be hereditary, which also reduced the need for expensive military bounties and unpopular conscription.16 This combination of powerful strategoi and local land allotments helped make soldiers more self-sufficient and better looked after by the authorities through the reduction of some bureaucratic delay, expenditure and abuses.17 There are some questions about the exact nature of service within the thematic armies. Could it be that the Opsikon was something of a standing army centred on the emperor or the position of the capital, while the other thematic armies involved slightly less permanent service, perhaps only mustered for specific actions? Certainly, being decommissioned when not on campaign would explain the seeming lack of complaint over reduced pay and rations.18
The size of these thematic armies is difficult to gauge, mainly because the sources record virtually nothing about the size of the Roman army of the seventh or eighth century. The only potential number given for the thematic army as a whole comes in 773, when Theophanes records that Constantine V ‘gathered the soldiers of the themata and the Thrakesians and joined the Optimatoi to the tagmata to a total of 80,000.’19 It is impossible that Constantine V congregated all thematic forces for a Bulgar expedition, stripping the Asian themes bare; however, this figure might represent the size of the entire Roman army, with Theophanes mistakenly associating it with Constantine’s expeditionary force.20 A deduction from pay roll figures has suggested that the Roman army of 641 was around 109,000,21 from which would have to be deducted the 15,000 of the seemingly lost army of Africa, along with further losses against various imperial enemies, to come to a total close to 80,000. Perhaps the only other numerical inference that can be made, beyond references to thematic detachments, comes in the origin of the thematic armies amongst the regional field armies, which we know the sizes for.
Such is the mess of the source material that it is impossible to give definite dates for the creation of the individual themes, although that did not stop some from trying. The appearance of ‘lands of the themes’25 in Theophanes saw some posit the existence of the ‘theme system’ as early as 622; however, this overlooks the pre-existing meaning of ‘theme’ as an army or military unit, rather than a geographic district, not to mention any anachronistic terminological use by Theophanes.
In dating the formation of the ‘theme system’, we are really left with inferences and the dates individual thematic armies are first recorded, some of which are much later than might be imagined. The Armeniac theme, taking in Pontus, Armenia Minor, eastern Paphlagonia and northern Cappadocia from its capital at Amaseia, is the earliest attested in 667. Its partner in the defence against the Umayyad Arabs was the Anatolic theme. It is first mentioned in 669, centred on Amorion and covering lands in Lycaonia, Pisidia, Isauria and much of Phrygia. Despite its central position in Bithynia, western Paphlagonia and parts of Galatia and its headquarters in Nicaea, the Opsikon theme is not mentioned until 680.26
However, it is the fourth of the earliest themes – the Thrakesian theme – that highlights some of the major issues with dating and even situating parts of the ‘theme system’. It would seem logical to think that it was stationed in Thrace, much like the Anatolic and Armeniac themes covered parts of Anatolia and Armenia respectively; however, while the ‘Thrakesian’ theme took its name from the Thracian field army, it was not stationed in Thrace, a disconnect that hinders its identification. It is not mentioned in the sources until 711, when the Thrakesian tourmarches, Christopher, was sent to Cherson by Justinian II.27 This saw it thought that the provinces of Ionia, Lydia and Caria that made up the Thrakesian theme were part of a larger Anatolic theme until being established as a theme of their own in the early-eighth century. However, the Thracianus exercitus recorded in the iussio of Justinian II in 687 is now thought to reference the Thrakesian thematic army, rather than the forces of geographical Thrace, which were instead attached to the Opsikon.28 It is likely then that the Thrakesian theme was created at around the same time as the Anatolic, Armeniac and Opsikon themes.29
Of course, exactly when that was is impossible to know. A more general suggestion for the dating of the themes is that the field armies were probably situated in the regions that form the thematic provinces around them in the last years of Heraclius’ reign (610–641),30 but little in the way of official establishment occurred until that of Constans II (641–668) or later. It could be that lulls in Romano-Arab conflict, such as that which came with the First Fitna (656–661), allowed the empire to give form to its ad hoc provincial and military arrangements. Indeed, Justinian II founded the Hellas theme (θέμα Ἑλλάδος/Thema Hellados) in 688–689 during the lull that came with the Second Fitna (680–692). It must be said that this Helladic foundation date is something of an extrapolation from a lack of direct source attestation. When the soon-to-be emperor Leontios was made strategos of Hellas in 695, this was the first mention of the Hellas theme, with it assumed that Justinian founded it while he was in the region in late 688.
For some added confusion, the Hellas theme was not actually called a ‘theme’ in the sources for another century. The term used for it was a στρατηγία/strategia, suggesting that it was under the command of a strategos, but it is unknown if there was any difference between a strategia and a theme. Could a strategia reflect the lack of strong Roman presence in parts of the territory of this new Hellenic province, with its strategos only having military authority? Could it be something of a stepping stone to a full theme? Or does it just highlight a lack of clarity of thematic terminology, with Hellas established as a theme in c.688 and the sources failing to mention/define it properly? There is also a lack of clarity on the land incorporated in the Hellas strategia/theme, but Boeotia, Attica, Megara, Euboea, Corinthia and the Argolid seem to have been part of it, with its capital at either Athens or Thebes.
As an addendum to the Hellas strategia, it is not clear what Justinian II did with the lands in Thessaly, Macedonia and around the Strymon that his campaign in 688–689 had helped secure. They would become themes in their own right over a century later after more Greek territory was reclaimed, but were they in any way attached to the new Hellas strategia? Were they something of a proto-theme or an even more ill-defined region under the command of the Thessalonian eparch? This would seem to be further evidence of the lack of clear ‘thematic’ organisation even by the late-eighth century.
Justinian II also seems to have extended the use of themes further west as well. Seals from his first reign31 record a certain Salventios serving as strategos, with the Sicilian theme being the only theme of comparable time frame. Arab sources also seem to back the presence of a strategos on the island between 687 and 695.32 Its capital was at Syracuse, with the Sicilian strategos having some control over the Italian duchies of Calabria, Naples, Gaeta and Amalfi.33 The establishing of a Sicilian theme may have been Justinian mistrusting the exarchs at Carthage and Ravenna politically, militarily and/or religiously, as well as thinking that the position of Sicily would allow a strong garrison force there to intervene in Africa and Italy. Justinian II or his immediate successors also appear to have organised Sardinia as something like a theme, but with the slightly lesser status of archontate. This is likely to have happened after the final demise of the Africa exarchate in 698, but suggests an added level of hierarchy to the ‘theme system’ by the eighth century.
There is another pair of ‘thematic’ provinces that appeared in the late-seventh and early-eighth centuries – the Karabisiani (Kαραβησιάνοι/Karabēsianoi) and the Kibyrrhaeots (Κιβυρραιῶται/Kibyrrhaiōtai). Both these establishments seem to be naval in nature, with the Kibyrrhaeots called a ‘nautical theme’ (θέμα ναυτικόν/thema nautikon) and the Karabisiani perhaps providing something of a prototype for it. However, the source material is unclear about their origins, organisation and even general make-up. The first definitive mention of the Karabisiani – the name derives from the Greek κάραβις, meaning ‘ship’ – comes in c.680 during a Slav attack on Thessalonica, but its formation could be linked to the growing Arab threat at sea, possibly after the Arab penetration of the Sea of Marmara in the 670s or even the aftermath of the Roman defeat at the Battle of the Masts in 655.34
The lack of clarity in the sources also provides considerable trouble in defining what the Karabisiani really was. Was it a provincial fleet stationed in the Aegean or a name for much if not all of the Roman navy, with bases in virtually every maritime province?35 There might have been some initial focus in the Aegean and southern Asia Minor, with a capital at Attaleia,36 with Justinian II seen settling Mardaites in southern Asia Minor as naval manpower. But this is by no means definite. Indeed, while it is sometimes named as the ‘Karabisiani theme’, this is incorrect. Whatever the Karabisiani was, it does not fit any of the admittedly nebulous ‘thematic’ organisation, lacking territorial divisions, land forces and civilian authority, remaining a purely naval organisation.37
The role of the Karabisiani is further muddied by the appearance of the Kibyrrhaeots. The first record of these ‘men of Cibyrrha’ comes with their involvement in the 697/698 expedition to Africa.38 Such a role in a high-profile expedition would suggest that the Kibyrrhaeot district was well-established by this date. Its location in either Caria or Pamphylia, in or near the initial Karabisian heartland, possibly makes the Kibyrrhaeots a core droungos subdivision of the Karabisiani. That their droungarios, Apsimar, could become the emperor Tiberius III may also hint at firm foundations and prominence.39 Justinian II’s bolstering of southern Asia Minor with Mardaite naval recruits could have formed the basis of an expansion of the Kibyrrhaeot droungos, seeing it promoted to a status lofty enough to even supersede the perhaps more unwieldy Karabisiani.40 However, if there was any such supplanting, it was more gradual than a single imperial directive. There is evidence for continued Karabisian activity on into the eighth century and possibly significant naval developments during the reign of Leo III. There would also be more themes added to the roster, but they were all from the later years of the Isaurian dynasty and beyond.41
‘Thematic’ Revolt?
Its continued expansion on into the eighth century would suggest that, despite its haphazard beginnings, the ‘theme system’ was an unmitigated success. Far from it. The improvised reorganisation might have helped restore some integrity to imperial territory, but it did not sweep away the empire’s internal or external problems: they helped put the brakes on collapse rather than reversing it. Externally, while some successful campaigning beyond the frontiers was facilitated, Slavs, Bulgars and Arabs continued to raid Roman territory with impunity, the thematic armies being largely incapable of stopping them. Even with the limited stability the ‘theme system’ brought, the empire still faced considerable internal problems – disgruntlement, revolt and sedition amongst the soldiery remained close to the surface, as it had done since the late-fifth century.42 Not even the reconquests of Justinian I had broken this trend of military unrest; indeed, his over-stretching of imperial resources exacerbated it.
Military revolt helped spark the first phase of the seventh-century crises, with Phocas leading the Danube army to overthrow Mauricius, while the Heraclian dynasty overthrew Phocas with the forces of the African exarchate. And while it was to rule the empire for a century, every emperor of the Heraclian dynasty faced some kind of unrest. Heraclius faced military opposition from supporters of Phocas, while there were suggestions of revolt amongst the army that was to be defeated at Yarmuk.43 The familial drama between Heraclius, Constantine III and Martina/Heraklonas was only ended by military intervention. Constans II faced usurpations in Constantinople and Africa. Even when thematic armies start to appear, there was no drop off in military unrest, with Constantine IV facing usurpation in Sicily, revolt from the Armeniac strategos, ‘protest’ from the Anatolic army, and undisciplined rout against the Bulgars. Justinian II was deposed not once, but twice by military revolt, either side of Leontios being deposed by a rebelling African expedition and Tiberius III being overthrown by a Bulgar-backed Justinian. In many ways, the seventh-century Roman Empire contained all the ingredients for the cascade of military revolt it played host to: repeated military failure and impotence, reduction in land and resources, and religious strife, all on top of a pre-existing tendency towards unrest. As will be seen, this period of military unrest, punctuated by revolt against the ruling emperor, was not going to disappear with the Heraclian dynasty.
But perhaps somewhat against expectation, while seemingly rife with opportunity to escalate such military unrest, the thematic armies and their powerful strategoi had little effect on the frequency of revolt.44 They did become focal points, but this was because they were the army and its leaders, which had been the centre of unrest for centuries, rather than any inherent issue with its specific organisation.45 There was also still military revolt from outside the thematic system. Aspects of the thematic system might provoke unrest, but many of them were also potential benefits as well. The contraction of the empire brought these armies closer to the centres of power, politicising them, but also bringing them closer to the influence of the emperor. It also brought the armies closer to each other, acting as each others’ deterrents, while alliances between strategoi could be a benefit and a threat to the empire. A strategos might be a powerful problem, but the concentration of authority in a single civilian/military governor could tighten imperial control if the strategos was loyal; hence the appointment of family members by Tiberius III and Leo III and the reuse of men of proven loyalty (and ability).46 Some care would have to be taken in such appointments, for while meritocratic competition could provide a small but effective and controlled leadership group, it could also promote in-fighting, resentment and paranoia both within and without.
Despite the power of these positions, it is difficult to gauge how much influence the strategoi had on imperial policy. As will be seen with Leo III during his time as Anatolic strategos, circumstances might dictate that a general in the field would have to act without reference to the reigning emperor. However, while the exarchs, strategoi and other officers would usually become the focal point of revolt, the influence of the ordinary soldiery should not be overlooked. Along with the clergy, they became increasingly prominent in presenting provincial opinion.47 Leo III would codify this ‘centrality of soldiers to both society at large and to the emperors’48 in the Ekloga, singling out the rank and file for more of a share in spoils.
While it may not have exacerbated it, the theme system did not eliminate military unrest. It was easy to blame unpopular emperors like Constans II or Justinian II for the revolts that they faced, but this does not account for the indiscipline and unrest faced by a popular and successful emperor like Constantine IV, or why the Heraclians in general continued to face problems in spite of being considered disciplinarians, without necessarily becoming unpopular for it.49 There were likely more systemic issues in play, reappearing even after sustained periods of military calm and success. But even with this continued undercurrent of unrest and the lack of clarity about what it was and who initiated it, it must be said that this ‘theme system’ worked better than could be imagined for an improvised reorganisation for a state clinging on for grim death.
