The Early Dynastic Palace at Hierakonpolis

The early DynasTic Palace aT hierakonPolis by Renée Friedman and Richard Bussmann Hierakonpolis is one of the central places of Egyptian state formation and plays a key role for investigating the emergence of Pharaonic kingship. The site was occupied during the 4th and 3rd millennia BC, when the idea of kingship formed in Upper Egypt, and later lourished as a regional centre and twin city of Elkab. The Narmer palette, found in a deposit of votive objects in the Early Dynastic temple area of Hierakonpolis, depicts a king for the irst time in a monumental fashion similar to representations current for the next three millennia. The investigation of palatial structures at Hierakonpolis presented below provides an archaeological view on the presence of kingship at the site, often overshadowed by the iconicity of the Narmer palette. Archaeological evidence of Dynastic kings and their Predynastic forerunners is abundant at Hierakonpolis.1 It includes (see Fig. 1): The Predynastic ceremonial plaza at HK29A in the low desert and the contemporaneous elite cemetery HK6 up in the Wadi Abu el-Sufian; the Early Dynastic palace and royal votive objects deposited in the temple area, located in the modern cultivation; the mudbrick enclosure (a.k.a. the Fig. 1 Map of Hierakonpolis with insets of locations implying the presence of central authority. (compiled by R. Friedman) 1 A still useful overview of the evidence is offered by AdAms 1995. 80 Renée Friedman and Richard Bussmann Fort) of king Khaskhemwy, used perhaps for the celebration of the royal sed-festival; and from the temple area, there are royal statues and stelae of the 3rd millennium, and blocks inscribed with royal names dating from the Middle Kingdom to the Ptolemaic Period. The Early Dynastic palace which will be the focus of this paper dates to a period when Hierakonpolis was of primary concern for emerging kings. The wealth of available material from the alluvial town mound (Kom el-Gemuwia) stands in contrast to the poor quality of the information on its contexts and the overall nature and relationship of the individual archaeological features. The original excavators (Quibell, Green and Fairservis) report a series of large structures and relevant inds, but their interpretations are dificult to advance without fresh data from the ield. The discussion below is mainly based on the published reports with a few comments added from re-documentation of excavated objects and access to some unpublished sources. Conclusions on the dating, use and precise function of the palace must remain tentative for the time being, but clariication of some details may help with situating the fragmented record into a more meaningful picture. Antecedents of authority If by ‘palace’ we mean an establishment implying the presence of some central authority either in its architecture or by evidence of centralised manufacture, such establishments have antecedents in the Predynastic occupation in the desert at Hierakonpolis (see Fig. 1). In the midst of the densest cluster of Naqada II settlement debris on the low desert, excavations have revealed a large compound possibly covering over one hectare which remained in use into at least Naqada IIIA, with later visitations. This was bounded on at least one side by a huge palisade wall of wood uncovered for over 50 m of its length (HK29B),2 which clearly extended further in both directions, potentially creating a compound that included the oval ceremonial centre of HK29A and its associated areas for the manufacture of stone vessels, beads and ine lithics.3 In addition, this compound may well have incorporated the so-called stone mound (HK34B), which was suggested to be the remains of cobble stone foundations for an administrative building.4 Although an informative Fig. 2 View of the niched-brick gateway of the palace on the Kom el Gemuwia in 1969 (photo courtesy of the Hierakonpolis Expedition archives) plan cannot be distinguished from its much disturbed remains, associated areas containing speciic types of pottery in vaguely deined room outlines suggest large storage areas and stand out from the mixed debris in the Predynastic town.5 Some idea of what an early Naqada II ‘royal’ residence might have looked like might been gleaned from the elite cemetery at HK6, in the wooden architecture of the mortuary complex encompassing the large and rich Tomb 16 with its surrounding burials of the human and animal associates.6 This suggests a rather unstructured arrangement surrounding a central court, which is perhaps not surprising. Augmenting these architectural traces, installations for the production of beer (HK11C Operations A and B, HK24AB) and the processing of meat (HK11C Square C3‒4) on an industrial scale can be associated with both the elite cemetery and the ceremonial / administrative compound in the Predynastic town. The productive scale of these installations implies the centralised command of substantial economic resources from early Naqada II times onward.7 How these earlier manifestations of authority were translated when the population moved out of the desert and nucleated in the loodplain town in the Naqada III period at the beginning of the state formation process remains to be explored. At present, aside from a temple whose nature and function remains a matter of debate,8 our evidence is limited to the mudbrick structure with the niched gate 5 6 2 3 4 HikAde 2011. FriedmAn 2009; Holmes 1992. HoFFmAn 1982, 130‒131; AdAms 1995, 36, ig. 9. 7 8 HArlAn 1985, 78‒85. FriedmAn et al. 2011. TAkAmiyA 2008; BABA and FriedmAn 2016; BABA, vAn neer and de Cupere 2017. mCnAmArA 2008. The Early Dynastic Palace at Hierakonpolis 81 Fig. 3 Detailed plan of the niched-brick ‘palace’ with wall phases indicated and probable location of Town houses 211 added (compiled from FAirservis 1986, igs. 7‒27; QuiBell and green 1902, pl. 68) located on the residual mound in the midst of the modern village, just to the north of the temple site. The palace-facade structure The palace-façade or niched gate structure and the area to the west of it were irst uncovered in 1969.9 Owing to the political situation in Egypt and other interests of the excavators, the structural remains to the east of the gate could not be investigated until 1981 when they were the focus of a lengthy campaign revealing several phases of use and construction.10 Selected areas in the eastern part of the excavated area were briely reinvestigated in 1988 in an attempt to resolve some outstanding issues, but the subsequent illness and premature deaths of both Michael Hoffman and Walter Fairservis meant that this work has remained unpublished and that detailed reports on the palace in general and the inds from within it were never prepared.11 When discovered, the elaborated niched gate stood in places to a height of 14 courses and retained a coating of white gypsum plaster in and out 9 10 11 Weeks 1971‒72. FAirservis 1986. Subsequent excavations near the structure have been undertaken by E.J. Walters. (Fig. 2). The projecting gate apparently stands at the northeast corner of a façade at least 40 m across, which was also provided with pilasters or buttresses, and probably continues further to the south (Fig. 3); the possibility that the façade continues to the north of the gate remains unexplored. It is the only known niched façade in a non-mortuary context and until recently was the only extensive architectural ensemble of Early Dynastic date that could be considered a ‘palace’. Despite its importance, the summary nature of its publication meant that little more could be said about it, other than to acknowledge its existence.12 The extensive recent excavations in the ‘labyrinth building’ at Buto13 now provide a plausible parallel against which the structural remains at Hierakonpolis may be contrasted and compared. In this contribution, the limited published evidence about the Hierakonpolis structure is briely reviewed and 12 13 E.g. kemp 2006. An exception to this is o’Connor 1992, who in an inluential paper proposed to view the palace structure as a temple based on the purported existence of a central sandy mound. This view can no longer be maintained. HArTung in this volume; HArTung et al. 2012; HArTmAnn 2016. 82 Renée Friedman and Richard Bussmann supplemented by the partial ield notes of the 1988 season14 along with a new study of the mud sealings recovered during the 1969, 1981 and 1988 seasons, which are amongst the limited number of objects that were retained from the excavations. At Buto, as discussed in the following paper, exemplary excavations have revealed three major construction phases along with various subphases. The irst main phase, datable to Naqada IIIC1 (beginning of the First Dynasty), shows a rather organic arrangement of rooms into units; the second, dated to Naqada IIIC2 (early First Dynasty) has rooms arranged around a central courtyard; and the inal construction phase in Naqada IIID (end of First Dynasty), when the area was completely rebuilt and a new, more formal structure was erected, is comprised of three distinct sections for which plausible functions have been proposed. It is with this last phase that the structural remains at Hierakonpolis seem to bear the closest resemblance. By contrast, the summary discussion and idealised stratigraphy of the Hierakonpolis ʻpalaceʼ (Fig. 4)15 suggest that after an initial phase of construction, which included the niched gate with façade and walls marked only Y on FAirservis’ 1986 plans (see Fig. 3), there were three to four subsequent modiications of greater or lesser extent. Walls designated as Y1, Y2 and Y3 were attributed to building phases within the use-life of the ‘palace’ proper, while X walls were attributed to later reuse and rebuilding. According to the excavator, the walls marked only Y retained (more or less) their original form and were used throughout the formal history of the structure.16 It is unclear on what basis a Y (with or without additional number) attribution was given to certain walls,17 since the levels reached during the 1981 campaign in the various excavation squares are nowhere provided, but were clearly not uniform. Unfortunately, taking these phase attributions at face value and using the Y walls (Fig. 3 in dark grey) to give an indication of the initial plan is not particularly illuminating.18 All one can assume is a structure with a façade at least 40m 14 15 16 17 18 Diary and recording forms of M.A. Hoffman and J.O. Mills; ceramic records of B. Adams; report of M. Hayes; 1988 report to the EAO, all unpublished and on ile with the Hierakonpolis Expedition. FAirservis 1986, 14‒15, ig. 28. FAirservis 1986, 14‒16. Various building techniques and brick sizes were noted but it is unclear how these relate to the wall phases, see FAirservis 1986, 5‒9, ig. 6. There are no Y1 walls marked on any of the plans in FAirservis 1986. Fig. 4 Idealised stratigraphy of the palace and provisional dates for the phases (after FAirservis 1986, ig. 28) long, with a somewhat asymmetrical projecting niched gateway, through which an opening led into a small court. A boundary wall along the north can be proposed, running from the north edge of the gate eastward for an unknown distance, but this wall seems to be of no greater thickness than many other internal walls. The southern boundary wall was not located and may have been destroyed by later levelling. Interior walls are poorly represented, but their alignment with the gate indicates a certain amount of planning from the outset, rather than organic structural development. The dating of this initial phase was always left vague and mainly circumstantial. Based on the complexity of the gate’s niched brickwork, an early First Dynasty date was suggested, citing parallels with the Neithhotep mastaba at Naqada and others in the reign of Aha.19 From the published evidence, it appears that the underlying stratum on which the gate was built (Level 7 on Fig. 4)20 should date to Naqada IIIA‒B, based mainly on the presence of net-painted cylindrical jars.21 The strata to the west of the façade were described as made of ine clay above which various layers were added to support a paving of ield stones. Artefacts found in the layers above this stone paving in 1969 were attributed to the Second to Third Dynasty, in line with the 19 20 21 Weeks 1971‒72. Phase 4 of FAirservis 1971‒72, ig. 19 is apparently equivalent to wall phase Z, stratum 7a on Fig. 4 here. See also FAirservis 1971‒72, ig. 17; for dating see HendriCkx 2006. The Early Dynastic Palace at Hierakonpolis pottery and other objects from the thick overlying stratum (stratum 3 of the 1971‒72 report), providing a terminus ante quem for the gate’s construction.22 The sealings apparently found directly upon the stone paving by the gate are re-examined below. Their date is dificult to narrow down to a speciic period within the Early Dynastic period, and they provide little help for dating the building. It is not clear whether or not the basal layers adjacent to the east face of the niched façade were reached during the 1981 campaign, thus there is no further direct conirmation for its dating. Nevertheless this early date can be indirectly supported by inds made in 1988 by the niched platform (Square 15N2W) at the eastern end of the excavated area, if this feature is considered part of the overall complex. Seemingly separated by a large plaza from the main complex of walls associated with the niched gate, at the eastern end of the excavated area there were further rooms fronted by an L-shaped platform of yellow compacted clay, which was apparently also niched (Figs. 3, 5). FAirservis reconstructed it as a loading dock for the magazines and administrative area to the east.23 Although there is no evidence for its function during the earliest phases, it is tempting to compare its location within the complex to the administrative area identiied at Buto, where sealings from the Naqada IIIC1 phase were found.24 FAirservis attributed all or part of the platform(s) to his Y2‒3 stage, although the foundations extended down to just above the water table of 1981, already suggesting this was a very ancient part of the palace complex. This was conirmed by excavations at the northeast corner of the platform (N‒N1) in 1988, which revealed a deposit of pottery (Feature 61) abutting the platform’s edge, near its base (Figs. 5‒7). This collection of pottery, which involved several whole or reconstructible vessels, can be dated to Naqada IIIB‒IIIC1, equivalent to the early part of the First Dynasty.25 A selection of the pottery types in this deposit is illustrated in Fig. 6.26 It included many straw tempered bowls with laring rims (Fig. 6e), often found stacked into one another, 22 23 24 25 26 Weeks 1971‒72. FAirservis 1986, 8, igs. 7c, 22‒23. See HArTung in this volume. The assemblage was attributed to Narmer-Djer by B. Adams in her 1988 summary of the ceramics. Drawings of the pottery that was retained from these excavations were made by D. Raue in 1999. Frequency data are taken from B. Adams’ ceramic inventory forms on ile with the Hierakonpolis Expedition. 83 and a large quantity of low walled oval platters (not drawn). Straw tempered pottery predominated, but one-third of the total assemblage was composed of marl wares, mainly jars, streak burnished bowls and just a few cylinders with decoration below the rim (Fig. 6a‒c). This was followed in frequency (5% of the total assemblage) by shale tempered restricted bowls, many with preiring potmarks on the exterior upper body (Fig. 6d). This fabric and shape has parallels at several sites dating to Naqada IIIA‒ C1.27 Untempered Nile silt was rare, and involved mainly streak burnished bowl fragments. The consistency of the assemblage suggests that the sealing (sealing 465 discussed below) attributed to this ind unit is intrusive. It was discovered in the sherd yard amongst pottery collected when the unit was cleaned prior to stratigraphic drawing. It could easily have fallen in from above, especially as sealing 464 and other uninscribed fragments were recovered nearby in a large circular feature cut into the north end of platform N1 (Feature 62) and sealed beneath a fallen wall (see Fig. 5).28 Immediately below this pottery collection, walls of an earlier phase were observed, which appeared to have been intentionally shaved down and illed in to provide a base possibly for the construction of the irst Y walls. Exposure was insuficient to determine the trajectory of these walls. Above the pottery cluster, a layer of dark brown midden, a maximum of 20 cm thick, contained pottery dated by Adams to mid to late First Dynasty, presumably equivalent to Fairservis’ stratum 5 and wall phase Y2. Built upon this stratum was the wall lanking the east side of platform N1, which Fairservis apparently attributed to his Y3 phase. However, it is not clear how this wall relates to those encasing the east side of platform N, which appears to be later. As will be discussed further below, the phasing of the walls in this area is extremely problematic, and it should be noted that during the 1988 excavations it was observed that the Y2 wall north of the N1 platform was actually built on top of it (the platform continued below the wall fall) and thus may postdate the formal use of the structure. The evidence for the Y2 walls must therefore be treated with care. However, the minimal depth of the deposit presumably associated with them, at least near the platform, might suggest a certain 27 28 See for example, kopp 2006, pl. 7.284. The information about the 1988 excavations beside the platform has been gleaned from the dairies of M.A. Hoffman and J.O Mills along with the unit/level forms on ile with the Hierakonplis Expedition. 84 Renée Friedman and Richard Bussmann Fig. 5 Detail of Square 15N2W and surroundings showing the location of the 1988 excavation features and results Fig. 6 Pottery from Square 15N2W, Feature 61 (drawings by D. Raue) The Early Dynastic Palace at Hierakonpolis amount of levelling of structural features in preparation for the Y3 construction in what appears to have been a major phase of re-construction. It is in the Y3 phase that the Hierakonpolis structure bears the most resemblance to Buto, but exposure is still insuficient and obscured by later constructions (Fig 3, walls shaded with light grey). Nevertheless, one can potentially recognize the three component parts as revealed in the Buto complex (see Fig. 3): 1) There are the so-called reception area and service rooms forming a block immediately to the east of the niched gate.29 2) A magazine or production area appended to the north of this. 3) To the east was an administrative area including the niched platform, though how it functioned at this time remains to be seen. This administrative sector was separated from the reception block by an open, later sand illed area, but appears to have included a stone-lined passage to a plaza. In 1988, a small gate, reportedly niched, was revealed, which apparently opened onto this open space between these two components,30 but unfortunately no plan showing its exact location or coniguration is currently available. As excavations were apparently undertaken by Fairservis in the southern halves of squares 16N2W and 16N3W, presumably the gate should be located in this vicinity.31 More detailed comparisons become more speculative. The Buto structure was entered through a nondescript opening in its west wall, taking the visitor along an indirect route to the main audience 29 30 31 Whether manufacturing can be said to have taken place in the north side of this block, paralleling Buto, is unclear. Manufacturing activities are attested by numerous tools for stone vessel making, spindle whorls for textiles and various lint tools, but these are reported in no particular concentrations and do not appear to be clustered in any particular phase. The rooms were reported as generally clean, although it is hard to believe the history of this structure was less complex than that at Buto, with rooms changing usage over time, see HArTung et al 2012, HArTmAnn 2016. 1988 Report to EAO. Excavations in 16N2W produced the feet of a faience igurine, probably a standing boy, according to the registration book. Further information about it is unavailable. Such igurines are notably absent from the palace proper. A concentration of them was found in 15N8W Room J2, outside and to the west of the southern end of the niched façade (FAirservis 1986, ig. 18), but their relationship to the palace is unclear. 85 hall near the centre of the reception block.32 At Hierakonpolis, access via the niched gate to what can be argued to be the equivalent hall (Fig. 3: Room A2, Room A7?, though almost entirely unexcavated), seems more direct. An opening in the southeast corner of the court just behind the gate seems to have provided bent access, but at no great distance; however, the original coniguration of the walls is far from clear. Alternative access through the long passage or court parallel to the southern façade can also be envisioned, potentially more in line with the layout seen in the token palaces of the funerary monuments at Abydos, especially that of Khasekhemwy.33 If we disregard the differences in the entrance ways and accept an audience hall in a central location (i.e. room A2) for the Hierakonpolis palace, there is a striking similarity with Buto in the overall layout of the room and surrounding features. This includes the broad courts/passages behind the respective west (HK) and north (Buto) façade walls, the complex of rooms to the left of the audience hall, and particularly the sunken features in both structures. At Buto, this subterranean feature is located near the entrance,34 while at Hierakonpolis, it is just to the left of the audience area. Here (Fig. 3: Room A5), Fairservis discovered a subterranean chamber, 1.95 m long and 0.63 m deep. In it was a stepped and sloping deposit of clay on which were placed more than 20 straw tempered platters, some containing animal bones including a bovid skull, a large vat, a stone vessel, copper objects and other items.35 At some point in its history, it apparently fell out of use and was covered over with a beaten clay loor.36 The function and purpose of these features at both Buto and Hierakonpolis remain a question. The sandstone and sand on the loor of Room A5 might suggest it was originally meant to hold water for bathing or puriication. Alternatively, it may have had a cultic function, as originally suggested by Fairservis; however, it contained none of the special pottery, such as jar stands, observed in conjunction with the subterranean feature at Buto.37 32 33 34 35 36 37 HArTung in this volume. Cf. BesToCk 2008, ig. 4. See HArTung in this volume. FAirservis 1986, 12, ig. 20b‒c, reproduced in moeller 2016, igs. 4.28‒29. Adams states the platters found in Feature 61 were similar to those found in the subterranean chamber, but as no detailed drawings of these platters are available, it is hard to determine whether this has any chronological signiicance. See HArTmAnn 2016, ig. 5. 86 Renée Friedman and Richard Bussmann Fig. 7 View of pottery deposit in 15N2W, Feature 61, during excavation in 1988 (photo by J.O. Mills, courtesy of the Hierakonpolis Expedition archive) The dating of the Y3 construction phase at Hierakonpolis has been highly inluenced by the sealing of Qa’a (sealing 309 discussed below). This was found in the north side of the complex in Room D in 17N4W (see Fig. 3) reportedly in situ in an ashy deposit with other sealings of similar or later date on a loor of phase 4 (i.e., the loor on which the Y3 walls were built and subsequent levels).38 Initially, the main Y3 phase of the palace was considered to date to the Second Dynasty based on the pottery and sealings found in 1969 and 1978.39 However, with the discovery of this sealing, its date was revised backwards, so that it was later suggested that the palace fell out of formal use at the end of the First Dynasty when the Qa‘a seal was deposited. After that, the structure was abandoned to so-called squatter habitation. Of course, the opening of a seal does not necessary correspond to the time of sealing itself. Although the type of pottery found in direct association with the sealings remains unknown, the idealised stratigraphy for this phase includes Meydum bowls, which, while present from the reign of Qa’a in an early form, do not become common until the late Second Dynasty.40 Further investigation in 1988 of Room D, where the seals were found, revealed several subphases of loors, platforms and silos, so the phase attribution of this inluential sealing should be treated with caution. A more detailed stratigraphic record is only available from the eastern part of the complex in the so-called administrative area. In 1988, excavations in square 15N1W in and around the circular silos in Room R9-R10 were continued down from the 38 39 40 Brookner 1986. FAirservis 1983. For pottery dating see HendriCkx et al. 2002; rAue 1999, 180‒183. level reached in 1981. Although exposure was limited, the records show that room walls (attributed to Y2‒3) had been built on 15‒20 cm of soil containing Meydum bowls (although there are no detailed drawings of the types), and interior lip bowls,41 which mark the late Second Dynasty at many sites, the geographically closest being Elephantine.42 Thus, the walls and silos in this particular room at least were potentially built in the late Second Dynasty, and preserve evidence of Third Dynasty activity as conirmed by the sealing (number 459) found within the silo in Room R10.43 Earlier walls and a hearth were found below the silos, but exposure was insuficient to determine the date. The walls of the rooms were later modiied (a low passage was cut between Rooms R9 and R10), but the initial construction does not appear ad hoc. While these chambers might be later additions to the complex and this limited stratigraphic window might not be applicable to other areas, it does suggest that at least some of the walls attributed to the Y3 phase were built in the late Second Dynasty. Considering Khasekhem / Khasekhemwy’s interest in the site of Hierakonpolis, it is perhaps not surprising that building works here were carried out during his reign. The unique coniguration of his ceremonial enclosure (the Fort) at Hierakonpolis, with its projecting gateway and nearly square dimensions (see Fig. 1),44 may well have been inluenced by or be a relection of this – by then rather ancient – palace structure. For how much longer the palace remained in formal use, and how one can determine this, are unresolved questions. The depression immediately to the east of the R series of rooms in 14N1W, called Room S, is probably the result of the sondage made by green in 1899 within one of the rooms of the town houses labelled 211.45 Whether the 211 structures are later constructions over this part of the palace or represent use or reuse of the structure cannot be determined,46 but from the dimensions it is quite possible that the west walls of the larger room are 41 42 43 44 45 46 Hays report 1988 on ile with the Hierakonpolis Expedition. rAue 1999, 181, ig. 37.2‒6. This seal was mistakenly read as the name of Semerkhet by the excavators, leading them to attribute the rooms to the late First Dynasty. FriedmAn 2007. QuiBell and green 1902, 18, pl. lxviii; AdAms 1974, 76‒77. Pottery illustrated in QuiBell and green 1902, pl. lxix.20‒ 23, from the ‘Town’ appear to come from in and partly below a room in House 211. Among them is an early form of Meydum bowl that could date to the late Second Dynasty. The Early Dynastic Palace at Hierakonpolis in fact the same as the east walls of the R series of rooms. If this is the case, it gives further indications of the layout of this eastern sector. Unfortunately, the seals associated with the building are too fragmented to allow a more deinite dating other than within the Early Dynastic period.47 Further evidence of late Second to early Third Dynasty activity was found in the adjacent square 15N2W on top of platform N1, which seems to have lost its original function at this time and was more or less at loor level. Walls were built on top of it and a circular pit (Feature 62) carefully cut into it was found to contain sealing 464, other uninscribed fragments and bread pots. The collapse of the wall built over the platform sealed this pit along with four successive living loors containing grinders and bread pots that were attributed to the squatter habitation. Whether this is actually the case, or whether this debris relects only a functional change of rooms as noted at various times at Buto remains to be seen. However, it would seem that at some point in the Third Dynasty, the palace was allowed to deteriorate. Exactly when the palace ceased to function as originally planned and was left to more organic reuse, as the many X phase cooking installations throughout and around it suggest, cannot be determined from the available records. While the nature of the evidence means we can only speculate, certain similarities between the Naqada IIID palace structures at Hierakonpolis and Buto suggest a vision of how a palace / administrative structure should be conigured and what features it should contain.48 The presence of the niched façade and respect for the probable antiquity of it and selected other features may have meant the builders of the Y3 phase walls had to modify this ideal. Only further excavations can tell whether this formalised plan represents an outgrowth of the earlier arrangement already present at Hierakonpolis or is a new relection of the organization of power 47 48 The best preserved example is CG14673 showing the signs j (M17), s (S29), S (N37), an inverted mA (U1), and n (N35). Three additional sealings associated with house 211 were transferred from the Faculty of Oriental Studies, Cambridge, to the Museum of Archaeology and Anthropology, Cambridge, and registered there under the numbers 2005.507 (former LE 156), 2005.513 (former LE 157) and 2005.514 (former LE 158). Although moeller 2016, 96‒102, sees residential, oficial and cultic activities present in the early palace, much like that known in later governors‘ residences, she states there is no hierarchy among the rooms and that the interior of the palace was as unstructured as the settlement surrounding it. The present study suggests that at least for the Y3 phase, this is not the case. 87 and authority. While this review has presented more questions than it answers, the quality of some of the archival material available for this paper suggest that future analysis of the full record by those in possession of it should be able to address some of the outstanding issues surrounding this important structure. Administrative practice The American team led by W. Fairservis and, later, M. Hoffman discovered fourteen clay sealings in the palace area: four in 1969, seven in 1981, and three in 1988. The objects were registered with the antiquities magazine at Elkab. Two of the 1969 and one of the 1981 sealings were later transferred to the Museum of Civilization in Cairo. Weeks and Brookner published ten of the fourteen sealings in a preliminary fashion.49 Some of the sealings discovered by J. QuiBell and F. green50 might relate to the palace but are excluded from the discussion below because their exact ind location is to date not yet fully established (but see above for house 211). The number of sealings is surprisingly low when administration is assumed to be the main purpose of the building. Perhaps the relevant administrative parts of the palace are located outside the excavated area or are enclosed in the unexcavated layers below water table. Five of the published items are deinitely peg sealings, one is a jar sealing, the rest could be either peg, jar, or fabric sealings.51 The corpus has not received much attention so far. It is worthy of being re-published in detail as evidence, however faint, of the administrative practices performed in the building and, more generally, outside the cemeteries of Abydos and Saqqara, from where the majority of known sealings derive. The 1969 group Sealings no. 15, 19 and 38 were found “lying within the gateway, above the stone loor, in addition to several fragments of stone bowls, lint knives, and an uninscribed slate palette”.52 They were registered on 8, 10, and 15 February 1969 as having been found in quadrants 17N6W and 17N7W. A fourth item, no. 39, was registered on 16 February but is neither mentioned nor published in the preliminary report. The register book does not offer information on the ind context and the association of 49 50 51 52 Weeks 1971‒1972; Brookner 1986. QuiBell 1900; QuiBell and green 1902. For typology and larger recently published sealing corpora, see von Pilgrim 1996, 234‒273; EngEl and müller 2000; grATien 2001; päTzniCk 2005; Wegner 2007, 299–361. Weeks 1971‒1972, 31. 88 Renée Friedman and Richard Bussmann no. 39 with the rest of the group is not self-evident. Quadrants 17N6W and 17N7W embrace the entire niched gateway and parts of the adjacent wall running southeast. Week’s description suggests that the sealings were found on the original loor rather than in one of the later layers associated with the secondary walls in the area. He classiies numbers 15 and 19 as “jar seals” and number 38 as a “document seal”. However, 38 is deinitely a peg sealing used for sealing a door or basket. The drawing of 15, which was transferred to Cairo, shows a spout-like feature characteristic of peg sealings. The sealing type of 19, also sent to Cairo, is obscure on the drawing, while 39 is deinitely a peg sealing again. Although the evidence is not entirely conclusive, the sealings might have originally been attached to the door that locked the niched gateway. Sealing 15, transferred to Cairo (Fig. 8): Find number 69-77, entry in register book on 8 February 1969, found in quadrant 17N6W according to register book or quadrant 17N7W according to the excavator, published by Weeks 1971‒72, 31, ig. 38d (not 39d). Fig. 8 Sealing 15 (after Weeks 1971‒72, ig. 38d) From the published drawing, the sealing is 3.8 cm high and 5.0 cm broad. Weeks does not suggest a translation of the inscription. The three signs in the middle might be mA (Gardiner sign U1), DA (U29) and the human eye (D4). päTzniCk reads the group wDA-mA as part of a name or title Hrj wDA.t mA.t “overseer of the magazine”.53 The translation suggests a rather clear cut administrative task, but it is neither evident what the individual words mean nor what the material correlate of an wDA.t-“magazine” would have meant in archaeological terms. Alternatively, the sign in the middle is a debased version of Hr (D2) and the group is to be read HrjwDA “controller”.54 53 54 päTzniCk 2005, cat. nos. 12, 20, 119, 156, 264. päTzniCk 2005, 102‒104. Sealing 19, transferred to Cairo (Fig. 9): Find number 69-103, entry in register book on 10 February 1969, found in quadrant 17N7W according to register book or in quadrant 17N6W according to the excavator, published by Weeks 1971‒72, 31, ig. 38 a (not 39a). Fig. 9 Sealing 19 (after Weeks 1971‒72, ig. 38a) The sealing is depicted as being 6.2 cm high and 12.0 cm long. Both inscriptions derive from the same part of the cylinder seal. The sealing pattern is unusual in the corpus from Hierakonpolis as the inscriptions do not refer to names or titles, but apparently to offerings alternating with representations of the standing king.55 The inscription on the left might read sT(.t) “beer (jar)” followed by the determinative of a beer jar and an ideogram stroke, unless this latter group represents a measurement ds 1 “one jar”. Alternatively, one could consider a reading sTj “smell, perfume”, or stj “to libate” as a description of an activity. The bowl represented above the group could function as the determinative of a preceding word or be another measurement.56 The signs behind the king could be Ss “robe, linen” or “alabaster”, or belong to an altogether different word ending in zS. The group on the right hand side is unclear. The signs to the very left might be part of a larger representation of the king seated on a throne.57 Overall, the pattern belongs with the group of the rather few, and probably prestigious, Festsiegel showing the king in some ritual activity. It is remarkable to ind an example outside the royal cemeteries. Perhaps the seal was used during a 55 56 57 Weeks 1971‒72, 31, ann. 9, says that kAplony doubted that the drawing was correct. kAHl 1994, 794‒795 with a range of possible readings of the sign W10. Most of them refer to substances or offerings. For recently excavated examples of the First Dynasty, see engel in: dreyer et al. 2011, 62‒72 and 76‒77, igs. 20 and 21 for a representation of the king seated on a throne. For previously known examples of Festsiegel, see kAplony 1963, nos. 211, 237 (king on throne), 364, 365. The Early Dynastic Palace at Hierakonpolis royal ritual performed in the palace, but there is no evidence to back up this hypothesis. Sealing 38 (Fig. 10): Find number 69-168, entry in register book on 15 February 1969, found in quadrant 17N6W, published by Weeks 1971‒72, 31, ig. 38c. 89 Third Dynasty, while the internal patterning of Aa1 may point to the later Second Dynasty as a date for the manufacture of the seal.61 Sealing 39 (Fig. 11): Find number 69-169, entry in register book on 16 February 1969, ind location within Hierakonpolis not known, unpublished. H: 4.9 cm, B: 6.2 cm, D: 2.4 cm. Munsell 7.5YR 3/1. The lower side is broken off. The central area of the reverse side is eroded. The preserved part is porous and not level. Perhaps it was pressed onto a mudbrick. At the bottom, a peg was impressed with a diameter of 2.0 cm. Fig. 10 Sealing 38 (photo: R. Bussmann, drawing: R. Bussmann and C. von Elm) H: 2.3 cm, B: 3.4 cm, D: 1.3 cm. Munsell 7.5YR 3/1. The upper, lower and left sides are completely preserved, the right side is broken off. A well rounded peg, circa 1.8 cm broad, and a cord wound around it are impressed on the reverse side. The two sealings on the front run parallel to the impression of the peg. From left to right, the top impression shows a circular sign with internal patterning, perhaps x (Aa1). The following signs might belong to a group Htp (R4), t (X1) and p (Q3). The remaining signs can be identiied as Hr (D2),58 s (S29), and t (X1). The signs might belong to one of the common names with the element Htp popular in Early Dynastic inscriptions, including those of Elkab.59 The inscription is preserved to a maximum height of 0.9 cm and a maximum length of 2.1 cm. The bottom sequence shows again Hr followed by s, nfr (F35) and x. The maximum preserved height is 1.0 cm. The maximum preserved length is 1.9 cm. As the signs Hr and s in both sequences overlap, they may represent the same sequence of the inscription. However, the surrounding signs differ and the impression does not stem from the same part of the cylinder. The group nfr and x features also on an impression from Elephantine. päTzniCk suggests a reading nfr-x.t-zA and believes that it is an epithet of local oficials.60 Palaeographically, the sign D2 seems to belong into the later First to early 58 59 60 The shape of the sign resembles outline b as deined by regulski 2010, 245. She argues that it is typical of the First Dynasty. regulski 2009. päTzniCk 2005, cat. no. 7. Fig. 11 Sealing 39 (photo: R. Bussmann, drawing: R. Bussmann and C. von Elm) The inscription on the front runs in a perpendicular direction to the impression of the peg. Due to the curved orientation of the surface, the signs in the middle part are not preserved and the surface has been rubbed off. In the broken-off area, several eroded impressions of a cord are preserved. The signs j (M17) and s (S29) are visible to the left of the eroded area. To the right, an individual is depicted seated on a stool and stretching his or her arm out to where originally an offering table must have stood. Recently excavated seals and seal impressions decorated with the offering table scene are dated to the irst part of the First and the late 61 regulski 2010, 244, 271. 90 Renée Friedman and Richard Bussmann Second Dynasty.62 The remaining signs are unclear. A thin line runs above and below the inscription indicating an original height of 2.3 cm. The maximum preserved length of the inscription is 5.0 cm, suggesting a diameter of the original cylinder of at least 1.6 cm. again at a later date. It is probably safest not to push the date of the layer back into the mid-First Dynasty and, from current evidence, rather to assume an occupation of this part of the palace during the Early Dynastic Period to early Old Kingdom, and perhaps late Second Dynasty as suggested above. The 1981 group The seven sealings, 306‒311 and 337, were found in room D within the palace (see Fig. 3); sealing 306 “on a habitation loor of the uppermost of the three Archaic levels in 16N-4W”, the others in “a single area of charred earth of the main habitation layer of Room D level (4) of the Archaic period in 17N-4W, approximately 5 meters north of the irst sealing ind”.63 Whatever the nature of the different levels deined by Fairservis really was, it seems that 306 belongs to a later layer, layer 3 in Fairservis’ terminology, than the rest of the group. Sealing 306 has been transferred to Cairo. Information on this object can be obtained from Brookner’s publication only. Brookner classiies the sealings as jar sealings and concludes that the “number of jar sealings in this small charred earth layer is the result of a simultaneous opening of jars, and casting into a refuse pit of their lids”. However, sealings 308, 311, 337 and possibly 310 are peg rather than jar sealings. The big sealing 309 might have been attached to a jar or to a sack or to a peg, and only 307 was squeezed into the mouth of a vessel. The sealings are coloured dark brown to grey with only a few reddish areas. They do not seem to have been exposed to great heat and were therefore mixed with the charred earth only after the latter had been burnt. The entire group may have been discarded simultaneously, for example during the clearing of an adjacent room, but the individual sealings could well have been broken at different times. For dating purposes, the name of king Qa’a on sealing 309 provides, at irst sight, a deinite date for layer 4. However, while the original cylinder was probably carved in the reign of Qa’a, it might have been used later, the sealing broken only after years of storage and the broken sealing discarded Sealing 306, transferred to Cairo (Fig. 12): Find number 81-17, entry in register book on 9 February 1981, found in quadrant 16N4W, room D, published by Brookner 1986, 24, ig. 17, ield registry page 6. 62 63 For Tell el-Iswid, see regulski 2014, 236, ig. 6. In Elephantine, a limestone cylinder seal with the offering table scene carved into it was found in the “Oststadt”, see päTzniCk 2005, cat. no. 391. Some of the seal impressions may also be part of the offering table scene, for example päTzniCk 2005, cat. nos. 425, 451, 497. The scene features also on other seal impressions from Hierakonpolis, see below sealing 311 and BussmAnn 2011. Brookner 1986, 24. Fig. 12 Sealing 306 (after Brookner 1986, ig. 17) Brookner suggests reading the words and phrases of the inscription as jmj-rA xtm.w “overseer of the sealers”, snTr “incense” and Hm “priest”. The group Hm might be part of a name rather than of a title.64 Sealing 307 (Fig. 13): Find number 81-126, entry in register book on 7 March 1981, found in quadrant 17N4W, room D, published by Brookner 1986, 24‒25, ig. 13, ield registry page 4. H: 5.8 cm, B: 2.7 cm, D: 2.8 cm. Munsell 2.5YR 3/1. The left and bottom sides are broken off. The reverse side has a coarse surface texture. The projecting bulb in the centre probably derives from squeezing the clay directly into the mouth of a vessel, without using a piece of linen or leather between the jar and the sealing. The upper inscription shows what probably is the lower part of the Neith standard (however, traces of the cross are not preserved), a ram (Brookner identiied it as an antelope) and a rectangular sign, perhaps S (N37). A thin line runs above and below the inscription. The surface behind the S-sign is squeezed and the two parallel curved signs and the sign with three triangles do not connect immediately to the proceeding ones. It is, in fact, not entirely clear whether these two signs are hieroglyphs or simply pieces of squeezed mud. A possible parallel is a seal from the enclosure of Khasekhemwy at Abydos whose inscription, according to peTrie and kAplony, reads Hm-Nt-$nmw “servant of Neith 64 kAplony 1963, 568, s.v. @m. The Early Dynastic Palace at Hierakonpolis and Khnum” followed by a seated individual before an offering table.65 The maximum preserved height of the inscription of 2.0 cm roughly represents the original height of the cylinder used. The maximum preserved length of the inscription is 1.6 cm. 91 height is 1.5 cm, the maximum preserved length 1.8 cm. The bottom inscription cuts into it. The signs are bordered by a thin line running above them. The maximum preserved height is 0.9 cm, the maximum preserved length 1.2 cm. The top inscription shows a crescent (perhaps the horn of a ram?) and a circular sign with crossed internal pattern (x or njwt (O49)). The inscription cuts into the inscription in the middle. It is preserved to a maximum height of 0.9 cm, and a maximum length of 1.6 cm. Fig. 14 Sealing 308 (photo: R. Bussmann, drawing: R. Bussmann and C. von Elm) Fig. 13 Sealing 307 (photo: R. Bussmann, drawing: R. Bussmann and C. von Elm) The bottom inscription shows the upper part of a H (V28), a D (I10), and a bird sign, perhaps the chicken quail (G34) or the vulture (G1). A thin line runs above the group. The maximum preserved height of the inscription is 1.4 cm, the maximum preserved length is 2.6 cm. It cannot be established whether both inscriptions derive from one and the same cylinder seal. Sealing 308 (Fig. 14): Find number 81-127, entry in register book on 7 March 1981, found in quadrant 17N4W, room D, published by Brookner 1986, 25, ig. 13, ield registry page 4. H: 3.5 cm, B: 6.0 cm, D: 1.4 cm. Munsell 2.5YR 3/2. All sides are broken off. The reverse side shows a slightly curved, eroded surface, perhaps the impression of a peg. The inscription in the middle runs parallel to the peg. It includes the Neith standard, x and Hr. The palaeography of the sign Hr (D2) seems to be typical of the First Dynasty.66 Its maximum preserved 65 66 peTrie 1903, 9, 29, pl. 12, no. 275, found in “house rubbish” (= kAplony 1963, 1159, ig. 509). regulski 2010, 245, outline b. Sealing 309 (Fig. 15): Find number 81-128, entry in register book on 7 March 1981, found in quadrant 17N4W, published by Brookner 1986, 25, ig. 13, ield registry page 4. H: 4.5 cm, B: 4.5 cm, D: 3.8 cm. Munsell 2.5YR 3/2. All sides are broken off. The reverse side shows the impressions of four rows of strings. The whole piece is rather heavy and could have been attached to a jar, a sack, or a peg. The irst and third columns of the inscription show a serekh with the name OA-a, particularly well preserved in the third column where the falcon on the serekh is, however, eroded. The signs in the middle column might be m (G17) and xnt (W17) suggesting a reading as jmj-xnt “chamberlain”, a title possibly referring to somebody responsible for clothing and dressing the king.67 klAesens suggests the reading zA-wr “great phyle” for a similar string of signs on a sealing from tomb 3505 at Saqqara.68 Perhaps both impressions derive from one and the same seal, but the measurements of the Saqqara 67 68 Jones 2000, 281‒282; for Second or Third Dynasty evidence, see the three ink inscriptions from the Step Pyramid published by lACAu and lAuer 1965, 17, no. 23, ig. 26, pl. 13 (= kAHl 1994, cat. no. 2305 and 2306). In emery 1958, 33, pl. 37.3 (= kAplony 1963, 1125, Abb. 244). 92 Renée Friedman and Richard Bussmann sealing are not provided. A third option is to understand the signs as @rw +r (M37) “Horus Djer”, in which case the names and titles of kings Qa’a and Djer would alternate on the seal.69 However, one would expect that both names were written either with or without serekh, and the names of the two kings do not feature together in other inscriptions. The reading as jmj-xnt is perhaps the most plausible. the peg impression. The signs are only fragmentarily preserved. The maximum length of the inscription is 1.4 cm, the maximum preserved height 0.9 cm. Fig. 16 Sealing 310 (photo: R. Bussmann, drawing: R. Bussmann and C. von Elm) Sealing 311 (Fig. 17): Find number 81-221, entry in register book in March 1981, found in quadrant 17N4W, published by Brookner 1986, 25, ig. 13, ield registry page 4. Fig. 15 Sealing 309 (photo: R. Bussmann, drawing: R. Bussmann and C. von Elm) A broad bar runs above the group demarcating the upper end of the inscription. The maximum preserved height of the inscription is 3.7 cm, the maximum preserved length is 4.6 cm. The original cylinder was probably about 5.0 cm high and thus larger than the other seals discussed here perhaps because it belonged to a royal rather than a local oficial. The inscription is cut by another impression at the bottom which is hardly preserved. Sealing 310 (Fig. 16): Find number 81-129, entry in register book on 7 March 1981, found in quadrant 17N4W, published by Brookner 1986, 25, ig. 13, ield registry page 4. H: 2.0 cm, B: 2.5 cm, D: 1.1 cm. Munsell 7.5YR 3/1. All sides are broken off. The reverse side bears the impression of a wooden peg of considerable diameter. The inscription on the front runs parallel to 69 Compare, for example, the list of royal names of the First Dynasty on a sealing from the tomb of Qa’a at Abydos: dreyer in: dreyer et al. 1996, 72, ig. 26, pl. 14b‒c. Fig. 17 Sealing 311 (photo: R. Bussmann, drawing: R. Bussmann and C. von Elm) H: 6.0 cm, B: 5.7 cm, D: 2.4 cm. Munsell 7.5YR 3/1. The bottom and right sides are broken off. On the reverse side, the upper part and the lower left hand side corner are broken off. The surface bears impressions of a string wound around a peg and from there falling down and of the wall which held the peg. Two inscriptions run almost parallel to one another and parallel also to the peg on the front side. The Early Dynastic Palace at Hierakonpolis The upper inscription shows an individual seated on a stool and stretching an arm towards a table with offerings.70 The stroke behind the wig might be a illing element or belong to the wig. The surface to the right of the seated individual turns at an angle. Although there are no signs of an interruption, the seal might have briely lost contact with the clay and parts of the inscription are lacking. The following group is composed of s, H, st (Q1) or rwd (O40), and the crossed sticks of the Neith standard. The group of staircase plus standard (of Neith?) occurs on a sealing from Abydos but the reading of it is uncertain.71 The remaining signs could be illing elements. The inscription is framed by thin lines running above and below the inscription indicating the original height of the inscription as 2.4 cm. The maximum preserved length of the inscription is 5.0 cm. The original cylinder, therefore, had a minimum diameter of 1.6 cm. The signs of the lower inscription are unclear. The group in the middle might belong to zS (Y3). A thin line runs above the inscription. The maximum preserved length of the inscription is 3.7 cm, the maximum preserved height 1.2 cm. There is no clear indication of whether the inscriptions derive from one or two different seals. Sealing 337 (Fig. 18): Find number 337, entry in register book on 7 March 1981, found in quadrant 17N4W, published by Brookner 1986, 24, ig. 13, ield registry page 4. Fig. 18 Sealing 337 (photo: R. Bussmann, drawing: R. Bussmann and C. von Elm) H: 3.6 cm, B: 4.4 cm, D: 2.3 cm. Munsell 7.5YR 3/1. Only the lower left part is originally preserved. 70 71 See comment on sealing 39 above. kAplony 1964, 23‒24, pl. 67, no. 2 at the bottom. 93 The reverse side bears the impression of a wooden peg of 5.0 cm in diameter and of a string wound around it. The inscription runs parallel to the peg impression on the front side. The signs mw (N35A) and H are clear but not suficiently speciic a sequence for translation. The remaining signs are unclear. A thin line runs along the bottom of the inscription. The maximum preserved length of the inscription is 2.5 cm, the maximum preserved height 1.4 cm. The 1988 group The three sealings of this group are unpublished. Their archaeological context is discussed above. Sealing 459 shows the impression of a string and a piece of linen and was attached either to the neck of a vessel sealed with a piece of linen or to a linen bundle or sack. The reverse side of sealing 464 bears impressions of ine strings, perhaps leaves of a plant or leather strips. The surface of the concave area is even and not impressed by linen. Perhaps the sealing sat on top of a piece of leather covering a small container. The surface of sealing 465 is partially eroded but the shape of the sealing suggests that it was attached on a peg. Of all seal inscriptions presented here only 459 and 465 show deinite use of a vertical separator. This feature was probably inspired by sealing patterns with a royal serekh which led to a vertical organisation of the inscription. The use of vertical separators seems to appear somewhere near the end of the Early Dynastic period and beginning of the Old Kingdom.72 Sealing 459 (Fig. 19): Find number AZ: R-10, entry in register book on 7 March 1988, found in quadrant 15N1W, within the silo in Room 10, unpublished. H: 1.7 cm, B: 2.2 cm, D: 1.4 cm. Munsell 2.5Y 2.5/1. The left and bottom sides are broken off. The reverse side shows impressions of three lines of strings and faint impressions of a folded piece of linen in the upper part. The inscriptions run perpendicular to the impressions of the strings. Several ingerprints disturb the inscriptions. The inscription on the left shows the sign for s and a vertical separator. The maximum preserved height of the inscription is 1.9 cm, the maximum preserved length 1.4 cm. The lower sign of the inscription on the right could be Sms (T18) “the follower”, a title well attested at Elephantine and dated by kAplony and päTzniCk not earlier than 72 kAplony 1963, 46. Renée Friedman and Richard Bussmann 94 to the Third Dynasty.73 In the next inscribed ield, separated by a vertical stroke, there are a rectangular sign and s, possibly again part of the title Sms. The maximum preserved height of the inscription is 1.9 cm, the maximum preserved length 1.4 cm. The preserved inscriptions stem from different parts of the cylinder. Fig. 19 Sealing 459 (photo: R. Bussmann, drawing: R. Bussmann and C. von Elm) Sealing 464 (Fig. 20): Find number 14-1, entry in register book on 23 February 1988, found in quadrant 15N2W, in pit Feature 62, unpublished. H: 2.1 cm, B: 2.6 cm, D: 1.2 cm. Munsell 7.5YR 3/1. The upper and lower sides are broken off. The left part of the reverse side is slightly sloped and smooth. Three thin strings, probably of leather or simple twisted grass, are impressed next to it. The right side is level but not smooth. The sealing is drawn here perpendicular to how it was originally used. It might have been placed over a piece of leather arranged over the mouth of a jar and held together with thin leather strips. Fig. 20 Sealing 464 (photo: R. Bussmann, drawing: R. Bussmann and C. von Elm) Two inscriptions run from right to left, perpendicular to the impressions on the reverse side. The s-sign of the lower inscription runs slightly into the upper inscription. 73 päTzniCk 2005, 107‒111. The adjacent sign might represent the kA-arms (D28) followed by t and, below, the head of a bird.74 The maximum preserved height of the inscription is 1.0 cm, the maximum preserved length 2.4 cm. Of the upper inscription, only part of an s-sign is preserved. Sealing 465 (Fig. 21): Find number 8-42, entry in register book on 20 February 1988, found in quadrant 15N2W, purportedly from the base of Feature 61, unpublished. H: 2.2 cm, B: 2.5 cm, D: 0.7 cm. Munsell 5YR 4/2. The lower and right sides are broken off. The reverse side shows a slightly curved, smooth upper part, probably the impression of a peg, and the impression of a string at the bottom. The uneven and eroded part in the middle probably derives from the clay being squeezed into the vacuum between peg and string. Fig. 21 Sealing 465 (photo: R. Bussmann, drawing: R. Bussmann and C. von Elm) If the inscription on the front is interpreted as representing two impressions, they would both derive from the same part of the cylinder seal. The upper inscription is bordered by two ragged lines. It shows a vertical separator and two arms in both squares bordering the separator. At the bottom is another simple line, probably part of the lower bordering line. The maximum preserved height of the inscription is 2.0 cm, the maximum preserved length 2.0 cm. The lower inscription overlaps slightly with the upper inscription. It shows the vertical separator and one pair of arms in each of the squares on both sides of the separator. The maximum preserved length is 0.7 cm, the maximum preserved length 0.7 cm. 74 Perhaps the name %kA, kAplony 1963, 640, ig. 881. The Early Dynastic Palace at Hierakonpolis Discussion Despite its small size, the sample of clay sealings from the palace of Hierakonpolis raises a few interesting questions.75 Palaeographically, the 1969 and 1981 groups have good parallels in comparable material of the First Dynasty, while the 1988 group seems to be later, perhaps Third Dynasty. Their later date ties in with the observations on inds and architecture in this area, which seem to post-date the structures near the palace gateway. However, more sealings from the site would be needed to determine how long seals were used, when sealings were discarded and redeposited, and whether the sealings follow a distinctive chronological and spatial distribution pattern across the palace area. A similarly small number of clay sealings was found in the late Predynastic and Early Dynastic layers of Tell el-Iswid, while Buto yielded a slightly more substantial series of impressions.76 The restricted evidence of early administration outside the royal and elite burials at Abydos and Saqqara77 suggests that the scope of administration and the role of palaces for maintaining administrative networks was limited in the Early Dynastic Period. The following discussion presents some thoughts of how one might approach this question. A irst aspect relates to the importance of writing for running a larger polity. The distribution of archives, including lists, tables, decrees and accounts, within and outside the political centre of the Old Kingdom78 could indicate that the use of documents was more wide-spread than the poor record of preserved papyri in the Early Dynastic Period relects.79 Tomb scenes showing how written documents are presented to the tomb owner add to the evidence on an iconographic basis. However, there is little hope to discover larger archives within early palaces when the latter are located in the lood plain or, as at Hierakonpolis, at or below groundwater table.80 75 76 77 78 79 80 The more substantial assemblages of Elephantine (päTzniCk 2005), Balat, Elkab (regulski 2009), and the Middle Kingdom forts and settlements (von pilgrim 1996; Wegner 2007) suggest that the corpus of Hierakonplis originally was larger, see QuiBell and green 1902 and BussmAnn 2011 for more sealings from Hierakonpolis. grATien 2001 offers a good overview of discussions of relevant sites. regulski 2014; kAplony, 1992; engel in: HArTung et al. 2012. regulski 2010, 73‒85. posener-kriéger 1975; 1976; 1986; 1994; 2004; posenerkriéger et al. 2006; TAlleT 2014. eyre 2013. Quirke 1996 laments the restricted evidence of palace archives from Pharaonic Egypt more generally. 95 A few clay sealings from Elephantine and at least one of the Third Dynasty from the temple area of Hierakonpolis were impressed on a papyrus roll.81 Although they represent a minority among the clay sealings overall, they are indirect evidence of papyrus being involved in administration from an early time onwards. In later corpora, the number of document sealings is similarly low,82 but written documents, like the Semna dispatches, clearly demonstrate that papyrus documents were exchanged between the periphery and the centre, at least from time to time, and not necessarily in the form of sealed documents. It would therefore be misleading to base too strong an argument about the use of papyrus in Early Dynastic administration on the sealing corpus alone. However, document sealings do bear on the question and have not been fully considered in this context. Whether or not the palace fulilled its original function in the Third Dynasty is currently dificult to say. Not all administrative practices require writing. Counting, storing, organising and negotiating hierarchies can be achieved eficiently by oral agreement and physical enactment. The management of people and goods in prehistory, including in long-distance trade, relied on visual and face-toface communication and is relevant up to the present day. The miniscule corpus of clay sealings from Hierakonplis and the lack of excavated documents should not be understood as proof of absence of administration on a larger scale. Only one impression of the corpus from the palace of Hierakonpolis includes a royal serekh. Among the sealings from the Early Dynastic eastern town at Elephantine, interpreted as the governor’s residence, and from the palace of the local governors at Ayn Asil, the number of sealings with royal names is also low.83 It is probably realistic to assume a similar proportion for the palace of Hierakonpolis. The few sealings with royal names found by Quibell and Green at Hierakonpolis are in line with this assumption. In stark contrast, all sealings discovered at the workmen settlement in Giza bear a serekh.84 The difference to Hierakonpolis may be a question of chronology in the sense that seals without royal names fell out of fashion towards the beginning of 81 82 83 84 päTzniCk 2005, 61‒62; BussmAnn 2011. von pilgrim 1996, 238; Wegner 2007, 300‒304. päTzniCk 2005, 63‒87, tab. 34; pAnTAlACCi 2001. nolAn 2010. 96 Renée Friedman and Richard Bussmann the Fourth Dynasty.85 In addition, the settlement at Giza was also occupied only for a short period of time, perhaps a generation or so, whereas the other sealing corpora accumulated over a century or more possibly witnessing changes in the administrative function of the sites at which they were found. An alternative explanation is to understand the differences sociologically as relecting different administrative networks. The Giza settlement, serving the organisation of the royal funerary cult, was embedded in a network of oficials directly related to the crown. Elephantine and Ayn Asil are served and administered by a more varied community whose members are a few royal and a broad spectrum of local oficials. The palace of Hierakonpolis may have belonged to the latter type, although this conclusion must remain hypothetical at this point. Conclusion Hierakonpolis offers an opportunity to explore how a royal palace was built into and transformed an existing community. The palace is located at “new Hierakonpolis”, the island in the cultivation possibly created by material washed down from the Wadi Abu Sufian. Perhaps it was designed together with the ceremonial area, usually interpreted as the temple of Horus, to form a new ritual landscape focusing on kingship. Unfortunately, the deicient excavation reports prevent an understanding of the stratigraphic relationship of the various buildings and their components. The temple apparently had its heyday in the late Predynastic and beginning of the Early Dynastic Period with an interlude under Khasekhem in the Second Dynasty.86 The palace might have followed a similar trajectory, built initially as an impressive royal foundation, used as such for a short period of time along with the temple and inally merging with local life once kings had 85 86 kAplony 1977; 1981 reviewed the clay sealing of the Old Kingdom, the vast majority of which bear a serekh and were found in royal contexts. Outside the centre, the evidence is more diverse, see Willems et al. 2009, 313‒315. BussmAnn 2010, 496‒498. The royal objects of the 6th Dynasty relect a different historical pattern of increased royal building activity throughout the country. disappeared from the site. The precise dating of this transition is dificult to establish, perhaps at some point between the Naqada III Period / early First Dynasty and the Second Dynasty or a bit later. With due caution, the palace may be reconstructed with a size of 80 × 80 m if the niched gateway was located in the centre of the front wall. In the late Naqada III Period, the palace of Hierakonpolis shows the strongest similarities to the palace of Buto. It is proposed here that it was divided into a reception area located behind the niched gateway, a residential area and an administrative area to the east. Material from the eastern part indicates a slightly later date (Third Dynasty) than the inds from other areas (First and Second Dynasties). One reason for this may relate to the topography of the site. There is a general rise of loor level and strata from the gateway eastward, possibly following a natural slope upward. However, at the eastern end of the excavated area, and just behind the niched platform, the slope abruptly falls off. The R series of rooms in this area are at a lowered level, and thus younger material still remained here, while later levelling, probably in the New Kingdom,87 removed these layers in the area of the niched gate. The uncertainties of interpretation should not overshadow the importance of the palace. Prior to the discovery of the niched gateway and the area behind it, Early Dynastic Hierakonpolis was known primarily for its temple, with a few houses scattered across the area delineated by the town enclosure wall. Interpretation of kingship at Hierakonpolis was largely restricted to its sacred aspect. The architecture and sealing corpus of the palace demon-strate that the site was designed on a much grander scale. Kingship was locally embedded in an apparatus of representation and power beyond the sphere of the gods. 87 See BunBury and grAHAm 2008. The Early Dynastic Palace at Hierakonpolis Bibliography AdAms, B. 1974 Ancient Hierakonpolis Supplement, Warminster. 1995 Ancient Nekhen. Garstang in the City of Hierakonpolis, Egyptian Studies Association 3, New Malden. BABA, m. and FriedmAn, r.F. 2016 Recent Excavations at HK11C, Hierakonpolis, in: m.d AdAms, m. midAnT-reynes, y. TrisTAnT and e. ryAn (eds.), Egypt at its Origins 4. Proceedings of the International Conference ‘Origin of the State. Predynastic and Early Dynastic Egypt’, New York, 26‒30 July, 2011, OLA 252, Leuven. BABA, m., vAn neer, W. and de Cupere, B. 2017 Industrial Food Production Activities during the Naqada II period at HK11C, Hierakonpolis, 3‒34, in: B. midAnT-reynes and Y. TrisTAnT (eds.), Egypt at its Origins 5. Proceedings of the Fifth International Conference ‘Origin of the State. Predynastic and Early Dynastic Egypt’, Cairo, 13‒18 April 2014, Leuven. BesToCk, l. 2008 The Early Dynastic Funerary Enclosures at Abydos, Archeo-Nil 18, 42‒59. Brookner, J. 1986 The Archaic Period Sealings, 24‒26, in: W. FAirservis, Excavations of the Archaic Remains East of the Niched Gateway. Season of 1981. With a Contribution by Jonathan Brookner, The Hierakonpolis Project III, Poughkeepsie, NY. BunBury, J.m. and grAHAm, A. 2008 There is Nothing Boring about a Borehole, Nekhen News 20, 22‒23. BussmAnn, r. 2010 Die Provinztempel Ägyptens von der 0. bis zur 11. Dynastie. Archäologie und Geschichte einer gesellschaftlichen Institution zwischen Residenz und Provinz, PdÄ 30, Boston, Cologne. 2011 The Seals and Seal Impressions from Hierakonpolis, EA 38, 17‒19. dreyer, g., engel, e.-m., HArTung, u., HikAde, T., köHler, e. C. and pumpenmeier, F. 1996 Umm el-Qaab. Nachuntersuchungen im frühzeitlichen Königsfriedhof. 7./8. Vorbericht, MDAIK 52, 11‒81. dreyer, g., BlöBAum, A. i., engel, e.-m., köpp, H. and müller, v. 2011 Umm el-Qaab. Nachuntersuchungen im frühzeitlichen Königsfriedhof 19./20./21. Vorbericht, MDAIK 67, 53‒92. 97 eyre, C. 2013 The Use of Documents in Pharaonic Egypt, Oxford. FAirservis, W.A. 1971-72 Preliminary Report on the First Two Seasons at Hierakonpolis, JARCE 9, 7‒27. 1983 The Hierakonpolis Project. Season January to March 1978. Excavation of the Temple Area on the Kom el Gemuwia, Hierakonpolis Occasional Papers in Anthropology 1, Poughkeepsie, N.Y. 1986 Excavation of the Archaic Remains East of the Niched Gate. Season of 1981. With a Contribution by Jonathan Brookner, The Hierakonpolis Project III, Poughkeepsie, NY. FriedmAn, r.F. 2007 New Observations on the Fort at Hierakonpolis, 309‒336, in: z.H. HAWAss and J. riCHArds (eds.), The Archaeology and Art of Ancient Egypt. Essays in Honor of David B. O’Connor, Supplément aux Annales du Service des Antiquites de l’Égypte Cahier 36, Cairo. 2009 Hierakonpolis Locality HK29A: The Predynastic Ceremonial Center Revisited, JARCE 45, 79‒103. FriedmAn, r.F., vAn neer, W. and linseele, v. 2011 The Elite Predynastic Cemetery at Hierakonpolis: 2009‒2010 Update, 157‒191, in: r. FriedmAn and p.n. Fiske (eds.), Egypt at its Origins 3. Proceedings of the Third International Conference ‘Origins of the State. Predynastic and Early Dynastic Egypt’. London, 27th July‒1st August 2008, OLA 205, Leuven. grATien, B. (ed.) 2001 Le sceau et l’administration dans la Vallée du Nil: Villeneuve d’Ascq, 7‒8 juillet 2000, CRIPEL 22, Villeneuve-d‘Ascq. HArlAn, J.F. 1985 Predynastic Settlement Patterns: A View from Hierakonpolis, Ph.D dissertation, Washington University, St. Louis, Missouri. HArTmAnn, R. 2016 Pottery from the Recent Excavations in the Early Dynastic Settlement at Tell el-Fara’in/Buto, in: B. midAnT-reynes and Y. TrisTAnT (eds.), Egypt at its Origins 5. Proceedings of the Fifth International Conference ‘Origins of the State. Predynastic and Early Dynastic Egypt’, Cairo, 13‒18 April 2014, Leuven. emery, W. B 1958 Great Tombs of the First Dynasty III. London. HArTung, u., BAlleT, p., Béguin, F., BourriAu, J., dixneuF, d., von den driesCH, A., FrenCH, p., HArTmAnn, r., HerBiCH, T., kiTAgAWA, C., kopp, p., leCuyoT, g., nennA, m.-d., sCHmiTT, A., senol, g. and senol, A. 2007 Tell el-Fara’in – Buto. 9. Vorbericht, MDAIK 63, 69‒166. engel, e.m. and müller, v. 2000 Verschlüsse der Frühzeit: Erstellung einer Typologie, GM 178, 31–44. HArTung, u., engel, e.-m. and HArTmAnn, r. 2012 Tell el-Fara’in – Buto. 11. Vorbericht, MDAIK 68, 83–114. 98 Renée Friedman and Richard Bussmann HendriCkx, s. 2006 Predynastic-Early Dynastic Chronology, 55‒93, in: E. Hornung, r. krAuss and d.A. WArBurTon (eds.), Handbook of Egyptian Chronology, HdO 83, Leiden. HendriCkx, s., FAlTings, d., op de BeeCk, l., rAue, d. and miCHiels, C. 2002 Milk, Beer and Bread Technology during the Early Dynastic Period, MDAIK 58, 277‒304. HikAde, T. 2011 Origins of Monumental Architecture: Recent Excavations at Hierakonpolis HK29B and HK25, 81‒108, in: r. FriedmAn and p.n. Fiske (eds.), Egypt at its Origins 3. Proceedings of the Third International Conference ‘Origins of the State. Predynastic and Early Dynastic Egypt’. London, 27th July‒1st August 2008, OLA 205, Leuven. HoFFmAn, m.A. 1982 The Predynastic of Hierakonpolis. An Interim Report, Egyptian Studies Association 1, Giza/Macomb. Holmes, d.l. 1992 Chipped Stone-working Craftsmen, Hierakonpolis and the Rise of Civilization in Egypt, 37‒44, in: r. FriedmAn and B. AdAms (eds.), The Followers of Horus. Studies dedicated to Michael Allen Hoffman 1944–1990, Oxford. Jones, d. 2000 An Index of Ancient Egyptian Titles, Epithets and Phrases of the Old Kingdom, BAR International Series 866, Oxford. kAHl, J. 1994 Das System der ägyptischen Hieroglyphenschrift in der 0.‒3. Dynastie, GOF IV.29, Wiesbaden. kAplony, p. 1963 Die Inschriften der ägyptischen Frühzeit, ÄA 8, Wiesbaden. 1964 Die Inschriften der ägyptischen Frühzeit. Supplement, ÄA 9, Wiesbaden. 1977 Die Rollsiegel des Alten Reichs 1. Allgemeiner Teil mit Studien zum Königtum des Alten Reiches, Brussels. 1981 Die Rollsiegel des Alten Reichs 2. Katalog der Rollsiegel, Brussels. 1992 Archaische Siegel und Siegelabrollungen aus dem Delta. Die Arbeit an den Siegeln von Buto, 23‒30, in: E. vAn den Brink (ed.), The Nile Delta in Transition. 4th‒3rd Millennium B.C. Proceedings of the Seminar held in Cairo, 21.‒24. October 1990, at the Netherlands Institute of Archaeology and Arabic Studies, Tel Aviv. kemp, B.J. 2006 Ancient Egypt: Anatomy of a Civilization, London. kopp, P. 2006 Elephantine XXXII. Die Siedlung der Naqadazeit, AV 118, Mainz. lACAu, p. and lAuer, J.-p. 1965 La pyramide à degrés V. Inscriptions à l’encre sur les vases, Cairo. mCnAmArA, l. 2008 The Revetted Mound at Hierakonpolis and Early Kingship: A Re-interpretation, 901‒936, in: B. midAnT-reynes and Y. TrisTAnT (eds.), Egypt at its Origins 2. Proceedings of the International Conference ‘Origin of the state, Predynastic and Early Dynastic Egypt’, Toulouse (France), 5th‒8th September 2005, OLA 172, Leuven. moeller, n. 2016 The Archaeology of Urbanism in Ancient Egypt. From the Predynastic Period to the End of the Middle Kingdom, New York. nolAn, J. 2010 Mud Sealings and Fourth Dynasty Administration at Giza, PhD Thesis, Chicago. o’Connor, d. 1992 The Status of Early Egyptian Temples: an Alternative Theory, 83‒98, in: r. FriedmAn and B. AdAms (eds.), The Followers of Horus. Studies dedicated to Michael Allen Hoffman 1944–1990, Oxford. päTzniCk, J.-p. 2005 Die Siegelabrollungen und Rollsiegel der Stadt Elephantine im 3. Jahrtausend v. Chr.: Spurensicherung eines archäologischen Artefaktes, BAR International Series 1339, Oxford. pAnTAlACCi, l. 2001 L’administration royale et l’administration locale au gouvernorat de Balat d‘après les empreinte de sceaux, 153‒160, in: B. grATien (ed.) Le sceau et l’administration dans la Vallée du Nil: Villeneuve d’Ascq, 7‒8 juillet 2000, CRIPEL 22, Villeneuved‘Ascq. peTrie, W.m.F. 1903 Abydos II (1903), EEF 24, London. pilgrim, C. von 1996 Elephantine XVIII: Untersuchungen in der Stadt des Mittleren Reiches und der Zweiten Zwischenzeit, AV 91, Mainz. posener-kriéger, p. 1975 Les papyrus de Gébelein. Remarques préliminaires, RdÉ 27, 211‒221. 1976 Les archives du temple funéraire de NéferirkarêKakaï (Les papyrus d’Abousir), BdÉ 65, Cairo. 1986 Old Kingdom Papyri: External Features, 25‒41, in: M.L. BierBrier (ed.), Papyrus: Structure and Usage, BMOP 60, London. 1994 Le coffret de Gebelein, 315-326, in: C. Berger, g. ClerC and n. grimAl (eds.), Hommages à Jean Leclant 1, BdÉ 106(1), Cairo. 2004 I papyri di Gebelein: Scavi G. Farina 1935, Studi del Museo Egizio di Torino Gebelein I, Turin. The Early Dynastic Palace at Hierakonpolis posener-kriéger, p., verner, m. and vymAzAlová, H. 2006 Abusir X: the Pyramid Complex of Raneferef. The Papyrus Archive, Prague. QuiBell, J.e. 1900 Hierakonpolis I, ERA 4, London. QuiBell, J.e. and green, F.W. 1902 Hierakonpolis II, ERA 5, London. 99 TAkAmiyA, i.H. 2008 Firing Installations and Specialization: A View from Recent Excavations at Hierakonpolis Locality 11C, 187‒202, in: B. midAnT-reynes and Y. TrisTAnT (eds.), Egypt at its Origins 2. Proceedings of the International Conference ‘Origin of the State. Predynastic and Early Dynastic Egypt’ Toulouse (France), 5th‒8th September 2005, OLA 172, Leuven. Quirke, s. 1996 Archive, 379‒402, in: A. loprieno (ed.), Ancient Egyptian Literature: History and Forms, PdÄ 10, Leiden. TAlleT, p. 2014 Des papyrus du temps de Chéops au ouadi el-Jarf (golfe du Suez), BSFE 188, 25‒49. rAue, D. 1999 Ägyptische und nubische Keramik der 1.‒4. Dynastie, 173‒189, in: W. kAiser et al., Stadt und Tempel von Elephantine 25./26./27. Grabungsbericht, MDAIK 55. Weeks, k.r. 1971‒72 Preliminary Report on the First Two Seasons at Hierakonpolis II: The Early Dynastic Palace, JARCE 9, 29‒33. regulski, i. 2009 Early Dynastic Seal Impressions from the Settlement Site at Elkab, 31‒48, in: W. ClAes, H. de meulenAere and s. HendriCkx (eds.), Elkab and Beyond. Studies in Honour of Luc Limme, OLA 191, Leuven. 2010 A Palaeographic Study of Early Writing in Egypt, OLA 195, Leuven. 2014 Seal Impressions from Tell el-Iswid, 230‒242, in: B. midAnT-reynes and n. BuCHez, Tell el-Iswid 2006‒2009, FIFAO 73, Cairo. Wegner, J. 2007 The Mortuary Temple of Senwosret III at Abydos, New Haven. Willems, H., vereeCken, s., kuiJper, l., vAnTHuyne, B., mAe., linseele, v., versTrAeTen, g., HendriCkx, s., eyCkermAn, m., vAn den Broek, A., vAn neer, W., BourriAu, J., FrenCH, p., peeTers, C., de lAeT, v., morTier, s. and de kooning, y. 2009 An Industrial Site at Al-Shaykh Sa’īd/Wādī Zabayda, Ä&L 19, 293‒331. rinovA,