n
PALACES AND PATRICIAN HOUSES IN THE MIDDLE AND LATE BRONZE AGES
Eliezer D. Oren
Introduction
It is difficult to identify palaces, patrician houses and governors' residencies at sites in Israel for a variety of reasons — among them the lack of written sources and, above all, the fragmentary nature of the building remains. It is evident, however, that major contributions to the formation of a 'Canaanite' architectural concept in Palestine were made by the neighbouring cultures of Egypt, and especially Syria, Anatolia, and Mesopotamia. The dominant influence on both private and public architecture was the traditional 'Oriental' house, which was built around a central, unroofed courtyard. This tradition characterized the eastern Mediterranean basin, and continued without significant change up to the classical period. Building remains of palaces in Israel exhibit an unbroken continuity from the Middle to the Late Bronze Age. Thus it is possible to speak of an architectural concept for buildings designed for the ruling classes of Middle and Late Bronze Age Canaanite cities. Identification of these buildings and their classification is based on their location in the city (near the gate or the temple, in the elite zones), as well as by the quality of the building materials and certain construction techniques. The fragmentary nature of the building remains also makes it difficult to elucidate the function of the various units - offices, storerooms, living quarters, servants' and service rooms. Nor can it always be established with certainty whether the buildings had a second storey. Construction details of roofing, windows, doorways, stairways, water supply and drainage systems, washing facilities, and lighting arrangements also cannot be reconstructed with any certainty.
Courtyard Palaces
The basic plan of a palace consists of a spacious, rectangular-shaped courtyard with rooms surrounding it on all sides or flanking it on two sides. The walls were relatively thick and constructed of mud brick on stone foundations. A characteristic feature of the palaces is a ratio of 1:1, or even 2:3, between the area of the courtyard and that of the rooms. Well-designed courtyard palaces appear for the first time in MB IIA occupation strata, and became the typical public building in Syria and Palestine from that time through the Late Bronze Age. The palaces occupied a considerable part of the urban area allocated for public buildings, and were usually situated near the temple; later, at the end of the MB IIC, they were erected near the city-gate. Since palaces extended over large areas and were located on sites that were continuously rebuilt, only a small number have been completely excavated (mainly in Syria). The high quality building materials used in their construction — hewn stones, orthostats, dressed stones — were in great demand by later generations of builders, thus the palaces were greatly destroyed by plundering. For these reasons reconstructions of the plans of these palaces often exceed the material evidence.
Megiddo is an excellent example of a prosperous Canaanite city with advanced civic architecture. It is possible to trace in detail the development of the palace plan in the Middle and Late Bronze Ages and to study its integration within the general layout of the city. Remains of palaces were uncovered in three areas: A A, BB, and DD; of these the most complete picture was revealed in Area AA.
In Area BB, west of the sacred precinct, fragmentary
105
HLIEZER D OREN
Kabri. Middle Bronze Age II Palace
remains of monumental buildings, apparently palaces, were unearthed in Strata XH-X. In this period (MB IIA-B), the palace and the temple formed a single architectural unit separate from the rest of the city. Towards the end of the Middle Bronze Age (Stratum X), a new architectural concept can be distinguished: the palace was moved from the temple area and henceforth became an integral part of the city-gate area.
In Stratum XII (MB IIA), the city plan was reorganized and the palace was established in the vicinity of the sacred area (Fig. I).1 The evidence from the excavations of G. Schumacher and G. Loud can help in reconstructing the western part of the monumental courtyard palace assigned to this
1. The shift in location of the palace from the sacred precinct to the area of the gate has also been noted at other sites as, for
example, Alalakh. Strata VTI-V.
stratum.2 The outer wall, which was 2 m. thick, was preserved in this section, with small halls and rooms built along it. The palace extended over an area of at least 1,000 sq. m. In Stratum XI (MB IIB) it was rebuilt (Building 5059) on foundations of field stones and the floors of the rooms were coated with thick plaster. It is possible to discern a basic plan consisting of a large courtyard with a beaten lime floor flanked by small rooms.3 In Strata X-IX the palace continued in use in its original plan, with only minor changes.4
In Area A A the palace (4031) of Stratum X (MB I1C) was erected in the vicinity of the city-gate, and served as the nucleus of the monumental buildings of the successive strata. It should be noted that in
2. See Megiddo II, Fig. 415.
3. See Megiddo II, Fig. 399; I. Dunayevsky and A. Kempinski; Megiddo Temples, EI 11, Jerusalem, 1973, p. 22, Fig. 15.
4. See Megiddo II, Figs. 400-401; Dunayevsky and Kempinski (above note 3), p. 24, Fig. 16.
106
PALACES AND PATRICIAN HOUSES
1 !—1
i ! I ! !
-J
I---/
2. Palace, Megiddo Stratum IX. Megiddo II, Fig. 381.
I 1 I I —I I__
10 m
I. Palace, Megiddo Stratum XII. Megiddo II, Fig. 415.
these strata very few changes were made in the city-gate area. Palace 4031 continued in use in Stratum IX, with only the raising of floors and slight repairs. The plan of Building 2134 of Stratum IX is thus an exact replica of the original plan of the palace of Stratum X (Fig, 2).s Buildings 4031 and 2134 are typical examples of MB Canaanite courtyard palaces, very similar in plan to the palaces at Tell el-'Ajjul, Aphek, and Tel Sera'. The dimensions of the later building are 22 x 25 m. and of the central courtyard, 9 x 12 m.; the latter comprising approximately one-fifth of the total area of the palace. A drainage system was found in the courtyard and a staircase leading to a second storey was found in the southwestern corner. The walls of the building on the west (and south?) were 1.2 m. thick, while on the north and
5. See Megiddo II, Figs. 380-381; A. Kempinski: Syrien and Palastina (Kanaan) in der lallten Phase der Mittelbronze IIB Zeit {1650-1570 V. Chr.), Wiesbaden, 1983, pp. 93-94, 169-172 and Plan I.
east they were as thick as 4 m! The massive walls on the north side, near the gateway and on the edge of the mound, indicate that the palace was also meant to serve as a fortress.
Following changes in the city plan in Stratum IX, the houses west of the palace were razed and the palace complex expanded into this area. The destruction of Stratum IX, attributed by the excavators to the campaign of Thutmose III, was apparently not complete in Area AA. Although the new palace erected here (Stratum VIII, Building 2041) (Fig. 3) was based on a different plan from that of the Stratum IX palace (Building 2134), the two shared walls on the northern and eastern sides and the floors were very close together.6 Building 2041 extended over an area
3. Palace, Megiddo Stratum VIII, Area A-A. Megiddo II, Fig. 383.
6. See Megiddo II, Fig. 382.
107
ELIEZER D. OREN
of 1,500 sq.m. (30 x 50 m.); the walls were from 2-4 m. thick, and the rooms had lime floors and sophisticated drainage systems. The central courtyard (2041) was entered through a wide opening. An inner courtyard (3091), with a shell pavement and a basin, seems to have had a series of doors that communicated directly with the important rooms of the palace. A group of rooms and courtyards on the eastern side were entered through a separate doorway in the south. The most luxurious wing was on the west side. It contained a forecourt in whose northern wall was a monumental entrance with basalt columns and piers. The entrance led to a group of rooms, one of which was presumably the throne room (cf. Ugarit, Fig. 10). An Egyptian lotus-shaped capital found in the eastern courtyard may have belonged to one of these columns.7 Beneath the floor of one of the small rooms (3100) on the northern side of the building, a hoard of gold vessels, ivory plaques, jewelry and ornamental objects was found.
m
4. Palace, Megiddo Stratum VIIB. Megiddo II, Fig. 382.
The palace of Stratum VIII continued in its original plan, including the forecourt (3091) and central courtyard (2041) and a well-built threshold between them, in the succeeding Strata VIIB-VIIA (Fig. 4).8 It should be stressed that there was only a slight change in the floor levels between Strata VIII-VIIB.
7. See A. Siegelmann: A Capital in the Form of a Papyrus Flower from Megiddo. Tel Aviv 3 (1976), p. 141.
8. See Megiddo II. Figs. 383-384.
The thick northern wall was destroyed and replaced by a narrower one, with buttresses built directly on top of the glacis of Stratum VIII. The principal change in Stratum VIIB occurred in the west wing. The courtyards and monumental entrance were replaced by a row of small rooms, with their doorways aligned along a central axis (3186) which led to a corner room (3103). Here were found a small raised platform and steps, which the excavators identified as a household shrine.
In Stratum VIIA, which came to an end at the beginning of the Iron Age, a special annex was built here consisting of three subterranean rooms (3073) 1.4 m. deep. The absence of any finish or plaster on the exterior surface of the building supports the assumption that these rooms were constructed within a foundation trench.9 In the rooms was found a collection of unique ivory objects whose date of concealment was established by an ivory pen box bearing the cartouche of Rameses III. The courtyards of the Stratum VIIA palace were paved with very large stones coated with plaster. The quantities of painted plaster fragments found indicate that the walls of the palace originally had painted decorations. The palace of Stratum VIIA was destroyed in a great conflagration, probably in the middle of the twelfth century B.C.
Remains of additional palaces were uncovered in Area DD (Fig. 5). Stratum VIII contained a
5. Palace, Megiddo Stratum Vtll, "Area D-D. Megiddo II, Fig. 4M.
9. T. Dothan: The Philistines and Their Material Culture, Jerusalem, 1982, pp. 70-76.
108
PALACES AND PATRICIAN HOUSES
Megiddo. Late Bronze Age Palace (Strata VIII-VII).
magnificent courtyard palace (Building 5020) that was similar in size and plan to Palace 2134 of Stratum IX in Area A A.10 The building included a central courtyard (11 x 15 m.) with a beaten lime floor on a gravel base, a table (for offerings, an altar?), and stone storage\nstai\ations.The palace apparently continued iw. use without chasv^. vs. Staatoxsv thirteenth century B.C.
One of the finest examples of a courtyard palace was uncovered by W.M.F. Petrie in his excavations at Tell el-'Ajjul." In a raised area in the northwestern corner of the mound were discovered a group of buildings that he identified as palaces. W.F. Albright, however, has shown that only Palace I should be
10. Megiddo II, Fig. 411.
11. See W.M.F. Petrie: Ancient Gaza, II, London, 1932, pp. 2-5, PI. XLV.
considered a palace. Palace 11 was apparently a patrician house (see below, p. 115), and Palaces III-V served as fortresses.12 Palace I (Fig. 6) was built on a foundation of large, well-dressed stone slabs, shaped tike orthostats^ ar^Taxlm.&tel\ (L7 ra.. bi^h and 0.2 m. truck. The slabs wete placed into wide
by brick walls about 2 m. thick. According to Petrie, the slabs lining the foundations were quarried from the city's fosse, thus providing the chronological link between the city's fortifications and the palace. The fragmentary remains of the southern half of the palace do not permit the reconstruction of the building. Petrie initially reconstructed the plan as consisting
12. See W.F. Albright: The Chronology of a South Palestinian City, Tell el Ajjul, AJSL 55 (1938), pp. 337-359.
109
ELIEZER D. OREN
Palace I, Tel! el- Ajjul. A. Kempinski: Syrien und Palästina in der letzten Phase der Mittel Bronze IIB Zeit, Wiesbaden, 1983, Plan 6.
of a square courtyard surrounded by rooms. It later became evident that the palace, measuring 35 x 50 m., contained a rectangular courtyard (25 x 40 m,) bounded by a row of rooms on the northern and eastern sides. The archaeological evidence indicated that the building was never completed, that there had
7.  Palace, Tel Sera' (detail). Plan by author.
been two phases of construction and that the floors had been raised. The main entrance of the palace cannot be located with certainty. The drainage system in the southeastern corner and the adjoining massive construction (a tower?) indicate that the entrance gate was in the southern wall. With a single exception (in Room OG), no doorways were found connecting the courtyard and the rooms. One of the rooms (MK) which had a plastered floor was a washroom.
The palace has been dated by scholars to the MB II, and its destruction, by fire, to the end of that period, during the expulsion of the Hyksos (Albright) or even earlier.13 In the opinion of this writer, the construction of the palace should be attributed to the end of the Middle Bronze Age and its destruction to the beginning of the Late Bronze Age.
In the southeastern corner of Tel Sera', in the area in which the public buildings stood in the Late Bronze Age, impressive remains of a monumental building were uncovered on bedrock. The building, only one corner of which has been exposed so far, belongs to Stratum XII (Fig. 7).14 Its foundations were built on an artificial platform, about one metre high, constructed of large fieldstones with a fill of pebbles and earth containing sherds dating from the Chalcolithic period to the Intermediate EB/MB. The walls, which were about 2 m. thick and preserved to a height of 2.5 m., were built of alternating courses of brown and white bricks thickly coated with plaster. The remains excavated so far include part of a courtyard with a row of small rooms on its eastern side. The rooms have door jambs and piers in the entrances and thick lime-plastered floors. The original structure is similar in plan to Palace I at Tell el-Ajjul. Four building phases, with changes including the raising of floors, were distinguished. The earliest palace has been attributed to the end of the Middle Bronze Age. The palace remained in use during the first phase of the Late Bronze Age.
On the acropolis at Lachish sections of a massive structure were uncovered in Stratum VIII. Most of it is still buried beneath the foundations of the fortified palace of Stratum V.15 Its walls, which are about 2 m. thick, were constructed on stone foundations and
13. See O. Tufnell: Tell el Ajjul, EAEHL I, Jerusalem, 1978, p. 57; A. Kempinski: Tell el Ajjul, Beth Aglayim or Sharuhen?, IEJ 24 (1974), pp. 145-152.
14. See E.D. Oren: Ziglag — A Biblical City on the Edge of the Negev, BA 45 (1982), pp. 164-165.
15. See D. Ussishkin: Excavations at Tel Lachish 1973-1977, Preliminary Report, Tel Aviv 5 (1978), pp. 6-10, Fig. 2.
110
PALACES AND PATRICIAN HOUSES
coated with thick plaster, as were the floors. Despite the very fragmentary state of the remains at Lachish, the thick walls, the use of piers, and other features can be compared to the palaces of Tell el-'Ajjul and Tel Sera'. The various phases of the Lachish palace belong to the MB NB-C; it was destroyed at the end of that period or at the beginning of the Late Bronze Age.
In the northwestern corner of Aphek (Area A) the remains of a huge building were discovered, of which only the central courtyard was exposed. The building extended over an area of more than 750 sq. m. Three building phases of the MB IIA could be distinguished.16 On the basis of the published data, the building can be classified as a typical courtyard palace, a large part of which was occupied by its central courtyard. The construction was of excellent quality: the walls were thick and laid on stone foundations, and extensive use was made of columns. The earliest palace (Phase c) contained a large courtyard with a row of column bases in the centre (to support the roof?). The latest palace (Phase a) contained two rectangular courtyards with thick lime-plaster floors and drainage channels. After this palace was abandoned, a new and larger one was built on the acropolis. It contained a central courtyard with two column bases more than one metre in diameter and was surrounded by a series of small dwelling and service rooms.17 This palace was attributed to the MB IIB (Fig. 8).
In Area IV at Shechem, east of the fortress temple, the Drew-McCormick expedition uncovered a large MB IIB-C building that G.E. Wright initially interpreted as a palace then later as a courtyard temple with casemate rooms (Fig. 9).18 In the absence of detailed excavation reports, it is difficult to discern the exact plan of each of the four phases of the structure (Temene 2-5) and their stratigraphical relationship. Nevertheless, the identification of the building as a temple on the basis of its similarity to Anatolian temples is untenable, and the original suggestion to consider it a palace seems more reasonable.19 The palace was bounded on the east and west by massive walls (Wall 900 and Wall D) that created an enclosed trapezoid-shaped area. It is possible, as was suggested by the excavators, that Wall D separated the palace
16. M. Kochavi: The First Two Seasons of Excavations al Aphck-Antipatris, Tel Aviv 2 (1975), pp. 17 f., Figs. 3, 6.
17. M. Kochavi: Aphek-Antipalris: Five Thousand Years of History, Tel Aviv, 1989 (Hebrew).
18. See Shechem, pp. 103 f., Figs. 64-70.
19. See G.E. Wright: BASOR 161(1961), pp. 33-39.
8.  Palace, Tel Aphek. plan by M. Kochavi, Aphek excavations.
9. Palace, Tell el-Balata (Shechem). Shechem, Fig. 64.
from the area of the city and even served as an enclosure wall.20 An examination of the building phases, especially Phases 2-4, suggests that the palace included an extensive courtyard (about 15 x 20 m.)
20. For a reconstruction see G.E. Wright: Shechem, tlAEHL IV, p. 1084.
Ill
ELIEZER D. OREN
Aphek. Late Bronze Age Palace.
flanked by a row of rooms on the north and east. Here, too, the courtyard occupied more than half of the building's area.
At Hazor, in Strata XVII-XVI of Area A, from the MB IIB-C, beneath the Israelite pillared building of Stratum VIII, the corner of a huge structure (Building 387) was uncovered that was, without doubt, a royal palace.2' The building's deep foundations and massive walls (more than 2 m. wide) indicate that it probably had an upper storey.
The function of the public building in Area F in the lower city of Hazor is still a subject of dispute. The building was originally identified as a palace, but its excavator, the late Y. Yadin, subsequently reconstructed it as a double temple.22
21. See Hazor III-IV, Plates V-VU; Y. Yadin: Hazor, pp. 124- J25.
22. See Dunayevsky and Kempinski (above note 3), p. 21, Fig. 14, contra Yadin: Hazor, pp. 96-98.
Recent excavations at Tel Kabri uncovered a section of an extensive and well-planned palace in Area D, dating to the MB IIB (17th century B.C.).23 Its plan, however incomplete, seems to have resembled that of the palace at Alalakh, and included a central courtyard flanked on the north, and probably on the south, by halls and subsidiary rooms (photo, p. 106). The ceremonial hall (611) to the north of the courtyard, adjacent to a large staircase, measured 10 x 10 m. and had a sunken jar in the centre. It appears that the walls were originally lined with orthostats above floor level, all of which have been robbed. The plastered floor of Hall 611 is frescoed with a rich
23. A. Kempinski: Four Seasons of Excavations at Tel Kabri, Qadmoniot XXIII (1990), pp. 37-38; A. Kempinski and W.D. Niemeier (eds.): Excavations at Kabri, Preliminary Report of 1989 Season, No. 4, Tel Aviv, 1990, pp. 43-46, XVI-XXVI.
112
PALACES AND PATRICIAN HOUSES
!0. Palace, Ugarit (Syria). C.FA. Schaeffer: Ugaritica IV, Paris, p. 1462, Fig. 21.
repertoire of geometric, floral and figurative motifs. Frescos are known from 18th century B.C. palaces at Mari and Alalakh. However, this unique fresco is best paralleled, stylistically and iconographically, by the contemporary Cretan-Theran Late Minoan IA wall painting tradition.
In the Middle and Late Bronze Ages, the cultures of Mesopotamia and northern Syria had a far-reaching influence on the development of the urban architectural tradition of Palestine. Although this survey cannot include a detailed comparative study, several examples — the palaces at Ebla, Mari, Alalakh and Ugarit — should suffice to reveal the
sources of inspiration for the architecture of Syria and Palestine.24
In the lower city of Tell Mardikh (Ebla) in northern Syria, part of a great palace (Royal Palace E) was uncovered in Stratum IIIA-B (MB IIA-B). The palace was built above an earlier public building in Stratum IIB2, dating from the end of the
24. See H. Frankfort: The Art and Architecture of the Ancient Orient (4th ed.), London, 1970, pp. 75-76; R. Naumann: Architektur Kleinasiens (2nd ed.), Tubingen, 1971, pp. 389 f.
113
ELIEZER D. OR EN
third millennium B.C.25 The palace had an extensive rectangular courtyard (8 x 15 m.) bounded on the north and east by rooms and halls, and on the west by a very thick wall (retaining wall? fortification?). The courtyard and the rooms had stone pavements set in clay, and thick plaster covered the stones. The rooms on the northern and eastern sides were panelled with orthostats, and the lintels and thresholds were built of smooth stones. Access to the central court was through a corridor paved with stone slabs and from there through two wide entrances. In the opinion of the excavators, the group of rooms south of the corridor forms the boundary of the southern quarter of the palace. The different building technique and absence of orthostats in the southern wing, however, cast doubt on the suggestion.
The monumental palace complex of the kings of Mari in the nineteenth-eighteenth centuries B.C. provides a detailed picture of the varied functions carried out in palaces in that era. The palace occupied an area of about 25 dunams (120 x 200 m.) and contained some 300 rooms, halls, and courtyards — including kitchens, bathrooms, a throne room, cult rooms, schoolrooms, workshops, storerooms, offices for clerks, and diplomatic, religious, and administrative archives. This was a well-designed plan: in Mesopotamian architecture the courtyard was not merely an open space, but a central, enclosed element in the palace complex in which a great variety of activities and functions were concentrated.26
Strata VII-VI at Alalakh provide us with valuable comparative data on public architecture from the Middle and Late Bronze Ages in Syria. The MB IIB 'Yarim-Lim' palace complex of Stratum VII contains all the characteristic components of the courtyard palaces, i.e. spacious courtyards surrounded by various rooms, thick walls, extensive use of buttresses, orthostat facing, etc.27 At Alalakh, as at other MB centres, the palace was adjacent to the temple and the two actually formed one architectural unit. The Alalakh palace comprised a ceremonial wing in the northwest, and residential and storage units in the southwest. The former, with its orthostat-lined walls, included a spacious courtyard, 9 x 21 m., that was surrounded by storerooms and other subsidiary chambers as well as a staircase leading to the second storey. The royal wing contained a large throne room
25. See P. Matthiae: Die Furstengraber des Palastes Q in Ebla, Antike Welt 13 (1982), Figs. 1, 7.
26. See A. Parrot: Mission Archeologique de Mari: Le Palais, Parts, 1958-1959.
and reception halls, the walls of which were richly decorated with fresco paintings.
The royal palace at Ugarit is one of the largest buildings of its kind from the Late Bronze Age in the Near East. It extends over an area of about 10 dunams and so far more than 100 rooms and halls arranged around ten inner courtyards, and a number of stairways leading to a second storey, have been uncovered (Fig. 10). The plan of the palace reveals that it was constructed in several building stages during the fifteenth-thirteenth centuries B.C. Some sections of the walls were built of well-dressed ashlar blocks with drafted margins and raised bosses. Wide grooves in the stone walls held horizontal wooden beams that gave the walls some flexibility, important during earthquakes. These last two features are also found at LB sites on Cyprus (Enkomi, Kition) and are especially typical of ashlar building in Iron Age Israel. The excavation of the palace disclosed sophisticated drainage systems and fine flagstone pavements in the courtyards. The spacious courtyard inside the gate to the palace contained a deep well. The elaborate courtyard on the southeastern side contained an ornamental pool to which water was conveyed through a network of stone channels installed beneath the palace floor. Important archives were found in the rooms of the palace, and one of the courtyards contained a kiln for firing cuneiform clay tablets.
A distinctive feature of Ugaritic palace architecture is the incorporation of a portico into the building units — elaborate entranceways, some of them stepped, with a pair of columns between stone piers. Buildings with this type of facade from the Iron Age are termed bit hilani in the literature, although the term was used primarily to designate the portico itself in the facade or in one of the wings of the building.28 At Alalakh Stratum IV (15th century B.C.) was found an excellent example of a courtyard palace which was entered through a monumental portico unit. The building, known as Niqme-pa's palace, has a composite plan extending over an area of more than 5000 sq.m., comprising ceremonial, residential and storage units. Its complex ground plan, monumental portico entrance, the use of orthostats, as well as the royal archives, all point to its role as the central palace complex of LB Alalakh.
Courtyard palaces from sites of Palestine and Syria exhibit a number of common architectural features:
0
27. L. Woolley: Alalakh, London, 1955, pp. 91-131.
28. See Naumann (above note 24), pp. 408-411; Frankfort (above note 24), pp. 151-152; 253, 276.
114
PALACES AND PATRICIAN HOUSES
thick walls, massive foundations, sometimes even a platform on which the entire building was constructed, paved courtyards, etc. The outstanding feature of the structure is without doubt the extensive courtyard, which occupied a major part of the total area of the palace. The palaces in the north and centre of Palestine are apparently earlier than those in the south, appearing in fully-developed form already in the first stage of the MB II. The resemblance of palace plans of Palestine and northern Syria attests to the adoption of Mesopotamian architectural concepts as a result of the extension of Mesopotamian culture into Syria and Palestine in the Mari period via cultural and commercial ties. This phenomenon is crucial to understanding the structure and social stratification of the urban population of Palestine in the Middle Bronze Age.
It should also be emphasized that, in comparison with the palaces of Palestine, those in Mesopotamia and northern Syria were extremely complex in plan. The courtyard served as an important element of the building, but the palace complex also contained numerous rooms and halls that served a variety of functions: throne room, reception halls, offices, storerooms, lavatories, etc. Palaces in Palestine were simpler in plan and it is difficult to determine the exact use of the rooms that surrounded the courtyards. The limited number of rooms which have been found around courtyards, and their modest dimensions (see the palaces at Tell el-'Ajjul, Megiddo, and Aphek), leave open the possibility that these remains may represent only part of the complex, or its nucleus. The term 'palace', as applied to the unparalleled examples found in Mesopotamia and northern Syria, is more appropriate for the complex of courtyards, halls and rooms of Building 2041 at Megiddo (Stratum VIII) than for the series of small rooms flanking the central courtyard of Building 4031 in Stratum X there.29
Patrician Houses
The multiplicity of terms found in the literature to designate the dwellings of the wealthy classes
— patrician houses, palaces, governors' residencies
— results primarily from the lack of exact criteria for defining the functions of the various quarters (bedrooms, living quarters, service rooms, lavatories, servants and domestic quarters, storerooms, etc.).
In this survey we will use a definition approaching that of R. Naumann, according to which a patrician house was more luxurious than an ordinary house and a palace was the largest and most magnificent building in a city,30 Patrician houses will be identified by their location in the general city plan; their proximity to the gate, palace, or temple; their dimensions and symmetry; whether their walls were thick enough to support a second storey; whether the plan included service rooms, living quarters, storerooms; the existence of a drainage system; the quality of the floors; and whether the choicest building materials were used. In contrast to the plans of temples and palaces, no uniform plan for patrician houses can be distinguished, even at the same site. Three buildings were erected in close proximity at Megiddo (3066, 3046, 3050) in the same period (Stratum X), but each exhibits a different plan.31 Accordingly, this survey presents only a sample and not a typology of patrician houses or a description of all such buildings so far discovered in excavations.
11. "Palace II", Tell el-'Ajjul. Ancient Gaza II, PI. XLV1.
Palace II at Tell el-'Ajjul is in fact a patrician house that was built directly above the northern wing of Palace I.32 Because the thin brick walls (approximately one metre wide) of Palace II were built in the Egyptian style — namely, without stone foundations — Albright suggested that it had been the residence of the local Egyptian governor at the beginning of the Eighteenth Dynasty (Fig. 11). The building has an inner courtyard (OG) with steps leading to a second storey. There were a number of small rooms alongside the courtyard, including a bathroom (OH) with a plastered floor and
29. See Megiddo II, Fig. 380; if this suggestion is correct, the fragmentary remains of rooms found to the west of Building 4031 are an integral part of the palace complex.
30. See Naumann (above note 24), p. 389.
31. See Megiddo II, Fig. 400.
32. See Petrie (above note II), PlaLe XLVI.
115