By the turn of the 9th century, the frontiers of the Roman Empire had contracted significantly since its heyday. Even with some limited recovery under the Isaurian dynasty and their immediate successors, imperial territory essentially encompassed only Greece, the Aegean and what is now Turkey.
However, that did not mean that the empire was in terminal decline - soldier emperors like Nikephoros II, John I and Basil II would see the Danube restored as the northern imperial frontier, and parts of the Levant and southern Italy reclaimed throughout the 10-11th centuries. And even in the early 9th century, the Romans were still able to demonstrate an ability to project power and influence beyond their borders, even into lands that they had never conquered.
This ability and the empire's (earned) reputation for technological superiority was demonstrated in a call for help from the Khazar Khaganate of the northern Caucasus.
Despite the occasional flare-up over control of the Crimean Peninsula, the Romans and Khazars had frequently been allies since the latter's appearance in the early 7th century.
The empire had found the Khazars a useful trading partner in bypassing the Mesopotamian and Iranian lands controlled by consecutive imperial enemies - Sassanid Persia, Umayyad caliphate and Abbasid caliphate. The Khazars also offered useful military aid to the Roman Empire against these foes, whether it be manpower joining a Roman expedition or diverting attention through fighting their own wars against those foes.
The extent of this Romano-Khazar cordiality can be seen in the future Constantine V marrying Tzitzak (rechristened Irene), a female relative of a Khazar Khagan in 732. This meant that Constantine's son and successor, Leo IV was half-Turkic, earning him the sobriquet 'Leo the Khazar'.
This alliance was called upon during the reign of the Amorian emperor Theophilos (829-842). At this time, the Khazars were faced new threats on more than one frontier, with pressure coming from the Magyars and possibly the Pechenegs too.
The early histories of such peoples are difficult to determine. While the early 9th century may be slightly too early to posit Pecheneg pressure on the Khazars, a century previously, a group of Magyars had moved to shores of the Don river. It has been conversely argued that this group was either a subjugated constituent tribe of the Khazar Khaganate or an equal ally and then opponent of the Khazars.
Trouble from the Magyars may have been exacerbated by the rebellion of three tribes of Kabars (Constantine VII, de adm. imp. 39-40 on Kabars) against the Khaganate who fled north-west to the Magyar lands of Levedia (subjugated to the Khazars or not). The potential or actual ramping up of tensions with the Magyar-Kabars may have been the catalyst of the Khagan's asking of the military architectural and of his ally, the Roman emperor in Constantinople. Certainly the location of the resultant Roman aid - the fortress of Sarkel on the bank of the Don - would seem to back the notion of Magyar-Kabar trouble being the source of Khazar concern. That this Khazar appeal was linked to Magyar activity is backed by both Constantine VII, who records the plea in the same section as he talks about the Magyars, and the 10th century Persian geographer, Ahmad b. Rustah, who has the Khazars, traditionally a semi-nomadic people, becoming more 'entrenched' in response to Magyar activity.
It has been suggested that the Magyars were not yet enough of a threat for the Khazars to feel the need to build such a large fortress as Sarkel. Instead, the Khazars could have been feeling some pressure from the southward expansion of the Rus along the river systems of Russia and Ukraine.
It could be that we need not identify a single enemy that encouraged the Khazars to fortify Sarkel and possibly other points within their sphere of influence. That several possible enemies can be suggested - Magyars, Rus, Pechenegs - as well as whatever internal divisions caused the Kabar 'secession' hints that the North Caucasus of the early 9th century were becoming a much more dangerous place, with traditional forms of 'policing' not being enough for the Khazars to retain control as they had done over the previous 200 years.
Whatever its catalyst, Theophilos' response to this plea for help from his Khazar allies was to dispatch the spatharokandidatos Petronas Kamateros with a group of engineers to oversee construction at Sarkel.
The exact date of this call for aid and the subsequent building of Sarkel is much discussed. However, whether it be in c.833 (Dunlop (1954) 186; Koestler (1976) 85), 838 (Erdal in Golden (2007) 85) or 840-841 (Golden in Golden (2007) 150), this is not a vast range of dates - covering only eight years, all of which were encompassed by the reign of Theophilos.
In recompense for this building aid, the khagan relinquished control of the Crimean city of Cherson to the empire. For his service, Kamateros was promoted to protospatharios; however, that was not the end of Is imperial service in connection to this building supervision. On his journey to and from Sarkel, Kamateros travelled sailed to via warship and then on to the month of the Don via transport ship. His time in the Crimea roused worries over the laxity of Roman control in the region (Constantine VII, de adm. imp. 42; Theophanes Continuatus III.28). Indeed, that Cherson, former imperial territory, was perhaps in the gift of the Khazar Khagan to give to the Romal emperor was in itself evidence that Roman control in the Crimea had been on the wane.
Upon hearing of this, Theophilos appointed Kamateros strategos of the newly created Klimata theme, later to become known as the Cherson theme. Again, the date of this foundation is not completely clear, with Kamateros in place by 840-841, possibly also with the position of genikos logothetes.
But what had Kamateros overseen the building of? The name ‘Sarkel’ was not a pre-existing one for the site of the new fortress along the Don - instead, it reflects that fortress' Khazar origins and perhaps the most prominent of the material used in its construction. That is because 'Sarkel' meant 'white house' in Khazar Turkic. Its whiteness stems from the limestone blocks used in its construction, for "since the place had no stones suitable for the building of the city, [Kamateros] made some ovens and baked bricks in them... making mortar out of tiny shells from the river" (Constantine VII, de adm. imp. 42; Theophanes Continuatus III.28)
Despite this name though, limestone was not the only material used in its construction, with flat and oval tiles, plaster, some Roman columns and a section of opus mixtum found on the site during excavations in the 1930s under M.I. Artamonov. Some of the bricks found at Sarkel also host Turkic tamgas, a version of a seal or stamp. This Turkic influence was not limited to the Khazars themselves, with the Sarkel fortress being garrisoned by Oghuz and Pecheneg Turks.
And Sarkel should not be viewed solely in a military sense. Its location on the Don controlled some of the overland routes to the Volga or was downriver of them, affording some control of east-west trade. Indeed, Sarkel’s position gave it access to the trade routes in all cardinal directions, whether it be east to the Caspian Sea and Central Asia, north through the Ukrainian and Russian river systems, west to the Danube and south to the Black Sea and the Caucasus. This made this Roman-built Khazar fortress into a hive of commercial activity throughout the 9th and on into the 10th century.
The potential benefits to fortifying the area, as well as trouble from neighbours, was recognised by the Khazars before the call went out to Theophilos for building help. That is likely seen in Sarkel not being the only and possibly not the first Khazar hill-top fortress built in the area.
There were also the fortresses of Semikarakory and Pravoberezhnoye Tsimlyansk on the opposite bank of the Don from Sarkel. Semikarakory was built with mud and some fired bricks, possibly suggesting that it was a Khazar solo project earlier than Sarkel, although there is some material – tiles, possibly thin brick, some plaster – could indicate Roman influence, such as imported material from the Crimea.
Tsimlyansk also had tiles, thin bricks and plaster, but perhaps most interestingly, there was some white limestone blocks, which had given Sarkel its name. As there is no mention of a second fortress being built under the guidance of Kamateros, could it be that the Khazars built Tsimlyansk after the Roman spatharokandidatos had returned home, using the techniques they had learned from him and the same materials, possibly left over from the building of Sarkel?
The suggestion that both Semikarakory and Tsimlyansk were destroyed during the 830-840s (Flerov (2018)) at least suggests that the area was indeed becoming increasingly dangerous and unstable. We might even speculate that it was the ability of whomever the threat the Khazars were facing to capture and destroy such fortresses that saw the khagan turn to the Romans to help build a better fortress.
Indeed, it is likely that a combination of foreign pressure and the commercial success of Sarkel and the Khazar position along various trade routes contributed to the khaganate’s decline. The flow of Arab silver and the fur trade attracted pressure from the Rus’ statelets and raiding parties from the north, while there must have been significant damage done to the infrastructure of the khaganate by the migrations of the Pechenegs and Oghuz (which was caused by Turkic intertribal warfare on the Central Asian Steppe over control of the trade routes).
It must be said that this was not an immediate decline – Sarkel itself did not fall until 965, when it was captured by Sviatoslav I, Prince of the Kievan Rus’ (the Khazar capital at Itil was sacked three years later sending it into terminal decline). Given the usefulness of its position, the Rus’ did not destroy Sarkel. Instead, it became known as Belaya Vezha, Slavic for ‘White Tower’ or White Fortress’. It would remain in Slav hands until the 12th century, when it was taken by another Turkish wave, the Cumans/Kipchaks, although by then, the Volga-Don trade route was declining in importance with the dropping in Arab silver production and the Rus’ focusing more on their Black Sea trade with the Roman Empire using the Dneiper and Dneister rivers.
The site of Sarkel is now submerged by the Tsimlyansk Reservoir, completed in 1952, while Tsimlyansk is also facing erosion from the encroaching waters of the dammed Don, although by no means to the flooded extent of the 'White House' of Sarkel.
Bibliography
Dunlop, D.M. The History of the Jewish Khazars. Princeton (1954).
Erdal, M. ‘The Khazar Language’, in Golden P.B. et al. (eds.). The World of the Khazars. New Perspectives. Leiden (2007) 75-108.
Flërov V.S. ‘Semikarakory: The Tower on the North Wall and a Relative Chronology of the Lower Don Khazar Forts’, Anthropology & Archeology of Eurasia 57 (2018) 190-214
Golden, P.B. ‘The Conversion of the Khazars to Judaism’, in Golden P.B. et al. (eds.) The World of the Khazars. New Perspectives. Leiden (2007) 123-162.
Koestler, A. The Thirteenth Tribe. London (1976).
Petrukhin, V. ‘Khazaria and Rus': An Examination of their Historical Relations’, Golden P.B. et al. (eds.) The World of the Khazars. New Perspectives. Leiden (2007) 245–268.
Wozniak F.E. ‘Byzantine Policy on the Black Sea or Russian Steppe in the Late 830s’, Etudes byzantines 2 (1975) 58-59
Zuckerman, C. ‘Two notes on the early history of the thema of Cherson’, Byzantine and Modern Greek Studies 21 (1997) 217-220.
Comments