This project was set up in order to collect new evidence (both archaeological and textual) on the character, origin and evolution of Islamic fortified settlements in Bilad al-Sham. The research focused on a single site, which can be considered highly representative in many respects: the castle of Shayzar.
So far, Progetto Shayzar started with a first season of fieldwork, which took place in March-April 2002. A specific study on the archaeological evidence of the site, which also included a survey of the written evidence, was carried out by Max van Berchem and Edmond Fatio and it was published in their Voyage en Syrie (Cairo, 1914, pp. 177-188). This important work provided a basis for setting up our project, and already addressed several research topics to be developed; their photographic and graphic recording of the site is also of great relevance to illustrate a number of substantial elements, which have now disappeared.
Shayzar lies on the western bank of the Orontes river, in central Syria. It can be regarded as one of the best surviving medieval landscapes (Fig. 1). It is exceptionally well preserved: no longer settled, it is basically untouched by recent urbanisation and modern restoration is fairly limited. Some of the buildings are preserved up to the level of the second storey, including the roof. At first sight already it appears clear that the castle today is the result of a number of different building phases spread along a wide span of time.
Fig. 1. General view of the castle from the E-NE. Complex CA1 can be seen on the far left.
Various elements indicate that Shayzar is an ideal choice to investigate a number of important research topics. According to the written sources discussed in the specific literature, the town of Sizara already appeared in the Amarna tablets; the Seleucids are reported to have reoccupied the town in the 4th century under the name of Larissa; it recovered its original name in the Roman period, while in the following Byzantine period it became Sezer. The precise location of this ancient settlement, however, has yet to be established. An early fortification is mentioned in relation to the Byzantine period in the second half of the 10th century; major building works are then attributed to the Banu Munqidh family after 1081. This phase should have come to an end with the earthquake of 1157, which was followed by an important restoration phase carried out by Nur al-Din. The castle then passed on to the Ayyubid and Mamluk sultans. A last restoration phase is attributed to sultan Baybars in 1261.
The above occupational history of the site clearly needs further investigation. However, it is clear that the archaeological evidence preserved on the site may shed new light on the evolution of settlement from Ancient Times to the Middle Ages, and especially on the origins and evolution of fortified settlements in the Near East.
Moreover, unlike other castles in the area, Shayzar was never settled by the Crusaders: its remains therefore represent an extraordinary archive for the local Islamic tradition of military architecture.
Some of the written sources dealing with the site are also quite exceptional: a version of the memoirs of Usama, a member of the Banu Munqidh family, survived and it provides a lively portrait of the time of the Crusades.
Methodology of the archaeological investigations
During this first phase, the field methodology consists in investigation methods that characterise Landscape Archaeology and, more specifically, the Archaeology of Masonry. It provides a detailed stratigraphic analysis of the various buildings and illustrates the sequence and the dynamics of the occupation of the site. At the same time, the data collected produces a series of documented references related to building techniques applicable to the study of other fortified sites in the area. For example, specific Masonry Types could be arranged in a precise archaeological sequence: Type 5A, of Period V, allows to date very precisely the introduction of smooth rustication combined with columns used as headers, i.e. the year 1233.
More traditional stratigraphic excavations could also be included in the field strategy, but only to solve specific questions arising from the investigations of aboveground evidence. They could also be employed to assist future operations of structural consolidation.
Furthermore, the archaeological and historical research can also be regarded as a first step towards a broader project aiming at the mise en valeur of the site. The detailed archaeological documentation collected will provide an essential database to address appropriate consolidation work.
Results
This first season of archaeological investigations carried out in March-April 2002 confirmed Shayzar as an ideal site for investigating the research topics envisaged by the project.
Operations on location can be summarised as follows:
- General archaeological survey of the site to identify the various periods of occupation and detail the general strategy of present and future research;
- Acquisition of data for the general 3D mapping of the site (D-GPS and Total Station survey);
- Archaeological study of one of the major architectural complexes identified on the site, the southern complex (CA1: CF 1, 2 and 3);
- Architectural survey and drawings of the CA1 complex; rectified photography of the relevant prospects.
Complex CA1
Results have shown the richness and the articulation over time of the archaeological deposits stretching along a very wide span of time, from Antiquity to the late Ottoman period. The sequence identified with the analysis of complex CA1 includes eight different periods. Period III and Period V can be dated in absolute terms, on the basis of inscriptions, to the Nurid and to the Ayyubid periods respectively. The other periods can only be dated in relation to Periods III and V.
Complex CA1 as it appears today consists of three different buildings: CF1, CF2, and CF3, well defined in the stratigraphic sequence and in terms of building techniques employed. Other structures from previous phases appear to have been included in these three buildings (Figs 2, 7).
Period I (in CF1 only): building of a monumental gate that was re-employed and transformed in later periods (Fig. 3, 4). The masonry is built in large ashlar blocks of stone, with dressed surface and regular dimensions (h 36-40; l 46-120; Masonry Type SHZ1). The masonry alternates stretchers and headers in the course on both faces of the wall, with no core.
Fig. 2. Complex CA1 from the E, at the edge of the artificial ditch that isolates the castle.
Period II (in CF1 and CF3): erection of walls related to at least two different structures which were later included in CF1, Period III, and in CF3, Period VI (Fig. 2, 6, 7). One of these walls is bonded to a glacis.
Walls of Masonry Type SHZ2 are made of both hewn and squared blocks (h 43-50 cm; l 39-120 cm) laid in parallel, approximately horizontal, courses; wedges are used in the joints but especially in the vertical joints. A large number of the blocks of this Masonry Type are re-employed. Columns which were transversally laid in the walls and visible in the external face are also used occasionally as lacing elements; differences in shape, size and material may indicate that they were spolia from pre-existing structures.
Fig. 3. Plan and sections of complex CA1.
Period III (CF1): building of CF1, making use of pre-existing structures of Periods I and II. CF1 is a two-storey building, with an entrance from the north: it gave access to the first storey of the building or to the second storey using a staircase on the eastern side (Fig. 3, 4). The masonry of Period III is characterised by the use on both faces of massive roughly hewn blocks (h 40-45 cm; l 50-73 cm), laid in courses, with abundant mortar and a very high number of wedges used either between the joints and on the surface (Masonry Type SHZ3). Structural elements, such as piers, pillars and brackets, are made of hewn blocks laid in mortar; a number of these blocks seem to have been re-employed, and the surface is re-dressed in several cases.
Fig. 4. Complex CA1, building CF1, from the N.
Period IV (CF2): first attempt at building CF2, only preserved on the northern front (Fig. 5). The external face of this wall is composed of hewn blocks of large dimensions, laid in mortar, with smooth rustication, bonded to a core of rubble and mortar; a significant quantity of small wooden wedges are used in the bed-joints (Masonry Type SHZ4).
Fig. 5. Complex CA1, building CF2, northern front, with the dated inscription.
Period V (CF2): building of CF2; CF1 was clearly still in use and was providing access to CF2 (Fig. 3, 6). The masonry is characterised by the use of squared blocks with smooth rustication on the external face (Masonry Type SHZ5); the internal face shows a well-dressed and even surface without rustication. Transversally laid columns, used to increase statics and stability, appear in the courses according to a specific building programme: in a layer just below the floor level and in correspondence with the brackets and the corbels of the internal vaulting. These columns are slightly projecting on the external face, and they cross the entire section of the wall, since they reappear on the inner face; they do not project from this inner face. Wedges are used to lay these columns in the courses. Not only these columns, but also a large number of blocks seem to have derived from a re-cutting and re-dressing of spolia material.
Fig. 6. Complex CA1, buildings CF2 and CF3, southern front.
Period VI (CF3): building of CF3, incorporating the whole of CF2. It is not clear whether building CF1 was still in use. Entrance was possible directly from CF3 (Fig. 2, 3, 6, 7). The masonry on the southern, eastern and northern external fronts is characterised by the use of squared blocks with smooth rustication on the external face, as in the case of CF2 masonry (Masonry Type SHZ5). Differences can be detected in the use of longer blocks on the external front (Masonry Type SHZ 5C), mainly in correspondence with internal floors, and in a more extensive use of columns, following the same programme already noted for CF2.
Periods VII-VIII: minor interventions, decay and abandonment of complex CA1.
Fig. 7. Elevation with stratigraphic analysis of CF2 and CF3, southern front, external.
Dated inscriptions allow relating two of the archaeological phases identified on the site with events mentioned in the sources. Period III can be related to Nur al-Din; at the entrance of CF1, one inscription is reported to have mentioned the name of Nur al-Din and the incomplete hijri date " five hundred and fifty
". At present, the inscription is badly eroded, but further examination will be attempted to document this very important piece of evidence which allows attributing Period III to the restoration of the site carried out by Nur al-Din after the earthquake of 1157 (and before 560/1163-4).
Period V can be ascribed to the year 1233: an inscription states that it was built by the Ayyubid prince of Aleppo, "al-Malik al-'Aziz Ghiyath al-dunya wa-l-din Abu-l-Muzaffar Muhammad" (613-634/1216-1237), on "the 21 Dhu-l-Qa'da of the year 630" (29 august 1233).
Because of its masonry, Period I may belong to the Antiquity; if this is confirmed by further evidence, it may clarify the location of the classical town. Period II as studied in complex CA1 may well be attributed either to the 10th century Byzantine fortification, or to the 11th-early 12th century fortification of the Banu Munqidh. If the latter is the case, the major destruction that the building of Period II has suffered may be attributed to the earthquake of 1157.
The post-Ayyubid phase is related to the restoration works carried out by sultan Baybars in 1261; it is likely that the building of CF3 (Period VI) should be attributed to the early Mamluk period, either to that of Baybars, or to the restoration carried out by Qalawun. An in situ inscription mentions a date related to this last major phase: the year 1290. . Only continuation of the work will allow placing the various identified phases in absolute terms.
Progetto Shayzar has just started. However, the research has contributed already at this stage towards establishing a sequence of the various masonry types observed in one of the major complexes on the site. Continuation of this work will allow to extend the same kind of analysis to the entire settlement: it will be possible to illustrate the dynamics of the occupation of the site organised in a chronological sequence, to understand the characteristics of the various castles that have been built on the site and the evolution of military architecture in the region.
This evidence will be essential to correctly address the mise en valeur project envisaged by local authorities on the site.
Cristina Tonghini
Università Ca' Foscari di Venezia
Dipartimento di Studi Eurasiatici