Byzantine Fire on the Water

The low state of medieval maritime technology ensured that battle tactics were just as basic. They had hardly progressed since Roman times. Confrontations at sea remained messy affairs that almost invariably devolved into unpredictable ship-against-ship mêlées. This helps explain why large-scale naval engagements were rare during the Middle Ages. Few naval commanders were willing to risk all in a single battle subject to so many uncontrollable variables. As on land, clashes at sea normally occurred only when one side or both could not avoid it.

The fact that there was no reliable ship-killing weapon compounded the uncertainty surrounding the outcome. The waterline ram or rostrum of the classical era was ineffective against the sturdier, frame-first hull construction which began to develop in the Mediterranean as early as the seventh century and found full implementation by the eleventh century. It proved utterly futile against the more robust ship architecture of the northern seas, even in Roman times. In his Commentarii de Bello Gallico (‘Commentaries on the Gallic War’), Julius Caesar said of the dense oak vessels of the Gauls, ‘Our ships could not damage them with the ram (they were so stoutly built).’ As a result, no warship in either the north or the south was known to have sported a ram by the seventh century. It was replaced on the Byzantine dromōn by a spur, a sort of reinforced bowsprit used to assist in seizing and boarding an enemy ship. The only weapon developed in the medieval period capable of destroying an entire vessel was ‘Greek fire’, a secret petroleum-based incendiary invented by a Syrian artificer named Kallinikos in the seventh century. Documentary and graphic sources indicate that it was spewed from specially constructed siphon tubes mounted on the bows of dromōns. Unfortunately its utility was extremely restricted. It had limited range and could only be deployed in calm or following winds.

Siphons for spewing ‘Greek fire’ were eventually mounted on protected platforms at the bow and possibly amidships. The parapeted forecastle (xylokastron) housed the main siphon, called the ‘raven’ (katakorax), while the castle amidships was the kastelloma. The aftercastle contained the kravatos, a structure to shield the kentarchos or captain.

The First Siege of Constantinople and the Advent of ‘Greek Fire’ (672–7)

Once Muawiyah had moved his capital to Damascus and consolidated his grip on power, he began preparations for an enormous expedition against Constantinople itself. In 672 he was ready. The caliph unleashed at least two separate fleets on the south coast of Asia Minor. Their activities must have kept the Karabisian fleet fully occupied. Both Crete and Rhodes were raided. One Arab fleet wintered in Cilicia (the southeastern coast of Anatolia) and the other in Lycia (on the south-central coast). Word of these incursions galvanized Constans’ son and successor, Constantine IV, into action. According to Theophanes, the emperor ‘built large biremes bearing cauldrons of fire and dromones equipped with siphons and ordered them to be stationed at the Proclianesian harbour of Caesarius [Constantinople’s Theodosian harbour]’. In 673 Muawiyah’s fleets surged into the Sea of Marmara and ravaged the Hebdomon district just southwest of Constantinople, then captured Kyzikos on the south shore of the sea. Here they established a base camp for incessant attacks on the city.

Constantinople would endure this maritime assault for the next several years, but the emperor was in possession of a terrible new weapon which would finally – and precipitously – end it. Residing in the city at that time was a Christian refugee from Heliopolis in Syria (modern Baalbek in Lebanon) named Kallinikos. Theophanes described him as an ‘architect’ or ‘artificer’ who had ‘manufactured a naval fire [or sea fire]’ which floated on the surface of the sea and could not be extinguished by water. Its precise ingredients were kept a closely guarded state secret and remain a mystery to this day. This has led to endless speculation through the ages and repeated attempts at replication. A similar Muslim concoction of the twelfth century was said to have included ‘dolphin’s fat’ and ‘grease of goat kidneys’. Early scholarly conjecture centred on saltpetre as the main component (as in gunpowder) or some form of quicklime, but recent empirical investigations, particularly by renowned Byzantinist John Haldon, have revealed that its primary ingredient was probably petroleum-based – most likely naphtha or light crude oil. The Byzantines had access to the oil fields of the Caucasus region northeast of the Black Sea where crude seeped to the surface. The theory is that Kallinikos may have distilled this into a paraffin or kerosene, then added wood resins as a thickening agent. The mixture was then heated in an air-tight bronze tank over a brazier and pressured by use of a force pump. The final step was the release of the flammable fluid through a valve for its discharge from a metal-sheathed nozzle, affixed with a flame ignition source. In a 2002 clinical test of this theory, Haldon and his colleagues, Colin Hewes and Andrew Lacey, were able to produce a fire stream in the neighbourhood of 1,000 degrees Celsius that extended at least 15m (49ft).

It was very probably a compound similar to this that Constantine caused to be loaded onto his dromōns in the autumn of 677. The fearsome new weapon was unleashed from swivel-mounted siphons in the forecastles with horrific results. Theophanes testified almost matter-of-factly that it ‘kindled the ships of the Arabs and burnt them and their crews’. To the Arab victims of his frightful invention, it must have seemed like some early version of ‘shock and awe’. The fact that they would have had no idea of how to combat the weapon must have compounded their panic. Water would have been ineffective. At that point they could not have known that the only way to extinguish the ‘liquid fire’ was with sand, vinegar or urine. The siege soon collapsed. What was left of the Arab armada withdrew, only to be severely mauled by a violent winter storm while passing abeam Syllaem in Pamphylia (on the south coast of Asia Minor between Lycia and Cilicia). Theophanes said, ‘It was dashed to pieces and perished entirely.’

The Second Siege of Constantinople and the Fall of the Umayyad Dynasty (717–50)

The continuing turmoil in Constantinople could not have gone unnoticed in Damascus. Earlier that same year Sulayman ibn Abd al-Malik assumed the caliphate and inaugurated his rule by propelling his brother, Maslamah ibn Abd al-Malik, into Asia Minor at the head of 80,000 troops, while a huge armada of reportedly 1,800 vessels made its way around the south coast. Constantinople was about to experience its most dire confrontation with Islam until its final fall over seven centuries later.

The details of the ensuing epic engagement are discussed in a separate section at the end of the chapter as an example of sea combat in the period, but it suffices to say here that it unfolded in a manner similar to the siege of 672–8, with much the same result. As the Arab forces approached Constantinople in the spring of 717, Leo the Isaurian, the strategos of the Anatolikon Theme, engineered a coup to replace the ill-suited Theodosios III on the throne. Under his inspired leadership as Leo III, the Byzantines then used dromōns spewing ‘Greek fire’ to break up an Umayyad attempt to blockade the Bosporus. The besieging Arab army fared even worse. A particularly harsh winter ravaged it with deprivation and disease. And the following spring offered little relief. Nearly 800 supply ships arrived from Egypt and Ifriqiyah, but their Coptic Christian crews switched sides en masse. Without the precious provisions which these ships carried, Maslama’s troops fell easy prey to the Bulgars of Khan Tervel, with whom Leo had formed a propitious alliance. The Bulgars butchered some 22,000 of the Arabs. Umar ibn Abd al-Aziz, the new caliph, had little choice but to recall his forces. It was a battered Umayyad army that retreated across Asia Minor in the autumn of 718 and only five vessels of the once massive Muslim armada managed to run the gauntlet of autumn storms in the Hellespont and Aegean to reach their home port.

It was a disastrous Muslim defeat, which should have put Islam on the defensive for decades to come, but inexplicably Leo chose this time to delve into the religious controversy that was to be the bane of Byzantium. In 726 he inaugurated Iconoclasm (literally, ‘the smashing of icons’) by ordering the removal of the icon of Christ over the Chalke entrance to the imperial palace in Constantinople. In 730 he followed up this action with an imperial decree against all icons. This polemical policy was to rend the fabric of the empire for the next fifty-seven years. It proved particularly unpopular in Italy and the Aegean areas. In early 727 the fleets of the Hellas and Karabisian Themes revolted and proclaimed a certain Kosmas as emperor. Leo managed to devastate and disperse these fleets with his own, again using ‘Greek fire’, the secret of which was apparently restricted to Constantinople at the time.

The episode, nonetheless, prompted the emperor to dissolve the troublesome Karabisian Theme and restructure the provincial fleets in order to dilute their threat to the throne. Leo placed the south coast of Asia Minor, formerly a responsibility of the disbanded Karabisian Theme, under the authority of the more tractable droungarios of the Kibyrrhaeot fleet, whose headquarters was transferred to Attaleia (present-day Antalya). Land-based themes, like the Hellas and Peloponnesos, were also allowed to maintain fleets of their own. These modifications to fleet organization were probably intended to help defuse naval power and make it more subservient to the emperor.

Despite their humiliating failure before the walls of Constantinople, the Umayyads took advantage of continued Byzantine upheaval both in the palace and in the Church to nibble away at the edges of the empire. A long period of raid and counter-raid ensued between Damascus and Constantinople, mostly involving either Egypt or Cyprus. But ultimately the Byzantines’ advantage in naval organization, possession of ‘Greek fire’ and virtual monopoly of such critical shipbuilding materials as wood and iron ensured they would prevail, at least in the eastern Mediterranean. The climax of the contest came in 747, when the Kibyrrhaeot fleet surprised an enormous armada from Alexandria in a harbour on Cyprus called Keramaia (exact location unknown). ‘Out of 1,000 dromōns it is said only three escaped,’ professed Theophanes. This was undoubtedly a chauvinistic exaggeration, but Umayyad naval power was evidently broken by the outcome of the battle and never again posed a serious threat to the Byzantine Empire. The Umayyad Dynasty came to an end just three years later when the Abbasids led by Abu al-Abbas as-Saffah crushed Caliph Marwan II at the Battle of Zab (Mesopotamia) in late January 750. The subsequent Abbasid Caliphate moved its capital from Damascus to Baghdad and focused its initial attention on the East.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.