A study of the life of Sinan reveals the story of a true creator, never satisfied with his achievements, constantly in search of new ideas and restlessly trying out innovative features. Sinan was already a mature person when he was appointed chief architect in his mid-forties, yet he was still to live through some fifty years of creative adventures, leading Turkish architecture to new heights.
The most important space unit in Ottoman architecture is the mosque. And it is in the mosques that we can best see how Sinan developed the Ottoman architecture of the interior space. The first mosque Sinan built after he was appointed chief architect is the Haseki (1539), which reflects the then traditional space arrangement, revealing no particular innovative trait. On the other hand, the Uskudar Mihrimah (1540?-48), which he began to build immediately after the Haseki, represents an important leap forward, employing three semidomes surrounding its main dome. Before even finishing this building, Sinan moved on to the Sehzade (1543-48). Sinan must have visited the Diyarbakir Fatih Pasha Mosque (1522) during the Irakeyn campaign and this work may have influenced the plan of the Sehzade, but its general outlook and design are most definitely Sinan’s.
The Sehzade could be considered as the most advanced stage of the square-based structure with a central dome surrounded by four semidomes, and a final statement on the subject but for the existence of a symbol like Saint Sophia which complicates this issue. Built in the 6th century, the building is covered with a 31 m wide dome and two semi domes which endow it with a truly magnificent appearance. It was indeed the subject of great admiration for the Ottomans, influencing their leaders as much as their architects, who almost idolised it. The Beyazid Mosque did not escape its influence (1506). We do not know why Sinan returned to the original Saint Sophia plan after the highly elaborate version he had developed with the Sehzade. A wish to use the innovations he had already tried out in his previous projects in order to show how it was possible to start out with a Saint Sophia plan and reach a completely different result must have played a role in this choice. Structurally, the Suleymaniye (1550-57) is conceived more consciously and rationally than Saint Sophia, which preceded it by a millennium. This is Sinan’s reworking of the design of Aya Sophia revised to suit the rules of Islamic architecture producing a concept of space totally different from that of its source of inspiration.
Sinan must not have contented himself with what he had accomplished in such a short period for we soon see him coming up with new ideas and experiments. This was a turning point in his life. He was not quite satisfied with the square-based dome model in particular, and was to consider the Sehzade, as his apprenticeship work, although it constituted the most accomplished version of this model. This may be in part due to the fact that in Islamic architecture the mosque plan tended to develop in width. A square-based dome with four semidomes resulted in a square planned mosque and not one developed in terms of width. Moreover, this plan now afforded little room for (an initial) innovation. Sinan finds answer to his quest for more width in the hexagon-based dome plan of the Edirne Serefeli Mosque (1447). Indeed, this mosque is a pioneering work in terms of the design of domed buildings. But Sinan was not very happy at his first implementation of the hexagonal model in the Besiktas Sinan Pasha Mosque (1555) where the effect of the triangular gaps remaining between the main and the adjacent domes was as negative as in the Edirne model. In the Topkapi Kara Ahmet Pasha Mosque (1558?-65?), he was able to cover the whole width of the space by adding trompes to the semidomes. But the death of the sponsor kept the building from being completed as planned.
Sinan’s quest continues in the midst of these experiments. This can be observed quite clearly in the Rustem Pasha Mosque (1562?), a new design which places the dome on an octagonal base. Some scholars trace this use of the octagon to the Kucuk Ayasofya (Sergius-Bacchus 572).24 In any case, the passage from the hexagon to the octagon seems only natural. Besides, Sinan had already supported the dome of the Hadim Ibrahim Pasha Mosque (1551) with eight arches. In the Rustem Pasha Mosque, the trompes set on the corners ensure the passage from the octagon to the square, the lateral galleries serving to widen the plan.
At about the same time, Sinan began working on the Edirnekapi Mihrimah Sultan Mosque (1562?-65?), a project which clearly indicates that he was in the midst of a serious experimental process. Indeed, as a square-based single- domed building, this mosque could only have been an attempt to bring a new interpretation to an old plan. This is why the mosque’s design is particularly bold. The dome is supported by a system consisting of four arches and four pendentives. The numerous windows piercing the walls under the arches create a lacework effect. These walls are totally free of any bearing function. The piers supporting the arches are unnoticeable from the inside. The effect of bareness created in the central area is also felt along the four sides and the plan has gained in width.
Also during the same period, we see Sinan returning to the hexagonal based plan, but with a protruding mihrab, in the Findikli Molla Celebi (1562 ?) and the Babaeski Semiz Ali Pasha (1565?) mosques, another indication that he is pursuing his great quest. In both mosques, the positioning of the hexagon ensures the widening of the interior space while creating a strong feeling of unity. Trompes on four of the corners of the hexagon cover the plan surface in width while the semidome covers the salient mihrab. These experiments may be recognised as part of an important preparatory design phase and quite likely involved some stressful moments for Sinan.
After these different experiments, Sinan must have judged the Rustem Pasha Mosque’s octagonal structure more promising than that of the Mihrimah Mosque at Edirnekapi. Indeed, however popular it may have been, the latter gave him some cause for concern regarding its supporting structure. He must also have been pleased with the protruding mihrab of the Molla Celebi. The tomb he designed for Sultan Suleyman during the same period (1567) is further indication that he had other plans at the back of his mind.
Here an octagonal outer wall and eight arches support two separate domes, but the most important feature in the design is the more rational use of six totally independent piers, a solution which had already been somewhat shyly implemented in the Kara Ahmet Pasha Mosque. The tomb’s plan bears a striking resemblance to that of Diocletian’s Mausoleum in Split, built in the late Roman era (316). We do not know if Sinan ever saw this monument. He may have come across similar works still standing in the region at the time. The Kubbet-us Sahra in Jerusalem (Dome of the Rock, 691) which was restored under his direction also has a similar structure. It was thus at the end of this long preparatory process, that the project of the Selimiye Mosque in Edirne came to life. Had Sinan sensed that another monumental mosque was going to be built? Or did he propose the project himself? We shall never know for sure.
Sinan thought it wiser to adopt the octagonal plan for the Selimiye (1568-75), which enabled him to reduce the bay of the arches. Such a bearing system was indeed necessary to support the huge dome (31.22 m.). The precedent of the Mihrimah must have made Sinan wary and had some influence on this choice.
On the other hand, the protruding mihrab model previously used in the Molla Celebi was deemed suitable for the new project. It not only emphasised the mihrab but also provided an additional support to the bearing system. The dome is carried by eight piers, two of them contiguous to the mihrab wall. Together with the trompes linking it to the corners and the semidome capping the mihrab, it encompasses the whole floor surface of the mosque, to which side galleries add extra width. Besides being the most admired achievement in Ottoman architecture, the mosque is also sinan’s own favourite work. Indeed, he considered it superior to Saint Sophia. In fact, the hexagonal and octagonal plans of sinan’s later works have little in common with Saint Sophia, be it in terms of space or in terms of their bearing system. The use of the octagonal structure on a monumental scale in the selimiye eclipses the symbolic predominance of Saint Sophia.
This achievement must have appeased Sinan, but it did not keep him from continuing to experiment. He seems to have kept a predilection for the hexagonal plan, probably because of the possibilities it afforded of having the lateral volumes integrated with the central area. Whereas with the octagonal plan, the row of two piers and three arches tended to divide the central area from the lateral ones, the hexagonal plan solved this problem and could be easily used to support domes of lesser width. Sinan therefore used the plan again, with a protruding mihrab wall in the Uskudar Atik Valide Mosque (1570- 79) and with a flat mihrab wall in the Kadlrga sokollu Mosque (1572). We cannot quite perceive the original unity of the space in the Atik Valide, because it was later widened (1583), but the perfect unity of space created in the Kadlrga sokollu is truly impressive.
Besides the hexagonal plan he had most often used, sinan returned once more to the octagonal plan in the Sokollu Mosque in Azapkapi (1577), in which he was probably trying to implement ideas he had already had on his mind when building the Selimiye. These basically consisted in centralising the dome by setting semidomes and trompes on every side of the octagon.
As for the interior space of the tomb he designed upon the death of Selim II (1577), it is quite special. The second dome is supported by piers situated at each corner of an octagonal plan, itself set inside a square plan. An intermediate space is thus created between the central area and the square outer wall. This intermediate space is wider and different from the one in the tomb of Sultan Suleyman. Supported by its own octagonal structure, the dome rises independently from its square shell. Such an effect had never been seen in sinan’s works before. Lateral passages surrounding the central area call to mind works of the early Byzantine period: San Vitale (547) in Ravenna, sergius Bacchus (572), as well as the Kubbet-us Sahra (Dome of the Rock, 691) and some Renaissance buildings.
Sinan’s last design in this long search in the perfection spatial arrangement is perhaps-although some scholars deny it was his work- the Nisanci Mehmet Pasha Mosque (1584). Here, an octagonal based central dome and its surrounding semidomes cover the whole space, the centralised plan gaining its full effect. Such a development could have been achieved by no architect other than Sinan.
From then on, Sinan, by now very aged, spends less and less time on project designs. Although he remains the chief architect of the empire, architects he has trained, such as Davut Aga, Dalgic: Ahmet Aga, sedefkar Mehmet Aga, pursue the work of the great master.