They began to blow with blacksmiths’ bellows into a furnace where there was fire, and from it came a great noise. There was also a brass [or bronze] tube, and from it came much fire against a ship, which burned quickly so that it all turned into white ashes…

This excerpt is from the Yngvars saga víðförla, a 12th-century saga attributed to the Icelandic Benedictine monk Oddr Snorrason. Although its plot revolves around the last Viking expedition to the Caspian Sea, led by Ingvar the Far-Traveled, what interests us here is the concise description it gives of a chemical incendiary weapon that the Eastern Roman Empire had invented and used at least four hundred years earlier, whose formula was a state secret: the so-called ignis graecus or Greek fire.

Byzantine documentary sources referred to it by various names, such as Roman fire, sea fire, war fire, liquid fire, sticky fire, or manufactured fire; all with the common term “fire”, since, as we saw in the initial text, it was based on the projection of a stream of flammable material (probably petroleum) that engulfed its targets in flames, primarily ships since it was mainly used in naval battles. This last detail will remind military history enthusiasts of other systems based on burning enemy ships, devised as early as Antiquity.

The most famous is the one invented by Archimedes in 214 BC to defend Syracuse against the attack of the Roman fleet led by Marcus Claudius Marcellus. It has also gone down in history under the name Greek fire, and although it has traditionally been said that it worked by reflecting sunlight onto the sails of the galleys through burning mirrors, practical recreations have shown that it would be almost impossible due to the time and ideal conditions needed to achieve combustion. Therefore, it was likely a different system, perhaps a cannon powered by steam.

The episode of Archimedes' mirrors in Syracuse, in a painting by Giulio Parigi
The episode of Archimedes’ mirrors in Syracuse, in a painting by Giulio Parigi. Credit: Public domain / Wikimedia Commons

The use of incendiary arrows had long been common, and it is known that as early as the 9th century BC, the Assyrians were using vessels filled with sulfur, petroleum, and bitumen-based inflammable compounds, something others quickly imitated. Thucydides records a step forward taken in 424 BC during the Battle of Delium between Athens and Boeotia in the Archidamian War: the use of a long tube on wheels that launched flames with the help of a large bellows. In the late 2nd century AD, a war treatise titled Kestoi, attributed to Sextus Julius Africanus, contains a formula based on resin, liquid asphalt, and quicklime, which was stored in a container and ignited using heat and intense sunlight.

However, as mentioned at the beginning, the Greek fire par excellence was the one invented in the Eastern Roman Empire. The chronicler John Malalas says in his work Chronographia (a history of the world from Ancient Egypt to the reign of Justinian I) that a philosopher from Athens named Proclus advised Anastasius I to use sulfur to set fire to the ships of Vitalian, a general who had rebelled against the emperor’s unpopularity due to the high taxes he had decreed and his Miaphysite faith (a doctrine that held Christ had one nature, both divine and human).

Nevertheless, based on the Chronographia of the ascetic monk Theophanes the Confessor, the invention of Greek fire is usually attributed to an architect named Kallinikos, around the year 670.

At that time, Kallinikos, an architect from Heliopolis, fled to the Romans. He had devised a sea fire that set the Arab ships ablaze, burning them with their entire crew. Thus the Romans returned victorious and discovered the fire of the sea.

The rivalry between Byzantines and Muslims from the 7th to the 9th century
The rivalry between Byzantines and Muslims from the 7th to the 9th century. Credit: Cplakidas / Rowanwindwhistler / Wikimedia Commons

Theophanes is not very accurate because in another passage he mentions Byzantine ships equipped with that weapon a few years earlier, which could suggest that Kallinikos only introduced an improvement and that Greek fire had been created by Byzantine technicians based on the scientific legacy left by the chemical school of Alexandria, something that could also have happened with Kallinikos himself, who in reality was not from Heliopolis but was Syrian. The error comes from George Cedrenus, an 11th-century Byzantine chronicler, who in his Compendium historiarum even claims that the architect’s family kept the formula secret, passing it down from generation to generation.

Nowadays, such prolonged secrecy is considered implausible. In any case, Constantine VII Porphyrogenitus (who, besides being an emperor, was a scholar and the author of the historical work Excerpta Constantiniana) confirms Kallinikos’s arrival from Egypt during the reign of Constantine IV, which lasted from 668 to 685. The empire was going through a delicate moment, as the Arab civil war had ended and the new Umayyad Caliphate had been established, whose leader, Muawiya I, turned against the Byzantines, who, weakened after long wars against the Sassanid Persians, had gradually lost their Levantine territories: Syria, Palestine, and Egypt.

The caliph besieged Constantinople in 672, and although this siege lasted several years in two phases (the first from 672 to 678, the second from 717 to 718), it ended in failure, thanks in part to the application of Greek fire against the Muslim fleet, with particular success in the Battle of Silea, which determined the final victory for the Christians and the Caliphate’s obligation to pay an annual tribute. This led to its repeated use in subsequent battles that allowed a Byzantine expansion between the 9th and 10th centuries, although it was never widespread.

Three dromons, typical Byzantine ships, sailing with sails and oars, artwork by Rafael Monleón
Three dromons, typical Byzantine ships, sailing with sails and oars, artwork by Rafael Monleón. Credit: Public domain / Wikimedia Commons

Civil wars within the empire itself provided a stage for Greek fire, especially in the revolt of the thematic fleets in 727 and the great rebellion of Thomas the Slav between 821 and 823. External enemies were not lacking either, such as the Russians (941-1043) and Bulgarians (970-971), who also had the opportunity to suffer the terrible effects of the secret weapon. So secret that the aforementioned Constantine VII Porphyrogenitus, in his De Administrando Imperio, urged his son and heir, Romanos II, never to reveal the formula, which, according to him, had been given to Emperor Constantine I by an angel for the benefit of Christendom.

However, as we said, it was an impossible task to keep it safe forever, and through bribing officials and capturing ships equipped with the weapon, Saracens and Bulgarians eventually got their hands on such a prized booty. Still, they were never able to reproduce the exact composition, having to settle for making their own version, which must have been less effective; nor were they able to copy the pressurized siphon launch system, resorting instead to catapults or hand-thrown bombs.

In this sense, some technical issues need to be explained, starting with the components. The only data in this regard are the ad hoc comments found in military manuals and documentary sources. One of them is the Alexiad written by Princess Anna Komnene about the defense of Dyrrhachium against the Normans in 1108:

This fire is made by the following arts: from the pine and certain evergreen trees, inflammable resin is collected. This is rubbed with sulfur and placed in reed tubes, and men blow on it with violent and continuous breath. Then, in this way, it meets with fire at the tip, ignites, and falls like a whirlwind of fire upon the faces of the enemy.

Hand grenades of liquid fire and caltrops dated between the 10th and 12th centuries and found in the Greek fortress of Chania
Hand grenades of liquid fire and caltrops dated between the 10th and 12th centuries and found in the Greek fortress of Chania. Credit: Badseed / Wikimedia Commons

It is not very explicit, and other writings do not help either. One has to deduce or assume these elements from the effects they produced, namely: they generated a great noise and much smoke; they called it liquid fire, so it was not a solid substance; it could burn on water and perhaps even ignite chemically thanks to it, extinguishing only with sand, vinegar, or urine; and it was launched both in vessels or grenades, and—most characteristically—fired with a siphon like a cannon. The thunder and smoke suggest an explosive discharge that led 19th-century scholars to propose saltpeter as the main component, thus originating a sort of archaic gunpowder.

However, the absence of references to the military use of saltpeter before the 13th century, both in the West and the East, seems to rule out that possibility. The inextinguishability with water led to speculation about a reaction between it and quicklime, which was indeed known and used by Byzantines and Arabs; but the Tactica (a military treatise written or commissioned by Emperor Leo VI the Wise) says that Greek fire was usually fired directly at the decks of ships, so it would only ignite if they were wet, which does not seem effective.

Another proposal was calcium chloride, which is obtained by boiling bones and, upon contact with water, releases a flammable compound called phosphine. However, it suffers from the same drawback as quicklime: it doesn’t work when dry. All of this draws attention to petroleum, which is abundant in the Black Sea region (under the control of Constantinople). In its crude form, it was called nafta by the Medes and medium oil by the Byzantines, according to the historian Procopius of Caesarea. Interestingly, the latter also referred to Greek fire as medium fire; significant, isn’t it?

An anonymous illustration from the Codex Vaticanus Graecus (1605) shows the use of a hand siphon from a siege tower
An anonymous illustration from the Codex Vaticanus Graecus (1605) shows the use of a hand siphon from a siege tower. Credit: Public Domain / Wikimedia Commons

Starting in the 9th century, the Abbasids commonly used hand bombs made of copper vessels filled with boiling oil, which they called naffāṭūn. Nafta thus emerged as the main component of Greek fire, although resins were likely added as extra fuel and thickener (hence the “sticky fire” mentioned at the beginning), along with other substances. Some non-Byzantine formulas included sulfur (a treatise by military theorist Mardi ibn Ali al-Tarsusi mentions it in the 12th century), charcoal, saltpeter, alcohol, tar, wool, camphor…

However, the most characteristic aspect of Byzantine Greek fire was its method of application. In its earliest form, a light catapult would launch a small jar (“composite pomegranate” it was called in Aragon), weighing less than nine kilos and capable of reaching distances of about four hundred meters, which contained burning material and caltrops. But the signature weapon was the cheirosiphōn or hand siphon, a type of portable cannon supposedly invented by Leo VI, used especially against siege machinery, as noted in Heron of Byzantium’s Parangelmata Poliorcetica, though Nicephorus II recommended using it against enemy military formations as well.

The Byzantine Empire’s dominance over the eastern Mediterranean relied on its formidable navy, which equipped its dromons (warships halfway between Roman biremes and later galleys) with ballistae and siphons for deploying Greek fire. The former were a type of catapult that worked as already described, though sometimes assisted by cranes. The siphons were bronze tubes, generally installed at the bow, though more could be added elsewhere if facing a more numerous enemy. These siphons fired jets of fire similarly to a flamethrower.

Model of a dromon equipped with a Greek fire siphon on its bow
Model of a dromon equipped with a Greek fire siphon on its bow. Credit: Bukvoed / Wikimedia Commons

The Alexiad describes such an attack by the dromon, noting that the siphons were shaped like lion heads and appeared like terrifying beasts spewing fire. The Wolfenbüttel manuscript, a 10th-century Latin text preserved in the German city of the same name, also describes it:

…having built a furnace right on the bow of the ship, they placed upon it a copper vessel filled with these things, lighting a fire beneath. And one of them, having made a bronze tube similar to what the rustics call squitiatoria, “jet,” with which boys play, sprayed it on the enemy.

Some modern reconstructions of siphons, created through experimental archaeology based on these descriptions, produced terrifying results: flames exceeding a thousand degrees in temperature with a range of up to fifteen meters. It’s easy to imagine the effect—not only material but also psychological—it would have on the enemy; they knew they could be burned alive because there was no way to extinguish the fire once it adhered to their bodies (even though Muslims protected themselves with felt and skins soaked in vinegar), as water only intensified it, contributing to its superstitious reputation.

Reproduction of a Greek fire siphon exhibited at the Thessaloniki Museum
Reproduction of a Greek fire siphon exhibited at the Thessaloniki Museum. Credit: Gts-tg / Wikimedia Commons

Despite this, Greek fire had its limitations. If the siphon’s range was those fifteen meters, it was quite short and would only be effective when closing distances before boarding. On the other hand, firing a siphon could be dangerous for the user if the wind blew against them or the sea was rough. It was safer to use on land, but its fame primarily came from naval warfare.

Returning to the chronological development, in the 12th century, the use of Greek fire declined, perhaps due to difficulty in sourcing the necessary raw materials; Anna Komnene mentions a case against the Pisan fleet, and then there are a few improvised cases during the defense of Constantinople in 1203, but little else. In the 13th century, its use reappeared and spread, but this time among the Muslims, who impressed the Christians during the Seventh Crusade: The tail of fire trailing behind it was as large as a great lance, recounted the Lord of Joinville in his memoirs. The Chinese and Mongols also adopted it with their own formulas.

Surprisingly, Greek fire survived until the 19th century, as it is documented that an Armenian named Kavafian offered an improved version to the Ottoman Empire; when he refused to reveal the composition, he was poisoned by imperial agents who, nonetheless, were unable to find what they were looking for. Perhaps this survival isn’t so surprising, considering that napalm is a descendant of it, as is the firing system of modern flamethrowers.


This article was first published on our Spanish Edition on August 9, 2024: Fuego griego, el arma incendiaria secreta del Imperio Bizantino que sobrevivió hasta el siglo XIX


  • Share this article:

Discover more from LBV Magazine English Edition

Subscribe to get the latest posts sent to your email.

Something went wrong. Please refresh the page and/or try again.