In constructing the shells of the Sydney Opera House, the Danish architect, Jørn Utzon claims having resolved a seemingly insurmountable structural problem while peeling an orange. He realised that the shells would be structurally viable if, like the sections of the orange they were parts that together formed a complete sphere. In order to build a major construction, architects are often compelled to address structural complexities, as well as difficulties relating to the terrain, building materials, access and climatic conditions. Through history, countless temples, tombs, fortifications, cathedrals, aqueducts and bridges have collapsed as efforts were made to extend the known boundaries and create larger and ever more complex structures, and not a few architects and construction workers have paid for their efforts with their health, careers, and occasionally, their lives.
In Portugal while attending a conference, I thought to take the opportunity to examine the so-called Castle of the Moors at Sintra, in the hills north of Lisbon. Someone had queried recently about how it was possible to build on a hillside with such steep gradients. While the site itself is of interest, it did not turn out to be architecturally remarkable (I found the adjacent Pena Palace considerably more remarkable, though not in a good way, it being a bizarre mixture of nineteenth century architectural poor taste and filled with an abundance of mostly hideous furnishings). The fortification walls and towers of the Moors' Castle had been constructed on the massive rock outcroppings that are a notable feature of the region, and although the castle appears quite impressive, particularly when viewed from a distance, its walls are not very substantial and were probably not very high, although it is hard to estimate as most of the upper parts appear to have been reconstructed in comparatively recent times.
The very function of a castle makes difficult topography almost a prerequisite, and castle builders were not always able to take advantage of such natural features as the rock outcroppings of Sintra. The ability of larger fortresses to stand often to considerable heights on extremely steep slopes, and not only to stand, but to withstand bombardment, weathering and earthquakes, is nothing if not impressive.
At Montfort Castle in the Galilee, sections of the rather low outer fortifications were, like those of the Castle of the Moors, built on natural rock. At Montfort they take advantage of a series of rock steps. But the main buildings of the inner or upper castle - the keep, the central two-storey domestic building that housed the knights' quarters, and the buildings that served as the administrative part of the castle and contained its ceremonial halls - were huge (between 16 to perhaps in excess of 30 metres tall), and massively constructed (from nearly three to as much as eight metres in thickness), and the incline of the slopes is extreme. They would have required very solid foundations, and yet, the small section of bedrock we exposed that supported the northern wall of the domestic building, had a foundation trench that was extremely shallow, no more than about twenty centimetres depth (I have seen similar shallow foundation trenches at Château Pèlerin ('Atlit), although there topography is not an issue, and the foundations at Vadum Iacob are frail and somewhat pathetic). Which raises the question - how were these sometimes feeble foundations able to support such huge structures and prevent their subsidence on a steep terrain?
I recall a previous visit to Portugal when I observed a technique employed to stablize fortifications at the convent of Christ at Tomar and it appears to be quite similar to what we have found at Montfort. The builders constructed a series of adjacent terrace walls below the exterior of the curtain wall, one to two metres high. They filled the space between these walls with rubble chips, probably waste from the quarriers and masons. At Tomar the thus-raised slope was covered with a stone revetment forming a sort of glacis that fulfilled a double function of supporting the curtain walls and hindering (perhaps not thwarting) attempts at undermining. The example at Karak in Jordan is perhaps one of the most remarkable of such revetments.
The problems that the unknown architects of the crusader castles faced were generally far more simple than those facing Utzon, but they nonetheless required intelligence and originality. Constructing a castle may have been less complex than formulating how to apply the physical qualities observed while peeling an orange to a vast and entirely unique structure like the Sydney Opera House, but it was considerable enough to make their often successful outcome an impressive achievement.