“Istanbul construction technology” may sound like a strange term at first; in general, the history of techno-cultural traditions associated with city names is not written in this way. However, if the city in question is Istanbul, a metropolis of antiquity, it is possible to talk of the Byzantine and Ottoman techno-cultural focus. Although influenced for centuries by various regions and traditions, Istanbul managed to create and maintain an autonomous construction tradition. Therefore, the city can be considered to be an extreme example within the context of construction technology.
In pre-capitalist economic systems the intensity of demand of one single city does not enable the sustainability of complicated and multicomponent construction activities. For example, in Europe of the Middle Ages, masonry structures, which rarely appeared in the cities, would be buildings that lasted for ages, like cathedrals and were ones that would take several consecutive construction seasons to complete. Since it was impossible for the established construction experts and workers to maintain their lives in such an unstable and inactive environment, they had to work by travelling constantly from one city to another. This socioeconomic predicament prevented medieval European cities from producing independent construction traditions, and gave the European masonry guilds an international characteristic. In addition, this mobility seems to have produced another important advantage. The mobility of the construction units enabled the spread and rapid development of architectural forms and techniques across the continent. In the Islamic Middle Ages, such an expansive labor mobility did not exist. As indicated in numerous inscriptions made by masters and architects in Anatolia, mobility still existed. However, there were no travelling guilds similar to those in Western and Central Europe in the Middle Eastern region. These European guilds would relocate in the aftermath of a conquest, either individually by choice or in a considerable number through forced migration.
In contrast, because Istanbul had been a large city since its establishment (it was the largest city in the world for a long time), it may have been the only settlement which continuously provided employment to a wide and varied architectural labor force. Even in the late Byzantine era, when economic opportunities became limited and structure dimensions narrowed, qualified staff for different types of construction were able to find employment here. At this time, the artisan foremen, for example in mosaics, who had to have higher qualifications, could only be found in Istanbul. Ottoman Istanbul continued this tradition in various sectors. For example, the construction of waterworks demonstrates an unbroken technological continuity that stretched from Rome until the nineteenth century. Even in the darkest periods, Istanbul was the only Mediterranean city with a fully functioning water system, and the interrelated expertise remained alive.
Rapid overcrowding in the Ottoman period and the increase in construction activities not only increased the number of construction workers in the city, but also kept them in operation. In this way, almost all construction practices that spring to mind when discussing classical Ottoman structures are those from Istanbul and its environs. Admittedly, although it is not possible to claim that these were only applied in Istanbul, when traveling from Edirne to the West, from İzmit to the east it is a fact that Istanbul practices moved away at varying rates. It was necessary to bring artisans from Istanbul when such crafts were required in other places. The opposite, namely bringing construction workers from other parts of the Ottoman state to Istanbul was only necessary for less skilled labor. Skilled labor could only survive in the intense architectural activity of Istanbul. The skills to construct the monumental masonry structures in Ottoman Istanbul are summarized below in order of implementation.
Pre-Construction Preparation Phases
A complicated and detailed series of activities, conducted in several stages, were carried out before construction. Some of the preliminaries to the pre-construction stage were technical in nature, while others consisted of the logistic preparations that ensured the provision of required material and labor force for a smooth construction process.
Technical preparationsPre-construction technical preliminaries began with the preparation of the topographic plan of the area where the construction was to take place. This process, called mesaha (survey) in Ottoman Turkish, was usually carried out by architects.1 However, in documents, in addition to the architects, there were skilled workers who were referred to as mühendis (engineer), or more infrequently as mesahacı (topographer); these skilled workers were responsible for making measurements.2 A special device made of silk, marked by knots at certain intervals, was used to measure horizontal length during the processes carried out for this purpose; in documents this instrument is referred to as a iki ucu mühürlü urgan (tether sealed on both ends). The urgan measured 75 tailor’s zira.3 For the measurement of vertical length, a havaî terazi (air balance) or just terazi was used. This instrument would be used after the route had been determined, something that was necessary during the construction of waterways.4 This device and practice was known in Iraq in the Middle Ages as well as in Vitruvius.
After measurements were taken of the construction site a project would be prepared for the area in question. In the sixteenth century the only equivalent in Turkish of an architectural project as a concept was the word resm. This word referred to every type of geometric measurement device used to measure two-dimensionally. During the conceptualization and design phases, the focus was mainly on the planning. There is no evidence showing that Ottoman architects in the sixteenth century made section or elevation view drawings or that they used these to shape the construction.5 In contrast, for important and complicated structures models were made.6 Such a model for Süleymaniye Mosque was shown in detail in a sixteenth-century Ottoman manuscript.7 It is clear that the model was not only a depiction of the structure before it was built, showing how it would be when completed, but it also functioned as a three-dimensional project to help in the technical stages of the construction; such models were made during the same period in Ottoman Turkey as they were, for example, in Europe.8
Ottoman architects benefited from a special modular system for making plans until the late eighteenth century. This type of system was also used by architects in Iran, Central Asia and India. The main part of the system was formed by squaring the project paper with vertical and horizontal lines. Ottomans used this squaring on a paper surface in the form of a low relief with a device known as a mıstar board. However, this technique was not applied in other Islamic countries9. The mıstar board was a plate upon which very thin silk thread was stretched horizontally to form squares. The paper to be squared was pressed firmly with the palm after it was placed on the board. As a result, the lines formed by the threads appeared on the paper as a low relief. The paper was rotated 90 degrees and pressed again to create the squares.
After the papers were squared, they could be drawn on. The drawing was created by following the lines with a straight edge. The advantage of using squared paper was that it enabled drawing without the need to compute scale. The architect determined how many zira each square equaled before drawing, and the scale was based on this. However, this method was sufficient only for small buildings that did not include complicated technical problems. How these kinds of structures would look in the third dimension was decided by decisions made at the construction site by the architect and usually after taking into consideration conventional practices and form patterns. Model making was a must for large mosques, as mentioned above.
Material and Labor SupplyMaterial and labor supply constituted the most important obstacle in efficient construction of Ottoman monuments. It was not possible to provide the necessary materials or labor from a large city like Istanbul when building large structures. Additionally, when such important structures are in question, there would be a rapid rise in the prices of materials in the city.10 In order to prevent this overpricing related to an increase in demand, which was described as undeserved gain, the central government both imposed sanctions and attempted to purchase materials from several different cities. The massive structures on the scale of Süleymaniye or Selimiye, as well as the much more modest vizier complexes, required the purchasing of materials and a supply of foremen and workers from a wide area. In both areas the central government and its representatives in the provinces devoted considerable efforts, even resorting to involuntary servitude.11 These stringent methods slowly disappeared after the seventeenth century.12
The main raw material required for construction was, naturally, stone. Used in structures in Edirne and Istanbul, organic limestone called küfeki was mined from quarries close to both cities and became the main material for centuries due to its easy availability and ease of cutting. Stone from local quarries in Anatolia and the Balkans were used; these had a limestone base and were easy to cut. There is no documentation to confirm purchases from the market to this end. Almost all of the monumental sixteenth-century structures in Istanbul were built with limestone mined from the state-owned quarries in today’s Bakırköy, Yeşilköy, Bahçelievler and Haznedar areas. Even for a massive construction like the Süleymaniye Complex, no other limestone source was used.13 Acemi oğlans were put to work in these quarries.14 However, there is no doubt that the stone required for smaller structures was mined and sold by private parties. Available documents indicate that there was, to some extent, standardization in stone size. However, the thickness and depth of the stones varied slightly in every structure. This is why as of the seventeenth century, only the length was mentioned as a unit and the measurement of volume or surface was not required to determine the price of the stone or the workmanship. The practice of stone cutting must have made the measurement of dimensions beyond the front length unnecessary; it is certain that only the front face of the stones was processed in the quarry and the reverse side was only roughly shaped. Naturally, processes such as fitting and trimming were necessary during construction.
Green dacitic tuff (odtaşı) was used in the bases of Istanbul structures and structures that needed to be heat-resistant, such as furnaces and fireplaces; this stone was mined in the Karamürsel region.15 The stone was obtained from state-owned quarries, but they were not operated continuously; rather they were operated based on the growing demand, with workers being sent from Istanbul. This business method, opening when necessary, was implemented in many places.16 However, as seen during the construction of Sultanahmet Mosque, the state no longer conducted mining but rather purchased stone from private parties.17
There were two sources for the marble required for construction in the sixteenth century: mining from quarries or reprocessing stone left over from other constructions. The Ottomans used white marble almost exclusively in structures. The first reason for this was, of course, the proximity of the white marble quarries, which were on Marmara Island, the largest consumer center. The history of this quarry from late antiquity until the reign of Selim II (1566-1574) is unclear.18 However, the oldest Ottoman document regarding Marmara Island is dated August 1570, and it is possible to determine that the quarries on the island ran continuously.19
It is thought that the technique used to mine marble and stone from the quarries did not change much from the Roman era. Open pit mining was preferred for production. The drift mining method was only used in Karamürsel.20 A number of methods were used, including tunnels, shaft mining and wedge cutting, according to the size and location of the piece to be extracted.21 For making blocks from the marble pieces, a method using a smooth blade or wire was used; friction was provided by fine sand, similar to the method the Romans had used. This was also a conventional method across the Eastern Mediterranean, but further east, for example in Iran, different methods were used.22 Another important characteristic of marble processing in the Ottoman state was the preference for production of semi-finished constructional components in the quarries, rather than pitch-faced blocks. Window frames, stairs, column caps etc. were made in the quarry according to the specified sizes and delivered to the construction location.23 These would be included in the structure after all the rough construction was completed and then decorated.
The most commonly used stone in Ottoman structures in Istanbul, after white marble was red cement conglomerates. These were extracted from the quarries in Mihaliç.24 However, extraction of breccia (a type of conglomerate) was not continuous and production was only carried out with workers and operation capital that were sent when necessary.25 In addition, some marble materials and all the granite and porphyry were taken from old structures.26 This, at that time, was a fairly common practice both in Europe27 and in the East. Some centers played a prominent role in this Ottoman practice. For example, Temaşalık (Kyzikos) in Edincik, Eski İstanbulluk (Aleksandria Troas) near Ezine, Ereğli of Marmara (Perinthos) and İzmit (Nikomedia) are frequently referred to in sixteenth-century documents. Collected marble and stones were always shaped after re-processing and, therefore, they changed their appearance. The columns were usually cut and shortened, or thinned and trimmed to adapt them to the new structures.28 Ottoman foremen probably used a type of lathe during the production of new columns and probably finished the old columns after placing them in the structure, trimming as needed.29
One of the main construction materials, bricks, is not known to have had a production area that was completely private in the sixteenth century. For example, some of the bricks needed for the Süleymaniye Mosque were purchased from the market, some were produced by private individuals and some were molded in state-owned brickyards.30 Although it was originally planned to purchase large-sized dome bricks, the most important material, the special molds that had been sent to Istanbul were reclaimed and they were made in “state-owned” batches.31 This did not change much in the seventeenth century; for example, the private sector and state production met the needs for the construction of large buildings like Sultanahmet külliye.32 Afterwards, all the elements required for bricks were provided entirely by the private sector. As Ottoman brickmaking technology remained almost the same until the mid-twentieth century, there is documentation of the process. This technique did not differ much from the method employed in Europe.33
Often used in construction starting from the sixteenth century, iron was processed in two different methods, according to the place where it would be used. Long, wrought iron rods, required for the strengthening of the main body of the construction, were produced in main manufacturing centers in the Balkans, primarily in Samakov; measurements would be sent for the construction.34 Tongues and headers, which were easier to produce, were also manufactured in these centers.35 Part of these were manufactured in forges in the construction area. However, this latter type of production style must have been smaller in scale compared to the first. Nail production was carried out in cities in the manufacturing centers of the Balkans or in cities like Istanbul and İzmit, where there was no iron manufacturing.36 In terms of nail production methods, Ottomans used contemporary European technology. This method consisted of passing the small iron pieces, heated to a very high temperature in the forge, would be beaten through holes on a metal plate.
Lead was used extensively in the construction of monumental Ottoman buildings for roofing and securing iron elements, such as clamps and tongues. Lead was brought from the production centers in the Balkans as ingots, as required. Since lead was considered to be a strategic raw material, trade could only be carried out on a limited scale; in the sixteenth and seventeenth centuries a form of state monopoly was established on its production and consumption. Thus, when the construction of massive structures was in question, the owners of the buildings had to make direct purchases, sending money to the production centers and organizing transportation from these centers to Istanbul. These building owners were either members of state or dignitaries, which enabled them to gain official permission.37 It was costly to arrange transportation by land, using donkeys or mules.38 The lead brought to the construction site was used by melting it in simple special furnaces and then pouring it into clamp and tongue molds or casting it into cladding panels.
Application and Foundations
Before the process of application, that is, determining the contours of the area on which the structure would be based, the site had to be prepared, particularly when constructing large buildings. It was necessary to level either by excavating or filling in the site as required. Like their contemporaries, Ottoman architects lacked the topographic techniques that would enable them to carry out the application on a non-planar surface. Ottoman builders used a rope grid for the application, as noted in the seventeenth-century text, Risâle-i Mi‘mâriyye.39 After this, excavations would be carried out to reach the solid ground on which the foundation of the construction was to be built. Under normal conditions, the foundations would be built as a strip foundation. Although it was thought that the piers of the large mosques rested on individual bases, there is no evidence to confirm this.40 Rather, existing examples demonstrate that Ottoman architects generally preferred strip foundations. Even in some centrally planned structures, such as important tombs of the sultans, there are examples which demonstrate that the entire foundation consisted of a base slab that went basically under the entire structure.41 In large structures, continuous foundations were generally stepped. In places where solid ground was not available, piled raft systems were utilized. Ottomans preferred using foundation piles that were short but spaced at regular intervals, like the Romans. It is known that 40,000 piles measuring 3.75-6.00 meters in length were used in the repairs of a bridge in 1552.42 A seventeenth-century document shows that iron wedges were placed at the end of the wooden foundation piles.43 Although the prevalence of this practice in the sixteenth century is unknown, it continued in the following centuries.
After the piles were erected, the heads of the piles were tied to one another to form a wooden grid. The function of this grid was probably to create a level ground or a zero-level elevation. The real foundation would be built based on this elevation. Evidence indicates that the wooden grid was not only used in piled foundations, but also in solid foundations. Such grids existed in the foundation of the Süleymaniye Mosque.44 As with the foundation piles, oak was the preferred material for the grids. When the grid construction was finished, a khorasan (concrete layer), measuring no more than 30-40 cm in thickness, was poured and the strip foundations of the structure would be built on top of this khorasan. Rubble masonry was applied to the foundations and “gallstone” was used in large Istanbul structures for this type of work. When the foundations reached the ground elevation, one more levelling would be carried out and the ground surface operations began.45
As in bridge construction, cofferdams were built if the construction was to be carried out directly in water. There is fairly detailed information on how such cofferdams were made during the construction of Sinan’s Büyükçekmece Bridge (completed 975/1567-1568). In Sâî’s manuscript Tezkiretü’l-bünyân, in which he described the life of Sinan, the construction of the bridge’s foundation is explained as follows46
... They built galleon-like cofferdams at each foot of the bridge and drained the seawater with pumps and piled poles, measuring the equivalent of two to three men’s height, with sledgehammers, tied block ashlars with strong iron clamps, pouring lead between these... (and thus completed the construction of the foundation).
The remarkable thing about this groundwork is the preference for using cut stone blocks and heading bonds in the bridge construction, although the foundation was made of rubble in land structures.
The first thing that had to be done when carrying out the steps to make a structure on the ground was, of course, the preparation of the structural framework. The carrier system could only be built by this means. Therefore, elevating scaffoldings were built around the structure. These scaffoldings surrounded the structure both internally and externally, had horizontal bars every three arş (Turkish yard) and, if one takes into account the materials ordered, the scaffoldings were suspended from wooden poles whose surface was roughly scoured.47 According to the information from the available inventory lists, Ottoman builders did not practice the method of knotting to build piers; this practice was very common and widely followed in Europe until the end of the 19th century. Rather than knotting, they made the connections with nails.
The transporter system elements for transferring the loads from the superstructure to the ground can be classified into two groups: walls and individual carriers. Several types of wall construction techniques could be seen in sixteenth-century Ottoman architecture. As they were economic, debris and pitch-faced walling were used in structures which were of modest size and scale. In this technique, known since the ancient Roman era, the stones, composed of sand, lime and brick-tile dust and small rocks, were bonded with a grout known as khorasan by the Ottomans. The use of the alternating bond construction method in the sixteenth century was only seen in masonry and second-degree structures of large külliyes which had been built with limited funds. However, this was a widely adopted method in the fifteenth century and created a visual diversity. Ottoman mixed brick and stone walls used a double bonding, like the opus mixtum of the Romans. In this type of bonding, the exterior wall consisted of regular horizontal stone and brick layers, but usually the inner wall was built with pitch or rubble stones. Together with khorasan, stone and brick chips would be used to irregularly fill in the section between the walls. Khorasan mortar was also used as the binder between the layers of stone and brick of the walls.48
In almost all prestigious constructions, adjacent coursed cut stone masonry techniques were applied. As with alternating bond, the wall was built in adjacent coursed masonry and rubble was used to fill the spaces between the walls alongside the khorasan. In large monumental constructions, both walls tended to be coursed in an adjacent manner. However, there are examples of construction where only the outer wall was built in this way, with the inner surface being composed of cheaper rubblework; this was the result of economic problems. Only five surfaces of the blocks used in the adjacent coursed cut stone work would be made with hammer-dressed stone. As the sixth surface was facing the interior filling part, it would be only roughly formed. The size of the blocks used varied. However, building blocks of small size were generally avoided. The length ranged from 0.80 to 1.00 m, with a height ranging between 0.30 and 0.40 m; measurements outside this range were rare.
One of the important features of practical Ottoman masonry was the absence of masons’ marks on the blocks. However, such marks were characteristic both in Europe in the Middle Ages as well as in Anatolian Seljuk architecture. The function of the marks was being able to estimate the amount of work carried out by each mason. It can also be considered a type of individual “domain” in a guild-type organization, that is, a “ritual” of the professional practice. The Ottoman building system did not allow stonemasons to be involved in any guild-like organization. In a system where the masons were casual laborers, there was no need for masons’ marks. However, a mason marking system made with paint was known to have been used in the seventeenth century, although it would be difficult to describe this as a mason mark in real terms. Because the stones were purchased from quarries operated by private individuals in that century, knowing the origin of the stone that was brought to the construction site was necessary, leading to the emergence of this system.49
From the end of the fifteenth century in adjacent coursed masonry, iron clamps and melted lead began to be used extensively in order to connect the stone blocks to one another.50 Clamps provided a rigidity to the stone masonry that could not be provided by masonry alone. Early clamps were slightly different in terms of size and manufacturing techniques as compared to those of later periods. A piece of iron would be made into pieces measuring 1.5 to 2 cm in width and 20 to 25 cm in length; after being forged and folded into two, the ends were twisted 3-4 cm upwards, giving the clamp a U-shape.51 In examples from the sixteenth and following centuries, however, iron rods were forged and shaped only once; they were not folded over and clamps were manufactured in sizes suitable for a variety of purposes. At this time, the clamp size increased. Although not reaching the diversity of forms that existed in the ancient Greeks, according to documents, a number of clamp types, described as small, large, medium, new or European clamp were available.52 However, it is not possible to understand from the terms mentioned in the documents which clamp style is being described.
In adjacent coursed stonework, iron pins, called dil (tongue), were used in addition to the clamps. These pins connected horizontal rows of stone to one another and strengthened the wall construction against transversal force.53 However, pin usage in all walls was not a very common practice in the sixteenth century. It only served the function of connecting the console blocks that constituted the entablature moldings in the main structure mass. In contrast, pin usage in minarets was very common. Minarets had to be well fortified against the transversal forces, as they had a delicate tower structure in diameter and height; moreover, Turkey is an earthquake zone.54 The practice of connecting the rows of stone to one another vertically with iron pins was used more frequently in the following centuries. For example, in Nuruosmaniye Mosque (1748-1755) all the exterior walls were clamped and pinned.55
In almost all types of masonry walls, there were bonding timbers which surrounded all of the structures at certain grades. Approximately 10 cm of the inner and outer surface of the wall structure used this bonding timber system; this usually consisted of two elements parallel to one another, forming a square or square-like section. The elements in question were connected to each other horizontally at regular intervals and with other elements that equally transected the wall. Also, examples indicate that the bonding timbers were attached at the corners with dovetails and nails. Therefore, it can be said that they functioned as a type of fitting within the masonry. This system was widely used in sixteenth-century construction. However, starting with Sultan Selim Mosque (completed in 929/1522-23) in Istanbul, rather than wooden beams, the buildings were fortified with two-parallel iron bars, on at least one elevation.56 This practice developed well in the next century, evolving into a system of iron reinforcements that fortified the structure in various forms; this matter will be addressed under a separate heading.
Structural elements, called monolithic bearings, consisted of buttresses and columns. It is possible that there was no difference, in terms of structure, between the buttresses and the adjacent coursed stonework. However, the situation that existed in large mosques has not been determined, as it is not possible to conduct more in-depth research than a surface survey. For example, were iron pins used in the piers? It is not possible to answer this question for sixteenth-century construction. However, it is clear that pins were not used for Cerrahpaşa Mosque (completed: 1002/1593-1594), the internal construction of which could be investigated after masonry surfaces were revealed after a fire. Nevertheless, it would be difficult to claim that iron pins were not used in the construction of buttresses on larger construction site.
The problems in the carrier system were largely solved in Ottoman classical architecture with the use of columns and arches. Ottoman columns were massive, monolithic structural elements. There is no example of building columns by placing one on top of another, as was done in antiquity. The use of the pedestal for columns was rare as well; the preferred method was to place the columns directly onto the base. One of the rare uses of pedestal columns can be seen in the interior of the tomb of Sultan Süleyman I (completed after 1566).
In all observable instances, it can be seen that bases of the columns have been filled with melted lead, while the column headers had iron pins and pin holes. This practice has been known since antiquity. However, an element that was unknown in antiquity which was widely used in Ottoman architecture was the connection brace, known as parazvane. Functioning as a concealer for the joint between the shaft of the column and the ground, these braces were usually made of bronze, more rarely brass.
It is obvious that during the construction of the structure, horizontal and vertical elevations needed to be continuously checked. In Ottoman architecture a plumb was used to ascertain the vertical level; this was common in other eras and locations. All known samples of plumbs are in the form of a spindle, the length of which was at least 15-20 cm; the connection to the spindle was made with a rope secured by a small jointed brass or bronze part. To inspect horizontal services was more difficult. For small surfaces, squares were sufficient. However, it is clear that this process could not be performed with a square. Certain levels of the structure, called tabaka (layers), functioned as elevations so that the levels of horizontal surfaces could be checked. It is thought that the cornice elevations in Hagia Sophia and large Roman structures also served as levels; this was also the case in the Ottoman structures.
In large mosques, the first layer was the elevation that began with the relieving arch on the exterior walls. The second layer was the level at which the exedral half-domes on the lower elevation were placed. The starting elevation of the main half-domes, the third layer, rested on the second level. The main dome rested on the fourth layer. The number of layers would increase or decrease according to the design of the construction. However, an essential feature of all the layers was that they would surround the structure. This allowed for all the necessary precautions to be taken to ensure that the horizontal surfaces on each layer were level. In order to inspect the level of horizontal surfaces in high areas, the only tool available for the Ottoman architects was the aforementioned “air balance”. After levels had been checked with this instrument, the construction of the next level could begin.
Generally, the preferred solution for every size of dome was building it as a bare shell with a plain pattern. Umbrella and polygonal umbrella domes were rarely built; however, when built, they were generally small in size. Within the walls of the outer dome of the tomb of Sultan Süleyman I, a ribbed masonry, rising to a certain elevation (approximately one-third of the height of the dome), can be seen; this is unique.57
No mold was used in constructing the umbrella or plain domes. This can be seen by looking at the domes, the inner surface of which was not plastered. This was also verified by the notes of an eighteenth-century traveler.58 There is no reason to think that the construction practice was different for domes that were plastered. At every stage of the construction of the dome, there was no need for a mold, due to the fact that there was a structure which could bear its weight. The only problem encountered in the construction, which had a geometric form or double curvature, was to create the proper spherical shape during the period when the bricks were laid. It is thought that very small domes were built to assist in this. In larger domes, the curvature was continuously checked while the bricks were being laid, using a rope the length of the radius; this would be tied to the center of the dome. As in large mosques the dome center could sometimes be 20-30 meters from the ground, a simple solution was found to determine this point: a pole which reached from the ground to the center of the dome would be erected, and a master rope or ropes would be tied to the pole.59 The same technique was used to determine the curvature of the pendentives that bore the weight of the dome, geometrically forming a section of the dome at a lower elevation.
Half-domes were impossible to be built without a mold. However, no data is available regarding how the Ottoman architects created these molds. Thus, it is unknown whether the mold was placed on bases that started from the floor, or rested on a cornice that ran along the layer from where the dome started. It is thought that the same technique used in supporting the main arches of bridges was adopted in half-domes and even in large vaulted arches, resting on the spaces created by the bases.
The construction of the domes was always carried out from outside. Holes as large as 20-25 cm, at a horizontal interval of 1.00-2.00, and a vertical interval of 1.00 mm were made on the outer surface during the construction of the dome; these holes were used to place the wooden elements that carried the base on which masonry workers stood. After the construction was completed, an element that was in the form of a truncated cone, known as a keystone, was placed in the small round opening that was at the top of the dome. This can easily be seen in the interior domes of the tombs of Sultan Süleyman I, Murad III (1954), and Mehmed III (1608).
Special thin square bricks were generally used in building Ottoman domes. Two sizes of brick, büzürg (large) and nime-i büzürg (half large) were used for dome construction.60 The size of the nime-i büzürg brick in the exterior dome of Sultan Süleyman I’s tomb was determined as being 5.0 x 24.0 x 49.0 cm.61 The large brick used in the main dome of the Süleymaniye Mosque was twice as large, namely 5.0 x 49.0 x 49.0 cm. The bricks were always positioned with their joints facing the center of the dome. This was necessary as the dome needed to support itself without the use of molds during the masonry process. A special empty earthenware container, called sebu (amphora) was placed in the masonry work.62 The reason behind this practice is a matter of debate. However, it is certain that the sebu were not used to lighten the weight of the dome during construction of large structures like the Süleymaniye Mosque; indeed, the usage of the sebu was very limited. As all the small domes comprising the roofs of rooms in a sixteenth-century structure like Hasan Paşa Han in Diyarbakır were bonded with special earthenware tiles,63 this suggests that the purpose was to provide relief. Since Ottoman domes were generally covered both internally and externally, it is not easy to observe the inner construction. Similar bonds can be seen in some other structures. This type of building technique is associated with the amphorae dome masonry used in Roman architecture. However, this comparison is quite misleading. The masonry in Hasan Paşa Han is not in the form of the San Vitale Basilica’s masonry in Ravenna, an early Byzantine structure. While in the Basilica of San Vitale the long axis of the amphorae followed the vertical curvature of the dome, the long axis of the tiles in the Ottoman example were directed towards the center of the dome.
Although the domes were always made with bricks, except in small structures, the pendentives were bonded with cut stone in large structures, such as sultans’ mosques. It is thought that this was helpful in preventing deformation.64
The use of iron “reinforcement” elements was widely used in an effort to strengthen the stone construction in monumental Ottoman buildings in Istanbul.65 The problems of gravitational forces in stone construction were understood by the Ottoman architects. Therefore, the use of some iron elements in sections and points which were important for the structure was regarded as necessary. The techniques that were widely known and implemented from the beginning of the fourteenth century consisted of clamps and pins, mentioned under the section “Carrier System”, where types of masonry are described. Clamps, thought to make important contributions to counterbalancing the tensile stress, had another function in construction. If there was not enough mortar in the horizontal joints of the adjacent coursed stonework to enable the blocks to be joined, as mentioned previously, this stonework was built with two walls. In order to prevent the deformation of the walls which resulted from pressure as the wall rose, clamps were used; these played an accelerating role since the hardening of the rubble stone-khorasan mixture that filled the gaps between the walls took some time.
In early Ottoman architecture, wood braces, interior and span braces could be seen in almost all structures; during the sixteenth century this was particularly true in structures that were modest in size and cost. The use of iron instead of wood was first seen in Edirne Üç Şerefeli Mosque (1438-1447), although only to a limited extent. In this structure, iron braces used in the arches over the revak (porticos), thought to be part of the original design, can be seen. Iron continued to be a valuable material that was utilized only in important structures, despite the widespread use of iron braces from this date; this usage was directly related to changes in the production technology of iron by the Ottomans, which dates to the early sixteenth century. Beyazıt Mosque (1501-1506) in Istanbul marks a turning point for the era. In this structure, both the revak and all bases of the interior were connected to one another and to the walls by iron braces. From the end of Bayezid II’s reign to the end of Koca Sinan’s era, Ottoman architecture developed the use of braces; this would continue until the eighteenth century. Beyazıt, Şehzade (1544-1548) and Süleymaniye (1550-1557) Mosques became the norms, setting the standard in terms of the positioning of the span braces in the interior of large structures. In all three examples, the bases supporting the main dome were fixed to the outer walls with double braces. They were also largest cross-section braces in the structure. The cross sections in Beyazıt Mosque, measuring 10 x 15 cm with a height of approximately 7.5 meters, are remarkable in their dimensions. The brace sections which reached 9 x 16.5 cm in Şehzade Mosque were quite large compared to latter applications. Secondary elements outside the main mass of the large mosques during the classical period and the cross-sectional dimensions of reinforcement braces were also proportional to their structural importance.
The reinforcement of the joist hanger level with the brace of exedral half-domes on the lower elevation of main half-domes was first done in Şehzade Mosque. All exedral half-domes were built with braces after the mid-sixteenth century. From the Classical Age of Ottoman architecture, the iron brace was handled cautiously in situations where problems were expected to occur. The most interesting example of this type of precaution can be seen in Edirnekapı Mihrimah Sultan Mosque (completed before 973/1565-1566). In this structure, the weight-bearing points of the two main supporting arches on the two sides of the mihrab axis were connected to one another with a double interconnected brace in the joist hanger elevation. The function of these braces was presumably to prevent any deformations that might occur in a perpendicular direction to the mihrab axis. It was thought that the bases supporting the main dome would not be deformed in a parallel direction to the mihrab axis, as the side naves in that direction would prevent this. The braces functioned well; no opening can be observed in the base of the mihrab wall in the southeast-northwest direction. However, very serious structural deformation exists in the perpendicular supporting arches of the main dome, which were not reinforced with braces.
In some of the structures, iron braces connected to one another were placed in the walls that provided structural fortifications. Their function was to connect one corner of the structure to the other. These kinds of elements could generally be seen on the mihrab walls in classical period structures, such as Piyale Paşa (completed in 981/1573-1574), Edirne Selimiye (completed in 982/1574-1575) and Nişancı Mehmet Paşa (completed in 997/1588-1589).
In Ottoman architecture, iron elements that supported the brick flooring, stairs and the roofing system constituted another group; these were known as kiriş (beam). It is known that in almost all large mosques the flooring of the mahfil (elevated lodge for the sultan), which did not have pendentives, was supported by a type of cable flooring in which iron beams were used. This can be observed in examples like Kadırga Sokollu (completed in 979/1571/1572) and Cerrahpaşa Mosques. The most complex example of this system can be seen in the ground floor of the Topkapı Palace Harem, Valide Sultan Quarter (post 1665), which was built with iron instead of wood on the order of the sultan when it was rebuilt after a fire in 1665. Here, a flooring vault can be seen; this was placed perpendicularly to a triple beam group that measures 3 m in width and with a cross-section measuring 10 x 8.5 cm. This consists of iron beams with 35-40 cm spacing and includes small brick drop vaults which were used to fill in the spaces. It is interesting that this and other iron floors in the Ottoman state were made due to a worry about fire; these date back further than any known European examples. Early Industrial Age floors in Europe were also built with the fear of fire in mind.
Either out of aesthetic concerns or because of the inability of iron to meet every need, it was uncommon to reinforce DÜZ TAŞ ATKI with secondary iron beams until the 1950s. In fact, this attitude still survives for important structures like Süleymaniye and the Edirne Selimiye Mosques. In the late sixteenth century, the rule was to support the TAŞ ATKI from below with a secondary element. In classical Ottoman architecture, the most common place where TAŞ ATKI and secondary iron beams could be found were door and window spans, stairs, landings and secret passages that emerged from the skirts of the dome skirt. The long edges of the profiles of the iron beams were positioned horizontally wherever they were used.
The most interesting part of the use of reinforcement metal elements in classical Ottoman architecture was the system of “bracing”. The braces surrounded the structure at one or more elevations and were part of the construction. Other architectural traditions and early Ottoman architecture had wooden examples of braces, either within domes or in the structural mass on which the dome was positioned. It took about a century for the wood to be substituted with iron in Ottoman architecture and for wood to be almost completely eliminated from masonry construction. At the end of the sixteenth century the same was true for bracing systems. Edirne Üç Şerefeli and Edirne Beyazıt Mosques (1484-1488) were the first examples where iron was used for bracing the domes. As noted above, these were leading structures. While the lower parts of the main wall were lined with bonding timbers as before, there is sufficient data indicating that an iron brace was included in the tensile region of the main domes. The details observed in Istanbul’s Sultan Selim Mosque, which was built in the first quarter of the sixteenth century, reveals that the main structure mass was first reinforced with double bracing on the same elevation. However, no information indicates whether a hidden bracing existed in the wall construction of Şehzade, Süleymaniye or Selimiye Mosques, which were built later.
To have a hidden bracing system surrounding the infrastructure on which the dome was positioned was rare in the sixteenth century. However, braces on the joist hanger elevation of the main supporting arches which supported the dome in some monumental structures functioned a surround. This practice emerged after the 1560s. Examples of this include Eminönü’s Rüstem Paşa (1562?), Babaeski Semiz Ali Paşa (after 1560), Kadırga Sokollu, Edirne Selimiye and Azapkapı Sokollu (completed in 985/1577-1578) Mosques, and Sultan Süleyman I’s tomb. It seems that the practice of surrounding both the construction of the exterior walls and the base system that supports the central dome separately with braces first appeared in the 1580s. Selim II’s Tomb, Kılıç Ali Paşa Mosque, Mesih Paşa Mosque (completed in 994/1585-1586), Nişancı Mehmet Paşa Mosque (completed in 997/1588-1589), Cerrahpaşa Mosque, Murat III’s Tomb are structures that were fortified in this manner.
The structural nature of domes made it necessary to create a tension area with the entire skirt structure, which formed an angle of approximately 38 degrees to the horizontal plane that passes through the center. However, materials such as stone or brick used in the construction of the dome were not suitable to answer this tensile stress. Thus, this area had to be fortified and reinforced in some way. The solution for the problem was the same throughout the eras: either the dome wall had to be thickened in the tension area or reinforcements had to be added in order to meet the tensile stress, thus compensating for the wall’s failure to meet the stress. For example, in Ancient Roman architecture the first method was preferred, while there are those who preferred the second method, using a wooden brace, for example the dome of the Florence Cathedral. In the mid-fifteenth century, Ottoman architecture tried to solve the problem of reinforcement by placing a brace in the construction; this was first seen with the main dome of Edirne Üç Şerefeli Mosque. This brace consisted of forged iron bars that were connected to one another. The thickness of the bars varied in every structure. Measurements, such as 7 x 7 cm in Edirne Üç Şerefeli Mosque, 95 x 7 cm in Şehzade Mosque, and 5-6 x 3-3.5 cm in Kılıç Ali Paşa Mosque are examples of the approximate upper and lower limits. In almost all of the mosques built during the 16th century, braces were passed through the windows along the skirt of the dome skirt in large tombs. The position and number of braces in the dome differed slightly according to the structure. Sometimes the brace was situated in the thick molding known as kubbenin kafa tahtası (head board), at the top of the windows. There are structures like Süleymaniye Mosque, where the dome was braced both from within the windows and at the kafa tahtası level as well. Sultan Süleyman I’s Tomb is the only known example from the sixteenth century to use braces on the line on which the dome is situated in the main structure. This practice became widespread in the eighteenth century.
Ottoman architects had extensive and unrivalled knowledge about using iron in reinforcements before the building of Sultanahmet Mosque (1609-1617). Thanks to this knowledge, although the Blue Mosque echoes the Şehzade Mosque along the main lines of the structural schema, it is also a structure that included all the developments from the previous five decades. Thus, this building was a peak of Ottoman architecture. Ottoman architecture maintained progress in using iron reinforcements and bracings from the seventeenth century to the early nineteenth century. However, with the start of the Industrial Era, new architectural-metallurgical breakthroughs which occurred in Western Europe left Ottoman technology behind the times; new techniques of iron use came to the Ottoman lands late, in the second half of the nineteenth century.
As a term, finishing elements included all the techniques that finished the rough structure. These can be divided into subgroups, such as paneling, roofing, carpentry, glasswork, water and technical fitting elements, including heating.
Paneling in Ottoman monumental architecture was used only for interiors. External surfaces were left bare, as alternating walls, cut stones or marble bonds. The implementations of decoration and damascene techniques on marble surfaces were always carried out after the rough construction was completed. It was possible to observe the marble covering of a flat surface and the subsequent decoration on the outer walls of Mehmed III Tomb, which is a partially completed structure.66 Tile coverings were made on the pediments outside the first row of windows in some large mosques and under the porticos in tombs. These were placed in the niche that were formed by carving the outer limestone surface, using a regular lime mortar, similar to marble inlaying.
The most common covering in the interior surfaces of the structures was plaster. First, the wall surface had to be chiseled so that the plaster would adhere better to the surface. Then, a mortar composed of wadding (flax fiber), lime, sand and water were applied as a thick layer of 3-4 cm on the surface in liquid form. Clay was sometimes added to this plaster in some structures. Another layer of thin and non-wadded mortar was applied on this rough layer of plaster. Special large headed plaster nails would be hammered into the surface at irregular intervals to ensure the adhesion of the rough layer of mortar to the dome, pendentives and ceilings; then plaster was applied to these. A thin plaster would be applied, a malakari decoration layer could be applied. The malakari layer was composed of a plaster-based mortar and low relief patterns were made by carving some spots with a small spatula while still wet. The upper surface of this was usually colored with black and red. For decorative purposes, interior muqarnas were considered to be part of the malakari decoration technique. These were made by giving tidier shapes with the malakari method on the rough underside of the bricks. However, marble and stone muqarnas on the crown gates of the structures were different. They were shaped in situ, like all other external decorations of the structure after the rough construction had been completed on the semi-processed surface, being carved like a sculpture.
Tiles were used as an interior covering, particularly seen in costly prestigious structures. Rough wall surfaces, which were directly chiseled, were covered with a thick layer of lime mortar. Other than in a few rare exceptions, the tile was in the shape of a square or rectangle and were standardized. The main reason for standardization was that the tiles were manufactured in a city (İznik) which was far from the regions where they were to be used. Thus, only the decorations would be different. Records confirm that these designs were custom-made, according to patterns sent from Istanbul, and would vary according to where they would be used in the structure. However, it should be noted that not all tiles were produced in this way; there was also a larger production area which reflected the preferences of the masters in İznik. As with the tiling technique used in modest structures today, the available material was adapted to fit, and the tiles were directly adhered to the walls.
The technical comfort elements of a classical Ottoman structure included heating and accommodation for fresh and waste water. Pits were generally used for heating, except when there was a complex stove-furnace system, 67 such as that found in Edirne Rum Palace. A sixteenth-century furnace in the structures, which is outside the boundaries of traditional architecture, was comprised of a conic smoke hood connected to a simple chimney. The smoke hood was a structural feature that was supported by a frame that was made of forged iron bars connected to one another. This frame could sometimes be covered by a brass sheet. Covering with tiles was seen slightly more often. Decorated plaster work was generally preferred due to its low cost. Whatever material the outer surface was made of, the inner surface would be coated with firebrick. This type of furnace was considered to be a luxury in this period and there were more modest types of stoves, of a prismatic shape, made of stone.
The heating system in baths were of a more complicated technical structure.68 The contribution of Roman baths is evident in these baths. The center of the bath system was the furnace. The numbers of stoves were dependent on the size of the bath. A copper water tank was heated by the furnaces. The smoke from the furnace or furnaces emitted partially from a chimney behind the storeroom. The hot air and vapor heated the structure by passing through the horizontal pipes (tüteklik) within the walls and the void (cehennemlik) created under the flooring, then passing through the small chimneys on the roof. It is not known how this advanced heating system was used in other buildings.
Two types of fresh water installation, fed by earthenware pipes and lead pipes, can be seen in structures that have classical Ottoman construction technology. However, pipe drains were mostly used in systems that provided water to the building from the main supply, whereas the waterworks, especially those with taps, were made with lead pipes. Lead pipes were made by folding the sheets on a cylindrical mold with a seam and welding the ends. Either the lead or the earthenware pipes were joined to one another with a special paste called lökün.69 The use of pipes was avoided in water piping, and special galleries made with bricks were preferred. Toilet bowls were placed directly on these large galleries and connected to the main sewer without leaving any space for interconnecting components.
The roofing material preferred in classical Ottoman architecture was lead. Roof tiles were always used to cover small scale and modest structures in every era in Istanbul. Instead of tiles, lead was sometimes used to cover the wood parts of masonry structures or important wooden buildings.70 Saturated felt roofing underlay was used if lead was the material used to cover wooden surfaces. The benefit of this underlay was to prevent the lead from heating so much from the sun that it could set fire to the wood underneath. For the lead covers of masonry surfaces like domes, special plaster nails were hammered in and then a 5-10 cm thick layer of hay and clay mix was plastered on top (a small amount of lime was also added). The lead, which was beaten into sheets, would cover this layer, stretching from the peak of the dome to the skirt. After the sheets in the adjacent lobes were hammered with lead nails, they were fixed by folding and soldering the junctions.
Until the mid-nineteenth century, in Ottoman monumental buildings of Istanbul glass was used in a type of stained-glass technique known as revzen. Both domestic and imported colored glass was used. The revzen was simple to make. First a mold would be prepared. Stained glass would be cut and placed into the molds. Then, a liquid plaster would be poured into the mold and left to set. However, since plaster alone could not sustain the weight of the glass, this stained glass had a wooden outer frame; ones with a wider surface even had fixtures composed of vertical and horizontal iron bars.
The finishing components to be placed last in the structure were wooden doors and window frames. These elements required a high level of craftsmanship and the craftsmen did not work on the construction site, but in workshops. When important structures were being built in small provinces, door and window frames were often made in Istanbul and then sent out.71
It is obvious that the technical knowledge Ottoman builders relied on was not the product of just one tradition, but instead was fed by many sources. In some areas, the data suggests the existence of a continuity from antiquity to the classical Ottoman period. For example, the urban water system was directly related to Roman practices in the same field, and these were preserved due to Istanbul being a Weltstadt that never lost its vitality.72 The same claims cannot be made in other areas. The similarity between basic Roman and Ottoman construction techniques could be explained by the mediation of the Byzantines. Yet, the cofferdam and piled footing used for construction in water or on weak soil had long been forgotten by the final centuries of the Byzantine Empire. Information about how Byzantine masters built structures in the thirteenth and fourteenth centuries, which required neither cofferdam nor piled footing, has been found. As no detailed historical panorama of Byzantine technology has been drawn, this needs to be cautiously explored. Nevertheless, functioning like a technical hub, similar to Venice, the role of the Istanbul Shipyard should not be overlooked. Even in the early nineteenth century, the expertise of shipyard officials was applied in such matters.
The role of Iran-Central Asia on Ottoman construction technology in the fifteenth century and later in supporting Anatolia in the early centuries with skilled labor is a question that needs to be investigated. The influence of the region on plaster work must have been considerable. It is, of course, more accurate to think the opposite could be true for rough structures. Iranian-Central Asian builders approached and carried out carrier and covering structural elements of buildings from a completely different understanding. Even the matter of basic construction did not concern them much. Some buildings were built directly on levelled ground. In metal use, which was of vital importance in classical Ottoman structures, no information exists which can be used for the sake of comparison with other Islamic traditions. In terms of the structural use of iron, Ottoman practices show similarities with Italian and, increasingly, Western European trends. We do not know whether this similarity is the result of any direct relationship. However, some elements such as kılıçlı bağlantı can be considered to be technical knowledge which Ottoman architecture owes to the Italians.73 In summary, there is still much to learn about the origins of the architectural technology used in the Ottoman state.
1 For a brief explanation, see: C. Orhonlu, “Şehir Mimarları”, Osmanlı İmparatorluğunda Şehircilik ve Ulaşım Üzerine Araştırmalar, ed. S. Özbaran, İzmir: Ege Üniversitesi Edebiyat Fakültesi, 1984, pp. 13-14.
2 For “Mühendis”, see: Câfer Efendi, Risâle-i Mi‘mâriyye: An Early-Seventeenth-Century Ottoman Treatise on Architecture, tr. and published H. Crane, Leiden: E. J. Brill, 1987, in the translation p. 96 and f. 78 in manuscript 78. For “Mesahacı”, see: TSMA, D.1797 and 977/ dated 1569-70 Edirne Selimiye Mosque Construction Cost Journal.
3 BOA, Mühimme Defteri, no. 22, order no. 397, p. 207, 6 Rabi al-Akhir 981/July 5, 1573; no. 23, row no. 275, p. 135, Rajab 28, 981/ November 24, 1573; no. 34, row no. 386, p.185, Safar 22, 986 / Nisan 30, 1578; no. 36, row no. 879, p. 333, 9 Rebîülâhir 987/ 6 May 1579.
4 Information regarding this operation exists in Sâî: S. Saatçi, “Tezkiret-ül Bünyan’ın Topkapı Sarayı Revan Kitaplığı’ndaki Yazma Nüshası”, Topkapı Sarayı Müzesi: Yıllık- 4, Istanbul: Topkapı Sarayı Müzesi Müdürlüğü, 1990, p. 73.
5 For a detailed study on Ottoman architectural drawings, see: G. Necipoğlu-Kafadar, “Plans and Models in 15th- and 16th- Century Ottoman Architectural Practice”, Journal of the Society of Architectural Historians, vol. 45, no. 3 (1986), pp. 224-243.
6 Necipoğlu-Kafadar, “Plans and Models”, p. 236 ed seq.
7 Surname, TSMK, H.1344, ff. 190b-191a.
8 J. S. Ackerman, The Architecture of Michelangelo, Harmondsworth: Penguin Books, 1970, p. 49.
9 From other Islamic countries, only plans from Iran and Central Asia can be found. Today, these can be found in Uzbekistan Academy of Sciences and the National Art Museum at the Victoria and Albert Museum in London.
10 For example, when Selimiye’s construction began in Edirne, material prices rose significantly: BOE, Mühimme Defteri, no. 7, row no. 1589, p. 564, Dhu al-Hijja 24, 975 / June 20, 1568.
11 There are many documents on compulsory civil service obligation. For example: BOA, Mühimme Defteri, no. 5, row. 1725, p. 620, Dhu al-Qidah 10, 973/ May 29, 1566 (The allocation of the population of Komat Village to the palace in Filibe. no. 26, row no. 565, p. 203, Jumada al-Awwal 21, 982/ September 8, 1574 (Christian Galley slaves used in Edirne Palace); no. 31, row. 126, p. 49, Jumada al-Awwal 4, 985/ July 20, 1577 (For the repair of Galata Palace Bolu müsellems’ post); no. 52, row no. 404, p. 161, Dhu al-Qidah 26, 991/ December 11, 1583 (Charging Tekirdağ nomads for the repair of Istanbul Fort).
12 For example, this kind of workforce was not utilized much in the construction of Sultanahmet Mosque. See Z. Nayır (Ahunbay), Osmanlı Mimarlığında Sultan Ahmet Külliyesi ve Sonrası (1609-1690), Istanbul: İstanbul Teknik Üniversitesi, 1975, pp. 88-93.
13 Ö. L. Barkan, Süleymaniye Cami ve İmareti İnşaatı (1550-1557), II vol., Ankara: Türk Tarih Kurumu, 1972-79, vol. 1, p. 351 ed seq.
14 Barkan, Süleymaniye Cami ve İmareti İnşaatı, vol. 1, pp. 358-359; vol. 2, p. 51, doc no. 100, p. 61, doc no. 121, 122.
15 For examples of documents, see: BOA, Mühimme Defteri, no. 26, row no. 280, p. 109, Rabi al-Akhir 4, 982/ July 24, 1574, row no. 362, p. 139, Rabi al-Akhir 19, 982/August 8, 1574. Also see K. Erguvanlı, Z. Ahunbay, “Mimar Sinan’ın İstanbul’daki Eserlerinde Kullandığı Taşların Mühendislik Jeolojisi ve Mimari Özellikleri”, Mühendislik Jeolojisi Bülteni, 1989, no. 11, p. 111; K. Erguvanlı et al., “The Significance of Research on Old Quarries for the Restoration of Historic Buildings with Special Reference to Marmara Region, Turkey”, The Engineering Geology of Ancient Works, Monuments and Historical Sites Preservation and Protection, ed. P. G. Marinos and G. C. Koukis, IV vol., Rotterdam, Brookfield : A.A. Balkema, 1988-90, vol. 3, pp. 631-638.
16 For example: Mihalıççık, Edincik, İzmit, Marmara Ereğlisi. Barkan, Süleymaniye Cami ve İmareti İnşaatı, vol. 1, p. 352 ed seq.
17 Nayır (Ahunbay), Osmanlı Mimarlığında Sultan Ahmet Külliyesi, pp. 97-98.
18 See: U. Tanyeli and G.Tanyeli, “Osmanlı Mimarlığında Devşirme Malzeme Kullanımı (16.-18. Yüzyıl)”, Sanat Tarihi Araştırmaları Dergisi, vol. 2, p. 4 (1989), pp. 23-24.
19 BOA, Mühimme Defteri, nr. 14, row nr. 506, p. 350.
20 Erguvanlı et al., “The Significance of Research on Old Quarries”, p. 636.
21 Erguvanlı et alç, “The Significance of Research on Old Quarries”, pp. 636-637.
22 This was observed by Nâsır-ı Hüsrev in Remle. Nâsır-ı Hüsrev, Sefername, tr. A. Tarzi, Istanbul: Milli Eğitim Bakanlığı, 1985, pp. 29-30.
23 For example, buttresses, stairs and flooring stones were ordered from Marmara Island for the construction of Atik Valide Mosque; BOA, Mühimme Defteri, no. 52, row no. 242, p. 100, Shawwal 14, 991/ October 31, 1583. Also see Barkan, Süleymaniye Cami ve İmareti İnşaatı, vol. 2, p. 39.
24 Erguvanlı and Ahunbay, “Mimar Sinan’ın İstanbul›daki Eserlerinde Kullandığı Taşlar”, p. 111.
25 Barkan, Süleymaniye Cami ve İmareti İnşaatı, vol. 1, pp. 353-354; vol. 2, pp. 39-40. Also the demand for the arches of Nişancı Mehmet Paşa Mosque: BOA, Mühimme Defteri, no. 60, row 547, p. 230, Jumada al-Awwal 18, 994/May 7, 1586.
26 Marble column demand from Bozcaada: BOA, Mühimme Defteri, no. 26, row no. 823, p. 284, Rajab 7, 982/October 23, 1574.
27 A decree on the incapacity of providing recycled columns for the Europeans from Eğriboz, Atina, Livadiya, İstefe: BOA, Mühimme Defteri, no. 33, row no. 357, p. 181, Shawwal 17, 985/ 28 December 28, 1577.
28 Barkan supposes that most of these are not collection centers, but quarries. Barkan, Süleymaniye Cami ve İmareti İnşaatı, vol. 1, p. 352 ed seq.
29 For example, this was the case in Selim II’s Tomb: U. Tanyeli, “Kanuni ve II. Selim Türbeleri Üzerine Bir Değerlendirme”, TAÇ Vakfı Yıllığı, no. 1 (1991), p. 85.
30 Barkan, Süleymaniye Cami ve İmareti İnşaatı, vol. 1, p. 381 ed seq.
31 Barkan, Süleymaniye Cami ve İmareti İnşaatı, vol. 1, p. 384.
32 Nayır (Ahunbay), Osmanlı Mimarlığında Sultan Ahmet Külliyesi, pp. 102-103.
33 For the construction method, see: Y. Kahya, “İstanbul Bizans Mimarisinde Kullanılan Tuğlanın Fiziksel ve Mekanik Özellikleri”( Ph D. Dissertation), İstanbul Teknik University, 1991, p. 6 ed seq.
34 For the iron manufacturing centers: Barkan, Süleymaniye Cami ve İmareti İnşaatı, vol. 1, pp. 361-369; vol. 2, pp. 122-148; G. Tanyeli, “Osmanlı Mimarlığında Demirin Strüktürel Kullanımı (15.-18.Yüzyıl)”( Ph.D. Dissertation), Istanbul Teknik University, 1990, pp. 6-17.
35 Barkan, Süleymaniye Cami ve İmareti İnşaatı, vol. 2, documents on pages 129-131.
36 Barkan, Süleymaniye Cami ve İmareti İnşaatı, vol. 2, p. 139 (For Samako); BOA, Mühimme Defteri, no. 10, row no. 482, p. 298, Shaban 7, 979/December 25, 1571 (Samako’ya); Ahmed Refik, Hicri On İkinci Asırda İstanbul Hayatı (1100-1200), Istanbul: Devlet Matbaası, 1930, pp. 181-182, doc no. 220: Shawwal 1169/ July 1756 (regarding İzmir nail manufacturers); For a document regarding the nail types required from Istanbul nail manufacturers see BOA, Cevdet-İktisat, row no. 866, Jumada al-Awwal 20, 1183/ 21 October 21, 1769.
37 BOA, Mühimme Defteri, no. 51, row no. 37, p. 11, Shaban 7, 991/August 27, 1583 (Siyavuş Paşa’nın Sofya’daki yapılarına gerekli kurşunu gönderilen adamına parasıyla verdirtmesi için Üsküp kadısına hüküm); no. 52, row no. 342, p. 135, Dhu al-Qidah 5, 991/November 21, 1583 (with the same purpose as Alacahisar, Üsküp, Kurşunlu, Şehirköy, İzenbol and Preznik qadis); no. 58, row no. 747, p. 293, Ramadhan 17, 993/September 12, 1585 (For Nişancı Mehmet Paşa’s mosques in Istanbul to Selanik and Sidrekapısı qadis ).
38 For the transportation of lead: Barkan, Süleymaniye Cami ve İmareti İnşaatı, vol. 1, p. 370 ed seq.
39 Câfer Efendi indicated this in “düzen ibi” in Risâle-i Mi‘mâriyye, p. 99.
40 Stated in this study: İ. H. Aksoy, İstanbul›da Tarihi Yapılarda Uygulanan Temel Sistemleri, Istanbul: İstanbul Teknik Üniversitesi İnşaat Fakültesi, 1982, p. 53 for Süleymaniye Mosque.
41 G. Tanyeli, “Kanuni ve II. Selim Türbeleri: Teknik Çözümleme”, TAÇ Vakfı Yıllığı, no. 1 (1991), p. 98.
42 Barkan, Süleymaniye Cami ve İmareti İnşaatı, vol. 2, the document on pages 254-255.
43 Barkan, Süleymaniye Cami ve İmareti İnşaatı, vol. 2, p. 275.
44 Aksoy, İstanbul›da Tarihi Yapılarda Uygulanan Temel Sistemleri, p. 52, figure 4.4., p. 53.
45 For the most detailed description of traditional Ottoman base construction techniques: Ahmed Efendi, Târîh-i Câmi-i Şerîf-i Nuru Osmânî, TOEM supplementary, Istanbul: Hilal Matbaası, 1335-37, p. 9.
46 Saatçi, “Tezkiret-ül Bünyan’ın Topkapı Sarayı Revan Kitaplığı’ndaki Yazma Nüshası”, p. 91.
47 For the wooden jetties used in the construction of Nuruosmaniye Mosque: Ahmed Efendi, Târîh-i Câmi-i Şerîf-i Nuru Osmânî, p. 36.
48 For khorasan see M. S. Akman, A. Güner and İ. H. Aksoy, “The History and Properties of Khorasan Mortar and Concrete”, II. Uluslararası Türk-İslam Bilim ve Teknoloji Tarihi Kongresi: Bildiriler = Turkish and Islamic Science and Technology in the 16th Century: Proceedings, Istanbul: İstanbul Teknik Üniversitesi Bilim ve Teknoloji Tarihi Araştırma Merkezi, 1986, vol. 1, pp. 101-112.
49 Nayır (Ahunbay), Osmanlı Mimarlığında Sultan Ahmet Külliyesi, pp. 97-99.
50 For the history of the use of clamps in Ottoman architecture, see: Tanyeli, “Osmanlı Mimarlığında Demirin Strüktürel Kullanımı”, pp. 45-51.
51 G.Tanyeli, A. E. Geçkinli and A. Ata, “Osmanlı Mimarisinde Kullanılan Demir Ögelerinin Üretim Teknolojisi”, VI. Arkeometri Sonuçları Toplantısı, Ankara: Eski Eserler ve Müzeler Genel Müdürlüğü, 1990, p. 111.
52 Çeşitler için: Tanyeli, “Osmanlı Mimarlığında Demirin Strüktürel Kullanımı”, p. 277 (Annex. D: Osmanlı Demir Teknolojisine İlişkin Terimler Sözlüğü’nde).
53 Ü. Kulaç, “Türk Taş Minarelerinde Döner Merdiven ve Metal Bağlantı Elemanların Yatay Yükleri Karşılamadaki İşlevleri”, I. Uluslararası Türk-İslam Bilim ve Teknoloji Tarihi Kongresi: Bildiriler = I. International Congres on the History of Turkish-Islamic Science and Technolog Proceedings), Istanbul: İstanbul Teknik Üniversitesi Bilim ve Teknoloji Tarihi Enstitüsü, 1981, p. 236.
54 Kulaç, “Türk Taş Minarelerinde Döner Merdiven”, p. 236.
55 Ahmed Efendi, Târîh-i Câmi-i Şerîf-i Nuru Osmânî, p. 22.
56 Tanyeli, “Osmanlı Mimarlığında Demirin Strüktürel Kullanımı”, p. 81.
57 Tanyeli, “Kanuni ve II. Selim Türbeleri: Teknik Çözümleme”, p. 100.
58 W. Eton, A Survey of the Turkish Empire, London : Printed for T. Cadell, Jun. and W. Davies, 1798, p. 236.
59 Eton, A Survey of the Turkish Empire, p. 236.
60 Barkan, Süleymaniye Cami ve İmareti İnşaatı, vol. 2, p. 158, no. 381; TSMA, D.42, f. 10b; aktaran: Nayır (Ahunbay), Osmanlı Mimarlığında Sultan Ahmet Külliyesi, p. 103.
61 Tanyeli, “Kanuni ve II. Selim Türbeleri: Teknik Çözümleme”, p. 100.
62 Barkan, Süleymaniye Cami ve İmareti İnşaatı, vol. 1, p. 72.
63 M. Sözen, Diyarbakır’da Türk Mimarisi, Istanbul 1971, p. 198.
64 R. J. Mainstone, Hagia Sophia: Architecture, Structure and Liturgy of Justinian’s Great Church, London: Thames and Hudson, 1988, p. 255, 256, picture 281.
65 Tanyeli, “Osmanlı Mimarlığında Demirin Strüktürel Kullanımı”, pp. 45-96.
66 Tanyeli, “Kanuni ve II. Selim Türbeleri: Teknik Çözümleme”, pp. 98-99.
67 See: S. H. Eldem, Köşkler ve Kasırlar, Istanbul: Devlet Güzel Sanatlar Akademisi yüksek Mimarlık Bölümü Rölöve Kürsüsü, 1974, vol. 2, p. 28 ed seq.
68 K. Klinghardt, Türkische Baeder, Stuttgart: Julius Hoffmann, 1927, p. 19 ed seq.
69 For lökün, see: C. E. Arseven, “Lökün”, Sanat Ansiklopedisi, Istanbul: Milli Eğitim Bakanlığı, nd., vol. 3, p. 1244.
70 For lead covering, see Y. Önge, “Klasik Osmanlı Mimarisine Şekil Veren Kurşun Örtü ve Kurşunculuk”, Önasya, 1970, vol. 5, no. 53-54, pp. 11-13; L. Ş. Merey, Osmanlı Mimarisinde Kurşun Kaplama Tekniği, [unpublished notes].
71 For example, the window and door frames of the mosque that was built by Zeyrek Ağa in Payas were sent from Istanbul. BOA, Mühimme Defteri, no. 58, row no. 212, p. 73, Jumada al-Awwal 17, 993/May 17, 1585.
72 See: U. Tanyeli, “16. Yüzyıl Osmanlı Su Teknolojisinin Kökenleri”, Istanbul, no. 7 (1993), pp. 99-105.
73 Tanyeli, “Osmanlı Mimarlığında Demirin Strüktürel Kullanımı”, p. 111 ed seq.