PAM II, Reports 1989-1990 (Warsaw 1991) CONTENTS HATSHEPSUT'S TEMPLE AT DEIR EL-BAHARI. EGYPTOLOGICAL DOCUMENTATION 1989-1990 Janusz Karkowski …………………………………………………… 7 TEMPLE OF TUTHMOSIS III, DEIR EL-BAHARI. REPORT ON THE 1990 SEASON Jadwiga Lipińska …………………………………………………… 12 CONSERVATION WORK ON KOM EL-DIKKA IN ALEXANDRIA IN THE 1989 SEASON Wojciech Kołątaj …………………………………………………… 15 EXCAVATIONS IN ALEXANDRIA IN 1989-90 Grzegorz Majcherek …………………………………………………… 19 POLISH-EGYPTIAN EXCAVATIONS AT TELL ATRIB IN 1990 Karol Myśliwiec …………………………………………………… 25 MARINA 1990 Wiktor A. Daszewski …………………………………………………… 31 THE ACTIVITIES OF THE POLISH-EGYPTIAN PRESERVATION MISSION AT MARINA EL-ALAMEIN IN 1988 Włodzimierz Bentkowski …………………………………………………… 38 POLISH-EGYPTIAN RESTORATION MISSION AT MARINA EL-ALAMEIN IN 1990 Jarosław Dobrowolski …………………………………………………… 44 DEIR EL-NAQLUN 1990 Włodzimierz Godlewski …………………………………………………… 48 DEIR EL-NAQLUN 1990: THE GREEK PAPYRI Tomasz Derda …………………………………………………… 54 ARABIC PAPYRI AND INSCRIPTIONS FROM NAQLUN, HERMITAGE NO. 89 0laf E. Kaper …………………………………………………… 57 DAKHLEH OASIS PROJECT: RESEARCH ON THE PETROGLYPHS, 1990 Lech Krzyżaniak …………………………………………………… 60 OLD DONGOLA 1989-1990 Stefan Jakobielski …………………………………………………… 65 THE OLD DONGOLA FORTIFICATIONS Włodzimierz Godlewski …………………………………………………… 74 NEA PAPHOS 1990 REPORT Wiktor A. Daszewski ………………………………………………….... 78 …………………………………………………… 85 …………………………………………………… 91 THE THIRD SEASON OF EXCAVATIONS IN NORTHEAST SYRIA, 1990 Piotr Bieliński …………………………………………………… 94 NEMRIK 9 Stefan Karol Kozłowski …………………………………………………… 102 LA REPRISE DES FOUILLES À M'LEFAAT (SAISON 1989/1990) Stefan Karol Kozłowski, Kazimierz Kuźma et Karol Szymczak …………… 112 THE FIRST SEASON OF EXCAVATIONS IN HATRA, IRAQ Michał Gawlikowski …………………………………………………… 119 PALMYRA Michał Gawlikowski THE ARAB CASTLE IN PALMYRA Janusz Byliński HATSHEPSUT'S TEMPLE AT DEIR EL-BAHARI EGYPTOLOGICAL DOCUMENTATION 1989-1990 Janusz Karkowski An Egyptological mission of the PCMA worked for a brief period in April of 1989.1 The main objective of this season was to trace the scenes of the Opet Feast from the southern end of the east wall of the Upper Court, as well as details of the upper, unpreserved scenes on fragments in the temple stores. In addition, the mission documented some of the unpublished figural graffiti of Greco-Roman times preserved on the walls and columns of the Ptolemaic portico and verified copies of the texts of the birth scenes in the northern middle portico and the texts in the royal complex to the south of the upper court. During the second season,2 conducted in 1990, studies in the temple stores were continued in order to provide new material for further restoration in this temple and for final publication. A separate study concerned Greek inscriptions left by pilgrims who visited the important sanctuary of Amenhotep, son of Hapu and Imhotep, which functioned in the temple ruins for over 500 years. The mission used transparent plastic film for tracings. The drawings were then re-traced and photographically reduced to produce 1 2 The mission comprised Miss Hanna Koczowska, artist; Dr. Janusz Karkowski, egyptologist, Director of the Mission; Miss Małgorzata Koperwas, archaeologist; Dr. Maciej Witkowski, egyptologist and Mr. Nur Abd el-Gaffar Muhammad, who acted as the inspector of the EAO. During this season, which lasted from 21 January to 3 April 1990, the mission comprised: Dr. Janusz Karkowski, egyptologist, Director of the mission; Mr. Mirosław Barwik, egyptologist; Dr. Nathalie Beaux, egyptologist; Mrs. Dorota Bielińska, archaeologist-documentalist; Miss Joanna Kusek, artist; Mr. Adam Łajtar, Greek epigraphist and Mr. Stefan Sadowski, photographer. The EAO delegated two inspectors : Messrs Reda Ali Suleiman and Nur el-Gafar. 7 © PCMA 2008 – digital reprint PAM II [= Reports 1989-1990] plates for the planned publication. Altogether over 120 full scale copies, averaging 1.10 x 1.50 m in size, were made. In choosing the scenes to be recorded the mission applied the following criteria: 1. Completing the documentation of temple elements which were recently restored. This included the walls of the upper portico, eastern, southern and northern walls of the upper court and the northern wall of the same court. Here, it may be signaled that some additional fragments were identified in the temple stores as coming from the southern wall of the upper court. They belong to the western jamb of the entrance to the room with the window. The tracing of all the decoration on these walls was completed. Fig. 1. New blocks from the walls of the solar chapel. 8 PAM II [= Reports 1989-1990] © PCMA 2008 – digital reprint 2. Recording the decoration of the parts of the temple which are closest to the rocks towering above the temple or which are excavated in the rock. Thus, the mission recorded the innermost part of Amun's sanctuary (chapel of Amun's cult statue and the southern chapel of the Ennead), as well as the solar complex. Tracing of the decoration in these rooms was completed. In the solar complex, a number of stored fragments could be ascribed to the western, unpreserved wall of the solar chapel (Fig. 1). In addition, the recording of the wall with niches in the upper court was started. All the scenes above the smaller niches were traced as well as the largest, unpublished scene to the south of the sanctuary entrance. During studies in the temple stores eight additional fragments were identified; they belong to the wall with niches and their exact location has been established (Fig. 2). Fig 2. New fragments from the wall with niches. 9 © PCMA 2008 – digital reprint PAM II [= Reports 1989-1990] Fig. 3. Two new fragments from the east wall of the niche of seven cows in the royal complex. Yet another task of the mission was the study of a group of texts and representations in the royal complex to the south of the upper court. This season the three versions of Chapter 148 of the Book of Dead were fully recorded together with loose fragments from the temple stores (Fig. 3). They were placed in Hatshepsut's hall of offerings on both sides of the false door, by the entrance to Tuthmosis I's offering hall and in the niche in the eastern wall of the vestibule of Hatshepsut's chapel. A preliminary study of the loose fragments of the unpublished Ritual of Hours from the ceiling of Hatshepsut's offering hall was done as well. All the texts were copied. Many fragments were assembled to form larger blocks. During the season it was possible to ascribe about 30% of the preserved blocks to the hymns of the particular hours of day and night. Already it is obvious that in the 10 PAM II [= Reports 1989-1990] © PCMA 2008 – digital reprint future not only the theoretical restoration of the oldest preserved copy of this important text will be possible, but also the practical reconstruction of the hall's roof. The last study completed during the season concerned Greek graffiti left on the temple walls by pilgrims in Greco-Roman times. These graffiti had been published by A. Bataille in 1952, but the present work proved that he omitted a number of important texts. Altogether Bataille's catalogue could be supplemented with 125 new texts. In addition, numerous corrections could be introduced to his readings of the texts. Among the unpublished texts the most interesting is the Hymn to Amenhotep son of Hapu, on the southern wall of the Ptolemaic Portico. Another unpublished text from one of the niches of the upper court dates from the time of Diocletian and shows that the cult of Amenhotep son of Hapu and Imhotep did not disappear in the 2nd century AD as Bataille believed, but survived until the end of the 3rd century. In some cases Greek inscriptions on loose fragments enable them to be placed in the walls. This primarily concerns fragments of the Ptolemaic Portico where, thanks to graffiti, five large fragments could be added to the cornice and column capitals. 11 © PCMA 2008 – digital reprint PAM II [= Reports 1989-1990] TEMPLE OF TUTHMOSIS III – DEIR EL-BAHARI REPORT ON THE 1990 SEASON Jadwiga Lipińska According to the programme, work concentrated on studying the broken temple elements stored, since the excavations, in a shed near the temple of Hatshepsut. Since 1978 a special team has been sent there regularly to deal with the finds from the ruined temple.1 The main task of this season was to complete the already assembled scenes and to find proof for linking the representations provisionally attributed together. A special study was undertaken on fragments of decorated doorways in order to determine their exact dimensions. This was especially important in cases when there were no traces left in the ruined building to allow a reconstruction of the plan. Comparatively few new assembled scenes were registered, as most of the newly identified parts of the decoration and inscriptions were joined to already registered scenes. In rough estimation, about 80% of the fragmentary material stored at Deir el-Bahari is now joined into 501 numbered scenes. 62% of these are now firmly attributed to particular walls of the temple, in many case allowing for the complete reconstruction of the iconography of the temple halls and rooms. In the Hypostyle Hall the wall decoration is ascertained for the western, northern and part of the eastern walls; only a few scenes 1 During the 1990 season the mission was headed by Assist. Prof. Jadwiga Lipińska and consisted of: Miss Janina Wiercińska, Mrs. Monika Dolińska and Miss Joanna Aksamit, archaeologists, Miss Kamila Baturo, archaeologist and documentalist, and Mr. Stefan Sadowski, photographer. The EA0 assigned two Inspectors: Mr. Reda All Sultan and Mr. Nur Abd el-Gafer, who greatly facilitated the work and were of considerable assistance to the Mission. Deep gratitude should also be expressed to Dr. Muhammad Saghir, Director of the EAO in Luxor and Dr. Muhammad Nasser, Director of the Gurna Inspectorate. 12 PAM II [= Reports 1989-1990] © PCMA 2008 – digital reprint Fig. 1. Plan of the temple. have been attributed to the southern one, probably destroyed earlier, and there is little hope to complete its theoretical reconstruction. The documentation of the reconstructed walls of this hall is now finished, and a set of original drawings, made to 1:4 scale, is now reduced to 1:12 for the purposes of publication. The decoration of the Bark Hall, consisting of two registers on each wall, is almost fully ascertained, with the exception of its southwestern corner. The studies resulted in establishing the fact that except for two normal doorways: one leading into the room from the Hypostyle Hall and one leading to the vestibule of the sanctuary, there were two other entrances, one in each of the side 13 © PCMA 2008 – digital reprint PAM II [= Reports 1989-1990] walls. This feature is unusual and there is no parallel arrangement known. This Hall has other unusual features, but their interpretation will be possible only after the drawings are completed and verified during the next season. The work on the decoration of the main sanctuary is not finished yet, but the drawings of the eastern wall and separate scenes from other walls will be completed shortly. The work on the reconstruction of wall decoration in a series of long, narrow shrines in the western part of the temple is under progress. Three of them (marked D, H, I) have almost been finished, with iconographic schemes established for all of their walls. For Room G one wall is still missing, and for rooms E, K and L only single representations have been attributed. With about 38% of the assembled scenes still awaiting attribution, there is hope to fill the gaps, and the work should be much speedier now, as a result of a process of elimination. It is planned to complete the first volume of the publication of the temple decoration soon, and it will contain the reconstruction and studies on the Hypostyle Hall. Another field of activity during the 1990 season consisted of extensive studies on the hieratic graffiti found on fragmentary columns and walls of the temple. In 1974 Dr. M. Marciniak published a volume on them, but left for publication countless smaller parts of the texts. With progress on the piecing together of the wall decoration, it was possible to join many hitherto scattered fragmentary texts and prepare another volume. Also, a number of the hieratic texts on broken columns were successfully connected; Mr. M. Barwik, who undertook this task, prepared a complete documentation of the subject. 14 PAM II [= Reports 1989-1990] © PCMA 2008 – digital reprint CONSERVATION WORK ON KOM EL-DIKKA IN ALEXANDRIA IN THE 1989 SEASON Wojciech Kołątaj Work conducted in the 1989 season is part of a broader conservation program undertaken by the PCMA with the Egyptian Antiquities Organization and aimed at preserving and exhibiting a large section of the ancient city centre of Alexandria, presently located in the centre of the modern town. The program covers several structures of both private and public nature, which preserve the urban layout of the 4th-7th centuries AD. However, in cases where archaeological studies have uncovered valuable relics from the 1st century BC to the 4th century AD; it is planned to exhibit these as well, without disturbing in any significant way the Late Roman and Byzantine systems. Conservation work is conducted on many sites, in conjunction with archaeological research as a rule, in order to preserve the uncovered relics of architecture left in situ. The specificity of Alexandria and the scope of our conservation program there1 draws from two elements. First and foremost, there is the nature of the damages resulting from extensive salination and humidity of the environment, seismic activity which has frequently destroyed the city in the past and, finally, many centuries of ransacking in search of building material. Secondly, keeping in mind that in the future the site will be made accessible to the public, it is necessary to ensure the safety of visitors. Basing their judgment on the principles contained in the Venice Charter and taking into account new trends in conservation reflecting the use of new substances and technologies presented in 1 The conservation mission included: Dr. Wojciech Kołątaj (director), Dr. Jan Borkowski; architect Mr. Mariusz Karst, Mr. Kazimierz Błaszczyk and archaeologist Mr. Grzegorz Majcherek. 15 © PCMA 2008 – digital reprint PAM II [= Reports 1989-1990] publications and at congresses, the competent Egyptian authorities have accepted and adopted the proposed general premises, of the program. In 1989 work concentrated in three areas. In the Theatre the re-erection of the southern wall of the complex was completed. The wall had fallen during an earthquake in the 8th century AD and was found lying in undisturbed order. Preserving the original courses of stones and at times lifting whole sections together with the plaster preserved on its face,2 it was possible to reconstruct this wall and protect it from humidity by introducing insulation layers and properly forming its top to ensure quick evaporation of rainwater. Missing blocks were replaced by new ones made to specification from the same kind of stone as the ancient ones. The place of destroyed bricks, impossible to retrieve from the debris, was filled with new ones, carefully chosen to fit into the whole as far as dimensions are concerned. No substances of any kind, either silicon– or resin-based, were used to reinforce the surface. To date, experiments with Ahydrosil Z as well as Silak and the Vacker preparation have shown that instead of attempting to protect wall surfaces endangered by salt efflorescence, it is necessary to build complete systems for protecting structures from seasonal dampness. It was also undertaken to clear, investigate and preserve the last three sections of the theatre corridor (ambulacrum), left unexplored when the theatre was excavated. Upon clearing two of the sections, measures were taken to protect their arcades and vaulting. The fissured structure was protected by injections under gravitational pressure of a mixture of hydrated lime, fine-grained sand and white Portland cement, taken in proportions 10:20:5. Destroyed bricks were replaced with ancient bricks retrieved from debris during excavations. In places where structural needs required the an2 The plaster was reinforced with a water solution of casein and polyvinyl acetate. 16 PAM II [= Reports 1989-1990] © PCMA 2008 – digital reprint cient substance to be rebuilt, new ceramic material was used, in size and resistance properties approaching ancient bricks. Effort was made to preserve even the tiniest fragments of ancient substance in situ. Work in the cisterns proceeded simultaneously. The building poses a serious safety problem. Having suffered from earthquakes and stone-robbing in medieval times – some 70% of the eastern facade and the four buttresses supporting it are gone – the structure is almost totally disintegrated. Work proceeded section by section, in trenches up to 7 m deep, with all possible precautions being taken. The first task was to ensure the stability of the most endangered northeastern corner of the structure. Stainless steel braces (diam. 20 mm), compressed as required by exerted forces, were used and appropriate anchors made.3 Only then was it possible to start work on the three buttresses. The buttresses were rebuilt on original foundations of the same kind of stone as the ancient substance. The still existing upper parts of the buttresses, which were deformed and hanging in the air, were recorded and dismantled. Restoration was subsequently completed using mortar which both in composition (lime, sand, brick aggregate) and material corresponds to ancient mortar. Parallel to this work we proceeded with the reinforcement of the structure of the inner wall which, unlike the facade, was bonded in an ash mortar. Also the reconstructed outer face is being anchored to the mostly original inner face of the wall. At present the wall is sufficiently protected not to disintegrate any further. However, more work in this area is planned, including a reconstruction of the fourth and last buttress. In the Baths the reconstruction of the foundations of the southern portico continued. In the future this will enable us to 3 The compression of the monument was done by Dr. Jan Borkowski according to a method and design of his own invention. 17 © PCMA 2008 – digital reprint PAM II [= Reports 1989-1990] conduct an anastylosis of the columns of this portico and to protect the extant vaults of the underground service area. As in the cisterns, work here is conducted in trenches up to 8 m deep under extremely dangerous conditions. The foundation has been investigated and preserved all along its length, that is, 26 m. The intercolumnia were determined, as well as the position and level of column bases. About 80 cubic meters of foundation wall was completed; only a small section will be visible in the underground service area, but the work was necessitated by the need to prepare a foundation for the planned anastylosis of columns. Two column shafts were prepared. One with a section angled 60° to the base requires three pins of stainless steel (diam. 30 mm) 0.40 m long to connect it directly to the bottom part with an accuracy of l-2 mm. The application of epoxy resins does not guarantee safety in view of the ageing process they undergo. Another three column shafts were prepared for anastylosis in the northern portico of the baths. One of these received an artificial stone addition to one-third of its height. A 9 m high tripod was constructed of 8" steel pipes. In the future it will serve in the anastylosis of columns in both porticoes of the baths and in the theatre street portico. Landscaping the slope surrounding the exposition continued throughout the season even though the removal of earth by mechanical means in the western part of the kom was stopped. Hopefully, this work shall start again in the coming season. 18 PAM II [= Reports 1989-1990] © PCMA 2008 – digital reprint EXCAVATIONS IN ALEXANDRIA IN 1989-90 Grzegorz Majcherek The Polish-Egyptian Conservation and Archaeological Mission at Kom el-Dikka, sponsored jointly by the EAO and the PCMA, and directed by Dr. Wojciech Kołątaj, worked from October 1989 till the end of July 1990. l The multiple tasks of this campaign could not have been accomplished without the indispensable assistance of the authorities of the Egyptian Antiquities Organization to whom we wish to express our thanks. Particular gratitude is due Mrs. Doreya Said, the Director of the Graeco-Roman Museum in Alexandria, for her constant help and support. Beside ongoing archaeological research integrated with the mission's restoration program, excavations were concentrated in the habitation quarter adjoining street R4 on the east (Sector W1N). The objectives of this campaign, which was a continuation of previous seasons, were twofold: to continue the investigation of Houses G and H, and to study and verify the stratigraphy and chronology of the area as worked out on the basis of excavations conducted so far. The continued occupation of the quarter from the Late Ptolemaic (c. 2nd-1st centuries BC) to the end of the Byzantine period (end of the 7th century AD) was fully confirmed. As a result of the present excavations, substantial remains of Late Roman building G were cleared and its plan and functions thoroughly studied. Several subsequent chronological phases covering the period from the 4th to 7th century AD were distinguished; their detailed description will be published in the final report. Fragments of the Early Roman structure fragmentarily l The archaeological research was headed by Grzegorz Majcherek who worked with a team including: Assist. Prof. Przemysław Gartkiewicz (architect), Assist. Prof. Barbara Lichocka, Dr. Barbara Tkaczow, Miss Jolanta Jabłonowska, Miss Iwona Zych (archaeologists) and Mr. Stefan Sadowski (photographer). The Egyptian Antiquities Organization was represented by Messrs Ala'a ad-Din Mahrous and Ahmad Mousa. 19 © PCMA 2008 – digital reprint PAM II [= Reports 1989-1990] preserved underneath were also unearthed. Further research here is planned for the coming season. Building G bears a general resemblance to the previously excavated House H, its plan however appears to be less regular. It was found that the walls of the building had been largely dismantled in the 8th-10th centuries AD, when they served as a ready source of material for the medieval structures built on a higher level. Owing to this, most of the plan could be traced only approximately in ghost walls. The walls that have survived are in their majority founded upon entirely new Fig. 1. Plan of the Late Roman houses G and H in sector W1 N 20 PAM II [= Reports 1989-1990] © PCMA 2008 – digital reprint foundations, which often have nothing to do with the earlier structures. The reason for the alteration of the old building plan, dated by coins and pottery to the turn of the 3rd and 4th century AD, must lie in the degree of destruction sustained by the Early Roman structures.2 The Late Roman occupation layer is separated from the earlier layers by a thick stratum containing ashes, fragments of polychrome plastering and architectural elements. Reused fragments of architectural decoration were also found in the foundations (rooms nos 2, 3, 10) and include such items as drums of engaged Doric columns covered with stucco.3 At the close of our work, the entire street facade of House G had been cleared, but its extent into the insula remained undetermined. On the remaining sides, both on the north and south, the building was separated from neighboring edifices by narrow alleys (some 1.50-1.70 m wide) perpendicular to street R4. The main walls of the house were built in a technique typical of the period and apparently derived from opus africanum; pillars constructed of large blocks, with spaces in between filled with small irregular stones in an ashy mortar. The western part of House G comprised four rooms, each provided with a separate entrance, the northernmost of which (G-8) had been explored already in 1989. The adjacent room (G-10) is the largest so far uncovered. Only the front wall with its thick lime plaster coating has been preserved. The debris overlying the latest clay floor contained numerous colored marble pieces of opus sectile pavement, which doubtless must have come from the destroyed upper storey. The finds from the occupation level: sherds of Egyptian Red Slip ware (A and B), Cypriot Red Slip and amphorae of Egyptian origin as well as several coins dated to the reign of Heraclius indicate that 2 The destruction of the eastern quarters of Alexandria is usually connected with the ransacking of the city by the Palmyrenians. Cf. A. Adriani, Repertorio d'Arte dell'Egitto Greco-Romano, Serie C, I-III, Palermo 1963-66, pp. 209, 211. 3 In general appearance these elements resemble the relics of a Doric style structure discovered beneath the nearby building of the National Insurance Company, cf. H. Riad, Vestiges d'un edifice ptolemaique, BSAA 1, 42 (1967). 21 © PCMA 2008 – digital reprint PAM II [= Reports 1989-1990] the building was destroyed probably around the middle of the 7th century AD.4 The two rooms located further to the south (G-2 and G-3) were also substantially damaged. The front wall was dismantled to the foundations, while the eastern walls have all but completely disappeared. Exploration in room G-13, which is side by side directly with G-8, uncovered a deep rectangular well. The shaft was cleared down to a depth of about 6 m, not reaching the bottom, and was found to have walls constructed of well-fitted dressed stone. Pairs of large blocks protruded from the middle of opposing walls at a point just below the preserved top and about 2 m down, directly below the first. These blocks may have served to mount some kind of water-wheel, equipped with ceramic pots (qawadis) attached with ropes. The proximity of this wall and a pair of pools discovered in room G-8 obviously indicates a relation between the two water facilities. The finds on the paved floor preserved around the well include coins of Tiberius II and Phocas and point to the turn of the 6th and 7th century AD as the period of use. A similar but fragmentarily preserved pavement was cleared in room G-11. The dating material from the fill contained two well preserved St Menas ampullae. The remains of a large lavatory were discovered in the spacious room G-12. An angled section of a conduit ran through the room and served presumably as a sewage tank, since no outlets were found. In the southern part of the room the conduit increased slightly in width and was covered with some large slabs. Two large blocks pierced with a rounded aperture c. 0.30 m in diameter were found sub situ. Several coins and a considerable amount of pottery sherds from the fill point to a surprisingly long period of use, from the 4th to 6th century AD. A small room located in the southeastern corner of the excavated area initially opened out onto the alley between houses G and C. In the 5th century some of the walls were replaced and the room enlarged. In summary, it appears that in the Late Roman period House G served needs as complex as the previously investigated House H. The 4 See the review of ceramic material in: G. Majcherek, Polish excavations at Kom el-Dikka, 1989-1990, Bulletin de Liaison du GIECE, 1991, in press. 22 PAM II [= Reports 1989-1990] © PCMA 2008 – digital reprint large rooms opening onto the street may have been used as shops, while the rooms in the back served both as workshops and domestic areas. Both houses were continuously occupied at least till the middle of the 7th century AD. In the southern wing of House H some remaining layers of Late Roman date were also explored. Room H-8b turned out to contain some well-preserved domestic facilities consisting of four containers set into the floor. Two of these were reused Baetican amphorae, while the others were bodies of Palestinian amphorae, which were an early 4th century AD version of the widely traded Gaza-type vessels. Some of the containers had their covers still in situ – limestone slabs with a funnel hole in the centre closed with stone stoppers. The fill in which the amphorae were embedded contained a great deal of pottery typical of the 2nd-3rd centuries AD, including two lamps: one with a relief representation of a sandaled foot, the other with an erotic scene on the discus. In compliance with the study program concerning Early Roman structures in the quarter, it was decided to explore the lower layers in room H-2. A small basin was discovered there, alongside the western wall. The rest of the room was taken up by a small staircase descending below the level of the basin and a small-sized well. It would appear that these facilities belonged to a private bath of the Early Roman villa. Another deep sounding was dug in the partly explored room H-8a. About 1.50 m below the 6th century street level, the remains of a clay floor covering the entire room space was noted. Close to the blocked entrance a small storage bin constructed of stone was discovered. Further exploration beneath it revealed the presence of a rectangular shaft 6 m deep. Although it was originally built as a well, the composition of the fill proved that in this period it must have served as a sewage tank. The clay flooring was laid upon an even layer of burning and ashes. It probably coincided with the destruction and final abandonment of the Early Roman house, tentatively dated to the late 3rd century AD. Apparently the structure was once more inhabited while still in ruins, with the aforementioned facilities being the first signs of renewed occupation. The Early Roman occupation level, comprising 23 © PCMA 2008 – digital reprint PAM II [= Reports 1989-1990] several lime floors was reached immediately below. The archaeological context of these layers included a considerable amount of ceramic material, bearing evidence to a long-lasting habitation in the period from the 1st to the 3rd century AD. Further exploration revealed a clear architectural sequence, visible in the walls of the house. Roman structures were built directly on top of the remains of Ptolemaic ones. The foundations of the Ptolemaic building were reached at a depth of 4.50 m below the street level, on top of a clay layer some 0.70 m thick. Much to our surprise the remains of yet another structure came to light under the deposit of clay. Although it is tempting to see it as the remainder of the first structure built in the area, its nature as well as precise dating is yet to be determined. Particularly important discoveries were made while studying the urbanistic development of the quarter. The sewage system uncovered in the alley separating houses H and G, proved the existence of the side street as early as the Early Roman period. It was presumably much wider at that time, reaching some 5 m, that is about half the width of the R4 street.5 Its presence throws new light on the previous findings and hypotheses concerning additional subdivisions of the great insulae defined by the basic urban street network.6Another channel was discovered on street R4 at a depth of 4.80 m. No precise dating material was available here: however, the level of the channel relative to the uncovered architectural structures indicates that it must have formed part of the urban sewage system of the Ptolemaic period. Its course along the facades of the houses presents evidence for the permanence of the street network of Ptolemaic date, existing in this area in basically unchanged form till the end of Antiquity. 5 6 An Early Roman street of similar course and dimensions was uncovered during the 1973 excavations below the Theatre Portico. M. Rodziewicz, Alexandrie III. Les habitations romaines tardives d'Alexandrie, Varsovie 1984, pp. 17-33. 24 PAM II [= Reports 1989-1990] © PCMA 2008 – digital reprint POLISH-EGYPTIAN EXCAVATIONS AT TELL ATRIB IN 1990 Karol Myśliwiec The Polish Egyptian archaeological mission sponsored jointly by the EAO and the PCMA pursued its work on Kom Sidi Youssuf at Tell Atrib (suburb of Benha) from 1 August until 6 October 1990.1 This campaign was possible thanks to the efficiency and friendly help of Dr. Ali El-Khouly (EAO, Cairo) and Mr. Muhammad Abdelhaq (EAO, Zagazig), to whom we wish to express our deep gratitude. The program of this campaign comprised: 1) archaeological exploration of the northeastern part of the area, untouched by excavations so far; 2) reaching the present level of the water table in the parts of the Ptolemaic town unearthed during previous campaigns (northwest of the excavated area); 3) further exploration of the area's northern zone, where our excavations in the preceding year did not reach deeper than the upper Byzantine level; 4) survey of deep-lying strata below the present water table level, using mechanical drilling to probe various parts of the excavated area. The northeastern part of the concession, comprising 4 squares 10 x 10 m, was excavated about 1.50 m below the present surface. 1 The mission included the following: Prof. Karol Myśliwiec, Dr. Barbara Ruszczyc, Dr. Aleksandra Krzyżanowska, Miss Anna Południkiewicz, Prof. Waldemar Chmielewski, Mr. Adam Łajtar, Mr. Włodzimierz Bogusz and Miss Iwona Zych, archaeologists, Mr. Janusz Wałkuski, architect, and Mr. Waldemar Jerke, photographer. The EAO was represented by two inspectors: Mrs. Somaya Abu Senna and Mr. Fawzi Abdinaim El-Barbari. 25 © PCMA 2008 – digital reprint PAM II [= Reports 1989-1990] Beneath a thick layer of black ashes followed by a stratum containing much pottery and brick rubble there were remains of Late Roman and Byzantine structures. The southern end of the quarter proved to be a center of lime production. Several large lime kilns were found on two different levels. Many fragments of painted plaster as well as fragmentary limestone blocks and terracotta tiles originating from the neighboring baths were found inside the kilns. A thick layer of lime rubble surrounded the kilns and extended up to the southern limit of a large mud-brick construction which occupies the northern end of the quarter. The structure embraces several small rooms, some of which may have served as storerooms, while others were probably inhabited. In one of these rooms a pottery deposit was found together with Late Roman coins and lamps of similar date. The deposit contained many globular pots and a portable cooking-stove. A second deposit came to light in another room on a slightly higher level. This was a set of large Egyptian amphorae and barrels dating to the Byzantine period. The extremely rich ceramic material belonging to the archaeological context of this stratum comprises several oil lamps made of marl clay and decorated on the discus with a stamped pattern of a frog with spread legs in various stylized versions. The most sophisticated of these reliefs shows two newborn children attached to a small circle which most probably should be identified as a picture of the vagina. At the western extremity of the Byzantine quarter there is a deep well built of red bricks. It is incorporated into a long and wide mud-brick wall constituting the east border of the Ptolemaic quarter. The well was explored down to a depth of c. 4 m below its top, water being pumped out while excavations proceeded. This exploration revealed a long period of use and multiple repairs made to the well in the Byzantine period. Owing to the particularly low water table during our excavation season (September-October), we could continue explorations in the previously unearthed Ptolemaic structures, reaching strata 26 PAM II [= Reports 1989-1990] © PCMA 2008 – digital reprint corresponding to the second half of the 4th and the beginning of the 3rd century BC. This early date is confirmed by the pottery and some small finds, among which two objects, so far unique, deserve particular attention. These are a terracotta lamp in the form of a small bowl with a partly contracted rim and an amphora handle with a round stamp containing just two large Greek letters: ΦΙ Several pottery wasters found in this stratum prove that the excavated quarter of Athribis was a center of ceramic production as early as the end of the Dynastic period and the very beginning of the Ptolemaic period. A large deposit of Early Ptolemaic (Ptolemy II and Ptolemy III) unfired ceramic vessels (Fig. 1) was unearthed below the walls of room 127, which were dismantled in the course of our work in order to study the pottery. The discovery of the deposit allowed us to establish a repertory of ceramic pots which were produced in the local workshops during the 3rd and in the first half of the 2nd century BC. Beside the most frequent forms, such as curved rim Fig. 1. Early Ptolemaic deposit of unfired pottery, found in Room 127. 27 © PCMA 2008 – digital reprint PAM II [= Reports 1989-1990] bowls and plates with flattened edge, there were also some closed forms, e.g. painted jugs and globular pots without handles, decorated with patches suspended from a white band, garlands and geometrical patterns, as well as large-size amphorae. As far as the identification of local ceramic production is concerned, one of our finds may be considered a genuine sensation, for it demonstrated that terracotta figurines were fabricated in these workshops as well. This unique object is a head belonging to an unfired figurine representing an elderly woman wearing a floral wreath on her head. Its realistic features and dramatic expression betray the activity of a local workshop with high artistic aspirations. This head was found beside a fragmentarily preserved kiln in the Early Ptolemaic stratum. Another original figurine of slightly later date (possibly late 3rdearly 2nd century BC) came to light in the vicinity of another kiln. It represented a seated dwarf with an animal tail and a disproportionately large phallus. Two lotus buds decorate his bald head. Several heads belonging to this type have already been found in our Early Ptolemaic strata but this is the first complete specimen of what was doubtless a fertility divinity of some kind, perhaps a magical symbol. This terracotta type must also have been produced in the local workshops at Athribis. A series of red-brick structures covered with a layer of plain white or polychrome plaster came to light below the Byzantine strata in rooms 160-162 (numbers corresponding to the Byzantine level). The evidence provided by the numismatic material points to the reign of Ptolemy VI as the date for these constructions. They comprise a long corridor leading to a square basin, which has a wide inner step and a miniature basin near its northwestern corner. Smaller basins of various shape occur in two groups east of the corridor. One of these groups is constituted by a pair of oval lavatories, which recall in shape those found previously a few meters from this place, close to the northwestern corner of the excavated area. It seems that 28 PAM II [= Reports 1989-1990] © PCMA 2008 – digital reprint all these structures belonged to a complex of baths built in the mid 2nd century BC. Some erotic figurines and relief fragments were found inside and beside these structures. Their iconography would suggest rather a frivolous function for the structure, maybe a "house of joy". This "recreation center" probably embraced the adjacent room 159, which was uncovered during previous campaigns and which is much larger than the workshops found in the southern part of the Ptolemaic quarter. The walls of this room were decorated with painted plaster. The archaeological context includes ceramic drinking cups with saucers and a few finely painted vessels. The relatively elevated stratigraphic position of the mid-Ptolemaic structures in this part of the excavated area, as well as the gradual sloping of Ptolemaic and Roman layers eastward and south- Fig. 2. Terracotta seal with relief decoration. Second half of the Ptolemaic Period. 29 © PCMA 2008 – digital reprint PAM II [= Reports 1989-1990] ward imply that at one time there was a natural hill here, occupied by buildings of a public or official nature. The workshop quarter was situated further to the south. Among the important finds of Roman date there is a headless statue representing a man standing beside a herm, which is crowned with the head of a bearded man wearing a headband. The composition and the modelling of the sculpture betray the rather clumsy hand of a local copyist. Objects of Ptolemaic and Roman date were found inside some of the walls, which had to be removed or cleared in order to facilitate further exploration. Among these there is a round terracotta seal which may be dated to the second half of the Ptolemaic Period (Fig. 2). Its flat surface bears a sunken relief representing a cupid riding a dolphin. The drilling carried out with a hand drill reached a depth of c 4 m below the lowermost of the excavated strata. An examination of the core samples demonstrated that this area was in use in the Late Dynastic Period (probably from the Thirtieth Dynasty on). There seem to be no artifacts of earlier date in this material. These early strata contain lots of pottery sherds, and in one of the examined spots there were also fragments of various stone objects (a stonecutting atelier ?). 30 PAM II [= Reports 1989-1990] © PCMA 2008 – digital reprint MARINA 1990 Wiktor A. Daszewski The 1990 season of excavations at Marina el-Alamein was carried out by an archaeological mission sponsored by the PCMA and lasted from 20 February until 22 March.l Work continued in the same area as in previous seasons, on the site of an ancient settlement where today a large tourist centre is under construction. Envisioned for a clientele in the thousands, the modern centre will include, beside a developed living complex consisting of villas, apartment houses and hotels, a yacht marina located inside the lagoon, the same which in Antiquity must have served as a harbor for coastal navigation between Alexandria and Cyrenaica.2 Polish excavations are concentrated on the necropolis in the southwestern part of the ancient settlement. To the extent allowed by circumstances the area covered by the ancient city was surveyed again and the monuments discovered accidentally in the course of construction works in 1986-1987 continued to be recorded. l 2 Prof. W.A. Daszewski was in charge of the mission, which included Prof. Zofia Sztetyłło, archaeologist and epigraphist, Dr. Maria Kaczmarek, anthropologist, Mr. Grzegorz Majcherek and Miss Iwona Zych, archaeologists, Mr. Jarosław Dobrowolski, architect and Mr. Stefan Sadowski, photographer. Mr. Kazimierz Błaszczyk participated in the work periodically. Following the excavations Messrs J. Dobrowolski and K. Błaszczyk, working under the terms of an agreement with the EAO, completed a full anastylosis of one of the tombs (T 1) and reinforced parts of the others. See J. Dobrowolski, p. 44ff. The EAO was represented by an inspector, Mr. Ramadan Musa. The mission would like to express its sincere gratitude for help received in the course of the work to Dr. Ali el-Kholy, Director General of the EAO for the Western Delta, Mr. Faisal Ashmawy, Director General of the Inspectorate of Marsa Matruh, and Mr. Ezzat el-Hammahmy, General Inspector of the EAO in el-Hammam. For the identification of the ancient centre on the basis of ancient literary sources, see W.A. Daszewski et alii, MDAIK 46, 1990. 31 © PCMA 2008 – digital reprint PAM II [= Reports 1989-1990] On the necropolis the mission continued the exploration of a complex of tombs located closest to the limits of the settlement. This complex was discovered and partly cleared during the previous seasons. The tombs were constructed on ground level and were made of dressed limestone blocks. As a rule they took on the form of large rectangular "boxes", around 2.80 x 2.50 m, equipped with two, three or four loculi for the dead. Crowning the structures was a kind of stylobate of two or three steps, constituting a base of sorts for the architectural monument built on top of the tomb – a massive column in the case of Tomb T 1, a pillar with corner pilasters in tomb T 1B, and finally a pillar with engaged columns in tomb T 1D. The capitals which topped the column as well as the pilasters were of a specific type recalling so-called Nabatean capitals.3 The loculi in the tombs were very much the same in size (about 2 m long, 0.60 m wide and from 0.90 to 1 m high), but they differed substantially in the number of dead laid to rest in them. This varied from one to several. During this season three loculi were investigated thoroughly: the northern loculus of T 1B and the middle and northern loculi of T 1D. In the loculus of Tomb 1B 13 skeletons were found. Anthropological examinations identified three males in the age groups 35-40(1) and 38-47 (2) and two females, 16-18 and 25-55 years old. The remaining skeletons belonged to one newborn and seven children from 18 months to five years of age. The disposition of human remains in tomb T 1D constitutes an interesting case. The central loculus, which had been discovered intact, closed with a slab decorated with a relief representation of the serpent Agathodaimon (see earlier report), was found to contain just one skeleton, that of a male 20-30 years old. In contrast, the adjacent northern loculus (the southern was not investigated this year) yielded remains of 22 persons including 5 males in the 35-55 age group, 2 females (40-50 years) and 15 children. 3 See W.A. Daszewski, EtTrav 15 (1991). 32 PAM II [= Reports 1989-1990] © PCMA 2008 – digital reprint Reconstruction work proceeding on Tomb 1 (excavated in the 1987/88 season) uncovered a third loculus beside the two already known, which had contained 11 and 16 burials respectively. This new loculus, discovered in the central part of the box-like tomb structure which served as a proper base for the columnar monument decorating the tomb, was found to contain two skeletons; the bottom one had been laid in a shallow pit scooped out in bedrock, while the upper one lay some 0.20 m above the first, on a layer of sand. It should be assumed that these two burials were the earliest, the tomb being literally built on top of them. A similar situation was encountered in tomb T 1B where beside the two side loculi there was a central one, originally just a trench in bedrock. The box-like structure of the tomb with its 5 m-high monumental pillar was built over this trench burial. The unidentifiable part of a tomb structure which was discovered last year joining at an angle the southeastern corner of Tomb 1D, upon closer investigation turned out to be a part of a different type of tomb in which the superstructure is connected with an underground burial chamber (Fig. 1). The part discovered at the end Fig. 1. Tomb T 1GH. 33 © PCMA 2008 – digital reprint PAM II [= Reports 1989-1990] of the previous season belonged to a monumental entrance which opened on a vaulted staircase leading to the hypogeum. The gate which was decorated with dentils and framed with relief pillars, was constructed of limestone blocks, as were the vault and upper section of the stairs. The lower steps were cut in bedrock. The underground chamber had nine loculi cut in the walls. Eight of the loculi were intended for burials, while the ninth was a shallow niche for grave furnishings. At the spot where the staircase opened into the chamber there was a square ventilation and lighting shaft (1.20 m to the side) cut from the surface. On ground level, this shaft was built up with limestone blocks (preserved height 1.20 m) and decorated with pilasters on the inside corners and half-sections of small Nabatean capitals on the outside. The full length of the tomb from gateway to the far wall of the chamber is 13 m; the hypogeum measures about 7 x 5.50 m. Particular loculi in the hypogeum contained from two to six burials. The finest burial was that of a male about 45-55 years of age whose remains had been placed in a rectangular lead coffin (1.83 m long, 0.34-0.42 m wide, 0.24 m high). Similar undecorated lead coffins are known from the Hellenistic necropolis of Hadra in Alexandria. The gifts that accompanied the man buried in the coffin included two terracotta unguentaria and Fig. 2. Terracotta oil lamp found in Tomb T 1GH. 34 PAM II [= Reports 1989-1990] © PCMA 2008 – digital reprint a glass bottle of excellent quality set next to the head. The glass receptacle still contained the dried remnants of some substance, presumably fragrant oils. At head and hip level, large holes had been ripped out of the coffin cover; apparently the tomb robbers expected to find jewelry on the head or neck and fingers of the buried man. This violation must have occurred relatively shortly after the burial itself. Evidence supplied by the finds from particular loculi – sherds, glass vessels and a terracotta oil lamp (Fig. 2) – indicated that the tomb was in use from the second half (or end) of the 1st century BC until the middle of the 1st century AD. After that it was filled completely with sand and remained untouched till the present day. A definitely more monumental burial structure was discovered in the western part of the necropolis, at some distance from the settlement. Its aboveground part was excavated in the course of previous seasons (Fig. 3). Built on a rectangular plan (17.60 x 10.80 m), the building had an Ionian column portico decorating its northern seaward facade and a large reception hall in the centre, furnished with two banquet couches of masonry. The rest of the building contained storerooms, lodgings and rooms for preparing banquets. This superstructure was connected integrally with an underground part. A monumental entrance framed by half columns and topped with an arch led from the southern end of the banquet hall to a small vestibule and a monumental staircase 17 m long descending 7 m into the rock. At the end of the steps there was a vertical shaft (6 x 5.50 m) which functioned as a sort of inner open court. On the surface the shaft was surrounded by a wall of limestone blocks which served as protection against sand. A rectangular opening in the southern side of the shaft led to an underground chamber measuring 7.25 x 6 m. Inside, the chamber had benches running along three sides and an altar set in the middle. It would appear that the chamber was never completed. Only one loculus was found, cut in the southwestern upper corner of the chamber. Tracing lines preserved on the ceiling of the chamber indicate where other loculi 35 © PCMA 2008 – digital reprint PAM II [= Reports 1989-1990] Fig. 3. Funerary monument S 6 in the southwestern part of the necropolis. were to be cut. Two burials were found in the loculus. Three other skeletons lay on a heap of debris in the corner of the room and on the bench nearby. The eastern part of the shaft has not been explored as yet. One cannot exclude that another chamber will be found here. The monumental scale of this complex is striking. The fineness of the superstructure and the dimensions of the underground part contrast with the unfinished state of the chamber, which was presumably intended as the main one, lying as it is on the main axis of the tomb. The total length of this tomb, taking together the superstructure and the underground part, is 42 m. It is undoubtedly one of the largest burial complexes of the Early Roman period in Egypt. Presumably, it was constructed in the 1st century AD and existed till the beginning of the 2nd century AD. At this time, it was apparently destroyed to some extent, the ruins subsequently serving as a place for various more or less accidental burials. Gifts accompanying the burials mentioned in this report were by custom extremely modest, remaining in contrast with the monumental character of the architectural complexes. Nevertheless, the 36 PAM II [= Reports 1989-1990] © PCMA 2008 – digital reprint few objects that were recovered are of value for they supply dating evidence. The box-like tomb structures may be dated on the basis of some terracotta oil lamps and glass vessels (mainly unguentaria and balsamaria) to the period from the end of the 1st century BC to the second half of the 1st century AD. A similar date can be assigned to hypogeum T 1GH. The monumental nature of the tombs on the southwestern necropolis at Marina el-Alamein corresponds to a peak in the development of the ancient town. At present some extensive living complexes are being excavated by the EAO in the area of the settlement. These are large prosperous houses with inner peristyle courts and reception halls. The rich architectural decoration of these houses is executed with great care as are murals, fragments of which have been preserved in situ. The first results of anthropological examinations draw an interesting picture of the population of this ancient centre. The sample collected in the necropolis is constituted by 78 skeletons. The mortality rate of children in the age group of 0-7 years is high (36%), typical of ancient societies. The mortality rate is almost identical for men and women in the age groups 20-35 (33%) and over 55 years (2.5% and 1.5%). However, in the age group 35-55 the mortality rate of males was six times that of females. The average age at death of males was about 40 years, of females around 35. It should be noted though that the sample is too small to permit any more general conclusions. Finally, it is worth mentioning that the average number of children borne to the women in this sample was three to five. Detailed findings concerning the morphological picture as well as pathology will be published separately by Dr. Maria Kaczmarek. 37 © PCMA 2008 – digital reprint PAM II [= Reports 1989-1990] THE ACTIVITIES OF THE POLISH-EGYPTIAN PRESERVATION MISSION AT MARINA EL-ALAMEIN IN 1988 Włodzimierz Bentkowski In 1988, under an agreement between the Egyptian Antiquities Organization, the PCMA and the State Ateliers for Conservation of Cultural Heritage (PKZ), a group of restoration specialists from the Zamość Division of PKZ, started conservation and restoration work at an ancient site discovered near el-Alamein in northern Egypt.1 Already in the summer and autumn of 1987 most of the discovered monuments were catalogued by a group of conservators from the PCMA and PKZ, this last group headed by architect G. Zamoyski. The site is located on the coast about 6 km east of el-Alamein. It stretches 1200 m east-west along the sea shore and 800 m inward from the coast. About 30 ancient structures were uncovered on the site. Preliminary investigations have shown that a settlement had existed here between the 2nd century BC and the 6th century AD. Among the structures uncovered so far are residential houses and public buildings, infrastructure and tombs. The condition of the ancient substance varies greatly. In most cases it is possible to trace fragments of walls, floors and fragmentary architectural details. The state of walls varies depending on the material used (dressed or broken stone, dried brick) and binding agents (limestone and clay mortars). 1 The group included architects Messrs Jan Radzik, Krzysztof Kamiński, constructor Mr. Andrzej Sawecki, conservator Mrs. Urszula Fidecka and the present writer. Working with us were Egyptian archaeologists: Messrs Mohammed Ashmawy, Abdelaziz el-Sayed el-Shennawy and Ezzat el-Hamahmy. The full report on preservation works in 1988 and the preliminary results of excavations in 1987-1989 will be published in: Marina el-Alamein, vol. 1, Archaeological Background and Conservation Problems, Warsaw 1991 (in press). 38 PAM II [= Reports 1989-1990] © PCMA 2008 – digital reprint In practice, we were limited by our means to a study of just three of the several discovered structures. We selected the following monuments: an underground cistern, a residential house and a monumental tomb. The choice was governed by the relatively good state of preservation, the most advanced stage of archaeological research, as well as the different function and architectural form of these structures. UNDERGROUND CISTERN (Fig. 1) The cistern, situated in the southeastern part of the ancient site, consists of a reservoir excavated in the rock and a partly vaulted entrance. It is filled with a deep layer of silt sand and the rocky roof is supported by four extant stone pillars has been damaged by seismic activity. The main problem for conservation in this case was to find a way to consolidate the rock in order to create conditions for completing research and making the site accessible to visitors. A special system of reinforced concrete binding joists was designed to strengthen the cistern roof; damaged sections of the roof should be underslung Fig. 1. Plan of the cistern. 39 © PCMA 2008 – digital reprint PAM II [= Reports 1989-1990] to this grid. The whole construction can be hidden under sandy soil such as covers the roof of the cistern already. The conservation program includes reinstallation of several sections of the corridor vaulting, bracing of the pillars and filling gaps with concrete. These measures will stabilize the structure, making future research safe. HOUSE H 9 (Fig. 2) House no. 9 is located in the central part of the ancient site. It is rectangular in outline (40 by 24 m), its walls preserved to the height of c. 2 m. The house has been cleared by Egyptian archaeologists, providing a complex of 14 rooms grouped around a central courtyard. The entrance to the building is on the western side. Four stone steps lead through a door to a vestibule (No. 1). South of it there is a latrine (No. 2) provided with a channel leading to a sedimentation tank located behind the outer wall of the house. On the northern side of the vestibule there is a storage (?) room (No. 3). Next is a kind of corridor (No. 4) leading to the courtyard (No. 5), the central part of which was probably left open. This court was lined on two sides by a portico and on the third by a vaulted (?) passage flanked by two pilasters. On the northern side, three medium-sized rooms (Nos 6, 7, 8) opened onto the court. In one of them (No. 6) fragments of a ceramic vessel were found forming part of a 'chimney' system. On the south the court is flanked by a large hall accessible through an opening with traces of a door (No. 9). In the next room (No. 10), which probably had a storage function, the floor was made of horizontally placed amphorae covered with sand. An opening closed by a stone slab gave access to two small storage cellars. Next to hall No. 9 there is a double-track stairway (No. 11) with its lowest two steps preserved. Next to it is a room (No. 12) situated in the 40 PAM II [= Reports 1989-1990] © PCMA 2008 – digital reprint southeastern corner of the house. Facing west, one can enter room No. 13 from the courtyard, while turning east through two openings one can enter the largest room of the house (No. 14). In one of these openings traces of a wooden doorframe were preserved. In the other there were remnants of wooden partitioning. The last space is a longitudinal room accessible from the previous one. The house was provided with a cistern under the court. Rainwater ran directly from the roof to the cistern. Water from the courtyard served to flush the latrine or was directed to a container disguised as a bench near the entrance. In the course of exploration many architectural details were found, including column drums, capitals, fragments of a tympanon, cornices and a frieze with triglyphs and metopes. Of the greatest interest were remnants of an aedicula – a shallow niche crowned with a conch and a tympanum supported on either side by two small columns (Fig. 3). Fig. 2. House H 9, ground plan. 41 © PCMA 2008 – digital reprint PAM II [= Reports 1989-1990] Fig. 3. Reconstruction of the aedicula found in House H 9. The walls of the building were covered with plaster. Some of the plasterwork was decorated with polychrome panels. Similar colors (blue, yellow, dark red and black) were found on the gypsum stucco of the courtyard columns, as well as on the stuccowork present on the remnants of the aedicula. Waterproof plaster lined the cistern, the latrine channel and the gutter. Beside preparing a full architectural record of the monument, our aim was to specify the required reconstruction work. In the case of House H 9 we decided to limit such work as much as possible. The structure we were dealing with here was only partly uncovered and its plan could not be traced in its entirety. However, some of the discovered elements needed to be protected and enhanced. Thus, we suggested such conservation work as was necessary to enable a clear and proper exposition of the structure in the future. Beside reinforcing and building up the walls in several places and making the openings more visible, we proposed a full reconstruction of a column from the portico and of the aedicula which could be placed in the wall of room No. 14. 42 PAM II [= Reports 1989-1990] © PCMA 2008 – digital reprint TOMB MONUMENT T 1 The monument discovered in 1987 (see W.A. Daszewski, p. 33ff. in this volume) is situated in the southwestern corner of the site. It consists of a pedestal (containing two loculi) and a column with base and capital. Some 80% of the structure is preserved undamaged. The monument was restored in 1990 (see J. Dobrowolski, p. 44ff. below). The three monuments discussed above are only a small part of those already discovered. All of them require urgent reconstruction work. Leaving them unprotected will lead to their complete disintegration in the course of just a few years. 43 © PCMA 2008 – digital reprint PAM II [= Reports 1989-1990] POLISH-EGYPTIAN RESTORATION MISSION AT MARINA EL-ALAMEIN IN 1990 Jarosław Dobrowolski The restoration mission1 operated in the complex of tombs excavated by Prof. Wiktor A. Daszewski,2 dating from the 1st century BC to the 3rd century AD (Fig. 1). All these monuments were built of soft local limestone and were discovered much destroyed, probably ruined by an earthquake and then damaged by extensive erosion. As the exposed stones are subject to further serious damage in the coastal climate of the site, it was found necessary to reassemble the excavated structures, thus also enhancing the touristic value of the site. A Polish-Egyptian mission sponsored by the EAO and PCMA was formed to perform the task. The main objective of the 1990 season was to reassemble a tomb registered as T l. The structure was excavated in 1987 and dated to the 1st century AD. A preliminary design for its restoration was prepared in 1988 by Dr. W. Bentkowski (see p. 41ff. above). It was established, both by excavations in 1987 and by additional research occasioned by the restoration works, that the tomb had the form of a huge squat column resting on a stepped stylobate which was mounted in turn on a cubic pedestal (Fig. 2). This large base contained two burial chambers (loculi) with entrances blocked by stone slabs, inside which multiple burials were found. The central part formed a stone casket built over a pit hollowed in bedrock. 1 2 From the PCMA: Jarosław Dobrowolski, architect (head of the mission) and Kazimierz Błaszczyk, building engineer. Representing the EAO were inspectors: Mr. Nader Ramadan Musa and Mr. Muhammad Abdel Hamid el Said. The photographic record was provided by Mr. Stefan Sadowski, a photographer in the PCMA. See above, p. 26ff. 44 PAM II [= Reports 1989-1990] © PCMA 2008 – digital reprint Fig. 1. Location of the restored tomb T l within the complex excavated by the Polish Mission. This compartment, which had been sealed by the construction of the superimposed pillar, was found to contain two skeletons (one in the pit and one above it). The column was a plain straight shaft, built of drums approximately 0.30 m high and consisting of four segments each, with a base and a capital of extremely simplified form. Lines traced on the upper surface of the capital suggest that a funerary monument (probably a statue) had once stood upon it, but no clues as to its form have been found. The column's simple base and capital were the only decorative elements of the tomb. The base of the structure and a part of the stylobate were preserved in situ, although some stones were missing and the surfaces were badly eroded. The column had fallen and its blocks were found lying over a neighboring tomb. In the first stage of restoration, these blocks (stored at the site since 1987) were gathered together and the drums were assembled on the ground according to their position recorded during the excavation; missing blocks were then added. Of the two surviving 45 © PCMA 2008 – digital reprint PAM II [= Reports 1989-1990] Fig. 2. Tomb T1 after restoration. A – new blocks, R – limestone rubble and mortar, C – bedrock, D – cardboard separation inserts, E – empty joints around edges. quarters of the column's capital one was discovered broken to pieces and was bound together with polyvinyl acetate in 1987. In 1990 the joints were found to be in satisfactory state and the block was classified fit for reinstallment; outer cracks were filled with mortar. Two missing quarters of the capital were supplemented by blocks cut at the site.3 Meanwhile eroded joints of the base of the monument were filled in with mortar, loose blocks adjusted and fixed, missing blocks restored. The interior of the central part of the base was cleared of the loose sand filling it down to bedrock and limestone rubble and mortar were put in its place, creating a solid foundation for the column. To prevent the weight of the column 3 By Messrs T. Kołaczyński and J. Sternak of the PKZ's Polish-Egyptian Islamic Mission in Cairo. 46 PAM II [= Reports 1989-1990] © PCMA 2008 – digital reprint from exerting pressure upon the much eroded outer walls, this concrete core was separated from the stone casing of the stylobate by board inserts and raked joints. Finally the column was reassembled using the simple method of wooden scaffolding and a pulley in a manner probably quite similar to the original construction. All the original blocks that could be installed were reintroduced into the structure. Newly added blocks amount to roughly 25% of all the stones. They were given a slightly different surface finish to distinguish them from the original ones. Total height of the monument after reconstruction is approximately 6.2 m, with the column itself 4.22 m high and 1.04 m in diameter. Apart from this main task, restoration works were performed on a neighboring tomb T 1A, a much damaged structure originally comprising four loculi in two levels, of which only the lower part was preserved. This tomb was protected in its ruined state by filling in eroded joints, adjusting and fixing loose stones, and supplementing a few missing blocks. A sand-retaining wall was built as a support for this tomb's foundation; it lies approximately 1 m higher than the neighboring T l. Some protective work was also performed on two other tombs, the full restoration of which is planned for the next season. 47 © PCMA 2008 – digital reprint PAM II [= Reports 1989-1990] DEIR EL-NAQLUN, 1990 Włodzimierz Godlewski In the sixth field season in Deir el-Naqlun (October 25 to November 14), a mission from the PCMA1 concentrated work on the excavation of Hermitage 89 and on a preliminary investigation of the wall surrounding the existing monastic complex of Deir el-Malak Ghubrail.2 HERMITAGE 89 Hermitage 89 lies immediately to the north of Hermitage 25, which was excavated in the 1989 season. Like the other hermitages, it was cut in a rocky ridge, which is the eastern edge of a plateau where the complex of monastic buildings had been built. This hermitage consists of six units, two of which were constructed entirely of mud brick, stone, bricks and sherds upon a courtyard in front of the original hermitage complex (Fig. 1). Notwithstanding, the plan of this hermitage does not fall into the same scheme as that of hermitages 1, 2 and 25 discovered so far. Only units A.1-2, constituting the core of the hermitage, may be considered similar to the units inhabited by the anchorites. Complex A.1-2 consists of two units. A large space A.1 (5 x 5.50 m) was 1 2 The mission led by Dr. Włodzimierz Godlewski included: Mr. Tomasz Derda, papyrologist, Mr. Jarosław Dobrowolski, architect, Mr. Tomasz Górecki, ceramologist, Mr. Piotr Jeute, archaeologist, Mr. Olaf E. Kaper, egyptologist and arabist, Mr. Adam Łajtar, epigraphist, Mrs. Bożena Mierzejewska, archaeologist, Mr. Bogusław Okupny, photographer, Mr. Muhammad Ahmed Halim, inspector of the Islamic and Coptic Section of the EAO in the Fayum. We are greatly indebted to EAO staff in Cairo and Fayum for their constant help during the season. J. Dobrowolski, Naqlun – Deir el-Malak Ghubrail: the existing monastic complex, Nubica 1-2, 1990, pp. 161-170. 48 PAM II [= Reports 1989-1990] © PCMA 2008 – digital reprint Fig. 1. Plan of hermitage 89. meticulously plastered and equipped with a storage bin cut in the floor in the southeastern corner, a bench along the western wall and a bed located just by the doorway. There were also several niches cut in the walls, the one in the eastern wall doubtless serving as a prayer niche. The other room (A.2) was much smaller and remained unplastered. Beside a niche in the eastern wall, it had a spacious niche, a small room in fact, in the southeastern corner, with a floor raised about 1 m above the walking level in this room. This niche presumably served as a place for sleeping. Both the equipment of the larger room and the various changes made in its interior are evidence for a rather long occupation of this complex. In the second phase, when the storage bin was filled in and covered with a floor, when a comfortable bed with a profiled headrest was added just by the southern wall and a hearth for cooking meals located in the northwestern corner, the complex seems to have functioned together with unit B.1, communicating through an enlarged window previously set in a niche in back of the storage bin. 49 © PCMA 2008 – digital reprint PAM II [= Reports 1989-1990] Unit B.1 is small (3.40 x 3 m), cut in rock and carefully plastered with mud and a layer of lime plaster. It had two benches by the northern and western walls and a prayer niche in the eastern one. The niche had a painted frame, of which only small fragments have survived. On the outside there was a band of floral decoration – a sort of wreath, with possibly palm leaves at the bottom. Inside the niche there presumably was a cross upon a podium: remnants of the latter survived as the sole evidence of the composition. It would appear then that in the second phase of the utilization of complex A.1-2, unit B.1 served as a sort of oratory. It was visited often by pilgrims who left numerous Coptic and Arabic graffiti giving names of visitors sometimes accompanied by their functions and their places of origin. All of the writings are undoubtedly Christian and seem to have been made in Late Medieval times. In the last phase of the hermitage's habitation a small vestibule B.2 was added in front of unit B.1; the new structure was built upon the original courtyard. Hermitage 89 had two other units, which are only just traceable. Unit D in the northern part of the complex was preserved at barely walking level: the kind of plaster floor in the interior suggests it had been a living space. The walls of unit C were of bricks and sherds set in mud, the roof was of reeds. Considering the character of the fill in this unit, it must have served as a domestic area, perhaps a pen: donkeys seem to have been kept here for some time. A large part of the original courtyard, presumably connected with the A.1-2 complex as well as possibly with units C and D, and at one time built over with units B.2 and C, has not been excavated, especially the southern and southwestern parts. Presumably part of the courtyard, just as the northern and southern ends of unit D, had been damaged. The fill of the rooms of hermitage 89 contained several objects connected with the dally life of the anchorites. These included fragments of clothes: both thick and thin tunics, and pieces of shoes, 50 PAM II [= Reports 1989-1990] © PCMA 2008 – digital reprint mainly sandals, with the upper soles decorated very carefully with impressed geometrical designs. A large portion of the finds are mats, ropes and baskets which may have been produced in the hermitage or may have simply been used here, for they were found mainly in unit C. Numerous wooden objects were either part of the spiritual life of the hermitage's inhabitants (fragments of crosses) or constituted elements of the equipment: pegs for setting in the upper parts of walls, pieces of tools. Apart from the pottery, on which below, written documents constitute the most important finds. Several were found written both on papyrus and paper, in three languages: Greek, Coptic and Arabic. The Greek and Arabic texts are especially interesting, being of the nature of contracts and agreements.3 Small pieces of a papyrus codex in Coptic additionally contain illustrations placed on the margins of the pages. The ceramic assemblage from Hermitage 89 falls into two groups: the pottery from unit C and the pottery from the storage pit in unit A.1.4 Typologically, these two groups are not homogenous, although both are the result of the same circumstances, that is both were deposited as rubbish in two separate spots, but at a similar time. In both groups domestic vessels, primarily amphorae and cooking pots, predominate. Storage vessels and tableware are much less frequent. Among the finds from unit C there stands out a three-handled pithos with painted decoration, not figural, of the type black and red upon white (slip). Of interest are also small tableware pieces representing local imitations of Late Roman wares, made of Nile silt (group K) and marl clay from the Aswan region (group 0). The contents of the pit in unit A.1 is much more interesting. Apart from numerous household pots, it included a group of about 3 4 See papers by Tomasz Derda and Olaf E. Kaper, below, pp. 54ff and 57ff. The information concerning pottery comes from Mr. Tomasz Górecki, who is responsible for preparing the ceramic material for publication. 51 © PCMA 2008 – digital reprint PAM II [= Reports 1989-1990] Fig. 2. Miniature amphora with painted and applied decoration. 40 vessels, handmade of Nile silt, decorated either with ornamental openwork, impressed or applied patterns, ornaments painted in white. The most frequently encountered form is an incense burner (about 10 pieces), next were small storage pots (four objects), small amphorae (three) (Fig. 2) and chalices (two). Apart from the pottery there were fragments of unfired vessels belonging, as far as form and material are concerned, to the same group; doubtless the vessels must have been produced and decorated in the hermitage or its immediate vicinity. As in unit C, a small number of tableware objects were found, all imitations of Late Roman groups K and 0. The dating of both chronologically coexistent dumps and their pottery contents is not certain. On the grounds of a few imported amphorae and a large assemblage of so-called chocolate amphorae 52 PAM II [= Reports 1989-1990] © PCMA 2008 – digital reprint of Egyptian production, the two deposits from the pit in unit A.1 and unit C may be dated to the end of the 6th and the 7th centuries. The group of handmade pots should be dated to not earlier than the middle of the 7th century in view of its similarity to Early Islamic unglazed pottery with openwork decoration. It would seem then that both deposits date to the beginning of the second half of the 7th century. The rare finds of pottery from the upper layers of the fill in units A.1, D and B.1 are similar to the pottery discovered last year inside Hermitage 25 and can be dated to the 9th-10th centuries. 53 © PCMA 2008 – digital reprint PAM II [= Reports 1989-1990] EL-NAQLUN 1990: THE GREEK PAPYRI Tomasz Derda Hermitage 89, described above by W. Godlewski (see p. 48ff.), was in use for a relatively long time. It is, therefore, hardly surprising that only some tiny scraps of papyrus were found in particular rooms and their identification as Greek is often only hypothetical. In room A.1, beside the scraps of two other presumably Greek papyri (inv. nos 49 and 90/90), there was a small piece of an account (inv. no. 51/90) written in chancery cursive of the 6th-7th centuries AD. In room C only one Greek fragment was found (inv. no. 13/90); it is a piece of another account written in upright chancery cursive. Of much greater importance was the contents of the storage pit in room A.1 which had been filled in and covered with a floor during a renovation of the hermitage interior in the 7th century (the date for the renovation is set by the finds, the papyri among them). Three papyri (inv. nos 87, 96 and 99/90) preserved in many fragments were found in this pit. There is complete certainty that even the smallest fragments belonged to one of these three. All three papyri are legal documents. The first one, inv. no. 87/90 (top and left margin are lost), is a document, most probably a receipt for the repayment of a loan, recorded by a notary of the Arsinoite nome; his subscription partly preserved at the bottom of the document cannot be identified with the subscriptions known from the other documents from this nome. The second document, inv. no. 96/90, preserved in two big fragments (bottom with the notary's subscription lost), is also a receipt for the repayment of a loan; the debtor is a certain Aurelius Makarios son of Jakob, from the village of Alexandrou Nesos in the Arsinoite nome, the creditor is a monk whose name is partly destroyed by a lacuna his function being called proestos Pyrgou, i.e., the abbot of a Pyrgos monastery. Although papyrological lexica do not men54 PAM II [= Reports 1989-1990] © PCMA 2008 – digital reprint tion the toponym Pyrgos, the reading seems to be fairly assured. The document begins with the indiction date (20th Phaophi of the 4th indiction), there is no invocation with the Christ formula. The third document, inv. no. 99/90, is preserved in many small, separate fragments. The only part which can be reconstructed at the moment is the top of the document (the first three lines) containing an invocation with the Christ formula EN ONOMATI TOY K(YPIO)Y KAI DECΠOTOY I(HCO)Y X(PICTO)Y TOY Θ(EO)Y (KAI) ΣWTHPOΣ HMWN, "In the name of our Lord, Jesus Christ, (our) God and Saviour", and the indiction date (15th Mesore of the 12th indiction). Many small fragments from the middle and bottom parts of the document seem to suggest that this was also a notarial document as the two texts described above, constituting most probably a receipt for the repayment of a loan. From the archaeological point of view, of greatest importance is the question of the date of these three papyri. As stated above, two of them (inv. nos 96 and 99/90) have their headings preserved; it is significant that there is an invocation with the Christ formula in the latter, while there is none in the first. The invocation as an element of legal documents was instituted by imperial decree of Mauricius; its force was felt in Egypt first in the year AD 591, and all known legal documents dated after September 591 have the invocation. There were in use, however, four main versions of the invocation and this is very important for setting a date for a document. First, under Mauricius (591-602) only the Christ formula was used; under Phocas (602-610) this formula was replaced by a new one invoking the Holy Trinity; under Heraclius (610-619; 629-641) a major division took place in which the Upper Egyptian nomes used the Trinitarian formula introduced under Phocas, while the Lower Egyptian nomes reverted to the formula of Christ that was in use under Mauricius; after the Arab conquest a slightly modified version of the Trinitarian formula was reestablished in the whole of Egypt (it is difficult to determine 55 © PCMA 2008 – digital reprint PAM II [= Reports 1989-1990] which version of the formula was in use during the Persian domination in the years 619-629, since we have only a few documents dated surely to that period). Papyrus Naqlun inv. no. 99/90 is dated 15th Mesore of the 12th indiction. This gives two possible dates: 8 September 593 or the same day in the year 638. We should, however, prefer the first date, if we assume that the papyri found in the storage pit in room A.1 were written more or less simultaneously. On this assumption P. Naqlun inv. no. 96/90, dated 20th Phaophi of the 4th indiction, might be dated either to 17 October 585 (i.e., to the last possible date before the decree of Mauricius) or, less convincingly, to the same day 15 years earlier. 56 PAM II [= Reports 1989-1990] © PCMA 2008 – digital reprint ARABIC PAPYRI AND INSCRIPTIONS FROM NAQLUN, HERMITAGE 89 Olaf E. Kaper During the excavations of Hermitage 89 (see above, p. 48ff.) a collection of Arabic texts and inscriptions was found. The following presentation of this find is based only on a preliminary investigation, principally concerned with the dating of the texts and inscriptions, as well as with the personal and geographical names occurring therein. N 88/90. Fragment of an Arabic papyrus. This is probably the oldest Arabic text found in this hermitage. Although papyrus as writing material continued to be used into the 14th century AD,1 the present document displays a variety of Kufic script which is characteristic of papyri from the 7th century AD.2 The fragment measures about 11 x 8.5 cm. On the recto (hor. fibres) only six fragmentary lines survive, including the right margin of the papyrus, and a part of a Greek numeral (ΣΝ, 250?) above. The first line contains the Basmalla, the remainder of the contents is unclear. The other Arabic texts are all on paper, and with but one exception these are preserved only as fragments. In all, the paper finds include around 15 fragments in Arabic. In the order of their registration, these include: N 36/90. Upper part of a document of 12 x 9 cm in size, possibly a letter. The remains of ten lines are preserved, none complete. Of these, one line of the text was written along the edge of the paper, after which the paper was turned 180° and one more line was added by the same hand. N 37/90. Small paper fragment (6 x 5 cm.) containing the beginnings of four lines written on both sides in the same hand. 1 A. Grohmann, Einführung und Chrestomatie zur arabischen Papyruskunde, Bd. I, Prag 1954, pp.3-4. 2 R.G. Khoury, in: Grundriss der Arabischen Philologie, Bd. I, Wiesbaden 1982, p. 263. 57 © PCMA 2008 – digital reprint PAM II [= Reports 1989-1990] N 38/90. Four very small paper fragments, one of which shows remains of the Basmalla. N 42/90. Small fragment with on the recto the remains of five lines of a list with accompanying numbers; the verso preserves only one line on the top, the rest of the paper is empty. N 43/90. This is a larger fragment, measuring 13.5 x 23.5 cm, of a contract concerning grain (kamh), signed by four witnesses. The signatures are preserved as well as parts of the last five lines of the text itself. The beginning and the end of all the lines have been cut off. The contract mentions the town of Fayum, and supplies the date of Friday, 14 nights before the end of the month Jumada al-akhira of the year 400 AM (approx. February 5, 1010 AD). The text of the fourth witness, moreover, seems to give the equivalent sun-date; the third day of the month Mecheir of the year 3(.)9, possibly the year 379. Four witnesses signed their names under the contract, three of them carry the name Muhammad. The first is called Muhammad b. Mahmud; the second Muhammad b. Ahmed b. Khaled; and the third Muhammad b. Maher (?). The contract was later torn up and its verso used for writing a letter. After writing, the paper was folded down to a size of about 3.5 x 3.5 cm; this text is completely preserved, although the ink has faded considerably. The addressee of this letter carried the name of Theodore, likely to be a Christian. The writer begins his letter with the Basmalla. No toponyms seem to have been contained in the letter, but perhaps the mentioning of the town of Fayum in the contract on the recto gives an indication as to the whereabouts of the sender of the letter. N 89/90. (three fragments). One is a small fragment, written on both sides in two different hands. Its recto has the remains of three lines, one of which contains the Basmalla. The other two fragments are even smaller, one of these preserves text on both, sides, written in two different hands. 58 PAM II [= Reports 1989-1990] © PCMA 2008 – digital reprint The many Arabic graffiti inscribed on the walls of room B (the oratory) of Hermitage 89, preserve the names of only five of the numerous visitors that left a written trace of their presence there. Their names are: Lutfallah from Fayum town (east wall), John (east wall), Sulayman (east wall), Abu al-Sakhra, the son of Sulayman (north wall) and As'ad the son of Saeed, son of Barsuma, son of Marc (south wall). No dates are preserved in these graffiti, but the names of Barsuma and Lutfallah mentioned therein suggest a date much later than the initial period of occupation of the cell. Barsuma becomes a common name only after the death of the saint of that name in AD 1317, and Lutfallah too seems to be a relatively late name. Most of the graffiti are short and contain prayers of the usual kind as inscribed by Coptic pilgrims. This type of inscription starts with the formula ya rab arham 'abdak (once ya aghfar li-'abdik) alkhati' ("O Lord, have mercy on (and forgive) your slave, the sinner...") or: adhkur ya rab 'abdak al-khati´ ("Remember, O Lord!, your slave, the sinner..."). Only one such text on the southern wall is longer than a few lines. Two graffiti mention the place of origin of the visitors as Fayum town, leading to the same conclusion that these texts were inscribed by pilgrims and therefore in the period following the occupation of the hermitage by monks. 59 © PCMA 2008 – digital reprint PAM II [= Reports 1989-1990] DAKHLEH OASIS PROJECT: RESEARCH ON THE PETROGLYPHS, 1990 Lech Krzyżaniak This season1 was a continuation of fieldwork carried out previously. In 1985, a method for recording petroglyphs in the Dakhleh Oasis was tested2 and in 1988 the first full programme of recording was initiated in a sample area.3 This area continued to be investigated in 1990. The investigation of petroglyphs is part of the research of the Dakhleh Oasis Project (DOP) Expedition organized by the Royal Ontario Museum in Toronto and directed by Prof. Anthony J. Mills. He was kind enough to invite a team from the PCMA to join in the project. Speaking for the petroglyph team, the present author wishes to express gratitude to Professor Mills for his initiative and kindness. The petroglyph research programme in the Dakhleh Oasis comprises the recording of their total in sample areas, improving the typology and chronology hitherto established for known drawings4 and using them as a source of information for studies of the way in which human attitudes toward animals developed in the oasis.5 This 1 2 3 4 5 The fieldwork was done by Assist. Prof. Lech Krzyżaniak (Poznań Archaeological Museum) and Dr. Karla M. Kroeper (Egyptian Museum in West Berlin). See L. Krzyżaniak, K. Kroeper, Report on a reconnaissance season of recording petroglyphs, December 1985, Journal of the Society of Study of Egyptian Antiquities, vol. XV/4, 1985, pp. 138-139. L. Krzyżaniak, Dakhleh Oasis Project: interim report on the first season of the recording of petroglyphs, January-February 1988, Journal of the Society of Study of Egyptian Antiquities, in press. H.A. Winkler, Rock Drawings of Southern Upper Egypt II, London 1939. L. Krzyżaniak, Petroglyphs and the research on the development of the cultural attitude towards animals in the Dakhleh Oasis (Egypt), Sahara, vol. 3, 1990, pp. 95-97. 60 PAM II [= Reports 1989-1990] © PCMA 2008 – digital reprint source is to be combined with the evidence yielded by habitation sites in this area, which are being investigated by a different team of the DOP. The aim of the research programme concerning the petroglyphs at Dakhleh is to record the total of these remains in the sample area – square E3 – situated in the eastern part of the Oasis. The tested area covers the better part of one square kilometer and in 1988 some 18 petroglyph sites were found here. In 1990, the team continued making tracings on transparent film and describing in words these remains; some were re-photographed as well. Eight sites were fully recorded during this season (Nos. 61-39/E3/1-8). Also, several other petroglyph sites were found in the area adjacent to the sample square; a photographical record was made of most of the finds. Almost all of the recorded petroglyphs seem to be dated to the time period between the Early Holocene (8th millennium BC) and the 3rd millennium BC. The majority depict animals and humans, sometimes in association. These associations seem to refer to different stages and forms in the development of the human attitude to animals in this area and turn out to be useful evidence in our studies on this subject. It is worth mentioning that the animal species depicted in the petroglyphs were identified in the field by the archaeozoologist of the DOP. Some of the petroglyphs clearly constitute prehistoric art of high quality. They are executed using different kinds of engraving and sunk relief; no paintings have been found. The recorded petroglyphs can be classified in three groups by subject and style. The first and most numerous group is composed of pictures of animals, only seldom in association with humans. The species depicted were identified as antelopes, giraffes and other unidentified quadrupeds, ostriches and schematically executed human figures, seldom in association with animals. In one case such an association shows a scene of giraffe hunting by what appears to be men with bows, assisted by dogs (Fig. 1). This group of petroglyphs 61 © PCMA 2008 – digital reprint PAM II [= Reports 1989-1990] Fig. 1. Eastern Dakhleh Oasis, site 61-39/E3-3, petroglyph No. 1. An engraved scene showing a giraffe hunt with what seems to be bowmen assisted by dogs. seems to have many traits in common with a style defined previously as the Earliest Hunters Style, which has been considered the earliest in the Oasis.6 The second group of petroglyphs is characterized by depictions of female figures, sometimes richly decorated (Fig. 2), seldom shown in association with humans or animals. In one such case, a scene shows an antelope apparently being pulled by a man towards a schematically depicted female figure (Fig. 3). This group is tra6 Winkler, op. cit., pp. 31-32. 62 PAM II [= Reports 1989-1990] © PCMA 2008 – digital reprint Fig. 2. Eastern Dakhleh Oasis, site 61-39/E3-2, petroglyph No. 1. An engraved group of decorated female figures and later (superimposed) sandals.An engraved group of decorated female figures and later (superimposed) sandals. ditionally considered as later than the one described above and is defined as "Early Oasis Dwellers".7 The third and latest group of petroglyphs in the local sequence is dated to the 3rd millennium BC. Many of them are represented by small finds excavated from well stratified contents of stone "rings” – remains of huts dated to the 5th-6th dynasty, judging from the numerous fragments of Egyptian pottery and seal impressions on 7 Ibid., pp. 27-30. 63 © PCMA 2008 – digital reprint PAM II [= Reports 1989-1990] Fig. 3. Eastern Dakhleh Oasis, site 61-39/E3-15, petroglyph No. 5. An engraved and stylized scene, apparently showing an antelope being pulled by a man towards a female figure. clay found in the same context. The depictions show figures of longhorned cattle, donkeys and bulls in association with men and scenes of hunting with dogs. The evidence yielded by petroglyph sites in Eastern Dakhleh seems therefore to point to different stages and forms in the development of the human attitude toward animals and their exploitation in the Oasis, such as hunting (killing, trapping, capturing) and husbandry (cattle, donkey). The fieldwork will be continued in 1992. 64 PAM II [= Reports 1989-1990] © PCMA 2008 – digital reprint OLD DONGOLA 1989-1990 Stefan Jakobielski In January and February of two consecutive years, 1989 and 1990, a team from the PCMA conducted respectively the 23rd and 24th campaign of excavations. 1 Fig. 1. Plan showing sites investigated in 1989-90. The work was carried out on chosen sites within the limits of our concession area (Fig. 1) and comprised: – continuation of excavations in the urban complex on Kom P, i.e., house PCH.1, conducted by Prof. W. Godlewski.2 1 2 The two expeditions headed by Dr. Stefan Jakobielski comprised: Prof. Włodzimierz Godlewski 1989, 1990; Dr. Bogdan Żurawski 1989, 1990; architect Mr. Jarosław Dobrowolski 1989, 1990; Dr. Małgorzata Martens-Czarnecka 1990; Dr. Krzysztof Misiewicz 1989; Mrs. Ida Ryl-Preibisz 1990; conservator Mrs. Ewa Parandowska 1990; anthropologist Dr. Maria Kaczmarek 1989. See his reports: Old Dongola 1988-1989. House PCH.1, PAM I (1990), pp. 14-16 and Old Dongola, House PCH.1, Nubian Letters 13 (Aug. 1989), pp.1-13. 65 © PCMA 2008 – digital reprint PAM II [= Reports 1989-1990] – – – – – – – – excavations of city fortifications in the riverside part of Kom B, commenced in 1990 by W. Godlewski, who described the results in a separate report (below, p. 74); continuation of excavations of the churches DC/EDC and the monastic complex DM on Kom D, supervised by the author of this report. The site has been the object of explorations since 1986; the 1990 season was devoted to restoration and preservation activities; recording architectural remains on Kom A and C in 1989 was the responsibility of Mr. J. Dobrowolski; survey of the cemetery grounds in the eastern part of the concession (1989), combined with excavations of chosen tombs, conducted by Dr. B. Żurawski; anthropological material from these graves, as well as those from Kom D was examined on the spot by Dr. M. Kaczmarek; archaeological and architectural survey of Kom H, chosen in 1989 as a site for future excavations; Mrs. E. Parandowska was charged with some emergency conservation work there in 1990. geophysical survey in the northern part of Kom R, cemeteries TEQ and RT and on Kom H (continuation of work from the 1988 season, carried out by Dr. K. Misiewicz); improving the aerial photography method using kites and its practical implementation in recording the sites in Old Dongola effected by Dr. B. Żurawski. KOM B Salvage excavations undertaken in 1988 on a low hill to the north of the urban agglomeration of Old Dongola, on the southern outskirts of the present village of el-Ghaddar, revealed the remains of 66 PAM II [= Reports 1989-1990] © PCMA 2008 – digital reprint Fig. 2. Churches on Kom D and the remains of monastic complex DM. two superimposed churches designated as DC and EDC 3 and a monastic complex DM (Fig. 2). Archaeological explorations of both sites were terminated in 1989 and the 1990 season was devoted to preservation and restoration activities. The work, supervised by Mr. J. Dobrowolski, included reinforcing sections of preserved paving of bricks, terracotta tiles and stone slabs, marking the position of missing granite column bases and church walls, partial reconstruction of these. The area of the monastic complex suffered extensive damage, both through rebuilding and intentional levelling as well as through flooding. Ever since the Post-Classical period (12th century) this particular spot served as the source of mud and sebbakh. The devastation is so extensive that even deciphering the complete plan of the monastic complex turned out to be impossible. Some chro3 The results of these works were published preliminarily by J. Dobrowolski, Church on Kom D in Old Dongola, Nubian Letters 9 (1987), pp. 1-5, id., On the first church on Site D in Old Dongola, Nubian Letters 11 (1988), pp.1-6, id., Kościół na Komie D w Starej Dongoli, Kwartalnik Architektury i Urbanistyki 3/1988, pp. 217-231. 67 © PCMA 2008 – digital reprint PAM II [= Reports 1989-1990] nological observations were made and four habitation phases of the complex distinguished (Fig. 2). The earliest phase (7th century?) is represented by a massive mud-brick wall, one meter thick and running E-W, which confined the inhabited area of the settlement on the north. This wall was later used as a foundation for the outer northern wall of the Monastery. From the next phase (7th-8th century) only a few structures have remained, including a wall constituting the southern limit of the area of the later rooms 4 and 5, the southwest corner of the room west of locus 11, fragment of walls to the west and south of rooms 13 and 14, as well as traces of Building 18. Evidence of a monastery having been erected here comes from the third phase, i.e., the beginning of the Classical Christian Period, the 8th-9th centuries. There is the refectory room (1) preserved only on the level of the stone slab paving; it is provided with some typical round benches used during meals. Adjoining the refectory on the east there was an unit composed of a vestibule(?) – 2, 2a, presumably subdivided into two parts by a wide arcade doorway. Preceding the entrance to the refectory building was a kind of small courtyard, possibly of earlier date. West of the refectory building, which was freestanding at the time, there were two cells (rooms 3 and 9) with their own yards (loci 4-5 and 8). The construction technique here is similar to the refectory walls, i.e., mud brick on a rough stone foundation. More to the west there was another yard (loci 6-7), which was presumably domestic in function. Although most of its original surface has disappeared In consequence of later levelling works, part of its equipment has survived in the form of a covered drain built of red brick and stones, which runs across the northern wall. Roughly the same period saw further building activity – the erection of rooms 11, 12 and 13 as well as extensive rebuilding of yard 6-1-7 and rooms 14, 16 and 17, but there is no indication as to the function of this part of the complex. By the end of the Classical Christian Period the monastery seems to have been destroyed by fire. Its eastern part with the refec68 PAM II [= Reports 1989-1990] © PCMA 2008 – digital reprint tory appears to have been used for some time following this disaster, but by the 12th century the area seems definitely abandoned. There can be no doubt that the DC Church, which most probably served previously as a monastic church, continued in use for a time. The grounds north and east of the church were used as a common cemetery. Some trial pits made here in 1989 dated the levelling of the cemetery, as of the entire area of Kom D, to the Late Christian Period, prior to the 14th century. Later still the DC Church was dismantled and building material removed to other sites. KOMS A AND C In 1989 some preliminary studies were made on Kom A, the urban centre of Old Dongola encircled by massive city walls. Building A.1, which is the only building on the Kom preserved above present ground level, appears to be a residential structure dating to the Post-Christian Period. Excavations were restricted to surface clearing and some trial pits dug in spots significant for distinguishing subsequent phases in the construction of the edifice, which was apparently erected in place of an earlier Christian structure (Fig. 3). The building technique of the oldest part of this edifice corresponds to solutions applied in Dongola in the Christian period in such buildings as the Royal Audience Hall converted into a mosque and churches of Classic Christian date. The continuity of build- Fig. 3. Sketch plan of House A.1, showing particular phases of construction. 69 © PCMA 2008 – digital reprint PAM II [= Reports 1989-1990] ing traditions would suggest that the structure in question originates from the earliest period of Islamic rule in Dongola. Dobrowolski, who examined building A.1, concentrated his attention on the characteristic shape of the corner entrance towergate. Analogous, although much later, are three gates in houses located in the area of the abandoned village (Kom C) further to the south, also measured in the course of the 1989 season. The gate belonging to House C.3 (18th century?) possessed a kind of machicolation – a round opening in the roof above the wide passage permitting an attack from above on the enemy attempting to force the entrance. In the western part of the abandoned village a large storied house C.1 (23 x 25m) was examined and measured. It belonged to a known and meritorious Dongolan family, the Suar ad-Dahab. Investigation of the building technique, spatial layout and interior equipment partly preserved in House C.1 brought new and interesting material for the study of the development of Post-Christian residential architecture in Nubia. SURVEY OF THE CEMETERY AREAS The survey covered three cemeteries: TEQ east of the northernmost part of Dongola town, TWH west of Kom H and TSJ lying close to the North Church (NC). The main goal was to fill out the chronological sequence of grave types known from Dongola, suggested by the earlier examinations of burial sites TEE, TNH TDC and RT (Fig. 1). Some of the graves excavated in the 1989 season deserve special attention. TEQ.1.7 is a more than 2 m deep slot-like trench roofed over with mud brick, containing a body shrouded in linen. Remnants of a rectangular mud-brick superstructure (mastaba) were still distinguishable on the surface. The dating of the grave to the Early Christian Period was confirmed by C 14 dating (second half of the 7th century). 70 PAM II [= Reports 1989-1990] © PCMA 2008 – digital reprint Fig. 4. Kom H and Cemetery TWH. Preliminary inventory of extant structures. Classic Christian tombs (burial chamber built of mud brick and barrel-vaulted, accessible by a small entrance from the shaft on its western side) are best represented by Tombs TSJ-1, -24 and TEQ.1.3. The latter served as a grave for ten individuals (men, women and children), buried most probably at one time. A sand dune accumulated on the western slope of Kom H has protected the tomb superstructures belonging to the THW cemetery. Twelve tombs were found here. They were made up of both fired and dried mud brick, coated with hard lime-and-gravel plaster. Originally, all twelve had superstructures in the form of mastabas shaped as crosses. All had lamp-boxes on the western side (Fig. 4). Curiously enough, bodies were buried in simple rectangular 4 The C14 sample gave a date of AD 860±40 years. 71 © PCMA 2008 – digital reprint PAM II [= Reports 1989-1990] trenches covered with sand, not in burial chambers of any kind. The excavated section of this cemetery can be dated to the first half of the 10th century. 5 KOM H In order to prepare the site for further methodical excavations, in 1989 we conducted a survey of Kom H located in the eastern part of the concession area. This large kom measuring some 100 m in diameter and encircled by cemeteries appears to form an isolated settlement, presumably a monastic complex fortified with a mudbrick enclosure wall, the whole built not later than the Classical Christian Period (Fig. 4). A detailed inventory of remains visible on the surface suggests that some of the buildings are preserved up to the vaults of the upper storeys, while red brick structures seem to have been dismantled down to the foundations. Murals and numerous graffiti of the Late Christian Period were observed on the walls of Unit S. Mrs. E. Parandowska, chief conservator of the National Museum in Warsaw, carried out emergency conservation work on this material in order to stop further decay of the painted and inscribed plaster.6 GEOPHYSICAL SURVEY Dr. K. Misiewicz from the Institute of History of Material Culture of the Polish Academy of Science continued in 1989 a geophysical survey of the area of our concession commenced in 1988. The main objective was to locate kilns on sites R and S alongside the Nile, presently covered by a sand-dune strip. A total of about 27,000 square meters (that is, about 30,000 measuring points) was survey5 A C14 sample from Tomb TWH.4 gave a date of AD 930±40 years. 6 The work and the results will be discussed in a separate report following the next campaign, in which we plan to subject Building S to methodical archaeological investigation. 72 PAM II [= Reports 1989-1990] © PCMA 2008 – digital reprint ed using the magnetic method. Magnetic anomalies were measured with two proton magnetometers PMP-5 of high sensitivity with a regular 1 m grid. Extant anomalies of the magnetic field suggest the presence of various objects underground, objects which when made of certain materials would disturb the natural magnetic value of the soil. It has been observed that the remanent effect resulting from burning (red-brick, slag, ashes) gives a very strong magnetic reading between 40 and 150 gamma. This phenomenon permitted nine clusters of kilns to be located very precisely regardless of their being 0.5 to 1.8 m below the surf ace. These results were confirmed archaeologically on site R.6. Here a trial pit 2 x 6 m was dug in a dune where no traces of any kind of human activity were to be observed on the surface. Upon removing 1.6 m of overlying dune, the tops of the walls of two kilns of red-brick appeared, their inner surface covered heavily with slag. Since they are of a different construction than the already known Dongolan kilns nearby, they appear to have had a different function. The site can be dated to the later Classic Christian period (10th-11th century). On the site of cemetery TEQ the magnetic method was instrumental in establishing individual concentrations of tombs and the extent of the burial ground. The results were extremely helpful in this case for the surface here has suffered extensive erosion and no tomb superstructures are to be seen. Some clearing was done on the site of the Rock Tombs (RT) in the north end of the concession area in order to verify the observations made on the basis of a geophysical survey conducted in 1988. It is certain at present that no other rock tombs existed here beside two imposing monuments RT.1 and 2, excavated in 1971 and 1981 respectively and dated back to the very beginning of the Christian Period in Dongola.7 7 S. Jakobielski, Polish excavations at Old Dongola 1978/79-1982 [in:] M. Krause (ed.), Nubische Studien, Mainz 1986, p. 308ff. 73 © PCMA 2008 – digital reprint PAM II [= Reports 1989-1990] THE OLD DONGOLA FORTIFICATIONS Włodzimierz Godlewski Preliminary investigations of the city fortifications located on the northwestern slope of Kom A were the prime objective of the mission during the 1990 season in Old Dongola.1 Relics of mud-brick walls have been observed on the surface around the perimeter of Kom A in an archaeological context suggestive of the late medieval period, most probably following the fall of the Nubian kingdom. In the course of removing deposits accumulated on the slope of Kom A it was found that post-Christian layers lay immediately on top of the remains of structures from the 6th-7th centuries. It would appear then that the structures in this area underwent total destruction sometime in the 13th-14th century with some of them being dismantled down to bedrock and the entire slope being exposed to erosion by wind and precipitation. A massive mud-brick wall 3.1 m wide was found to run a straight course from the northwest to the southeast for over 32 m. The northern end of it adjoined a rock outcrop now largely eroded away, which must have been significantly higher at the time when the defenses were built and must have been of strategic importance as the northernmost point of Kom A. At the other end, the wall, which is preserved only in foundations not exceeding 0.7 m in height at any point and often destroyed down to bedrock, disappears entirely and seems to have no relation with the stone wail built nearby. Excavations cleared a roughly triangular area west of the mudbrick wall which clearly served as its northeastern boundary. The structures identified there all touch upon the mud-brick wall. They 1 The excavations on Kom A were headed by the present writer with the help of Mr. Jarosław Dobrowolski (architect). For the results, see also W. Godlewski, The fortifications of Old Dongola. Report on the 1990 season, Archéologie du Nil Moyen, 5, 1991, in print. 74 PAM II [= Reports 1989-1990] © PCMA 2008 – digital reprint seem to have been constructed simultaneously with the wall and appear to be connected with it in a functional, if not structural sense. They obviously constitute part of an architectural project, strategic in nature. It consists of a network of mud-brick walls forming units which were subsequently filled in with debris in order to create a flat building site. Both the rampart wall and the platform structure are dated mainly on the grounds of pottery recovered during the excavations. The entire assemblage belongs without exception to the early period and should be dated in all probability to the 6th-7th century AD. East of the northern end of the rampart wall, just off the rocky outcrop constituting its termination, two circular mud-brick structures filled with gravel were cleared. Both were on bedrock level, at a distance of 1 m from one another. The southern structure was round, roughly 0.80-0.95 m in diameter, with a slightly flattened eastern side. It consisted of upended mud-bricks laid directly upon bedrock; the surviving remains are some 0.4 m high. On the east there was an opening (9 x 14.5 cm.). Outside it, the rock was hollowed out in a 0.4 m long, narrow channel, the width of which matched the opening. This furnace was found filled with alternating layers of iron ore and charcoal. It would appear that the smelting process in this structure had been abandoned at an early stage. A C14 dating of the charcoal -from the southern furnace narrows the date to the very beginning of the 6th century (AD 510).2 In this respect the Dongolan structures, only the second of their type to be known from the area of the Sudan, are later than the furnaces uncovered in Meroe. 3 2 The dating was carried out by Prof. M. Pazdur from the C14 Laboratory of Gliwice Polytechnic (Gd.-5753). 3 P.L. Shinnie, F.J. Kense, Meroitic iron working, Meroitica 6, 1982, pp. 17-28; R. Tylecote, Metal working at Meroe, Sudan, Meroitica 6, 1982, pp. 29-42. 75 © PCMA 2008 – digital reprint PAM II [= Reports 1989-1990] The rocky outcrop which terminated the northern end of the wide mud-brick wall appears to have had a stone structure reinforcing it on the east and south. The structure or rather its foundations were constructed without the use of mortar, directly upon the sloping rock surface. The material used was a rosy granite and included numerous shattered column drum fragments, capitals and bases. Only the exterior face of the wall was laid in straight courses. The mentioned granite foundation structure obviously served as the foundation or lower part of a spacious tower which must have once stood upon the rocky eminence, doubtless forming part of the fortification system mentioned earlier. The entire assemblage of architectural decoration fragments recovered from the relics of the granite foundation structure appears to be homogenous as far as material is concerned, as well as in style and chronology. One may easily conclude that they all originated from some monumental building which suffered a violent and presumably intentional destruction. Considering the evidence of church building in Dongola presently available, one may suppose that the architectural granite fragments came from a basilica destroyed by Arab troops during their raid on Dongola in AD 652.4 This may have been the Church of the Stone Pavement I.5 In the eastern part of the excavated sector, where surface relics of mud-brick walls were considered as post-Christian at first, investigations reached bedrock, revealing structures belonging to four different stages in the construction of city fortifications. The limited area covered by the excavations precludes as yet a detailed characteristic of particular levels. Undoubtedly the oldest structure is a massive tower which must have constituted a corner of the extensive fortification system of 4 W. Godlewski, The Cruciform Church site in Old Dongola. Sequence of buildings from the 6th to the 18th century, Nubica I/II, 1990, pp. 523-524. 5 Ibidem, pp. 519-523. 76 PAM II [= Reports 1989-1990] © PCMA 2008 – digital reprint Dongola. Only the western facade of this structure has been cleared so far and even that not along its entire length, so not much may be said of its plan. It would appear that the tower was an elongated structure with a rounded northern end, which projected well beyond the line of the walls. It cannot be excluded that it had once been part of a gateway. The tower was founded on bedrock. Mud brick was used to construct the core, while the exterior was faced with large irregular stone blocks with smaller rock chips filling the spaces in between. The tower was evidently a structure of considerable height for the preserved relics of the mud-brick core rise to a height of 6.75 m, while the stone facing, which was doubtless the object of plunder in modern times, reaches 5.1 m at the most. 77 © PCMA 2008 – digital reprint PAM II [= Reports 1989-1990] NEA PAPHOS 1990 REPORT Wiktor A. Daszewski Another season of excavations in Nea Paphos by a mission sponsored by the PCMA started on 3 September and ended on 5 October 1990. Work progressed in all three of the sectors covered by the mission's investigations, that is, in the so-called Villa of Theseus, the so-called House of Aion and the Late Hellenistic house (Fig. 1).1 In the area of the Villa of Theseus, that is, the Late Roman residential structure, investigations focused on the northern end of the eastern wing. In this part of the villa Roman walls have all but totally disappeared making the plan of the residence difficult to follow. On the other hand the earlier structures are quite clearly visible. Large segments of walls cleared in previous seasons were found grouped on either side of a longitudinal street of Hellenistic date. They formed rows of rectangular rooms, the axes of which ran at right angles to the axis of the street. Particular rooms were equipped with drainage channels which emptied into the main sewer running down the street below its surface. At the time that the Villa of Theseus was built this street had disappeared completely under new constructions and the earlier walls were used in part as foundations for partition walls inside the villa. Some of the rooms, both 1 The mission comprised: Prof. Wiktor A. Daszewski, Prof. Zofia Sztetyłło, Dr. Evdoxia Papuci-Władyka, Mr. Henryk Meyza, archaeologists, Dr. Stanisław Medeksza, architect and Mr. Bogusław Okupny, photographer. Participating on a volunteer basis were students from the Institute of Archaeology of Warsaw University and the Academy of Fine Arts in Warsaw, as well as foreign students from Cyprus, the United States and Great Britain. The mission wishes to express heartfelt thanks to the authorities of the Antiquities Department of Cyprus for their friendly aid in the course of the work, as well as for help in the reconstruction and anastylosis of the monuments. We are especially grateful to Dr. A. Papageorghiou, Director of the Antiquities Department, and to Dr. D. Michaelides, the archaeologist in charge of the Paphos district. 78 PAM II [= Reports 1989-1990] © PCMA 2008 – digital reprint Fig. 1. Plan of the Polish excavations. VT – Villa of Theseus, HA – House of Aion, HH – Hellenistic House. on the eastern and western side of the street, still had great pithoi sunk in the ground. The content of two such pithoi on the western side of the street was investigated in a previous season, but did not supply any conclusive dating evidence. Investigations were continued this year, the object being a large terracotta container sunk into the ground near the outer wall of a room on the eastern side of the street. The results were very interesting. It turned out that the container was a kind of cylinder without bottom. Its flat floor, some 0.50 m below the preserved top of the container, was made of potsherds set flat in a layer of clay and bearing evidence of burning. Inside the container, in a fill of rather homogenous grey silty soil mixed with ashes there was a pottery deposit including plates, cups, bowls, kitchen pots, fragments of Megarian bowls, a stamped Rhodian amphora and three terracotta oil lamps. The entire group can be dated to the period from the end of the 3rd to the early 1st century BC. The 79 © PCMA 2008 – digital reprint PAM II [= Reports 1989-1990] stamp on the amphora belongs to Diodotos, dated to 210-175 BC. By the same token there can be no further doubt concerning the dating of the dwellings on either side of the street the later Hellenistic period. Within the House of Aion work proceeded to the northeast of rooms discovered so far, particularly in the vicinity of rooms 7 and 8. The debris inside room 7 included many bits of painted plasters from the walls and ceiling. The decoration was figural, vegetal and geometric; the figural parts were of outstanding quality. Substantial fragments of the decoration were pieced together from preserved bits in the course of the season; an example of this work is a male figure in the dress of a kitharodos, perhaps a representation of Apollo (Fig. 2). Excavations were carried out to the north of this room, on the other side of its northern wall. Two other rooms were found to be adjacent to it. Their floors were on a much higher level and they were separated by a N-S wall without foundation. They can be dated quite precisely by numerous coins and pottery of the second and third quarters of the 4th century AD. In the northwestern part a stratigraphical trench reached bedrock at little over 2.10 m below the preserved tops of the walls. It was found that the earliest habitations belonged to the Ptolemaic period, presumably the beginning of the 2nd century BC. These Hellenistic structures subsequently served as foundation for the northern wall of room 7. Nevertheless, the area must have been occupied at an even earlier date, even before anything was built here. Sherds lying immediately upon bedrock are proof of it: fragments of a black-figured Attic vase and a hydria of the White Painted VII type and perhaps also White Painted II. The most interesting find from the trench was a terracotta oil lamp discus from the middle of the 1st century AD, decorated with a very rare representation of the nymph Amaltheia nursing a small Zeus. The trench east of room 8 of the House of Aion uncovered a large part of yet another room lying on a level some 0.50 m higher 80 PAM II [= Reports 1989-1990] © PCMA 2008 – digital reprint Fig. 2. House of Aion, Room 7. Reconstructed fragment of painted wall decoration. 81 © PCMA 2008 – digital reprint PAM II [= Reports 1989-1990] Fig. 3. Plan of the Hellenistic House. than the rest of the house. This newly discovered room had a plain mosaic floor, similar to that in room 8. It was made up of large tesserae measuring about 2x2 and 2.5 x 2.5 cm. The greatest effort was concentrated upon continued excavations of the so-called Hellenistic House (Fig. 3), which lies immediately to the south of the Villa of Theseus. The building represents a type of house with a central peristyle court around which the main reception and household rooms were planned together with presumed lodgings. The northern wing of this structure was destroyed completely when the later villa was constructed, the latter being located on a slightly lower level. Owing to the insignificant differences in the original elevation of the area and the leveling work that took place, the stratigraphic sequence of archaeological layers was turned upside down with Hellenistic layers connected with the construction and occupation of the Hellenistic house being found above the Late Roman levels in the Villa of Theseus. The previous seasons saw the clearing of the entire eastern wing of the peristyle (that is, its preserved part) and some of the rooms adjoining it. The rooms in the northern part of the wing were found to differ from those in the other end. The northern rooms all had traces of fine mural decoration consisting of painting in the so-called structural style, which corresponds to the First Pompeian Style. The rooms in the southeastern corner of the portico were of 82 PAM II [= Reports 1989-1990] © PCMA 2008 – digital reprint a totally different character, having some connection with the use of water as the waterproof floor and drain in one of the rooms would suggest. The room next to it on the east was a latrine of fair size indicating that it was intended for a number of users at a time. The latest excavations continued the investigations of the western wing of the peristyle begun last year. What became apparent was the fact that the Ionian column portico of the eastern and southern wings was completely different, both in size and nature, from the western one. New fragments of columns and pillars were discovered there. The capitals represented the specific variety of the Corinthian order that is typical of regions which came under Ptolemaic influence. One such capital of large size (top span is 1 m) was discovered last year and this season supplied new fragments. The stylobate of the western colonnade is wider than that of the eastern one and draws from the size of the columns themselves. A sounding made in the middle part of the western stylobate brought some surprising results. It demonstrated that the stylobate did not have any foundations, at least in this section. Opening off the western portico was a large room (no. 10) more of which was uncovered this year. Us dimensions 11 x 6.50 m show it to have been a reception hall with an axis that is perpendicular to the portico. A wide entrance with a span of 3.10 m opens out on the portico. The mosaic floor in this room is made up of irregular chips of a creamish-white limestone and of black basalt(?). The decorative motif is extremely simple: a thin, about 10 cm wide black band running parallel to the room's walls on a uniform whitish ground. The materials used (irregular bits of stone) as well as the technique, coloring and design place this mosaic among Hellenistic floors known from other regions of the Mediterranean, especially from the nearby Cyrenaica and from Greece proper. The Paphos mosaic should be considered a work of the 2nd century BC. Traces of stucco relief pillars are visible on the preserved parts of the walls of room no. 10, while the fill yielded many fragments 83 © PCMA 2008 – digital reprint PAM II [= Reports 1989-1990] of painted plaster. The designs included red and black bands, green and yellow lines. A much richer wall decoration, and much better preserved as well, was found in room no. 11. Three subsequent layers of plastering were recorded on its walls, the paintings being done in the al secco technique. On the two earlier layers of plaster the decoration took on form of large rectangular fields of uniform color, alternately red and black. Their sides as well as the corner of the room were emphasized by vertical bands of either red or green color. White and red bands separated the fields from each other. The latest layer of plastering was a whitish plaster with a red line running about 0.30 m above floor level. The ceramic material accompanying the relics of architecture indicated that the building, constructed in the later Hellenistic period, continued in use right down to perhaps the second half of the 1st century AD, when it was destroyed in an earthquake. The ruins served subsequently as a ready source of building material, an activity which apparently continued up to the middle of the 2nd century AD, as indicated by ceramic evidence from the fill of the dismantled walls. It was at this time that the Villa of Theseus started taking shape; many fragments of architectural details found previously reused in the walls of the Roman residence can now be surely traced to this Hellenistic House. 84 PAM II [= Reports 1989-1990] © PCMA 2008 – digital reprint PALMYRA Michał Gawlikowski This was the third consecutive season spent in the downtown area of ancient Palmyra, to the north of the Great Colonnade and halfway between the Tetrapylon and the Funerary Temple. We continued work in two neighboring blocks, one of them residential and the other public in character, the latter being occupied by a secular basilica converted into a church. Besides, we have started comprehensive investigations in the Arab Castle high on the hill northwest of the ruins.1 We decided to continue this year the excavations of the house partly cleared in 1988 (Fig. 1).2 The entrance opened to the east into the street giving access on its other side further north to the Byzantine churches, while the street on the western side of the house passed close to the Great Colonnade, along the basilica excavated last year. In 1988 we were able to clear about 400 m2 of the surface of the house, i.e., the northern and western wings opening into the centre courtyard. Upon clearing the other dependencies of this courtyard, it was found that originally it had been linked to the adjoining peristyle court the columns of which still stand further 1 The mission included, beside the present writer, Dr. Maria Krogulska, Messrs Grzegorz Majcherek and Janusz Byliński, all archaeologists, and two architects, Messrs Marek Barański and Jan Kempa. We have met, while in Palmyra, with the most efficient and generous help of our friends of many years, whose collaboration has proved as precious as always. It is for me a pleasure to acknowledge their part in the success of our mission. Mr. Khaled As'ad, Director of Antiquities in Palmyra, was with us as often as he could afford and let us feel his support at all times; his associates Ahmad and Ali Taha assisted us daily in the field, sharing our labors and smoothing our way throughout. Needless to say, there is no forgetting the friendly support of Dr. Ali Abu Assaf, Director General of Antiquities, and of Dr. Adnan Bounni, Director of Excavations, who have made our project possible. 2 See PAM I, 1990, p. 37-40. 85 © PCMA 2008 – digital reprint PAM II [= Reports 1989-1990] Fig. l. Plan of the house after the 1990 season. Younger walls are shaded. to the south; once this was cleared too, we found that several passages led into yet another colonnaded courtyard. One should count the total surface of the house well over 1200 m2, of which nearly 900 m2 have been excavated already. It was established, as a result of the former season that this house was built in the later half of the 2nd century AD, and was used without interruption, but not unchanged, down to the 8th or even early 9th century, when it was finally deserted. The stucco decoration dating back to the foundation of the house collapsed only then, covering with its debris an accumulation layer including pottery, glass and coins, all confirming a late date for the abandonment. We can add now more evidence to the same effect. The original house, no doubt the residence of a very affluent family, underwent over time a series of partitions, apparently between relatives, as the 86 PAM II [= Reports 1989-1990] © PCMA 2008 – digital reprint resulting units continued to share a common courtyard and an entrance from the street. One of the latest occupants signed his name on a floor in Kufic script as (probably) Hassan. This determination of the continuity of urban life in this part of Palmyra appears to be the most important result of our research so far. The northern courtyard (loc. 13) measured originally c. 10 by 14 m. It opened to the north and to the west into separate porticoes of three columns each, giving access to several rooms (loci 1-11). The details of this arrangement were described in my first report. This time we excavated the court itself and its dependencies on the eastern and southern sides. Two columns on the northern side were found fallen and one of them was re-erected. To the south, there was a single room 5 by 6 m (loc. 12), accessible from the courtyard and from the adjoining western porch (loc. 10) until the latter passage was blocked. More Islamic evidence were found in the fill close to the floor of this room, while much higher there were elements of two stucco cornices once adorning the walls at two different levels. The original courtyard had no other dependencies and extended east right to the street wall. The entrance was through a narrow corridor (loc. 16), in line with room 12 just described. Immediately to the right and left of the street door, there were two passages: one narrow in the corner of the courtyard, probably meant to serve a future eastern wing, the other, 1.40 m wide, leading south into the main part of the house through room 17. The eastern wing was not added until the Islamic period. It consists of two rooms set higher than the original floors, but even with two platforms marking this later level in the courtyard. Room 15 had two doors from inside and a door opening into the street, while the original entrance was blocked at its inner end and transformed into a shop. The new entrance was abandoned in its turn and still another opened in the adjoining room 14. A shallow basin appearing on the present surface marks this last access to the house, above 87 © PCMA 2008 – digital reprint PAM II [= Reports 1989-1990] a typically Ummayad red-painted plaster floor. The former vestibule, room 15, was then transformed into a stable, as suggested by a manger near the blocked street door. In the middle of the courtyard (loc. 13), a sewage collector was found, cut into the rock and corbel-vaulted, now filled with clay. It was joined by an open drain coming from the northeastern corner of the courtyard, where there was a well 4.50 m deep, set into a platform which had blocked one of the intercolumnia of the northern portico. This well, as the one of the same depth within the western portico, collected underground water at a surprisingly high water table level. The courtyard neighboring to the south (loc. 22) commanded a reception suite. It was entirely paved, though not at one time, and four of its columns were still standing on its eastern side, bearing the architraves. The restoration team of the Palmyra Museum, now ably headed by Mr. Ali Saleh Taha, has reinforced them, replaced the capital on the fourth standing column, and erected the one surviving column on the southern side. The peristyle is irregular, due to the oblique course of the wall dividing this court from the dependencies of the one described above. It is not clear why this aberration arose, leading the architect to leave unfinished the northernmost column engaged in this wall which was built around it. Close by, there is another well, 8 m deep, going certainly back to the 2nd century; it is provided with a monolithic basin above the pavement. There are two reception rooms on the western side of this courtyard. The huge room 20, measuring 8.00 by 5.50 m, opened under the colonnade and was provided with niches and symmetrical doors right and left, leading to the neighboring rooms. The stucco fragments found there are clearly the most elaborate of the whole house and include many interesting motifs such as olive trees, vine scrolls, human heads (destroyed by iconoclasts) and parts of armor (Fig. 2). The room was used into Islamic times, as the rest of the 88 PAM II [= Reports 1989-1990] © PCMA 2008 – digital reprint Fig. 2. Stucco fragment from Room 20. house, and coin finds tend to place the blocking of the side doors in the beginning of the 7th century. The parallel room 19 is smaller, but it was preceded by a porch of two columns, later walled in front and equipped with a tannur. The eastern wing of this part of the house still awaits excavation. We also conducted this year a limited program of stratigraphic investigation behind the apse of the basilica excavated in 1989. The results are interesting: it appears that the apse was not part of the original 2nd century building, as was thought earlier because of the style of the decoration of its arch, but that it was built especially for the church, while borrowing the voussoirs and the imposts of the arch from another monument. Pottery evidence points to the 4th century as the date of the church and this is confirmed by a coin of Constantius II as Caesar (324-337), found on the floor upon which the apse was erected. For this purpose, the shops lining the street alongside the original basilica were dismantled, their doors blocked, and the whole sector transformed first into two, then into one big room accessible from the porch in front of the church. The dependency of the church was destroyed and filled up in the late 6th or early 7th century, but the aisle itself survived until the 8th century, as described in my 1989 report. 89 © PCMA 2008 – digital reprint PAM II [= Reports 1989-1990] It is very tempting to link the transformation of an earlier hall of probably commercial use into a church with the inscription3 relating the restoration in 328 of the portico in front of the building by a city official called Flavius Diogenes. If this association is correct, this text would mark the installation (impossible before 324) of the first church in Palmyra, which would have been the see of Bishop Marinus mentioned as one of the Fathers present at the Council of Nicaea. 3 J. Cantineau, Inventaire des inscriptions de Palmyre, vol. III, 27. 90 PAM II [= Reports 1989-1990] © PCMA 2008 – digital reprint THE ARAB CASTLE IN PALMYRA Janusz Byliński The investigations of the Arab Castle began this year1 after a preliminary prospection in 1989 which yielded encouraging results. The season was devoted to surveying and measuring the castle in order to prepare detailed plans and sections. Some limited clearing work was carried out, recovering evidence useful in establishing the date of the castle. The castle was built on a rocky hill overlooking ancient Palmyra from the northwest. The building stands on an outcrop of rock at the top of the hill, falling steeply both to the southwest and southeast. The location of the stronghold on this irregular surface inevitably determined its plan and internal arrangement. The whole structure is surrounded by a moat, which was cut in the rock and made deeper by an embankment supported by a retaining wall. The construction of the castle is commonly ascribed to the Druze emir Fakhr ed-Din Ibn Ma'an (1595-1634), but after last year's prospection it became obvious that at least two phases of construction could be discerned in the building, and that the pottery sherds found in several spots inside date back to the medieval period (13th14th century). Further research this year fully confirmed this assumption. The castle underwent a number of extensions, reconstructions and repairs while in use. The earliest stronghold was roughly triangular in shape and remains the core of the building as it is seen today. New segments were added at least twice, and some of the earliest towers were engulfed. Doors and windows were closed by new walls, passages closed, and some of the machicolations, stripped of their original function, became level with new courts. In some 1 The group working in the castle included Messrs Janusz Byliński, arabist and archaeologist, Jan Kempa, architect, and Ahmad Tana representing the Palmyra Museum. 91 © PCMA 2008 – digital reprint PAM II [= Reports 1989-1990] towers new doors were pierced to facilitate communication with the new segments. Although, there is nothing that could help say how much time passed between these consecutive phases of construction, two different kinds of fabric can be easily distinguished: irregularly hewn stones laid in courses leveled with mortar (almost exclusive in the earliest phase), and regular blocks (at least some of them may have come from dismantled tombs or city walls), prevalent in the later phase. The exact delineation of the different phases has been left for the next season when appropriate soundings will be made. Pottery samples were collected both inside the castle and on the slopes of the hill. Both sets of finds are homogenous, containing sherds of glazed vessels characteristic of the late 13th and 14th century and some pieces that might date back to the early 13th. Unfortunately, the fragments collected are too small to allow a full assessment of shapes and designs. Nonetheless, the variety of types is striking, although all of them seem to be Syrian products and no imported ware has yet been found. On the other hand, a number of clay pipes found all in one room of the castle are evident proof of occupation in the Ottoman period. Measurements taken during this season are sufficient to draw a plan of the castle's perimeter and of the rooms and halls on the two floors of the northern part. Levels have been established and marked in the whole structure. Some architectural elements were cleared as, for instance, the flight of stairs leading to the central segment of the southern part and an outer draining channel parallel to the southeastern wall in the corridor leading to a large hall in the northwestern corner of the castle. It is this hall that was chosen for a fuller excavation this season, mainly in view of the installations it contained, which were likely to determine its function and its relations to other parts of the castle. The floor was covered with a thick layer of dust, rubbish and animal dung, only an outline of two circular structures being vis92 PAM II [= Reports 1989-1990] © PCMA 2008 – digital reprint ible on the surface. At first glance they seemed to be related to the processing and storage of food. Eventually, we uncovered a circular stone enclosure with a grooved stone in the center, which in all probability was a grain mill; it is in the southern corner of the hall. In the eastern corner we cleared the fill from a sunken circular silo and from another half-circular silo contiguous to it. Both containers were set within a roughly quadrangular structure made of stone, red baked bricks, mud-bricks and some reused parts of basalt grinding stones. The structure was partly destroyed. Under the highest floor there is another, on which the mill was built, and which was laid on top of a fill of crushed mortar, pebbles and rubbish (including pottery sherds) used for leveling the irregularly sloping bedrock. The two containers just mentioned were built lowest down, on an irregularly surfaced foundation made of several layers of lime mortar. This foundation is visible in the corridor adjacent to the mill, the floor of which is on the lower level. The layer of animal dung and rubbish above the floor associated with the mill contained pottery sherds, clay pipes, fragments of cloth and leather shoes, wooden spoons and numerous fragments of broken grinding stones. Many sherds certainly belong to the 13th-14th century, but a fragment of an underglaze-painted, frit-body vessel of the Raqqa type may date back to the early 13th century. Excavations were conducted also in the walled terrace of tower VII (part of the earliest structure, now inside the castle), where an arched imprint in the wall plaster suggested a possible place of worship. The dust and rubble removed, it became clear that the terrace had been resurfaced after blocking the stairca- 93 © PCMA 2008 – digital reprint PAM II [= Reports 1989-1990] THE THIRD SEASON OF EXCAVATIONS IN NORTHEAST SYRIA, 1990 Piotr Bieliński In 1988 a team from the PCMA started excavating two sites situated within the Hassake Eastern Dam Reservoir Area. One of these was Tell Djassa al-Gharbi, the other Tell Abu- Hafur.1 Our explorations in this region were part of the International Salvage Program of the Hasake Dams Area, which is organized and supervised by the Syrian Directorate General of Antiquities and Museums. In 1990 the third season of these excavations took place. We wish to express here our gratitude to the Directorate General of Antiquities for the help and constant support offered us throughout the campaign. In particular we owe sincere thanks to Dr. Adnan Bounni, Director of Excavations, Mr. Jean Simon Lazar, Director of Antiquities in Hasake, Mr. Hussein Youssuf who was working with us as a representative of the Directorate, and finally Dr. Antoine Suleiman. The campaign started on September 22 and lasted until November 6. This was our last season of digging since the water from the Khabur began filling the Hasake Eastern Dam basin on the day we finished excavations and by now both sites have been covered by the waters of the new lake. The mission was composed of four archaeologists headed by Dr. Maria Krogulska at the beginning of the season and later on by the present writer. 2 TELL DJASSA AL-GHARBI The investigations on this site were carried out in four separate sectors. Two of these: the so called step trench (squares A-J) and 1 2 See PAM 1, 1990, pp. 17-25. The members of the staff were: Mrs. Dorota Bielińska and Messrs Andrzej Reiche, Rafał Koliński and Tomasz Skierniewski. For the last week we were aided by Dr. Franciszek M. Stępniowski who was a member of this team in the 1988 season as well. 94 PAM II [= Reports 1989-1990] © PCMA 2008 – digital reprint - Fig. 1. Tell Djassa el-Gharbi. Plan of the site. trench K-L (squares K-L), situated on the northern slope of the tell, were already the object of explorations during previous seasons. (Fig. 1) The two new trenches (squares M-N-0 and sounding P) were opened in the eastern part of the site. Within the long step trench (squares A-J) this year's main effort centered on further exploring layer II. In this layer, which is of Early Dynastic III date, we continued the exploration of some mudbrick structures discovered last year. The eastern architectural unit, called Building I, was the object of our particular interest. (Fig. 2) There, inside a large room with two false arches (locus 5), some limited soundings were dug. It was established that originally this 95 © PCMA 2008 – digital reprint PAM II [= Reports 1989-1990] room was nearly 4 m high and that it was rebuilt at least once. A thick deposit (0.80 m) of floors and accumulated debris suggests that this room was in use over a long period of time. Another room, most probably also belonging to the same Building I, was excavated this year south of locus 5. This new room, labeled locus 6, was very similar to the adjoining locus 5. It was of similar dimensions and had two intersecting arches arranged similarly. It seems now quite evident that the part of Building I described above must be at least partly contemporary with the huge platform of mud brick exposed already in 1968 on the northern slope of the tell. The results of a careful examination which took place during this season suggest that this large platform, over 10 m wide, was not a homogenous structure, but that it was composed of at least three parts. (Fig. 2) It seems also that the innermost part of the platform may be slightly younger than the rest of it. The inner face of the platform was found this year near the northern limits of trench K-L and another fragment of this strange structure was unearthed in a small sounding on the eastern slope of the mound (trench P). The remains discovered there have confirmed the supposition that the platform protected not only the northern slope of the tell, but the eastern one as well. It seems that this structure as a whole had a rather irregular character and was composed of several separate segments made of mud bricks and loose clay. During this season layer I was exposed in the small trench K-L, running parallel to the main step trench. (Fig. 1) Fragments of two large rooms with clay benches, storage jars and fireplaces were found there. At least three main phases of development could be observed here, closely corresponding to the earlier stages of layer I as distinguished within the main step trench. These stages are representative of the late phase of the Early Dynastic III period. Structures belonging to the same stages of layer I were discovered also in the new trench M-N-0, located on the eastern slope of the tell (Fig. 3). They consisted of the remains of at least three separate dwelling 96 PAM II [= Reports 1989-1990] © PCMA 2008 – digital reprint © PCMA 2008 – digital reprint 97 PAM II [= Reports 1989-1990] Fig. 2. Step trench (squares A-E). Constructions of Layer II. Fig. 3. Layer I house found in squares M-0. units. The largest fragment excavated within this trench measured c. 7 x 5m and consisted of four rooms. The rooms contained numerous domestic installations such as: fireplaces, querns, mortars, a storage jar, tannur and small basin. The exterior walls of this house were constructed of two rows of large mud bricks, while the partition walls were much thinner. The walls (preserved to a height of c. 1.20 m) were reinforced with internal buttresses and plastered with a thick layer of white plaster. Of the other dwellings not much can be said as only small sections were exposed within the boundaries of this sector. In the same M-N-0 trench, just below the lowest floors of houses belonging to layer I we found the top of layer II with fragments of the upper parts of an arched mud-brick structure. It closely resembled buildings discovered in layer II of the step trench A-J. Last year Layer III was found only in the main step trench, where its top was reached immediately beneath the foundations of arched structures belonging to layer II. Our information about this 98 PAM II [= Reports 1989-1990] © PCMA 2008 – digital reprint layer and its dating in particular was fairly sketchy. To gain more data, during the 1990 season we opened a larger sounding within the main step trench. The architectural remains of layer III exposed there consisted of fragments of two rooms with numerous adjacent floors. What was even more important was the pottery found in these rooms for it helped solve the question of the dating of layer III. Like both the lower levels of Layer I and the whole of layer II, also layer III turned out to be of Early Dynastic III date. We were not so lucky with the dating of the lowermost occupational layer found on Tell Djassa al-Gharbi, i.e., Layer IV. In spite of our efforts (sounding in the northern part of step trench A-J), this layer did not yield any diagnostic potsherds which would have allowed us to set a more precise date than the Early Dynastic period in general. This last season of excavations was important for a more complete understanding of the character of the 3rd millennium BC settlements on Tell Djassa al-Gharbi. The remains discovered in layer II can be considered as particularly interesting. Such elements as the substantial platform and buildings composed of tightly packed arched structures are considerably different from the architecture of other later 3rd millennium BC settlements excavated in this part of Syria. The function of the platform, as well as that of the arched structures, will certainly need further clarification. It is to be noted, however, that similar arched structures have been found on Tell Abu Hadjarat within the same Hassake Eastern Dam Basin3 and on Tell 'Atij where a large platform was also uncovered. 4 3 Personal communication by Dr. Antoine Suleiman. 4 ` campagne de fouilles a` Tell M. Fortin, Rapport préliminaire sur la premiere 'Atij, Syria 65 (1988), pp. 139-71; idem, Rapport préliminaire sur la seconde campagne de fouilles a``Tell 'Atij, Syria 67 (1990), pp. 219-250. 99 © PCMA 2008 – digital reprint PAM II [= Reports 1989-1990] – TELL ABU HAFUR On Tell Abu- Hafur the explorations were carried out in all three sectors begun in 1989. The first, designated Area A, is situated on the southeastern slope of the main mound. During this season our investigations within Area A were rather limited. In this sector, where last year a habitation quarter composed of at least five separate dwelling houses was unearthed, this year we made only some minor soundings, verifying some architectural elements and taking some additional measurements. In most parts of this sector we succeeded in uncovering the architectural remains of layer A, which like the three upper layers discovered here before, also represents the Early Dynastic III period. Within Area A we also excavated a fragment of a small lane which existed there in periods represented by layers 3 and 4. This lane was set perpendicularly to both streets cleared in this sector in 1989. This year's investigations in Area A confirmed the continuity of occupation in this part of the site during most of the Early Dynastic period. In the other sector, the long step trench set on the northern slope of Tell Abu- Hafur (Area B), our efforts centered on two problems. The first was continued exploration and more precise dating of structures discovered in layer VI of this trench. The second task was the further investigation of a huge mud-brick wall found within layer VIII of this trench. To achieve the first of our aims we extended the step trench to a width of more than 5 m, making it possible to clear a larger part of the structures belonging to layer VI. Some more small rooms belonging to the same structure were discovered, but what was more important was the diagnostic pottery material discovered in them. It is now absolutely clear that layer VI in Area B of Tell Abu- Hafur belongs to the period of ED III occupation at the site. The substantial mud-brick wall of layer VIII belongs to the same period as well. To reconstruct its original position and extent, three small test trenches were opened on the north100 PAM II [= Reports 1989-1990] © PCMA 2008 – digital reprint ern side of Area B and two others on its southern side. Each of the trenches measured c. 3 x 2 m. Within three of them the topmost remains of this huge wall were discovered. It would seem that this wall which is at least 6 m wide and c. 10 m high protected the ED III settlement on Tell Abu- Hafur against flooding from the nearby wadi, as the remains of this wall were found only on the slope on the northern side of the tell. The third sector of Tell Abu- Hafur subjected to exploration was the small mound called Tell Abu- Hafur East. The aim of this season's explorations was to study further the stratigraphical sequence established there last year, when numerous remains of the Mitanni period were found. For this purpose, three small soundings were made on the western slope of Tell Abu- Hafur East. Some fragments of a fortification wall erected on foundations made of stone blocks were unearthed. The wall itself was constructed of mud bricks. These fortifications represent a 2nd millennium BC occupation of the site and can be attributed to the Mitannian period. In a layer overlying this, some remains of a Neo Assyrian structure were found, most probably also of a defensive nature. This discovery has extended the stratigraphical sequence of the site, since in the previous year we did not find any layers which could be dated to a period later than the 2nd millennium BC. 101 © PCMA 2008 – digital reprint PAM II [= Reports 1989-1990] NEMRIK 9 Stefan Karol Kozłowski In the years 1985-1989 an expedition of Warsaw University directed by the present author conducted extensive archaeological research on Nemrik 9, a site in northern Iraq (Dohuk district) lying some 50 km northwest of Mosul. The investigations were sponsored by the PCMA and the Iraqi Service of Antiquities to whom we owe many thanks for their help and support. The present report covers not only fieldwork, but also some lab analyses. as well as detailed studies, either published already or currently in print. The site or at least its later phases date to the Pre-Pottery Neolithic. It is a large multi-phase settlement covering a maximum area of 2.5-3 hectares.1 It lies on the Tigris, just 2.5 km from the modern river bed, on the second terrace some 65-70 m above the present level of the river. The terrace itself is cut by numerous wadis flowing generally from the north to the Tigris and gathering water from the not too distant highlands of the Kurdish Mountains. The Nemrik site lies on a sort of peninsular terrace confined on the east and west by wadis. CHRONOLOGY The dating of the Nemrik site is based on a long series of C14 analyses conducted in the Gliwice Laboratory by M. Pazdur. Disregarding dates that are aberrations, the beginnings of the settlement should be put at the close of the 9th millennium b.c. and its end in the middle of the 7th millennium. 1 For a full description of the site and contour map, see: S.K. Kozłowski, K. Szymczak, Preneolithic site Nemrik 9 [in:] Researches on the Antiquities of Saddam Dam Basin Project, Bagdad 1987; S.K. Kozłowski (ed.), Nemrik 9. Pre-pottery Neolithic Site in Iraq. General Report – Seasons 1985-1986, Warsaw 1990. 102 PAM II [= Reports 1989-1990] © PCMA 2008 – digital reprint STRATIGRAPHY 2 Generally, the stratigraphy of the Nemrik site should be divided into two main layers differing from each other in terms of lithology and in genetics to an extent. The lower part of the profile, directly on top of clayey virgin soil with mole holes is made up of loess-like deposits, mainly of aeolian origin. In these formations three occupational levels can be distinguished, marked by the occurrence of objects (flint, stone, less frequently bone and clay), charcoal and ashes, animal bones and shells, architecture and burials, as well as clay floors covering more or less continuously the space between the houses. The upper part of the profile is made up of a thick layer of stone evidently of anthropogenic nature. It constitutes the remains of a stone pavements between the houses, repeatedly repaved. This layer contained the same kinds of finds as described above for the lower part of the profile, with the exception of clay floors. It is possible to distinguish two settlement phases in this upper layer. SETTLEMENT PHASES It is clear from the above that five settlement phases have been distinguished. Separating the phases are periods when obviously there was no permanent settlement on Nemrik. The first three phases which are less well known can be dated from the close of the 9th to the end of the 8th millennium, the last two (IV and V) to the end of the 8th and the first half of the 7th millennium. The latter two also provided most of the small finds as well as architecture and burials to mention the layout of an early Neolithic village. 2 S.K. Kozłowski, Nemrik 9, a Pre-pottery Neolithic Site in Northern Iraq, Paléorient 15/1, 1989 and S.K. Kozłowski, op. cit. 103 © PCMA 2008 – digital reprint PAM II [= Reports 1989-1990] ARCHITECTURE 3 All in all over 20 architectural objects were discovered and investigated completely or in part. They were sunk as a rule into the surface currently in use, sometimes to a depth of as much as 2 m, at other times much less. In the first four settlement phases the structures were as a rule round or oval, mostly of a dwelling nature (particularly structures Fig. 1. Nemrik 9, phase IV. 3 S.K. Kozłowski, A. Kempisty, Architecture of the Pre-pottery Neolithic Site in Nemrik 9, Northern Iraq, World Archaeology 21, No. 8, 1990; S.K. Kozłowski (ed.), Nemrik 9, Pre-pottery Neolithic Site in Northern Iraq. House 1, 1 A and 1 B, Warsaw, in press. 104 PAM II [= Reports 1989-1990] © PCMA 2008 – digital reprint with a diameter of 5 m or more), sometimes tombs or domestic areas. The fifth phase features subrectangular houses. In the earlier phases the prevailing structures are small and built in the pisé technique. In phase III we find walls constructed of large cigar-like mud bricks dried in the sun. The structures are larger now, reaching 7-8 m in diameter. In many houses some interior "architecture" has survived in the form of benches, pits and post holes with massive pillars in the pisé technique appearing in phase IV, a feature that is unique in the entire region. Clay plaster is common and in phase V floors are painted red. No hearths or door openings have been found. OTHER OBJECTS A large refuse pit is connected with phase V. BURIALS In the earlier phases the dead were buried in graves located among the houses. In phase III mass burials were made under house floors, but also in between the houses. In phase IV or V, a small burial ground was used in the southern part of the village. The inhumations are almost all contracted and lying on one side. Partial burials were encountered as well. SETTLEMENT LAYOUT (Figs 1 and 2) The thoroughness of the investigations of phases IV and V permitted an almost complete, if not fully complete reconstruction of the layout of the settlement, a fact that is unprecedented in the research on this region to date. In both phases a large percentage of the space was paved with broken stone pebbles. Large stone objects such as grinders, mortars, polishing slabs etc. were located on these pavements. A few 105 © PCMA 2008 – digital reprint PAM II [= Reports 1989-1990] houses, circular in phase IV and rectangular in phase V, were sunk into these pavements. At the southern end of the site there was a small burial ground and in phase V a large refuse pit was located in the centre of the settlement. It would appear that in the last two phases of the settlement it consisted of no more than a few buildings. In each of the houses the space on the clay benches, regarded as beds, was barely enough for ten people. Therefore, there is reason to believe that the Early Neolithic population of the village did not exceed several dozen in the last two phases. Fig. 2. Nemrik 9, phase V. 106 PAM II [= Reports 1989-1990] © PCMA 2008 – digital reprint FLINT INDUSTRY4 The flint industry is extremely numerous and features singleplatform cores for obtaining bladelets by the pressure technique. These bladelets were frequently sectioned. The main raw material was a local tertiary flint; imported obsidian is encountered rarely as is a chocolate-coloured flint which was imported in the form of ready blades used subsequently in the production of sickle-blades. Among the retouched tools retouched blades predominate together with perforators, retouched flakes and end-scrapers, all forms making up the Near Eastern standard of the time. Rare picks are distinctive as well as rather more frequent tanged points, both rhomboid and leaf-shaped. These two groups and backed bladelets as well set the industry apart, entitling it to be called Nemrikian. El Khiam points are infrequent. What is more important still is the fact that this industry practically does not undergo any evolution over 1500 radiocarbon years. Therefore, in northern Iraq, we do not observe a clear bipartite cultural sequence such as the classic Levantine sequence (Sultanian – Tahunian). GROUND STONE INDUSTRY5 The Nemrik collection of ground and pecked stone implements is extremely rich (more than 3000 objects) and varied. It consists of grinders, mortars, polishing slabs, polished axes, various pounders, stone "pegs", bolas, "maceheads", flint choppers etc. some rare stone vessels and finally jewellery. Being as it is so rich and varied, the assemblage permits a reconstruction of a full model of this industry for the whole region. 4 S.K. Kozłowski, K. Szymczak, Flint industry from House 1-1A-1B at the Prepottery Neolithic site in Nemrik 9, Northern Iraq, Paléorient 15/1, 1989. 5 R.F. Mazurowski, Ground and Pecked Stone Industry in the Pre-pottery Neolithic of Northern Iraq, Warsaw, in press. 107 © PCMA 2008 – digital reprint PAM II [= Reports 1989-1990] It also develops in time, the process being clearly visible upon analysis of particular settlement phases. As a result the regional model mentioned earlier is enriched by the factor of variability in time. This in turn allows for the verification of the debatable dating of Iraqi sites, for instance, a significant rejuvenation of the Jarmo sequence or suggestions of a mixed, multi-phase character of the assemblages from Zawi Chemi-Shanidar. BONE OBJECTS Bone objects are rare and fairly standard (awls, perforators, chisels). The needles are distinctive – flat and equipped with a large hole and an oblique engraved ornament. CLAY OBJECTS The so-called tokens are small, handmade and weakly fired objects taking on geometrical shapes, such as pancakes which are the most numerous, cones, balls and rarely cylinders. Accompanying these there are rare animal figurines (boar and pig, ram, cattle?). OBJECTS OF ART Deserving special attention are about 20 pieces of stone sculpture, originating from at least phase III or even phase II of the settlement at Nemrik. The objects include small (up to 20cm in length) representations of bird heads, a snake, an unidentified mammal, a lioness, buffalo hoof, and finally a male head and a figure of a woman. All are homogeneous in style and technique and of high artistic standard. The technique comprised piquetage modeling of a pebble using a flint tool for details, and finally very precise polishing. The only Near Eastern analogies for the Nemrik series is the sculpture of the Natufians, which is, however, earlier. In most cases the sculptures are damaged and only in one we can be fairly sure 108 PAM II [= Reports 1989-1990] © PCMA 2008 – digital reprint that the piece has survived in almost original form and place (the floor of a burnt house 2A). It would appear that these were household deities. DATA ON THE ENVIRONMENT AND ECONOMY The original landscape of Nemrik as well as the economy of its inhabitants can be fairly precisely reconstructed on the basis of several factors: the analyses presently being completed on pollen, dendrology, plant macroelements, malacology; comparative data obtained in interviews of eldest members of the local community, analyses of the remains of the original landscape; Layard's descriptions, and finally geomorphological and pedological data. It has already been noted that the settlement was located between the highlands and the valley of the great river, on a flat terrace cut by wadi. It appears highly feasible that it was located purposefully at the junction of two or even three biotopes: the steppe (the terrace), the forest (the highlands) and perhaps even the river valley, although water quite certainly came from the nearby streams. Thus, it can be said that the site was surrounded by what amounted to a steppe parkland with clumps of trees including poplars/willows and ashes. Tamarisks were among the trees which grew in the river valley, while the highlands were covered with a sparse mixed oak forest. The fauna differed in each of these zones. In the steppe zone there were chiefly antelopes and gazelles, as well as equids, the river valley had boars among others, while in the highlands there lived fauna of the mixed forest including deer, boars, beavers, bison, aurochs, bears and wild goats. Lions and Indian buffalo were also to be found in the district. There were fish in the river, crawfish and turtles in the streams. The animals mentioned above (without the fish) constituted the mainstay of an intensive hunting economy of the Nemrik 109 © PCMA 2008 – digital reprint PAM II [= Reports 1989-1990] dwellers. Even though domestic animals such as sheep, goats, cattle and pigs were known to these people since at least phase III, they never seemed to be prevalent in the diet. Hunting and the beginnings of husbandry were accompanied by a gathering economy (seeds of wild grain, lentils, vetch, shells of Helix salomonica and Unio tigridis among others). There is no such certainty as far as cultivation of grain is concerned, although pollen analyses seem to answer this question, at least for the later phases. Almost no evidence of fishing was observed. This data, the source for which are as yet incomplete analyses, shows the Nemrik environment to have been extremely rich and used in an optimal fashion with the economy taking advantage of all the available resources. At the same time it initiated a neolithization process that was most probably contemporary to the one in the Levant. CONCLUSIONS The importance of the Nemrik site for learning about a significant part of the prehistory of the Near East at a turning point in its development cannot be overestimated. Thanks to fieldwork (more than eleven months in the field when summed up), it was possible to gather extensive data, permitting the conclusions to be well grounded, even if not in harmony with textbook elaborations. Foremost, there is the well dated and clear stratigraphical sequence which in many details differs from sequences known from the Levant and the Zagros. It has turned out that the phenomenon of Pre-Pottery Neolithic is far more complex than previously thought with various local solutions to be considered (for instance, the Nemrik flint industry lasting throughout the history of the settlement in unchanged form, a different sequence of the architecture, different architectural details, on original art, partly local ground stone industry etc.). The layout of the settlement was found 110 PAM II [= Reports 1989-1990] © PCMA 2008 – digital reprint to be different from what was known on the subject from Jericho soundings. As a matter of fact, in view of the Nemrik finds, many sites considered typical should be re-dated or at least their homogeneity put in doubt. This is a sobering reflection to realize once again that we are still far from understanding the phenomenon of the Early Neolithic in the Near East. 111 © PCMA 2008 – digital reprint PAM II [= Reports 1989-1990] LA REPRISE DES FOUILLES A M’LEFAAT (SAISON 1989/1990) Stefan Karol Kozłowski, Kazimierz Kuźma et Karol Szymczak Ce site déjà classique, identifié et sondé en 1954 par R. Braidwood (Braidwood 1954, Dittlemore 1983), est situé à 35 km à l'Est de Mossoul et à 90 km au Sud-Est de Nemrik (voir, p. 105). II a été récemment l'objet d’une fouille de sauvetage, occasionnée par les travaux de l’autoroute voisine et menée en 1984 par M. Matti Baba Altun du Departement des Antiquités de Mossoul. A notre tour, ayant achevé la fouille de Nemrik, nous avons repris le chantier en 1989.1 LE SITE Le petit tell de M'lefaat, aujourd’hui fortement réduit et perturbé par des travaux industriels, repose sur des dépôts loessiques et des graviers tertiaires accumulés sur la rive droite de la rivière Khazir, affluent du Grand Zab. II n’a que 2 à 3 m de haut, et 90 m environ de diamètre, culminant à 13.50-14 m au-dessus de la rivière. Les travaux de l’autoroute ont totalement détruit les zones nord et centre du tell, primitivement presque circulaire, et décapé plusieurs dizaines de centimètres de la partie sud. Des constructions de l’époque assyrienne, ainsi que de la dernière guerre mondiale, ont également endommagé le site. LA STRATIGRAPHIE La partie méridionale du tell exhibe deux niveaux d’argile rapportée, chaque fois en une couche étendue, d’une quinzaine de 1 Membres de la mission: Stefan K. Kozłowski (directeur), Kazimierz Kuźma, Karol Szymczak et Jolanta Kamińska. 112 © PCMA 2008 - digital reprint PAM II [=Reports 1989-1990] centimètres d'épaisseur au maximum, interprétée comme la surface d’une place de village, autour de laquelle plusieurs maisons ont pu être dégagées. La couche argileuse inférieure est recouverte par un dépôt de terre grise, alors que la couche argileuse supérieure, creusée de petites fosses rondes ou ovales, utilisées comme foyers, épouse le relief du tell et présente la même morphologie. On y distingue des structures consistant en des pierres fichées dans l’argile et des dallages isolés, recouverts par un niveau très érodé contenant peu de silex, mais où des outils lourds en pierre, plus on moins en place, sont assez abondants. Les maisons répérées aux deux niveaux, entourant la place centrale, sont circulaires ou ovales, enfoncées dans les couches inférieures. Elles ont été comblées par les terres en provenance des couches érodées du sommet du tell. Le matériel archéologique y est souvent très menu et dut être récupéré par tamisage à l’eau. A l’état actuelde nos recherches, la stratigraphie du site peut se résumer par le tableau suivant: Structures d’habitat Sol d’argile Couche supérieure 1 2 3 - - - - - supérieur Couche inférieure 1A-3A 4 5 6 7 8? inférieur LA CHRONOLOGIE Quatre datations absolues ont pu être obtenues, gràce à M.F. Pazdur du laboratoire de Gliwice, que je remercie pour son obligeance. La plus ancienne (12 760 ± 660 b.p. – Gd 4483) provient des charbons de bois recueillis parmi les matériaux de remplissage de 113 © PCMA 2008 - digital reprint PAM II [=Reports 1989-1990] la maison 4 (phase ancienne du site). Cette datation est probablement trop élevée. La datation suivante, obtenue également à partir des charhons de bois de la phase ancienne, cette fois de la maison 3A, indique 10 290 ± 180 b.p. – Gd 6149. Les dates Gd 6150 (10 890 ± 140 b.p.) et Gd 4465 (10 850 ± 200 b.p.) ont été obtenues à partir des charbons recueillis dans des foyers au fond des petites fosses rondes de la couche argileuse supérieure. L’occupation du site de M'lefaat est ainsi située au cours du e 9 millénaire avant J.-C., avant celle de Nemrik. L’intérêt de ce site consiste donc à illustrer la séquence culturelle du Nord iraquien à l’aube des temps épi-paléolithiques et au début du Néolithique, à l’époque où la tradition natoufienne se développe en Palestine et celle de Shanidar dans le Zagros, l’une et l’autre totalement différentes de ce qu'on trouve à M’lefaat. Encore une autre tradition est signalée par Watkins à l’Ouest de l’Iraq, à Qermez Dare. L’ARCHITECTURE A ce jour, 10 maisons circulaires ont été identifiées, dont deux maisons superposées 1 et 1A dégagées par R. Braidwood dans le secteur nord du tell. La première mesurait 2.7 m à 3.5 m de diamètre et s’integrait dans la couche argileuse supérieure. Les fouilles iraquiennes de 1984 ont livré 4 maisons sur deux niveaux (maisons 5, 6, 7 dans la couche inférieure, et la maison 2, plus grande, dans la couche supérieure), toujours dans le secteur septentrional. Nos fouilles de 1989 et 1990 ont mis au jour 5 nouvelles maisons : maison 3 dans la couche supérieure et les maisons 3A, 4 et 8 dans la couche inférieure, la position stratigraphtque de la maison 8 demeurant incertaine; elles se trouvent toutes à la limite sud-est du tell. Nous ne donnons ici qu’une brève description de quelquesunes de ces structures. 114 © PCMA 2008 - digital reprint PAM II [=Reports 1989-1990] La maison 3 est ovale, large de 7.8 m et de 6.2 m. Les murs sont en pisé, épais de 40 cm ou plus, conservés sur 70 cm de haut. A l’intérieur, nous avons relevé deux niveaux successifs du sol, des banquettes, des traces d’enduit sur les parois. La maison 4, à moitié conservée, était circulaire, avec le diamètre de 6 m environ. Les murs atteignent 75 cm de haut avec une épaisseur de 45 cm et portent aussi des traces d'enduit. Un seul niveau du sol présente au milieu une banquette en pisé, concave, haute de 10 cm. La maison 8 est de forme ovale, 4 m sur 4.7 m. Ses murs n’ont que 10 à 16 cm d’épaisseur, atteignant la hauteur de 1.40 m; ils sont construits en briques crues façonnées à la main, en forme de cigares allant jusqu’à 70 cm de long. Deux banquettes diamètralement opposées (nord et sud) garnissaient l’interieur; sur l’une d'elles était fixée une meule. L’INDUSTRIE LITHIQUE Les silex décrits ici ont été recoltés en 1989 dans le remplissage des maisons 3 et 4 (sans tamisage). Ils sont au nombre de 583 pièces, ce qui est modeste pour un site iraquien de cette période. L’étude des matériaux de 1990, recueillis au cours du tamisage systématique à l’eau, n'est pas encore terminée. Les éclats dominent, faisant 61.7% de l’ensemble; la plupart sont corticaux ou sub-corticaux. Les lamelles, représentant 23.7% de la collection, sont le plus souvent très petites, régulières, intentionnellement sectionnées; leur morphométrie rappelle fortement les lamelles de Nemrik. Il y a aussi 7.5% d’outils retouchés et 5.7% de nucléus. La plupart de nucléus sont coniques, avec un seul plan de frappe, ayant produit des lamelles régulières. II en existe d'autres, non coniques avec également un plan de frappe, globuleux ou discoïdes. Tous ressemblent de près ceux de Nemrik, tout en étant beaucoup plus petits. 115 © PCMA 2008 - digital reprint PAM II [=Reports 1989-1990] L'outillage est surtout composé des lames et d’éclats retouchés (70% d’outils). On remarquera quelques pièces denticulées et des raclettes avec retouche dorsale ou ventrale. Les lames peuvent être retouchées avec grand soin ou au contraire d’une façon irrégulière; certaines portent des traces d’utilisation. Parmi les armatures microlithiques (15% de l’outillage), il y a des lamelles à bord abattu, à base oblique et retouche ventrale, ainsi que des microlamelles retouchées. Mentionnons enfin de rares perçoirs, des grattoirs courts et une troncature retouchée. Les fouilles de 1990, comme les anciennes recherches américaines, ont également apporté des pointes d’el-Khiam, des triangles allongés à base concave, des lamelles à dos avec troncature retouchée et des lamelles à bord abattu. L'industrie de M’lefaat ressemble beaucoup à celle de Nemrik: même morphologie et la façon d’exploitation des nucléus, même morphométrie, même technique de fractionnement des lamelles, le répertoire similaire de l’outillage. Les différences consistent principalement en pauvreté numérique des silex à M’lefaat, comparée à leur abondance à Nemrik. Proportionnellement, à M’leffat il y a plus de lames que d’éclats, et les armatures y sont plus nombreuses, alors que manquent les pointes et les piques, caractéristiques pour Nemrik; ces disparités suggèrent un mode de vie différent. Nous avons en outre retrouvé plusieurs dizaines de gros outils en roches locales diverses, comme il y en a aussi à Nemrik: grandes meules de grès rosé, broyeurs ronds ou ovales, mortiers en calcaire et pilons cylindriques façonnés à partir des galets. Citons encore quelques bolas de silex, plusieurs boules perforées et hachettes polies en roche verte. LA FAUNE La fouille a rendu en abondance des restes d’animaux, généralement bien conservées, mais déterminées à ce jour seulement en 116 © PCMA 2008 - digital reprint PAM II [=Reports 1989-1990] partie. Pour les mammifères, une gazelle et un carnivore ont été identifiés; en revanche, il y a de nombreux os d’oiseaux et des poissons d’assez grande taille. Des moules et des éscargots complètent cette liste provisoire qui évoque plutôt l’économie d’un village de chasseurs et de pêcheurs. De nombreuses graines recueillies par flottement indiquent la pratique de la cueillette. CONCLUSIONS 1. M’lefaat est un village du 9e millénaire av. J.-C. Du point de vue chronologique et économique, c’est un village épi-paléolithique, contemporain du Natoufien du Levant et de Shanidar dans le Zagros. Par rapport à Nemrik, il est plus ancien, et peutêtre précurseur. 2. II s’agit d’un village véritable, occupé à deux reprises par une population sédentaire pendant des périodes assez longues (mais encore mal déterminées). 3. Pour la phase plus récente au moins, on est en droit de parier d’une place centrale recouverte d’une couche d’argile, entourée des maisons circulaires, munies de banquettes. Les murs de ces maisons étaient en pisé ou en brique crue. Quelques trous de poteaux ont été observés pendant les fouilles iraquiennes. Contemporaines des maisons natoufiennes en pierre, ces habitations semblent les plus anciennes constructions connues en pisé et en brique crue. Elles ressemblent de près aux maisons de Nemrik. Des meules, des mortiers, des foyers enfoncés, ainsi que des aires de galets étaient installés sur la place centrale. 4. L’industrie lithique en silex représente la phase la plus ancienne connue à ce jour de l’horizon mésopotamien. Les grattoirs denticulés, les lamelles retouchées, la manière de débitage, ne différent guère du matériel de Nemrik. Il convient de signaler la rareté des points d’el-Khiam et 1’abondance des pointes et lamelles à bord abattu, connues à Nemrik (en compagnie des “pointes de Nemrik”) depuis la fin du 9e millénaire. Il semble bien y avoir 117 © PCMA 2008 - digital reprint PAM II [=Reports 1989-1990] une filiation entre les industries lithiques de M’lefaat et de Nemrik, tant en silex qu’en d’autres roches. 5. Le Proche Orient a connu, au cours du 9e et du 8e millénaires, trois grandes traditions culturelles: - la tradition du Levant, avec le Khamien qui succède au Natoufien; - la tradition mésopotamienne, avec les industries de M’lefaat auxquelles succède le Nemrikien, et avec un prolongement possible vers l’horizon de Jarmo; - la tradition du Zagros, bien représentée à Zawi Chemi. 6. A partir de la seconde moitié du 9e millénaire av. J.-C. existent déjà en Mesopotamie certains traits technologiques et morphologiques que l’on retrouvera dans le PPN du Levant (brique, industrie microlithique en silex avec les pointes de Khiam). II est désormais raisonnable de supposer qu’un certain nombre d’éléments constitutifs du PPN du Levant, surtout l’architerture d’argile, tirent leur origine de la province mésopotamienne. 118 © PCMA 2008 - digital reprint PAM II [=Reports 1989-1990] THE FIRST SEASON OF EXCAVATIONS IN HATRA, IRAQ Michał Gawlikowski Following the kind invitation of Dr. Moayed Saeed Damerji, Director General of Antiquities and Heritage, an archaeological mission worked in Hatra from March 10 to April 14, 1990.1 We found there very favorable conditions of work and rest, and most of all the efficient, friendly cooperation of our Iraqi colleagues, Sd. Manhal Jaber in Mosul and Sd. Hikmat Bashir el-Aswad in Hatra itself. It is our pleasant duty to acknowledge their very valuable help and to thank them most cordially for all they have done. We also extend these thanks to the appointed Representative of the Directorate General, Sd. Mohammad Ahmad Baud, who assisted us daily in our work. As settled with the Directorate General during my previous visit to Iraq, the activities of the mission were to include the recording and study of the defense walls of Hatra on one hand, and on the other, soundings in the main temple court in search of traces of the original settlement. However, during the first days of our stay we found what seemed to be the line of the oldest walls of Hatra, halfway between the great temple and the walls standing on the southern side of the city.Accordingly, I requested permission to check these observations through digging. I can say now that our expectations have been fully confirmed. Meanwhile our recording of the standing walls started at the southeastern corner of the fortifications and went along the eastern front for about 500 m up to the East Gate, recently restored. The main features of this imposing defensive system can be studied already from this sample, but we intend to proceed further with the recording. The defenses are built in depth for about 30 m, beginning with an earthen wall and ditch, and including the main wall with towers 1 The mission included Messrs Janusz Byliński and Tomasz Herbich (both archaeologists), Adam Dolot (architect) and the present writer. 119 © PCMA 2008 – digital reprint PAM II [= Reports 1989-1990] and parallel inner wall, both in mud brick on stone foundations and both over 3 m thick. The original height of the walls can be estimated at 10 m, judging from the remains of steps leading to the top and allowing for a protective parapet along the walk. We have assured ourselves of this and some other points through several soundings on the inside of the wall. It is known from published inscriptions that both the northern and eastern gates were built by Lord Nasru, who is also certainly responsible for the whole rampart. This enterprise took place about AD 140, i.e., shortly after the same Lord Nasru had completed the temenos wall by AD 138, and before AD 152, when the wall was already in use (see inscriptions H 272 and H 336). Later additions to the rampart, consisting mainly of massive stone towers intended for housing catapults, seem to have been executed after the unsuccessful siege by Septimius Severus in AD 198/199. The wall was subsequently repaired and reinforced until it finally confronted the Sassanian troops of Ardashir in AD 240. The question immediately arises where is the wall, which had resisted Emperor Trajan in AD 117. Hatra was at the time "neither big nor rich" as Cassius Dio says, hardly a fitting description of the magnificent city of barely twenty years later. With this question in mind, we explored the walled area and noticed a clear straight line dividing the low ground to the south from the maze of small tells marking the densely built quarters around the Great Temple. This line runs about 230 m from the south wall and about 320 m from the temenos, meeting on the east a wadi which runs in front of the Great Temple. All tombs on the southern side are to be found beyond this line which represents, as we supposed, a trace of the defenses of the 1st century Hatra. We opened several trenches along the line of the supposed early wall and found it immediately (Fig. 1). It was built in a straight line seven or more bricks deep, i.e., about 3 m, on rubble foundations. Our trenches made it possible to follow the wall for about 200 m of its length. The preserved height varies from one to six bricks above 120 PAM II [= Reports 1989-1990] © PCMA 2008 – digital reprint Fig. 1. The localization and course of the 1st century defense wall. the base. The wall was dismantled in antiquity and houses were built on top of it. Later on it was subject to heavy erosion which has caused it to disappear entirely. Close to the wadi the wall turns at a right angle to the north. The corner was reinforced by a massive tower of mud brick on a base of good ashlar masonry some 6.6 m wide on the southern side and projecting on either side for 3.5 m from the face of the wall. While not dated exactly, this structure is in fact the oldest extant monument of Hatra, marking the limits of the archaic city in the 1st century AD. The extension of these defenses on the other sides cannot be traced with equal assurance on the ground. We consider the results of this season as highly encouraging and hope to continue our research in Hatra in the future. We would like to spend the next season excavating the South Gate and recording the remaining parts of the cleared rampart. If accepted, this project would eventually provide us with data for a comprehensive publication of the Hatra defenses. 121 © PCMA 2008 – digital reprint PAM II [= Reports 1989-1990]