By Michael S. Fulton
In July 1188, Saladin and his army arrived at Saone (also known as Sahyun and Qal’at Salah al-Din), one of the largest castles held by the Crusaders in Syria. Despite the castle’s size and strength, it fell to the Muslims after a siege that lasted only a few days.
The castle of Saone sits on a spur in the Syrian Coastal Mountains (Jabal Ansariyya), about 25 km inland from the Mediterranean coast and 70 km south of Antioch. The oldest surviving part of the castle is a central Byzantine citadel, which dates back to at least the tenth century, when the castle was recaptured from the regional Muslim ruler by Emperor John Tzimiskes. Saone was back under Muslim control by the time elements of the First Crusade arrived in Syria in 1097; the castle fell under Frankish (crusader) rule at some point following the Europeans’ capture of Antioch the following year. Saone had become the seat of a Frankish lordship, part of the Principality of Antioch, by the second decade of the twelfth century, at which point it was held by a certain Robert, who was to be succeeded by his son William. Aside from this, we know very little about the history of Saone and its lords until Saladin arrived with his army in 1188.
Saone sits in a strong topographical position. The castle spur is sandwiched between valleys (or wadis) to the north and south, each more than 200 m wide, which join to the west of the castle and continue out to the Mediterranean coast. The walls of the castle crown this spur and were developed in such a way that three distinct wards were created. The Byzantine citadel sits at the centre of the castle in the middle of the central ward; to the west, a long bailey surrounds the remainder of the castle spur; to the east, a series of walls create another ward and secure the approach from the source of the spur.
The most impressive element of the castle’s defences is an incredible ditch, approximately 100m long (north to south), about 27m deep and approaching 20 m wide in places, which separates the castle from the remainder of the spur to the east. When the rock was removed to create this chasm, a single, needle-like pier of natural rock was left in place near the north end. This was meant to support a removable bridge that could be accessed from a gate in the outer eastern wall, allowing the inhabitants of the castles and approved visitors to move over the great ditch.
The situation in the Latin East changed dramatically when, in July 1187, Saladin defeated an army of crusaders at the Horns of Hattin, all but wiping out the army of the kingdom of Jerusalem, the largest of the three remaining Frankish principalities. Saladin spent the rest of the year taking possession of the urban centres of Palestine, and initiating siege activities against the main inland castles. With most of Palestine under his control by year’s end, Saladin devoted much of his attention to Frankish Syria in 1188. Gathering his forces near Crac des Chevaliers in late spring, the Muslim army marched through the Frankish county of Tripoli and then northward up the coast towards the principality of Antioch, capturing most of the port towns along the way. When he reached Latakia, Saladin turned inland, bringing him to Saone.
Accounts of the siege
Three contemporary accounts of the siege have survived; all were written from the attackers’ point of view and all three authors would appear to have been eyewitnesses. ‘Imad al-Din al-Isfahani acted as Saladin’s personal secretary, having joined the sultan’s service following Saladin’s seizure of power in Damascus, and may have held more influence than any other non-military figure in Saladin’s administration. Baha al-Din ibn Shaddad, who first met Saladin while serving as a diplomat from Mosul, had only recently joined the sultan’s entourage, but was appointed judge of the army and became one of Saladin’s close confidants. Ibn al-Athir was not a member of Saladin’s administration; he was instead a client of the Zankid dynasty of Mosul – it had been a branch of the Zankid family that Saladin had usurped when he took power in Syria. He appears to have joined Saladin’s army after completing a pilgrimage, most likely inspired by the recapture of Jerusalem the previous year.
Saladin’s army arrived at Saone on 26 July 1188. According to Baha al-Din, six trebuchets were set up early the following day. These would appear to have been erected on the spur to the east of the castle, across the great ditch, as this is where ‘Imad al-Din and Ibn al-Athir claim Saladin positioned himself. From this location, the Muslims’ artillery would have been able to provide cover and inhibit Frankish efforts to actively defend the eastern battlements of the castle. The sources are in agreement that al-Zahir Ghazi, Saladin’s son, took up a secondary position across one of the valleys, where another one or two trebuchets were set up. It was from here that al-Zahir’s contingent allegedly damaged a section of wall.
On the morning of 29 July, Saladin ordered a general assault. With archers and artillery providing cover, Muslim assault forces were able to gain control of a section of the castle’s outer wall about an hour into the attack. The Frankish defenders were compelled to abandon the compromised ward of the castle and fall back on the citadel, from where they negotiated their surrender. Despite this relatively concise narrative, the lack of specific details has left it unclear exactly what happened (and where) on the final day of the siege.
The sources are in general agreement regarding the course of the siege, perhaps not least because, in addition to all three probably having been witnesses to these events, both Baha al-Din and Ibn al-Athir almost certainly had access to a copy of ‘Imad al-Din’s account, the first of the three to be completed.
There are, however, slight differences between the versions of what transpired. ‘Imad al-Din’s narrative, composed in verse, with lines often carrying a double meaning, is the most sensational. He presents the damage inflicted by the artillery of al-Zahir Ghazi’s Allepan force as almost a distinct episode during the siege. By comparison, Baha al-Din appears to link the area where this damage was inflicted to the later general assault, which led to the stronghold’s capture.
A feature in all three accounts is a certain corner of the castle. ‘Imad al-Din specifies that this was at one end of the ditch and had not been sufficiently fortified or completed by the Franks, and that it was here that the Muslims initially broke in. Ibn al-Athir echoes these remarks but omits where this corner was located. Baha al-Din claims that al-Zahir Ghazi’s artillery targeted a certain corner or salient of the castle, though he does not identify this as a naturally weak point. Although Baha al-Din does not specify where the decisive attack was made, his remarks relating to the damage inflicted by al-Zahir’s engines imply that this was where the besiegers first entered the castle.
Previous theories about the siege
Serious study of crusader castles began in the late nineteenth century. Emanuel Guillaume Rey, who visited Saone in 1864, provided Ibn al-Athir’s account of the siege in his study of the castle. Max Van Berchem, who also visited Saone before the end of the nineteenth century, took into consideration all three accounts. Although he noted that it was unclear where the suggested breach was made, he concluded that the main attack came from the east, while al-Zahir Ghazi provided a distraction. Van Berchem, and everyone since, placed al-Zahir on the far side of the northern valley.
A new theory was put forward in 1968 by Gabriel Saadé, which emphasized the role of artillery. Leaning on Baha al-Din’s account, he argued that al-Zahir Ghazi’s artillery, which targeted the north wall of the western ward, created a breach that was then stormed. He proposed that the corner at the end of the ditch, which features in ‘Imad al-Din’s account, was associated with the western ‘ditch’ separating the western ward from the central portion of the castle.
This theory went largely unchallenged until the early twenty-first century, when a team of Franco-Syrian historians and archaeologists, having examined the standing masonry of the castle more closely, argued that it had instead been Saladin’s artillery, not that of al-Zahir, that opened a breach in the castles’ defences. This explanation, written up by Jean Mesqui, places the breach at the north end of the outer eastern wall, where there is evidence of considerable rebuilding.
There are a few issues with these theories. First, the central ward of the castle is separated from that to the west by a drop in the rock level, rather than a ‘ditch’, as Saadé suggests. Critically, all three contemporary sources provide a detailed description of the castle and its various walls, but all three note only one ditch, which cannot be mistaken for anything other than the great chasm at the eastern end of the castle. The theory articulated by Mesqui faces no such issue and fits more naturally with the account of ‘Imad al-Din. Whereas the south-eastern corner of the castle sits on a roc-cut scarp, the roc-cut of the ditch was not wrapped around the north-eastern corner of the castle, allowing for a difficult, though conceivable, climb up to the walls above.
The glaring issue with both theories, however, is the role, and implicit power, attributed to the besiegers’ artillery. Both proposals suggest that the attacking engines were capable of breaching the walls of Saone in a period of no more than 48 hours. Even a cursory examination of the physics is sufficient to impress the issues with such theories; however, it is just as useful to ask if artillery could be used with such effect here, why was it not employed with equally devastating results elsewhere? If such engines capable of achieving what has been suggested were in fact used at this time, why do we not see an abrupt shift in the whole art of siege warfare and castle construction?
A new interpretation
The simple answer is that mechanical artillery was nowhere near as powerful in the late twelfth century as has been suggested, and it was not the primary cause for the capture of Saone in 1188. We need to look no further than other sieges that Saladin undertook in 1188 to see the relatively underdeveloped nature of this technology at the time.
Although some details of the siege remain obscure, it is possible to explain why certain embellishments appear in the eyewitness accounts. The suggestion that artillery was capable of breaching the walls of the castle seems to originate in ‘Imad al-Din’s flowery rendition, which describes the wall opposite al-Zahir’s artillery as bowing to the besiegers; consider that his next thought claims that almost all of the castles’ defenders were killed during this barrage of stones and arrows. ‘Imad al-Din employs similar exaggerations in his descriptions of other sieges. Baha al-Din may have been inclined to adopt this portion of his contemporary’s account precisely because it was associated with al-Zahir Ghazi. By the time that he was composing his history, Saladin was dead and Baha al-Din had entered the service of al-Zahir Ghazi – it was in his interest to shower extra praise upon his patron. When considering Ibn al-Athir’s version of events, it follows ‘Imad al-Din’s account quite closely, but is far more sober; the colourful language and evocative exaggerations have been removed, as well as any suggestion that artillery breached a portion of the castle’s defences. Unlike his contemporaries, Ibn al-Athir was not beholden to Saladin or his heirs – he had no incentive to add or preserve embellishments that might add glory to figures who led this siege.
It seems these exaggerations relating to artillery, effectively disregarded by Van Berchem a century ago, have been responsible for most of the misunderstandings relating to the siege since. If we disregard them, as Ibn al-Athir chose to do, the siege is far less spectacular. Saladin established himself in a primary position on the spur to the east of the castle, where he erected artillery to inhibit the garrison’s ability to interfere with his preparations. Meanwhile, al-Zahir Ghazi took a secondary force to draw attention away from the east end of the castle and force the defenders to divide their manpower. Following two days of preparations, the main assault force gathered in the ditch below the eastern wall, then began to climb up to the northeastern corner of the castle. From the top of the spur to the east of the ditch, Saladin’s archers and artillery pinned down the defenders who might have hoped to prevent the Muslim attackers from reaching the base of the castle’s walls. The sources hint that this corner of the castle, or perhaps the ditch below it, was not sufficiently fortified or otherwise complete, facilitating the massed frontal assault launched by the besiegers. To the west, al-Zahir probably launched a diversionary attack to keep the Franks from concentrating their forces. Within an hour, Saladin’s men had gained control of a section of the outer wall, leading to the collapse of the Franks’ defence.
Although far less sensational than certain readings of the eyewitness accounts and the theories proposed by modern historians, this is precisely the same siege plan that Saladin employed against the nearby castle of Bourzey (20 km to the northeast) less than a month later.
Michael S. Fulton is an Assistant Professor at the University of Western Ontario and an Instructor of medieval history at Wilfrid Laurier University. His books include Artillery in the Era of the Crusades, Siege Warfare during the Crusades, and Contest for Egypt.
Top Image: Photo by Anas Al Rifai / Wikimedia Commons