“Greek fire” was a flaming mixture fired from the ships of the Eastern Roman (‘Byzantine’) Empire from the 7th century. The fire would cling to flesh and was impossible to extinguish with water. It was one of the most terrifying weapons ever made.
Voluminous literature has been composed on the subject of Greek fire. Despite numerous attempts to analyze it deeply enough and describe in full detail, the weapon has remained mysterious in certain aspects, and especially in such as the precise chemical composition of the incendiary mixture used by the Byzantines (which was the Greek fire proper) as well as the details of the equipment used for discharging it. This ignorance of ours clearly proves that the Byzantines managed to keep the secret from their friends and enemies equally effectively.
Origins of Greek fire
The invention of Greek fire is commonly ascribed to a certain Kallinikos, an architect from Helioupolis in Syria.Theophanes the Confessor reports on this event:
At that time Kallinikos, an architect from Hellioupolis in Syria,took refuge with the Romans and manufactured a naval fire with which he kindled the ships of the Arabs and burnt them with their crews. In this way the Romans came back in victory and acquired the naval fire. The development is said to have taken place at the beginning of the 670’s, when the Byzantines had to struggle for survival in their mortal combat against the constantly growing power of the Arab Caliphate. The mention proves that the chronicler regarded the invention of the new weapon as a technological and military breakthrough of the utmost importance because it greatly contributed to checking the Arab expansion into the Byzantine territory. It should be admitted, of course, that the utilization of various incendiary substances and devices in warfare was not the idea of the sole Byzantines.
However, it ought to be added that it was they who were the first to take advantage of such weaponry on a larger scale and more efficiently. Although Theophanes’s fairly laconic mention does not let us conclude whether Kallinikos either worked out the composition of the incendiary substance itself or contributed to developing an efficient way of firing it at their enemies or, ultimately, achieved both, it could be inferred from the words of the chronicler that Kallinikos’ merits in this respect were significant enough to perpetuate the tradition that the creation of naval fire should be associated with his name.
Composition of Greek fire
We have already written that Byzantine attempts to keep the composition of naval fire secret were so successful that all today’s historians can only surmise what it was. In the history of the research into the issue three main standpoints concerning the composition of the substance can be distinguished.
a. Saltpetre theory
– Followers of the first hypothesis are inclined to promote the thesis that the basic ingredient of Greek fire was saltpetre and consequently regarded the mysterious deadly substance as an ancestor of gunpowder. This conclusion was drawn from descriptions which referred to a loud, thunder-like audible effect of using the weapon, accompanied additionally by clouds of thick smoke, and also on the fact that the flame of Greek fire could be projected from a siphon at a distance. This view, however, has two main shortcomings. First of all, we have no evidence whatsoever that saltpetre had appeared in Europe and the Near East by the 13th century. Neither was it mentioned in Arab sources. What is more, the physical properties of the final product received from it (which by the by are commonly known because they were well described) are dramatically different from those that we know about from extant Byzantine and other sources.
b. Quicklime theory
–The supporters of the second theory, which has evolved out of the observation that Greek fire was impossible to extinguish using water, claim that the main ingredient of it was quicklime. They maintainthat its contact with water caused a sudden reaction, which determined the power of Greek fire. The drawback ofthis view is the fact that quicklime-based compound substances, in order to acquire its incendiary properties, had to come in contact with water. Byzantine sources point out, though, that the use of naval fire did not necessarily have to involve meeting this conditio sine qua non. What is more, experiments conducted in modern times have shown that, in the open sea, the reaction of quicklime and water has no effect of a potent enough explosive power. Yet another idea linking the explosive reaction of naval fire to its contact with water was based on the view that its main ingredient was calcium phosphide. It is known that during its reaction with water phosphines, which ignite instantaneously, are released. Nevertheless, experiments carried out with this substance have shown that it does not cause effects in any way similar to those described in historical sources.
c. Petroleum theory
– Ultimately, the followers of the third view believe that the basic ingredient of naval fire was crude or refined oil (in other words petroleum). It should be admitted that this theory finds most support among contemporary researchers.
The hypothesis is strongly corroborated by the fact that the Byzantines had access to crude oil sources. They were located on the north-east coast of the Black Sea and Constantine Porphyrogennetos, for instance, mentioned that they could be found in Tmutorakan. Oil flowed out naturally to the surface, where its volatile components began to evaporate into the air, as a result of which the remaining substance became viscid and difficult to set on fire. According to J. Haldon and M. Byrne, the Byzantines must have acquired the skill to collect the substance before the evaporation of its volatile and inflammatory ingredients, determining the usefulness of this raw material as the basis for the production of Greek fire. Additionally, the scholars who promote the last theory also point out to the fact that such crude oil, both in its natural and refined form, when heated and fired under high pressure gives similar effects to those described in historical sources.
It is equally important that the followers of the last theory support their point of view by means of a linguistic argument. Notably, they emphasize a direct connection between the nomenclature used to refer to crude oil and Greek fire in question. The researches notice that one of the names given to the flammable substance called popularly Greek, liquid or naval fire was also Median fire.
Since Procopius of Caesarea reports that crude oil, usually called naphta (νάφϑα), was sometimes termed Median oil, the epithet directly adjoined to the terms referring to both the substances (i.e. the adjective Median) appears to connect them and prove that the weapon’s chemical composition included petroleum termed either Median oil or naphta.
Last but not least, it seems that the final proof substantiating the theory lies in the below-quoted fragment froma ninth-century Latin manuscript preserved in Wolfenbüttel in Germany, which reads:
The material of the fire of the Tyree boys: naphta, tow, pitch, a fire arrow. Naptha [is] a species of balsam originating in Babilonia [Egypt] in humid places, which colloquially we call marisci, and it seems to swim there upon the water like fat. Also, there are two kinds of balsam. One originating from Mount Sinai, exuding from rock, whence “rock of oil”; the other twigs, which mixed together produce an inextinguishable fire. For when the Saracens proceed in war to a naval battle, having built a furnace right at front of the ship, they [Saracens?] set on it a copper vessel full of these things, having put fire underneath. And one of them, having made a bronze tube similar to that which the rustics call a squitiatoria, “squirt”, with which boys play, they spray [it] at the enemy.
Greek fire surely included other substances on top of petroleum. It is maintained that additional components of naval fire were plant resins. They were admixed as a thickener of the oil itself and, additionally, as a medium maintaining and enhancing the power of the flame, making it also more sticky. In order to prove the last opinion the military treatise entitled Praecepta militaria, written after 963 AD and ascribed to emperor Nicephor Focas, is quoted, where (in connection with Greek fire) appears the term πῧρ κολλυτικόν, which means adhesive or simply sticky fire.
Flame-throwing devices
Source-based research and quite recent experiments based on them have led to the conclusion that any flame-throwing device probably consisted of a few basic elements. The first of them was a siphon (σίφων), a kind of a pump, whose role was to increase the pressure of crude oil which was stored in an airtight container kept hot with a little brazier-like boiler termed propyron (πρόπυρον) located underneath.
The substance, when heated to a proper temperature and pressurized adequately, was fired in a given direction with a nozzle covered in bronze and mounted on a swivel, i.e. strepton (στρεπτόν). At the moment of discharge it was ignited at the mouth of the nozzle by a source of flame, for instance by a lamp. The flames released from the device are assessed to have reached over 1000°C. The range of the weapon is thought to amount to 15 metres.
High pressure and temperature accompanying the discharge of naval fire made the procedure of using it dangerous and even life-threatening. For this reason heat shields known as skoutaria sidera (σκουτάρια σιδηρᾶ) or boukolia ( βουκόλια) were used to protect those operating the Byzantine flame-throwers. The fact that historical sources donot provide any information about accidents might indicate that the shields made an effective cover and also that the operators attending to the equipment (σιφωνάτωρ) were well (if not perfectly) trained.
Devices characterised above were used mainly on Byzantine warships. This way of deploying them led to the fact that one of the terms most commonly used in the present day’s scholarly and popular dissertations to refer to the weapon was naval fire. However, there also existed manual flame-throwers, i.e. kheirosiphnes (χειροσίφωνες), whose use on battlefield is confirmed for the 9th and 10th century AD. They were described by emperor Leo VI.
Ultimately, it is known that the Byzantines would sometimes resort to throwing or shooting at their enemies earthenware pots, usually termed khytrai (χύτραι), which were filled with a fluid concoction that, as we surmise, was the same or similar in composition to Greek fire.
(Source: “Naval fire/liquid fire. Byzantine ‘miracle’ weapon and the question of its familiarity to the Bulgarians between the 7th and 11th century”, by Mirosław J. Leszka & Maciej Kokoszko, 2012)
Abstract An incendiary weapon used by the Byzantine Empire (Eastern Roman Empire) was “Greek Fire”. During the Byzantine Empire “Greek Fire” was typically used in naval battles and gave the Byzantines great technological advantages and was responsible for many key military victories. The Byzantines would apply “Greek Fire” using a pressurized nozzle or siphon to project the liquid mixture onto the enemy. The chemical composition of “Greek Fire” is not known and it is lost in time. The exact formula therefore, remains a matter of speculation. After an extensive chemical literature search, this article will attempt to solve the riddle regarding the chemical composition of “Greek Fire”. Three substances (Potassium nitrate, Calcium oxide and Petroleum oil) would comprise the majority of the “Greek Fire” composition. The resultant mixture of Potassium nitrate, Calcium oxide and Petroleum oil would be mixed thoroughly until a heterogeneous liquid suspension mixture was obtained and the Byzantines would place this mixture in a pressurized nozzle for warfare application. In combination, this heterogeneous liquid suspension mixture would give rise to the properties observed and written in history for “Greek Fire”.
(Source: “The Mystery of “Greek Fire” used by the Byzantine Empire: A Chemical Perspective”, by Brett Cohen, 2017)
Research-Selection for NovoScriptorium: Anastasius Philoponus
Novo Scriptorium 
I find questions such as the composition of Greek fire to be fascinating; it’s something we might not get a definitive answer for, but the act of investigating is, itself, very much a worthwhile exercise. My guess is that it was probably oil-based with resin, and the nature of the apparatus for throwing it suggests the same. But there’s no empirical proof, and in that circumstance the answer can only be expressed as a range of possibilities (as you’ve outlined in the blog). That’s OK: sometimes history can’t yield definitive answers. Sometimes it yields more questions – but that’s good too, it forces us to think. For me, all the possibilities as to what Greek fire was point to the fact that those in the ancient world were as ingenious and inventive as we are – as always, the human mind and the nature of the human condition are a constant through history.
LikeLiked by 1 person