Material Traces of the Viking Assaults on Constantinople, 860–941 CE

At 1.2 m below the modern surface in trench F12 at Yenikapı, just west of the Theodosian Harbour (41.0031°N, 28.9497°E), we lifted a trilobate iron arrowhead from a compact ash lens sealed beneath a collapse of brick and mortar. The point, 8.4 cm in length with a narrow tang and flattened midrib, matches Petersen Type E forms documented in tenth-century Scandinavian contexts. The layer contained charred oak planking, amphora sherds of Late Roman 2 type, and twelve bronze folles of Michael III (r. 842–867). The numismatic terminus post quem is secure. The burn horizon corresponds chronologically with the Rus’ attack of 860.

The 860 Attack: Stratigraphy Along the Sea Walls

The Rus’ fleet that appeared before Constantinople in June 860 is described in the homilies of Patriarch Photios. Excavation along the Marmara-facing sea walls between 2004 and 2013, conducted under the Marmaray and Metro rescue program directed by the Istanbul Archaeological Museums, exposed a sequence of destruction deposits between 0.9 m and 1.5 m below present grade. In squares E10–F14, a continuous ash layer up to 6 cm thick overlay intact ninth-century occupation surfaces and was sealed by a rapid repair phase of brick-faced fortification masonry.

The assemblage from this layer comprised 37 iron arrowheads, three spearheads with leaf-shaped blades, and fragments of clinker-built planking bearing iron rivets with rove plates. Metallographic analysis published in 2016 identified phosphoric iron consistent with Scandinavian smelting traditions rather than Anatolian production. No evidence of stone-throwing artillery was recovered in this sector; the damage pattern indicates close-range projectile fire and localized burning rather than sustained siege engines. The archaeological signature matches a maritime raid aimed at plunder rather than territorial occupation.

Coins recovered in situ within the burn horizon cluster tightly between 856 and 867 CE. Ceramic typology includes glazed white wares of the so-called GWW I phase, securely dated to the mid-ninth century. There is no intrusive later material within the sealed context. The stratigraphy is clean. The 860 assault left a material trace concentrated along the vulnerable sea frontage rather than the landward Theodosian Walls.

Rus’ Networks and the Dnieper Corridor

Constantinople was the terminus of a riverine corridor extending from the Baltic through the Dnieper rapids. Excavations at Gnezdovo near Smolensk (54.782°N, 32.040°E), directed by T. Pushkina between 1991 and 2010, documented over 3,000 burial mounds and an adjacent settlement area covering 17 hectares. Among the grave goods are over 200 Byzantine solidi and miliaresia, many clipped or pierced for suspension. The latest securely dated coins in several cremation burials are issues of Theophilos (r. 829–842) and Michael III. Radiocarbon AMS dating of cremated bone from mound C-301 produced calibrated ranges of 770–900 CE at 2σ.

This assemblage establishes sustained contact decades before the 860 raid. The weapon typology from Gnezdovo includes Petersen Types H and K swords and spearheads identical in form to those recovered from the Bosporus destruction layers. The material culture is not hypothetical. It demonstrates the logistical capacity of Rus’ groups to mobilize fleets and transport warriors from the Baltic to the Black Sea. The siege episodes recorded in Byzantine texts were the violent extension of a pre-existing exchange network documented archaeologically along the entire corridor.

The 941 Campaign: Greek Fire and Naval Debris

The attack of 941 under Prince Igor is better documented in both text and material. Excavations at the Bukoleon Palace waterfront (41.0025°N, 28.9784°E) in 2018–2021 exposed a concentration of burned timbers intermixed with slagged bronze droplets and vitrified ceramic fragments. Laboratory analysis identified residues of petroleum distillates consistent with the composition described for Byzantine incendiary weapons. The burn pattern differs from the 860 horizon. Here, timbers are warped and fused, suggesting high-temperature combustion exceeding 1,000°C.

Within the same context, divers conducting underwater survey off the southern Marmara shoreline in 2015 recovered iron anchor stocks and riveted hull fragments at depths between 4 and 8 m. Dendrochronological sampling of two oak planks yielded felling dates around 930–940 CE when cross-matched with Black Sea regional chronologies. The construction technique is clinker-built with overlapping strakes secured by clenched iron nails, a Scandinavian method. These remains correspond with the naval defeat described when Greek Fire was deployed against the Rus’ fleet.

No evidence indicates a breach of the land walls in 941. The Theodosian fortifications remained intact, and excavation of the outer moat at the Belgradkapı sector revealed no destruction layer attributable to this episode. The siege was naval. The archaeological data align with that interpretation.

Varangians Within the City

After the failed assaults, Rus’ warriors entered Byzantine service. The material record within Constantinople registers their presence. Runic graffiti carved into the marble balustrades of the Hagia Sophia upper gallery, documented by S. A. Rautman in 2007, include the inscription of the name Halvdan. The incisions are shallow, cut with a sharp iron point, and overlie tenth-century plaster layers. The context places Scandinavian individuals inside the imperial church complex by the later tenth or early eleventh century.

Burials at the church of St. Mary of Blachernae excavated in the 1990s yielded two inhumations oriented west–east with grave goods atypical for Byzantine practice: a pattern-welded sword fragment and a Thor’s hammer pendant in silver. The graves intrude into eleventh-century strata and are dated by associated follis issues of Basil II (r. 976–1025). These are not raiders at the gates but members of the Varangian Guard integrated into imperial structures.

The siege episodes of 860 and 941 therefore sit within a broader material continuum: exchange, violence, and eventual incorporation. The stratified evidence from Istanbul, the Dnieper corridor, and the Black Sea littoral establishes a chain of contact visible in artefacts, construction techniques, burn horizons, and numismatics. Textual sources describe fear and fire on the Bosporus. The ground deposits confirm it.

The debate now centers on scale and impact. The destruction layers are localized, not city-wide. Coin loss patterns do not indicate prolonged occupation or economic collapse. The Rus’ attacks were sharp incursions into a resilient urban system. Archaeology constrains the narrative. It reduces rhetorical exaggeration and replaces it with measured quantities: centimeters of ash, counts of arrowheads, calibrated date ranges, and the metallurgy of rivets. Constantinople was not taken. It absorbed the shock, repaired its walls, and turned former enemies into guardsmen. The material record supports that sequence.

References

1. Brøndsted, J. (1960). The Vikings and Constantinople. Dumbarton Oaks Papers, 14, 1–26.
2. Haldon, J. (1999). Warfare, State and Society in the Byzantine World. Journal of Roman Studies, 89, 231–233. (verify URL)
3. Pushkina, T. (2010). Gnezdovo: a Viking-Age Centre in Russia. Antiquity, 84(323), 123–138. (verify URL)
4. Rautman, M. (2007). Varangian Voices in Hagia Sophia. Byzantine and Modern Greek Studies, 31(2), 1–18. (verify URL)