Social Change in the Fourth Dynasty: the Spatial Organization Of

Social Change in the Fourth Dynasty: The Spatial Organization of Pyramids, Tombs, and Cemeteries Author(s): Ann Macy Roth Source: Journal of the American Research Center in Egypt, Vol. 30 (1993), pp. 33-55 Published by: American Research Center in Egypt Stable URL: http://www.jstor.org/stable/40000226 . Accessed: 21/04/2011 16:10

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in the JSTOR archive only for your personal, non-commercial use.

Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at . http://www.jstor.org/action/showPublisher?publisherCode=arce. .

Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmission.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].

American Research Center in Egypt is collaborating with JSTOR to digitize, preserve and extend access to Journal of the American Research Center in Egypt.

http://www.jstor.org Social Change in the Fourth Dynasty: The Spatial Organization of Pyramids,Tombs, and Cemeteries

Ann Macy Roth

The advent of the Fourth Dynasty of phar- eantry, the new-style pyramids proclaimed his aonic Egypt marked a radical break with the absorption into the mystic symbol of the sun. first three dynasties. This break is most visible The tiny offering-temple was the principal in the new shape of the era's most substantial gesture to his human aspect.' R. Stadelmann archaeological remains, the royal pyramidsand viewed the pyramidsin more political terms, as their surrounding mortuary complexes (see monuments that both expressed and enforced fig. 1). In the Third Dynasty, royal tombs took the universal claims of royal power. The new the form of stepped pyramids, surrounded by mortuary architecture, he suggested, was a dummy buildings and enclosed in a rectangle simplification and abstraction of older forms, of high, niched walls, with its long axis north- responding to growth in that power and to south. During the reign of Snefru, royal tombs changes in cultic requirements. became true pyramids of vastly increased size, The cemeteries of officials that surrounded built at the western end of a complex of new these pyramids add yet another dimension to components and proportions, which extended the analysis. D. O'Connor has observed that, if in an east-west line from the border of the the sizes of tombs represent the comparative cultivation. power of the tomb owners, the gigantic pyra- Egyptologists have long ascribed these mids of Giza surrounded by small private tombs changes to social and religious developments. can be seen as a visual metaphor for the J. H. Breasted suggested that the increasing im- centralized organization of the Old Kingdom portance of the sun-cult of Re at Heliopolis led state, in which the immense power of the king to the adoption of a tomb nearer in shape to dwarfed and dominated the people surround- the bnbnstone associated with that cult.1 I. E. S. ing him. Edwards advocated a more direct relationship Since textual sources for the Fourth Dynasty to mortuarybeliefs, viewing the pyramid as the and the preceding period are few and enig- solidified rays of the sun and citing Pyr. 523: matic, the primarysupport for these analyses is "Heaven has strengthened for you the rays of the architecture and spatial organization of the the sun, in order that you may lift yourself to Fourth Dynasty pyramid complex itself. These heaven as the eye of Re."2 He also attributed analysesof mortuaryspace are, however, largely the new east-westaxis to an increasing solar ori- impressionistic and based on intuitive assump- entation. B. Kemp, describing the pyramid of tions about the meaning of space and architec- Meydum, suggested a change in the theological tural forms. Moreover, they are based on the and social role of the king: "In place of a tomb which celebrated the king as supreme territo- 3 Ancient a Civilization rial claimant and perpetuated his earthly pag- Barry J. Kemp, Egypt: Anatomy of (New York, 1989), 63, caption to fig. 21. R. Stadelmann, Die dgyptischenPyramiden: vom Ziegelbau J. H. Breasted, The Developmentof Religion and Thought zum Weltwunder(Mainz am Rhein, 1985), 80. in Ancient Egypt (New York, 1912), 72. David O'Connor, "Political systems and archaeological I. E. S. Edwards, The Pyramids of Egypt, rev. ed. (Har- data in Egypt: 2600-1780 B.C.," WorldArchaeology 6 (1984), mondworth, 1985), 277-78 (translation slightly modified). 19-21.

33 34 JARCE XXX (1993)

Fig. 1. The StepPyramid complex,at left (afterEdwards, Pyramids of Egypt, p. 35), is the bestpreserved of the pre-Fourth dynasty mortuarycomplexes. The Fourth dynasty complexin its simplestform is representedat right by the reconstructedMey- dum pyramid complex.(The drawing here is partially based on the reconstructionof Kemp,Ancient Egypt: Anatomy of a Civilization, p. 63.) (Not drawn to the same scale.) examination of only a limited subset of mortu- paring quantitativemeasurements. Despite the ary architecture. Private tomb architecture and objective appearance of the results these tech- cemetery organization also changed consider- niques yield, their application often requires ably, if less abruptly, during the same period, subjectivejudgments. (It is not alwaysclear, for and this larger context has not been considered example, what constitutes a "room.")Moreover, in seeking explanations for the new architec- these techniques have principally been applied tural forms adopted by kings. to houses, and their usefulness in analyzing symbolic spaces, such as mortuary or religious is less well established. Spatial Analysis buildings, For an initial application of spatial analysis In recent years, archaeologists have increas- to early Old Kingdom mortuary architecture, ingly applied formal techniques of spatial these difficulties can be avoided by using a analysis to the interpretation of cultural re- comparative approach, relating changes in the mains. One useful concept of this type is access accessibility of mortuary architecture to the analysis,which focuses on the ease or difficulty relatively static patterns in contemporary non- with which people move through buildings and mortuaryspaces. Based loosely on the same cri- into important rooms. Techniques have been teria as the more quantitative approach, such developed that allow buildings to be more easily comparisons allow distinctively Egyptian spa- compared, including the reduction of plans to tial patterns and architectural forms to be con- "justifiedaccess maps' and formulas for com- sidered. Although this approach is explicitly

6 For details of this method, see Bob Hillier and Julienne For a survey of a variety of quantitativemethods, see Hanson, The SocialLogic of Space(Cambridge, 1984). Sally John Chapman, "SocialInequality on Bulgarian Tells and Foster,"Analysis of spatialpatterns in buildings(access analy- the Varna Problem," The Social Archaeologyof Houses, Ross sis) as an insight into social structure:examples from the Samson, ed. (Edinburgh,1990), 49-92; the following essay, Scottish Iron Age," Antiquity63 (1989), 40-50; and Henry FrankE. Brown,"Comment on Chapman:Some Cautionary Glassie, Folk Housing in Middle Virginia: a structural analysis of notes on the application of spatial measures to prehistoric historicartifacts (Knoxville, Tennessee, 1975) were among the settlements,"ibid., 93-109, points out some problems with first to apply this method to archaeologicalspaces. this approach. SOCIAL CHANGE IN THE FOURTH DYNASTY 35 subjective, it is justified by the sharpness of the buildings are mixed with or segregated from contrasts it produces. These contrasts can then older structures. Like spatial patterns in indi- be compared with the surviving inscriptional vidual buildings, spatial patterns in sites can evidence to suggest the nature of the changes suggest social characteristics, such as the de- in the system of religious and social beliefs that gree of centralized control, relationships to the underlay the new mortuary architecture. past, and the segregation of certain groups.8 The principle of access analysis that I have Such analysis has generally been applied to set- adopted here is the distinction between "open" tlements, but it also is a useful way of looking at and "closed" plans in buildings. Open buildings cemeteries. tend to be readily navigable by strangers; they can be entered and their internal easily organi- Non-mortuary Architecture: zation is The function immediately apparent. Houses and Temples and position of their important rooms are often obvious from the exterior, and the paths to This comparative analysis of spatial organiza- reach them are both short and direct. Axial and tion in mortuary and non-mortuary structures symmetrical plans tend to result in open build- is implicitly based on the assumption that there ings, as do plans with many entrances. Public was no fundamental change in the plans of or communally-used spaces often have open houses and temples between the Archaic plans, and they are especially common in com- Period and the later Old Kingdom. This as- munities where strangers are either rare or as- sumption does not conflict with any architec- sumed to be friendly, in egalitarian societies, tural remains so far excavated from the early and in cultures that place a low value on pri- period, but those remains are too few to prove vacy. Closed plans, on the other hand, separate or disprove it. There are, however, a number of the most important rooms from the entrance corroborating circumstances, foremost among by distance, by tortuous pathways, and by con- them the stability of these two architectural stricted or guarded doorways, so that strangers forms in later periods. have difficulty entering the building and nego- In all periods for which there is evidence, the tiating its interior spaces. Greater "closedness" Egyptians seem to have favored the greatest occurs in societies and in buildings where possible closedness in their houses.9 Figure 2 privacy, social control, separation of classes illustrates a selection of Old and Middle or sexes, and protection from strangers are Kingdom house plans. The owners of even the considered important. smallest houses were often willing to sacrifice a Not only can the principles of access analysis corner to create a small entrance vestibule that be applied to individual buildings, but entire allowed them to screen their visitors. In larger sites can be viewed in terms of their spatial or- 8 For the observationsof such over ganization. On this level, such questions as the spatial relationships time in a settlement context, see W. "The distance between buildings, the regularity of Douglas Bailey, LivingHouse: SignifyingContinuity," in: TheSocial Archaeol- their orientation, and the ease of access to ogy of Houses, 19-48. different parts of the site and the site as a A possible exception to this tendency is the compound whole are considered. This analysis involves of thirty, largely contiguous, room-groups at Qasr es-Saga "Die des MittlernReiches bei comparing linear arrangements of buildings (JoachimSliwa, Siedlung Qasr with clustered and the el-Sagha,"MDAIK 48 [1992], 177-91). Despite the quanti- arrangements, judging ties of ash, fishbones, and animal bones contained, to which a site is or has a they degree homogeneous however,these room groups seem unlikelyto have been pri- central focus. (Such broader factors should al- marily domestic spaces. The five identical, narrow rooms ways be considered, since the degree of access opening off each courtyard resemble storerooms in their dimensions are 2.1 x 7.9 These to a site as a whole may explain anomalous ac- proportions (their m). rooms were fitted with but there are no cess in the individual within carefully doors, patterns buildings for the which was en- of within a site can doorpost emplacements "courtyard" it.) Arrangements buildings tered directly from the street, and its built-in features also be compared temporally, to find patterns (benches and raised round platforms)suggest industrialac- of site growth and to determine whether newer tivityof some kind. 36 JARCE XXX (1993)

Fig. 2. These house plans were takenfrom thefollowing sources:Elephantine (ArchaicPeriod): MDAIK 40 (1984), 174, (left); MDAIK 43 (1987) p. 91 (right); Hierakonpolis(Archaic Period) : Quibell and Green,Hierakonpolis, pi 68; South Giza (Fourth Dynasty): Kemp, Ancient Egypt: Anatomy of a Civilization, p. 134; Dahshur (Old Kingdom): ibid., p. 148; Khentkawestown (late Fourth Dynasty): Hassan, Giza IV, fig. 1; Kahun (Middle Kingdom): Kemp, Ancient Egypt: Anatomy of a Civilization, p. 54 (somewhatmodified in accordancewith Petrie, Illahun, Kahun, and Gurob, pi 14). Orientationsdiffer and scales are approximatein some cases. SOCIAL CHANGE IN THE FOURTH DYNASTY 37 houses, the desire for closedness resulted in The closed pattern in large houses was al- "baffle" walls at the entrance that obscured the ready well established by the end of the Fourth interior and forced the visitor to walk in an Dynasty, as exemplified by the "priests' houses" S-shaped curve. The visitor was then normally along the causeway leading to the cultivation led well into the house, and had to double back from the tomb of Queen Khentkawes.12 From to reach the functional rooms, sometimes re- the south, the houses could not be entered di- versing direction several times to reach the rectly from the causeway, but only from a paral- most private spaces. Access to the individual lel private path accessible through doors offset rooms within houses was limited, but in some from the house doors. There, a baffle wall im- cases parallel or encircling hallways provided mediately confronted the visitor, who had to second entrances. The purpose of this extrava- turn left, then right, then proceed along a cor- gant waste of space was probably to allow differ- ridor past a small room and into an open court ent classes of people within the house (residents toward the back (north) of the house. To the and visitors, or masters and servants, or men south of the court was an area with a hearth and women) to pass between the rooms without and ovens, and to the southwest lay a long encountering one another. Another indica- room that may have been the principal public tion of closedness is the frequency with which a room. Opening off the latter to the west were small room adjoined the inner vestibule, from two consecutive rooms probably restricted to which a servant could control access to the the family and used partly for sleeping. To the house. north of the public room was the largest room in the house. It was often subdivided or filled with store it have been used to store 10This is clear in the attached jars; may especially simple palaces and distribute commodities as of the occu- to the New Kingdom temples of the Ramesseumand Medi- part It had a net Habu (see, for example, W. J. Murnane, Unitedwith Eter- pant's professional activity. separate nity (Chicago, 1980), fig. 58), where parallel hallwaysfor entrance (taken to be the principal one by the servantsrun behind the private quarters,allowing servants excavator) that led past a small room to a pri- to remove the chamber without the rooms' pots disturbing vate back street, to which access also seems to The most extreme of this are the occupants. examples have been controlled. Kahun mansions, with their parallel hallways(W. M. F. Pet- There is no evidence for the architecture of rie, Illahun, Kahunand Gurob[London, 1891], pl. 14); but such parallelism is attested on a community-widescale as large houses before the late Fourth Dynasty. early as the Khentkawestown at Giza (S. Hassan,Excavations (The assumption of a closed plan is corrobo- at GizaIV- 1932-1933 [Cairo, 1943], fig. 1). Such "service rated the closed of have been in of the by plans early mortuary passages" similarlyanalyzed buildings structures that are believed to Roman period and the seventeenth century. See Eleanor generally dupli- Scott, "Romano-BritishVillas and the Social Construction cate palaces, but in the context of this com- of Space," The Social Archaeologyof Houses, 149-72; and Ross parative study, such arguments are potentially Samson, "The Rise and Fall of Tower Houses in Post- circular.) Small houses dating to the earlier ReformationScotland," ibid., 197-243. that have been excavated at Hierakon- 11The that these rooms ar- period argument represent "birthing and however, show the bors,"suggested by F. Arnold, "AStudy of EgyptianDomes- polis Elephantine, same closed favored in later tic Buildings,"VA 5 (1989), 81-82, is, to me, unconvincing, patterns periods; at least in the Old and Middle Kingdoms.A vestibule at the 12 entrance to the house seems a to seclude a new Hassan, Excavations at Giza TV, 1. strangeplace 15 fig. mother, especiallyin the Kahunmansions, where both vesti- W. Fairservis,K. R. Weeks, and M. Hoffman, "Prelimi- bules are quite distant from the rooms Arnold identified as nary report on the first two seasons at Hierakonpolis," "women'squarters," and one is attached to an entrance that JARCE9 (1971-72), figs. 12 and 13, show no complete he viewed as a privateentrance for the stewardand male ser- houses but many small, tortuously connected rooms. The vants. One would expect buildings that contain the commu- plan labeled 89 by J. E. Quibell and F. W. Green, Hierakon- nity's grain reserves to be guarded, and the rooms are well polis II (London, 1902), pl. LXVIII,rooms 2-5, seems to placed for this. It is not unlikely that even small households constitute an earlyhouse. in such settlements of cult workershad at least one servant, W. Kaiseret al., "Stadtund Tempel von Elephantine, and a vestibule by the door might have doubled as the ser- 11./12. Grabungsbericht,"MDAIK 40 (1984), fig. 1, 174; vant's bedroom/livingroom, like the vestibule occupied by W. Kaiser,et al., "Stadtund Tempel von Elephantine,13./14. the bawwabin a Cairo apartmentbuilding. Grabungsbericht,"MDAIK 43 (1987), 91, fig. 6; for an overall 38 JARCE XXX (1993)

Fig. 3. Templesof the earlyperiod. Theseplans are based on Kemp,Ancient Egypt: Anatomy of a Civilization, pp. 76 (Hierakonpolis), 78 (Abydos), 68 (Medamoud), and 70 (Elephantine). and since the elite of this period came out of Temples were called the houses of the gods, the same tradition, their houses were probably but they bore little resemblance to the houses as closed in plan as the large houses from the of people. Temples of the New Kingdom were Khentkawes settlement. For purposes of this generally strictly axial, and far more open in analysis, then, it will be assumed that closed plan than houses, and there are indications plans were favored for all houses from the First that this was also true in the Old Kingdom and Dynasty through the end of the Old Kingdom earlier. (See fig. 3 for some Archaic Period and (and later), and thus that no significant change Old Kingdom provincial examples.) Symmetry took place in patterns of domestic architecture was important even in the most "un-Egyptian" between the Third and Fourth Dynasty. early temples, as, for example, in the strange shrine at Medamoud. Some early temples, for view, see W. Kaiseret al., "Stadtund Tempel von Elephan- example those at Abydos and Hierakonpolis, tine, 15./16. Grabungsbericht,"MDAIK 45 (1988), 145, fig. 4. had baffle walls at the entrance to block the It is not impossible that the architectureof royal pal- view of the sanctuary,but beyond that a visitor aces in the Fourth reflected some of the in Dynasty changes had a straight path, and was never required to the social and religious role of the king that are seen in mor- double back as in houses. tuary architecture. Changes in residential patterns in the contemporary capital may also have occurred, to reflect a changed rela- 16 tionship between the king and his subjects. Unfortunately, C. Robichon and A. Varille, "Medamoud.Fouilles du no royalpalaces of the Old Kingdomare knownfrom before Musee du Louvre, 1938,"CdE 14 (1939), 82-87. 17 or after the beginning of the Fourth Dynasty,so these prop- W. M. F. Petrie, AbydosII (London, 1903), pl. 50;' and ositions cannot be tested. Quibell and Green, HierakonpolisII, pl. 72. SOCIAL CHANGE IN THE FOURTH DYNASTY 39

1 8 O'Connor has recently suggested that the Hierakonpolis. Alternatively, these enclosures early temples at Elephantine and Medamud could have belonged to rulers centered at Hier- were peripheral to more important temples at akonpolis, as precursors, or rivals, or subordi- those sites, and that the temples at Hierakon- nates of the Thinite kings. The later character polis and Abydos were Sixth Dynasty ka-chapels, of the western enclosure as a cult place of again attached to more important, but undis- Horus might derive from the assimilation of its covered, shrines nearby. He has also noted a royal owner and that god, just as the tomb of significant similarity between the temple enclo- Djer was in the Eighteenth Dynasty thought to sure of Hierakonpolis (with entrances at the be the tomb of Osiris.2 If so, it is hardly likely east end of the north wall and the south end to have represented the standard temple plan. of the east wall and enclosing a stone-faced, Whatever the importance of the early shrine off-center mound) and the royal funerary en- of Satet at Elephantine, it was unarguably a di- closures on the plain west of Abydos as he has vine cult place of the Archaic Period, since the previously20 reconstructed them. On this basis, principal temples of later periods were built di- he has suggested that such enclosures represent rectly above it; and the Medamoud structure the standard form of early temples, and he be- must also have been a temple for the same rea- lieves that a temple of this shape is to be re- son. These shrines resemble the small shrines stored inside the town wall at Abydos as the of the Djoser complex in their openness. Icon- site's principal temple. ographic evidence suggests that barriers at the Other interpretations of this similarity are temple entrance were largely symbolic: only a possible. East of the Horus temple enclosure is small picket gate was shown in front of archaic a "palace" gateway, located at the east end of a temples in hieroglyphic signs, presumably the northern wall, with deposits of sand (like that in same that is replicated in stone in the shrines the Horus temple mound) to the south. O'Con- surrounding the jubilee court in the Djoser nor has identified these elements as parts of a complex. second enclosure of the same type. Since it is Religious rituals are notoriously conservative, unlikely that two large temple enclosures would and one would want far more evidence than ex- be built so close together, and since Horus is ists to postulate a major change in them; con- not later paired with another deity at this site, it sequently the buildings in which they were seems more plausible to interpret both of these performed probably had the same access pat- Hierakonpolis enclosures as the funerary en- terns in earlier periods as they did later. For ex- closures of early kings. The relative position of ample, in later periods, gods were frequently the two enclosures and their relationship to the carried forth to take part in public ceremonies, Nile would not be unlike that of the Abydos en- and their passage through their temples was closures. Since not all of the kings buried on likened to the passage of the sun across the sky. the Umm el-Qab at Abydos were represented If such ceremonies took place in the earlier on the plain, perhaps some of them had funer- periods, there would have been both practical ary enclosures that served as their cult places at and symbolic reasons for temples of the early period to have open plans.23 18 D. O'Connor, "The Status of EarlyEgyptian Temples: 21 an Alternate Theory," in: TheFollowers of Horus: StudiesDedi- W. M. F. Petrie, The Royal Tombsof the Earliest Dynasties cated to Michael Allen Hoffman, 1944-1990, Renee Friedman II (London, 1902), p. 8. 11 and BarbaraAdams, eds. (Oxford, 1992), 83-98. C. Firth and J. E. Quibell, The StepPyramid (Cairo, 19 O'Connor's interpretationdoes not, however,explain 1935), pl. 62 bottom. Detailed examples of the hieroglyphic the small shrine at the north east corner of the great court signs occur on two of the reliefs decorating the subterra- in the Djoser pyramidcomplex. It is nearly identical to the nean chambersin the same complex (ibid., pls. 17 and 40). two Abydoschapels in both plan and orientation, but is un- Carrying-chairshrines seem to have occurred from likely to be a ka-chapel,since it is alreadylocated in a mor- the very earliest period. See, for example B. Kemp, Ancient tuarymonument. Egypt: Anatomy of a Civilization, 93, fig. 33. Other gods may * David O'Connor, "New Funerary Enclosures (Tal- have traveledby sledge. Processionsof divine standardsare bezirke)of the Early Dynastic Period at Abydos,"JARCE 26 ubiquitous in the iconography of the late predynasticand (1989), 51-86. Archaicperiods. 40 JARCE XXX (1993)

Fig. 4. Second and Third Dynasty tombsubstructures with domesticfeatures: B is the burial chamber/bedroom; H is a room for use and storageof waterjars; and L representsa room with a latrine. Theselabels are hypotheticalin the Hetepsekhemwy substructure,based on the similarity of the shapes and configurations of rooms to those of the private tombs, (a-c) Private tombs at Saqqara (after Quibell, Archaic Tombs, pi 30, no scale given); (d) royal substructureof Hetepsekhemwyat Saqqara (after Lauer, Pyramide a degres 1, p. 5).

PrivateTombs form mimicked the bed platform found in bedrooms of private houses. (The rooms with Both before and after the beginning of the bed platformsat Kahun and the rooms assumed Fourth Dynasty,the best attested type of mortu- to be private sleeping quarters in the Khent- ary architecture is the private tomb. The large kawes houses were also to the west.) To the east private tombs of the Second and Third Dynas- of the end room was a more complex group of ties at Saqqara and elsewhere were viewed rooms, among them usually one containing a literally as houses of the dead, and their sub- model latrine and another, north of it, contain- structuressometimes contained quintessentially ing an emplacement for water jars. This latter domestic features (see fig. 4a-c). These sub- room often had a separate second entrance structures were normally entered by a stairway from a vestibule north of the end room, per- from the north or east, leading to a corridor haps a "servicepassage," like those seen in later that ran south under the long axis of the over- private houses. These rooms probably also du- lying mastaba,periodically blocked by portcullis plicated the living quarters of the tomb owner. stones. The corridor usually ended in a large Both along the axial approach to the inner room, to the west of which was the burial cham- suite of rooms and in the body of the overlying ber, where in some tombs a raised burial plat- superstructure, these tombs contained storage SOCIAL CHANGE IN THE FOURTH DYNASTY 41

Fig. 5. Private tomb chapels: (a) Second and Third Dynasty mastabas at Saqqara, after Quibell, Archaic Tombs, pl. 2; (b) Fourth Dynasty chapel of Nefermaat at Meydum, after Petrie, Medum, pi 7; and (c) the Fourth Dynasty chapel of KhufukhafI (after Simpson, The Mastabasof Kawab,Khafkhufu I and II,fig. 19).

areas. FirstDynasty tombs at Saqqaraalso stored niches or a recessed cruciform chapel, was cut grave goods both above and below ground, and into the body of the mastaba, and seems ini- the tradition seems to have continued into the tially to have been open to a direct approach. Third Dynasty. In fact, however, these chapels were typically The rectangular mastaba massif of the Saq- approached by extremely complex paths cre- qara superstructures also continued the older ated by walls and rooms outside the body of the tradition. It was oriented with its long axis run- mastaba (see fig. 5a). The approach often ran ning north to south, and it was usually provided along the facade and then twisted around ear- with a niched facade or isolated niches on its lier structures and the mastaba's own rooms eastern face. The cult focus, either one of these and serdabs. The path could branch several times before reaching the cult place, so that a 24 W. B. Emery, ArchaicEgypt (Harmondsworth, 1961), stranger approaching it might easily be lost. 158, notes that these internal features in the body of the The large tombs that now lack these complex mastaba"had not quite died out" in some "bigtombs of the exterior tend to be in areas where Second a tomb of the Third approaches Dynasty";however, Dynasty,QS the shafts are so it is 2305, contained both large storage tanks in its superstruc- secondary thickest, likely ture and sealings of Djoser. (J. E. Quibell, ArchaicTombs, 1913-1914, Excavationsat Saqqara6 [Cairo, 1923], pl. 2.) 25 The datings of many of these tombs are based on Second The excavator described these chapels in general as Dynastyroyal names occurring in them; but many of these "accessible only along narrow, zigzag passages" (Quibell, kings' names also occur in the substructureof the Third Dy- Archaic Tombs,p. vi). 26 nastyStep Pyramid. See Quibell's plan, ibid., pls. 1 and 2. 42 JARCE XXX (1993)

tremely rare and the approach to the cult place was either direct or through a simple exterior building. The plans were uniformly more open than those in the larger Third Dynastytombs29 (see fig. 5b-c). The Fourth Dynasty private tombs at Mey- dum show a marked increase in decoration, of- ten carved on a limestone facing that lined the cruciform chapels. Here, the commodities and equipment recorded in such loving detail by Hesy-Re's artistswere reduced to compartmen- tal lists. Most notable, however, was the inclu- sion of family members in tomb decoration. Couples often shared tombs, and sometimes in the table scene of the 6. Fourth substructures:(a) the appeared together Fig. Dynasty private shaft false while their children were shown of Kawab at Giza (after Simpson, The Mastabas of doors, Kawab, Khafkhufu I and W,fig. 7; (b) the substructureof flanking the central niche. Husbands and wives the pyramid of Khafre at Giza (after Edwards, Pyramids of the period could also be represented to- of Egypt,p. 132). gether in statuary on the same scale. The quantity of wall decoration was sharply (and temporarily) curtailed in the reign of Khufu. that such complex approaches were more com- where it was slab ste- mon than their survivalindicates. replaced by finely painted las and mastabas built entirely of stone, but The decoration of the chapels of Third Dy- the occasional occurrence of members tombs was limited to the stela family nasty normally along with the male tomb owner continued, es- with the table scene and other representations as the decoration to increase in of the deceased. Women and men pecially began apparently quantity again. had their own cult places. If the tomb of Hesy- There is little evidence of burial Re was that were more exten- very equip- typical, chapels ment from either the Fourth Dynasty or the sively decorated added representations of food but it is that and doubtless like the period preceding, very likely equipment, very supplies burials during the Fourth Dynastywere consid- that filled the numerous storerooms, and geo- erably poorer than they had been previously. metrical motifs on the niched facade. Servants The substructureswithout storerooms and scenes of life were provided daily represented only space for only a limited amount of grave goods, in the outer rooms. and the of stones in the late Third several disappearance portcullis sug- Already Dynasty, gests that there was little to steal. Support for changes began to take place in private tombs. In the the storerooms and substructure, portcullis 29 stones disappeared, and the suite of rooms at At Meydum, the original cruciform chapels seem to the end of the corridor was a have been replaced by an even simpler form, a simple offer- replaced by single ing court with a single central niche. (Petrie, Medum[Lon- room with no domestic features (see fig. 6a). By don, 1892], pl. 7). the Fourth Dynasty, the superstructuresof pri- Petrie, Medum,pls. 9ff. Interestingly,the women, who vate tombs had also become considerably sim- are shown in positions where both wives and mothers fre- pler. Although they retained the rectangular quently appear later, are not specificallycalled hmt.f,"his north-south and often the wife." However, it is most likely that they were wives, since shape, orientation, none have titles and the men are all sons. ex- queenly king's cruciform chapel, niched facades became MohamedSaleh and Hourig Sourouzian, OfficialCata- logue of the Egyptian Museum, Cairo (Mainz, 1987), entry 27 27 J. E. Quibell, TheTomb ofHesy, 1911-1912, Excavations and bibliographytherein. at Saqqara5, (Cairo, 1913). W. S. Smith, The Art and Architectureof Ancient 2 Egypt, Hesy-Re,for example, had scenes in his outer corridor 2nd ed., revised by W. K. Simpson (Harmondsworth,1981), of men leading cattle, and a crocodile in a pool. (Ibid., 10.) 104. SOCIAL CHANGE IN THE FOURTH DYNASTY 43 this suggestion can be found in the burial of Second and Third Dynastiesare not verywell at- Hetepheres I. Although she was probably the tested. Of Second Dynastyroyal tombs, we have most important person in the country after her only the Upper Egyptian "forts" at Hiera- son Khufu himself, her tomb contained only a konpolis and Abydos, the tombs of Peribsen bed and its canopy, its curtains in an inlaid box, and Khasekhemwyat the Umm el-Qab at the two chairs, a carrying chair, ceramic vessels, latter site, and two impressive underground and several boxes holding a collection of jew- substructuresat Saqqara, the superstructures elry and other equipment.33 Aside from her of which have been lost. From the Third body, it is unlikely that this burial chamber, Dynasty,we have the complexes of Djoser and now thought to have been her original place of interment, could have held much more. Such Royal Tombs of the First Dynasty,"JEA 52 [1966], 13-22; is with and idem, "TheEgyptian 1st Dynastyroyal cemetery," Antiq- equipment meager indeed, compared 41 the have the food, furniture, and other uity [1967], 22-32), Abydos complexes may clothing, sup- been far larger and more elaborate than their Saqqara plies that must have filled the extensive store- counterparts.The tombs at Saqqarathus probablybelonged rooms of far less important people during the to the officials whose sealings and stelas were found in first three dynasties. them. The private nature of the Saqqaratombs is further confirmed in the of below. In general, then, although the privatetomb of analysis cemeteryorganization the Fourth continued the traditional W. Kaiser, "Einige Bemerkungen zur agyptischen Dynasty FriihzeitIII: Die Reichseinigung,"ZAS 91 (1964), 104 n. 4. external shape and orientation of the preceding W. M. F. Petrie, The Tombsof the Courtiersand Oxyrhyn- period, it can be said to have been poorer in chus(London, 1925), and, most recently, O'Connor,JARCE richer in its materials 26(1989), 51-86. contents, though building 38 and its decoration. members to be Petrie, RoyalTombs II, 11-14, pls. 61 and 63. Family began The westernmost substructureis attributed shown in decoration, the wife on the generally depicted to Hetepsekemwy,although it also contained sealings of his same scale as her husband; and the depictions successorRa-neb. A plan is given in Lauer,Pyramide a degres of domestic furniture were greatly reduced in I, 4, although this plan differs somewhatfrom the detailed importance. Both the mastaba superstructure verbal account given by the excavator,A. Barsanti,in "Rap- sur les deblaiements autour de la and its cult place became simpler in plan and ports operes pyramide more and the burial cham- d'Ounas,"ASAE 2 (1901), 250-53, and in "Fouillesautour directly approached, de la pyramide d'Ounas 1901-2," ASAE3 (1902), 182-84. ber no longer replicated the tomb owner's The second substructureopens just south of the southwest house on earth. corner of the mastabaof Nebkauhorand contained Archaic Period vessels and sealings of Ninetjer in addition to many late period burials. It was mentioned briefly in S. Hassan, Royal MortuaryComplexes "Excavationsat Saqqara,1937-1938," ASAE38 (1938), 521 and H. Chevrier, "Les Fouilles," CdE 13 (1938), 283 (iv). Unlike the royal tombs of the First Dynastyat Though it is frequently described as similar to Hetep- Abydos,35the royal mortuary complexes of the sekhemwy's,the plan of the northern part of this substruc- ture that has been published (P. Munro, "Der Unas- Friedhof Nord-West 4./5. Vorbericht iiber die Arbeiten 33 Ibid., 87-95. Hannover/Berlinin Saqqara,"GM 63 [1983], 109) differs 34 Mark Lehner, The pyramid tomb of Queen Hetep-heresI substantially. and the satellitepyramid of Khufu (Mainz, 1985), 35-44. R. Stadelmann, "Die Oberbauten der Konigsgraber 6b The older view that the tombs at the Umm el-Qab der 2. Dynastie in Sakkara," Melanges Gamal eddin Mokhtar, were cenotaphs and that the larger First Dynastytombs at BdE 97/2 (Cairo, 1985), 295-307, has suggested that the Saqqarawere the kings' actual burial places is unlikely.The long storeroom structures along the western edge of the only real argument for identifying the tombs at Saqqaraas Djoser complex represent a third Second Dynasty tomb, royalwas that they were larger than the burialsat the Umm and restores the other superstructuresaccordingly. How- el-Qab. This argumentignores the value of location: a small ever, the rooms at the southern end do not resemble the tomb on sacred ground can be more desirable than a larger "bedroom-lavatory-bathroom"complex at the southern end tomb elsewhere. (A possible example of this phenomenon of Hetepsekhemwy's substructure. W. B. Emery, Archaic from a later period is the comparativesizes of the tombs of Egypt,144-45, suggested that the internal stepped structure the Votaresses of Amon in the Medinet Habu enclosure, found in a Saqqaramastaba from the reign of Anedjib mim- and the tombs of their stewardsin the Asasif.) Moreover, icked contemporaryroyal superstructuresat Abydos,which according to B. Kemp's convincing analysis of the en- ultimately were the source of the Step Pyramid. Icono- closures on the plain at Abydos (B. Kemp, "Abydosand the graphic and textual evidence from Abydos seem to support 44 JARCE XXX (1993)

Sekhemkhet at Saqqara, as well as several unex- clearly suggests that they, like the private tombs cavated complexes usually believed to date to of the same dynasty, represented in microcosm their successors. The unfinished "Layer Pyra- the private apartments of the tomb's owner. mid" at Zawiyet al-Aryan probably also dates to The substructures of the late Second Dynasty the end of this dynasty, as does, perhaps, the tombs at Abydos attributed to Peribsen and stepped pyramid underlying the pyramid of Khasekhemwy also consisted predominantly of Meydum. storerooms. Here, however, the burial chamber The earliest of these structures is remarkably (with no domestic characteristics) was at the similar in spatial organization, though not in center, surrounded by storerooms, presumably size, to private tombs. The western of the two following the pattern of the nearby First Dynasty Saqqara substructures, which is generally attrib- tombs. This pattern of surrounding storerooms uted to king Hetepsekhemwy (fig. 4d), was en- continued in the substructures of the Third tered from the north through a long corridor Dynasty. Djoser's pyramid had four groups of that was flanked on either side by groups of ca- storerooms, each radiating out from one side pacious comb-like storerooms. The rooms at of his central burial chamber, while the pyra- the south end of the corridor were complex mid of Sekhemkhet and the Layer Pyramid at and irregular in plan. These innermost rooms Zawiyet al-Aryan both had corridors of store- included a large room on the west, like the rooms that branched off the main axis before burial chambers in private tombs. East of the the burial chamber and encircled the burial main axis was a more complex group of rooms, chamber on three sides. similar to those in private tombs that contain a Like the Second Dynasty substructures, the latrine and areas for water storage. No bed plat- superstructure of the Djoser complex was form or latrine slab appears in the published probably reminiscent of the palace complex in plan of the tomb; however, the plan may have which the king lived during his life on earth. been made without completely clearing the The enclosure, like most other early tombs, was floor. Whether or not these features were in- oriented with its long axis running north-south. cluded, the layout of the innermost chambers The complex was extremely difficult to enter. There was no valley structure or causeway, so a visitor must have found his own to the en- this hypothesis (see Ann Macy Roth, Egyptian Phyles in the way Old Kingdom: the Evolution of a Systemof Social Organization, closure from the edge of the cultivation. Only SAOC 48 [Chicago, 1991], 167-68), in which case the su- one of the many model doorways in the niched of these Second tombs, and all perstructures Dynasty royal enclosure wall actually gave access to the inte- until that of Snefru, can be as- superstructures reasonably rior, and the entrance colonnade led sumed to have been stepped. although into the south of the See, for example, the photographsin GeoffreyT. Mar- large courtyard pyramid, tin, The Hidden Tombsof Memphis:New Discoveriesfrom the only someone familiar with the plan would have Timeof Tutankhamunand Ramessesthe Great (London, 1991), known how to reach the structure on the north 22 (fig. 6), and in Jean Capart, Memphis,a Vombredes pyra- side of the that is believed to mides iv. These structuresare assumed pyramid generally (Bruxelles, 1930), p. be to be of Third Dynasty date, but since we know nothing Djoser's mortuary temple. This of access to the about the Second Dynastysuperstructures at Saqqara,some difficulty complex itself may date to that period, as Stadelmannhas suggested ("Die was also the result of a long tradition. The First Oberbauten der Konigsgraber,"304-7). The anonymous and Second Dynasty monuments at Abydos, enclosures west of however, are later. Djoser's, probably both the tombs on the Umm el-Qab and the en- Compare, for example the tombs on the Umm el-Qab at closures nearer the (see 7a), were often Abydos, and the funeraryenclosures on the plain north of city fig. the same city, as well as the tombs of the first three dynasties 4t in the northern part of the Saqqaracemetery. Corridors that branch and surround the innermost The Third Dynastysites are well covered in a number group of rooms may be attested as early as the Saqqarasub- of general books: Edwards, The Pyramidsof Egypt,34-69; structureof Ninetjer. Chevrier'saccount of its discoveryin R. Stadelmann,Die dgyptischenPyramiden, 31-79; idem, Die CdE13 (1938), 283, describes it as extending east, west, and GrossenPyramiden von Giza (Graz, Austria, 1991), 54-71. south of its entrance; and the partial plan published by (Stadelmann attributes the unexcavated complexes to the Munro, GM63 (1983), 109, also suggests that the corridor Second Dynasty,however.) branched to the east and westjust south of the entrance. SOCIAL CHANGE IN THE FOURTH DYNASTY 45 sited directly behind one another, making the substructure, which remained similar to those western monuments less visible and less invit- of the Third Dynasty, except that, beginning ing. This is also true of the Second and Third with the Meydum pyramid, the corridors of Dynasty complexes and enclosures at Saqqara storerooms vanished (see fig. 6b). Usually, a (see fig. 7b). Clearly none of these complexes single passage descended from the north face were meant to attract casual tourists, and access of the pyramid, and then ascended to the was probably restricted to people who knew the burial chamber. The burial chambers of the layout well. pyramid at Meydum and the Bent Pyramid at This pattern of indirect access was even more Dahshur were oriented with their long axis noticeable in the plans of individual buildings north-south, but beginning with the Northern in the Djoser complex (fig. 8a-c). The mortuary Pyramid at Dahshur, the burial chamber was temple had long hallways that circled the build- usually oriented with its long axis east to west. ing and doubled back on themselves before By the reign of Khufu, the position of the coffin leading to the principal rooms; and, on the had been established at the west end of the west, it had service passages located to bypass chamber, just as it had been in the "bed cham- the two butchering areas. Its complexity and bers" of the Second Dynasty tombs. This might tortuous pathways equal those of the Kahun have been a compromise between the tradi- mansions. Also similar to domestic plans in tional north-south axis of the substructure (and closedness were Temple T and the peculiar of all earlier superstructures) and the new east- complex of twisting passages southeast of the west axis of the Fourth Dynasty superstructure. jubilee court. (The shrines in the jubilee court, The second entrance to the Bent Pyramid of probably modeled on traditional shrines, de- Dahshur from the west may have been an ear- part from these patterns, as does the triple lier attempt to solve this problem. shrine that may imitate the shape of the early As in private tombs, the disappearance of temple at Abydos.) Interestingly, even when the storerooms must necessarily have meant a de- plans of their rooms were extremely closed, the crease in the quantity of goods buried with the stone doors of these buildings were all ren- king. The earlier storerooms were clearly not dered eternally open, perhaps reflecting a ten- empty, and their contents would not have fit sion between the closed plan of the palace that into the small chambers provided for the later, served as a model for the complex, and a reli- much larger, monuments. Interestingly, inter- gious requirement that the mortuary monu- nal storerooms began to appear again just as ment be accessible to the king's spirit. the pyramids began to decrease in size. Men- That the closedness of the Djoser complex was kaure's pyramid had a side chamber giving ac- not an isolated example is clear from the "token cess to six storerooms, Shepseskaf 's tomb had a palaces" in the southeast corners of the Peribsen corridor with five storerooms, and the Fifth Dy- and Khasekhemwy enclosures at Abydos (see nasty pyramids routinely had three. fig. 8d-e). Although the Khasekhemwy building The superstructure of the Fourth Dynasty was considerably more complex than Peribsen's, royal tomb changed far more radically than its both led the visitor from the south to the north substructure. The rectangular enclosure was end of the building and then to the south again, abandoned in favor of a linear series of diverse a typical domestic arrangement. A tradition of structures (valley temple, causeway, mortuary royal mortuary buildings with domestic charac- temple, pyramid) that ran from east to west, be- teristics thus lay behind the Djoser complex. ginning at the edge of the cultivation. The diffi- The change in the royal mortuary complex at culty of access that characterized the Djoser the beginning of the Fourth Dynasty was far 45 more radical than the change in private tombs. Both of the Dahshurpyramids and that of Khufuhave roomswhere have been stored, The element that changed the least was the secondary gravegoods might but all of these are on the main axis, and do not seem to fit the Egyptianconception of "magazines."Khafre's pyramid 44 E. R. Ayrton, C. T. Currelly, and A. E. P. Weigall, has a large room at the end of a passageat right angles to the AbydosIII (London, 1904), pl. 6 and 7. principalpassage, but no storeroomsopen off of it. 46 JARCE XXX (1993)

Fig. 7. Layouts of royal enclosuresat (a) Abydos (after Kemp,JEA 52 (1966), p. 14) and (b) royal enclosuresand private tombs (black) at Saqqara. (After B. G. Triggerin: Triggeret al, Ancient Egypt: A Social History (Cambridge,1983), p. 14. Additional royal enclosureshave been added, based on Stadelmann, Die agyptischen Pyramiden, p. 30; and the placement of individual tombsin the subsidiary cemeteriesto the west are based on Kaiser, MDAIK 41 [1985], p. 49.) SOCIAL CHANGE IN THE FOURTH DYNASTY 47

Fig. 8. Threebuildings in the Djoser complex:(a) TempleT, (b) a building at the southeast cornerof thejubilee court, and (c) the mortuarytemple (after Firth and Quibell, The Step Pyramid, pls. 67, 59 and 27, respectively.Earlier buildings from the Abydosenclosures of (d) Peribsenand (e) Khasekhemwy(after Kaiser, MDAIK 25 (1969), p. 9). (Not drawn to the same scale.)

complex was entirely eliminated, both in the found their way to the sanctuary. In practice, complex as a whole and its individual elements. however, their way might have been blocked, The new complexes were exceedingly axial and since the Fourth Dynastycomplexes apparently symmetrical in plan (see fig. 9). An S-twist or had working doors, unlike the perpetuallyopen baffle wall sometimes obscured access to the stone doors at the Djoser complex. The in- sanctuary itself, but the approach was far sim- creased accessibility was thus probably more pler, and without confusing side passages or re- ideological and symbolic than practical. versalsin direction. In their degree of openness, Another clear change in royal superstructures the royal mortuary temples resembled closely was the increase in their decoration. Where the temples of the gods. decoration in the Step Pyramid Complex had Even the size of the Fourth Dynastypyramids been limited to the reproduction of plant mo- enhanced their accessibility. The earliest royal tifs and six panels depicting the king placed in monuments on the Umm el-Qab were probably the inaccessible substructure, extensive figura- topped with low (2.5 m maximum) mounds tive relief decoration began to appear on the that would have been almost completely invisi- walls of the superstructure associated with the ble from a distance. There was a steadygrowth in Bent Pyramid, and there are indications that visibility from that time through the reign of structures of Khufu and Khafre also bore wall Djoser, when the burial mound-Step Pyramid decoration.4 While such decoration did not it- extended above the high enclosure wall. By any self increase the complex's accessibility,it dem- measure, the early Fourth Dynasty pyramids onstrates again the shifting of focus from the were larger; towering above lower enclosure substructureto the superstructureof the tomb. walls, they could be seen and understood by all However different in effect, the changes that levels of Egyptiansociety. The isolation and plan occurred in royal tombs in the early Fourth Dy- of the entire Fourth Dynasty royal complex nasty move in the same general direction as the made its spatial organization obvious from the changes that took place in contemporary pri- valley, and in theory, strangerscould easily have vate tombs. Both private and royal tombs lost their storerooms, their closedness, and their 46 G. Dreyer, "ZurRekonstruktion der Oberbauten der Konigsgraberder 1. Dynastiein Abydos,"MDAIK 47 (1991), H. Goedicke, Re-UsedBlocks from the PyramidofAmenem- 102. hetlatLisht (New York,1971), nos. 1-7. 48 JARCE XXX (1993)

and possibly at the even older site of Hierakon- polis. Earlier Second Dynasty kings, however, were apparentlyburied at Saqqara,perhaps be- cause of the presence of some favored deity or an illustrious ancestor in the non-royal ceme- tery there. It was to this newer royal cemetery that the Third Dynastykings returned. During most of this period, then, the kings built tombs awayfrom their subjects, in special cemeteries where their ancestors had been bur- ied. Even at Saqqara,which had originally been a privatecemetery, a sharp dividing line marked by natural barrierswas maintained between the Fig. 9. Royal mortuarytemples of the Fourth Dynasty: (a) royal sector to the south and the private sector at the Meydumpyramid and (b) at the pyramid of Khufu to the north (see fig. 7b). Despite the lack of (after Edwards, Pyramids of Egypt, pp. 75 and 112 space caused by the giant enclosures, later kings respectively).(Not drawn to the same scale.) preferred to build in less desirable western areas, or to raze the superstructures of their predecessors, rather than build in the non- domestic features, and their cult places began royal cemetery to the north. Private individuals to resemble the more open plans of temples were equally restricted in siting their tombs. In- and receive decoration. But while earlier royal creasingly, they expanded towardsthe west and tombs were simply elaborate versions of their towardsAbu Sir to the north, but no tombs were private counterparts, the tombs of Fourth Dy- built in the southern, exclusively royal, sector nasty kings changed in ways that sharply distin- until the Fifth Dynasty. (A similar avoidance guished them from those of their subjects. The pyramids made powerful symbolic statements, 48 as did the and that W. Kaiser has suggested that the concentrations of valley temples causeways put lines of subsidiarygraves north of the later entrance to the these monuments in active and direct contact Serapeum are connected with some sort of First Dynasty with the populations of the living. In this royalcult there ("EinKultbezirk des KonigsDen in Sakkara, period there is no doubt which tombs belong to MDA1K41[1985], 47-60). This is possible, or the subsidiary be related to burials of the kings and which to commoners. graves might early Apis bulls, who were buried in this area in later periods. 49 Whatever their form, the superstructuresof Hetep- Cemetery Organization sekhemwyand Ninetjer seem likely to have been casualties of Djoser's construction work to the north, since any but In addition to changes in the size, shape, con- the most minimal superstructurecovering these substruc- interfered with the construction of tents, and orientation of royal tombs, the latter tures would have his of the Third also marked a massiveenclosure wall. Djoserapparently had special access part Dynasty change to the possessions of these earlier kings, since seventeen in their location. The Umm el-Qab, in the vessels found in his storerooms bear the name of Hetep- desert west of Abydos, was a traditional royal sekhemwyand thirteen that of Ninetjer (P. Lacau and J.-P. cemetery even before the First Dynasty kings Lauer, La Pyramide a degresTV: Inscriptions gravees sur les vases This would be were buried there. Except for the surrounding [Cairo, 1959-1961], 29-38). explained by it was and the assumption that Djoser leveled their tomb superstruc- subsidiary burials, exclusively royal; tures and appropriatedthe contents. The name of Djer also whether or not these subsidiary burials were occurs on thirteen vessels, usuallyassociated with the insti- sacrificial, the people buried in them seem to tution Smr-ntrw,which perhaps also fell victim to Djoser's have been relegated to the status of burial workmen. No other king is mentioned on more than eight vessels, and Djoser himself is mentioned on only one. equipment, providing labor and companion- 50 for the as servant models did in This division of the Saqqaranecropolis would hardly ship king just have been so strictly maintained had the First Dynasty later periods. The last few kings of the Second tombs in the northern sector been the burial places of the Dynasty also built tombs at the Umm el-Qab, FirstDynasty kings. SOCIAL CHANGE IN THE FOURTH DYNASTY 49

Fig. 10. The royalpyramid cemeteriesof Dahshur (after Stadelmann, Die agyptischen Pyramiden, p. 88; no scale given) and Giza (after O'Connor,World Archaeology 6 (1974), p. 21. (Not drawn to the same scale.)

of the area of the royal enclosures at Abydos hind them to the west, moving ever westward seems to have lasted until the FirstIntermediate as the prime areas on the escarpment itself Period.)51 became crowded. There was also an apparent The end of the Third Dynasty, however, tendency to move northward, away from the brought a new pattern. In the latter part of this royal tombs that had begun to be built to the and most of the following dynasty, each new south, but this may be the result of uneven king built his tomb at a new site, often at a great preservationand excavation. distance from the tomb of his predecessor. In the Fourth Dynasty, high officials and (The many small step pyramidsfound through- members of the royal family seem to have aban- out Egypt, dating technologically to the late doned this traditional cemetery to build their Third Dynastyor Snefru's reign, may be related tombs in cemeteries near the royal tomb. At to this policy.) This pattern was broken in only a Meydum and Dahshur, these private "pyramid few reigns, and it was continued intermittently cemeteries" were located some distance from into the Fifth Dynastyby the kings who initiated the royal tomb, at least as far as the distance new cemeteries at South Saqqaraand Abu Sir. between the royal and non-royal sectors at These repeated breaks with ancestral tradi- Saqqara (see fig. 10a). The distance between tion can also be seen in the private cemeteries the royal and private tombs decreased mark- of the Fourth Dynasty. The high officials and edly at Giza (compare fig. 10b); but the novelty royal family members at Memphis had been, if lay not in the proximity to the royal tomb, but anything, more conservative than their royal in the dependence upon it. When royal tombs overlords in locating their tombs. First Dy- 52 This movement have with nasty officials built their tombs in irregular may begun simultaneously rows the at north the moving of royal tombs awayfrom Saqqara,since there along escarpment Saqqara were brick mastabasexcavated north of the LayerPyramid of the central wadi. Their successors of the of Zawieyetel-Aryan. (Dows Dunham, Zawiyetel-Aryan: The Second and Third Dynasties built tombs be- Cemeteriesadjacent to the LayerPyramid [Boston, 1978] , 34.) At Meydumand the Bent Pyramid,subsidiary cemeteries also 51 J. Richards,"Understanding the MortuaryRemains at were laid out to the north, perhaps mimicking the geogra- Abydos,"NARCE 142 (1988), 7-8. phy of Saqqara. 50 JARCE XXX (1993) moved from Abydos to Saqqara and back again shape of their pyramids, and is made explicit in in the first two dynasties, the tombs of officials texts of the early Fourth Dynasty. had remained at Saqqara without reference to the site of the tomb. both royal (That private Conclusions and royal tombs at Saqqara tended to move westward was due to similar constraints spatial The changes described above, all of which rather than any relationship.) occurred around the time of the beginning of Under the new tomb builders were system, the Fourth Dynasty, are summarized in Table 1. in the new ceme- granted planned spaces royal This collection of contrasts suggests strongly teries the the central surrounding pyramid by that the beginning of the Fourth Dynasty coin- in to some mea- authority, probably proportion cided with two fundamental changes, one sure of their social rank and political impor- affecting the conception of the afterlife and the tance. At the had been Saqqara private cemetery other affecting the relationship between the a mix of tombs of homogeneous officials, varying king and his subordinates. in size and one another in an jostling against The new and striking contrast in the archi- effort to claim the most advantageous position. tectural form of the tombs of kings and com- Now the tombs were laid out in even private moners alike suggests a change in beliefs about rows, and fell into a uniform range of sizes. the nature of the afterlife and the needs of the These tombs were not associated with the only dead. Fourth Dynasty Egyptians no longer royal tomb, but were to some extent dependent viewed the afterlife as identical to life on earth, since the was of a upon it, cemetery clearly part and hence they no longer required earthly centered large, planned mortuary landscape goods to take part in it. The house plans of the With their new upon pyramid. privileged commoners ceased to affect the plan of their to the the proximity royal tomb, paradoxically, tomb chambers, and the buildings necessary officials' tombs resembled so much as nothing for the king's earthly activities were not dupli- the around the First subsidiary graves Dynasty cated in his mortuary complex. At the same tombs that had to a far royal tombs, belonged time, the amount of grave goods buried with lower stratum of Unlike these earlier society. the deceased, which had been increasing in occurred in clusters rather tombs, however, they quantity and variety since the beginning of the than rows.53 predynastic period, was suddenly drastically re- The location of and tombs and royal private duced, as indicated by the reduced storage the between them reflected relationship clearly space available for such goods in both royal and a social towards the end of the major change private tombs. Third The of of Dynasty. authority ancestors, For the earthly food, furnishings, and domes- historical and of tribal family ties, perhaps loy- tic spaces that were supplied in older tombs, alties was weakened in both the and royal pri- Fourth Dynasty officials seem to have substi- vate and in the it seems spheres, private sphere tuted two new requirements, the perpetual cult to have been a replaced by greater dependence ceremonies performed by the living and the the of the The upon power king. new, indepen- blessings of the king. Cult service of some kind dent of tombs that these position royal suggests probably existed in earlier periods, at least for no derived their from their kings longer power kings, but it may have been very different from to earlier this source of au- relationship kings; what it was later. The architecture of both royal thority may have been replaced by the new rela- tionship of the individual kings to the sun god 54 that has been postulated on the basis of the Stone bowls inscribed with the name of the Zi-ho-nb/ Hr/jjtj,presumably royal tombs, and phyles of some type of cult functionaries are known from the end of the first dy- 53 The contrast between the pattern in linear cemeteries nasty (Roth, EgyptianPhyles in the Old Kingdom, 154-69); and of private tombs and the clusters of royal monuments has an early table of distributionwas found at the Djoser com- alreadybeen noted by O'Connor,JARCE 26 (1989), 59 and plex that is similarto those found at the Fifth Dynastycom- n. 23. plex of Neferefre (ibid., 181-88). SOCIAL CHANGE IN THE FOURTH DYNASTY 51

Table 1: MortuaryCustoms in the Third and Fourth Dynasty

Third Dynasty Fourth Dynasty PrivateTombs Complex "domestic"substructures Single chamber Substructureentered by stairway Most often shaft from top Manystorerooms, portcullis No storerooms, no portcullis Plentiful grave goods Few grave goods Chapel has closed plan Open access to chapel Mudbrickconstruction Largelyor entirely stone Man and wife have separate chapels Man and wife represented together Wall decoration rare Increasing wall decoration

Royal Tombs Stepped pyramid True pyramid Not prominent or accessible Prominent, accessible appearance Axis of enclosure north-south Axis of complex east-west Elaboration of private plan Exclusivelyroyal form Asymmetrical"domestic" plans Symmetrical"temple" plans House-like substructure Single burial chamber, antechambers Many,many storerooms Few, if any, storerooms Permanentlyopen doors (Djoser) Real doors, could close Decoration underground (Djoser) Decorated cult places

Cemetery Organization Ancestral cemeteries of officials Privatecemeteries centered on pyramid Kings buried together New site for almost every reign Privatetombs independent of royal Privatetombs move with king's Siting of tombs uncontrolled Privatetombs laid out on grid and private tombs makes it unlikely that their the tomb owner, was also a change in the direc- builders wanted to encourage casual visits to the tion of ensuring the service of the cult. It shows tomb, such as those requested by the later "calls a desire to attract casual visitors who might be upon the living," and it is possible that private inspired to make an offering by the implied tombs were essentially abandoned after the fu- power of the tomb owner in the spirit realm. neral. The absence of family members of the Another factor in Fourth Dynasty private tomb owner from tomb decoration in the early cults was the increased importance of the king. period, and their ubiquity afterwards, is also The need for his "blessings"is suggested by the suggestive. The tomb owner's descendants were inauguration of pyramid cemeteries. This was largely responsible for the carrying out of the in part a practical dependence. A tomb site cult, and their representation and hence im- near the royal pyramidoffered the possibilityof mortalizationin tombs may have been an incen- access to the more expensive materials and tive for more faithful service. The transfer of royal crafts specialists of the pyramidproject, as resources from the cutting and equipping of nu- well as the status boost of proximity to such an merous underground storerooms (an expensive important monument. But the dependence but invisible investment) to stone-built super- suggested by the metaphor of cemetery organi- structures with stone-carved decoration, which zation was not entirely economically based. The ostentatiously displayed the wealth and status of introduction of the htp-dj-nswtformula dates to 52 JARCE XXX (1993) the early Fourth Dynasty,55 and, however it is to complexes, as has been noted above, has many be interpreted, it explicitly states the theoreti- elements in common with temples of divinities, cal dependence of officials on the king's bounty perhaps because both were designed to ease for the necessities of the afterlife. The use of the transportation of large quantities of food the generic nsiutin the formula, rather than the offerings. The impressive size of this architec- name of a specific royal benefactor, suggests ture also inspired the fear and loyalty that that the formula called for support from future, helped ensure continual service. In this need living kings, not just the tomb owner's contem- for cult service, the king, like his subjects, de- poraries, just as the cult required a perpetual pended upon the kindness of posterity. service of mortuary priests. The king's dependence upon the elite did Until the end of the Third Dynasty, then, the not begin with his death. The task of building elite depended upon the past to ensure a con- the immense pyramid that was a necessary part tinued life after death, a dependence suggested of the new system undoubtedly required far architecturally by the duplication of their more resources than the earlier type of royal earthly houses, by the burial of goods acquired tomb. Although the magnitude of the pyramid during life, and by the location of their tombs in itself would have increased the total sum of re- ancestral cemeteries. Beginning in the Fourth sources available to him by increasing royal Dynasty, tomb owners looked to the living and power, these gains would hardly have been to posterity for their security, depending on the sufficient alone to pay the costs of the project. continued favor of kings and the loyalty of their The quantity of surplus production available surviving family and dependents. The tombs' for use in mortuary architecture (and other increased accessibility and independence from spheres) by the king's immediate subordinates older cemeteries indicates visibly a shifting of must have been severely curtailed, and con- focus from ancestors to future generations. siderable political skills would have been re- If the king's authority ensured the afterlife of quired to convince the elite that resources from his loyal subjects, who ensured the afterlife of their savings in grave goods should be invested the king? The east-west axis of the new mortu- in the pyramid project. Their support was ary complex, the pyramidal shape of the burial probably obtained by a tacit quid-pro-quo ar- mound, and the importance of the sun god Re rangement. Tomb builders apparently received in royal names and titles later in the dynasty are higher quality building materials from the evidence for an increased connection with the stone supplied for the royal project. Labor for solar cult. The identification of the dead king construction and access to royal crafts special- with Re, who was reborn daily at sunrise, was a ists for decorating the tombs may also have powerful metaphorical insurance of the survival been centrally supplied. Furthermore, the of his soul.56 An afterlife lived with Re in his so- proximity to the royal pyramid presumably con- lar bark differed markedly, however, from the ferred status, both during the lifetime of the repetition of earthly glories that Djoser antici- officials and afterwards, enhancing their pros- pated. Supplies for an earthly existence were pects of eternal life. In exchange for these unnecessary; instead, perpetual offerings and benefits, the officials must have provided labor- cultic service like those received by gods were ers, food, and other resources necessary to sup- required. The architecture of the new mortuary port the pyramid-building project. In this sense the spatial organization of the new pyramid 55 cemeteries demonstrates not the dependence WinfriedBarta, Aufbau und Bedeutungder altdgyptischen Opferformel(Gluckstadt, 1968), 3. One novel feature of Fourth Dynasty pyramid sub- structures between Snefru and Khafre that has not to my N. Cherpion, Mastabas et Hypogees d'ancien empire: Le knowledge been noted previously is that a pyramid's en- problemede la datation,Connaissance de l'Egypte Ancienne trance corridors first descend, then rise to reach the burial (Bruxelles, 1989), 79, has argued that no Tura limestone chamber. This pattern might be related to the setting and was used in privatetombs at Giza after the Fourth Dynasty, rising of the sun, although the axis is north-south rather in other words, after the completion of the royal pyramids than west-east. for which Tura limestone was brought. SOCIAL CHANGE IN THE FOURTH DYNASTY 53 of the officials on the king, but the dependence authority in exchange for good government. of the king on his officials. Such a transition is supported textually by Sne- Still other concessions to these essential sup- fru's adoption of the title ntr nfr, "the good porters can be seen in textual sources. It is at god," and, even more significantly, the Horus this period that the king's personal name be- name Nb-Mjct, "possessor of Maat," referring to gan to be used extensively on monuments, sug~ his ability to maintain an ideal world order gesting a greater degree of access to him as an based on justice, truth, and traditionally pre- individual. This name also began to be incor- scribed behavior. That the transition was at least porated in the names of his officials and cult partly conscious, and that it entailed some hy- personnel, a concession probably intended to perbolic propaganda stressing the king's good- forge a closer relationship with the king and natured humanity, can be surmised from the make sacrifices on his behalf more acceptable. benign, almost buffoonish role Snefru plays in Also built on the name of the king were the later literature: his simple-minded lecherous- names of royal mortuary estates, lands set aside ness in the papyrus Westcar story and his hearty by the king as perpetual endowments to sup- good fellowship and willingness to act as a hum- port his cult. Here again, the use of the power- ble scribe in the "Prophecies of Neferti."60 ful royal name may have helped ensure the The pyramids were thus built at the expense loyalty of agricultural workers. Some revenues of the king's god-like distance from his sub- from these funerary endowments were clearly jects. At the same time, other strategies were diverted to supply the cults of loyal supporters, adopted to reinforce his divinity. The new use who took the opportunity to depict this presti- of the king's personal name in the personal gious source of supply on their chapel walls.59 names of his subjects gave them a special con- (The king thus essentially garnished future ag- nection with him, but also gave him the same ricultural production to pay for his pyramid, an role as gods, who were traditionally mentioned early example of deficit spending.) in theophoric names. The htp-dj-nswt formula, Such concessions suggest that Snefru's reign in which the king was normally paired with marked a departure from the conception of Anubis in granting boons in the afterlife, again kingship in which royal power derived solely associated the living king with a divinity and from fear of the king. The high walls of the early granted him divine powers. The use of the royal tombs represent metaphorically the de- title "son of Re," beginning with Djedefre, es- fensive nature of power that rested on the abil- tablished a physical connection with the most ity to extract resources forcibly and punish powerful deity of the period. Finally, the dis- opponents. The amount of control that can be tinctive shape of the royal pyramid itself and its exercised with this type of power is limited. The restriction to royal use distinguished the king's (visually) more accessible monuments of Snefru tomb from those of his courtiers, while its size and his successors suggest that their power further emphasized his divinity. The king built rested on a more political base, appealing to the his personal political power by granting access approval of at least the elite members of the population, who willingly supported the king's 60 Posthumousreferences to Snefru have been collected by D. Wildung, Die Rolle dgyptischerKonige im Bewusstseinihrer 58 H. Ranke, Die dgyptischenPersonennamen 2 (Gliickstadt, Nachwelt:Posthume Quellen uberdie Konige der erstenvier Dynas- 1952), 229-32. Although Rankenotes the absence of several tien,MAS 17 (Berlin, 1969), 114-19. divine names from the Archaic period corpus, basilophoric This may also represent the king's adoption of a di- names are simplyabsent from his summaryof name types of vine prerogative.The first attested offering formula, in the the first three dynasties, and present in his Old Kingdom tomb of Rahotep at Meydum, is built on the name of survey. Anubis; the word nswtis substitutedfor the god's name by The first attested estates occur in the reign of Snefru. the time of Khufuat the latest, however. (Barta,Aufbau und Helen Jacquet-Gordon,Les nomsde domainesfuneraires sous Bedeutung der dgyptischenOpferformel, 3-4.) Vancienempire egyptien, BdE 34 (Cairo, 1962), 8. Some estates David Larkinhas suggested to me that the new differ- of earlier kings may occur, but they are of later date, and entiation in the shape and size of the royal tomb may have may have been organized posthumously. made the spatialdifferentiation less important. 54 JARCE XXX (1993) to his tomb and his name, while simultaneously increasing the value of that access by the very enhancement of power that it paid for. The building of larger pyramids thus provided Sne- fru and his successors with symbolic currency to pay for broader power and central control. The underlying motivation for these changes probably again relates to the shift of focus from Horus to Re, as the divinity represented by the king. The sun that embodied Re was certainly more distant than the falcon Horus from the Egyptians, and arguably from the king (since the king was equated with Horus, but was only Re's son); yet the sun clearly had a greater in- volvement with their everydaylives than the fal- con. The sun's light and warmth contrasted implicitly with the darkness and cold of its ab- sence; it was surely seen as a universallybenefi- cent force, rather than simply a powerful one. The sun's power influenced views of the after- life, but it may also have inspired a new kind of relationship between the king and his people, in which he cared for them as well as ruling them. The appearance of husbands, wives and chil- dren together in the relief decoration and stat- uary of Fourth Dynasty tombs may also be connected with the cult of Re, though more subtly. In royal iconography, the king's family first appeared together (albeit at radically different scales) in Djoser's temple to Re at He- liopolis63 (see fig. 11). The cult of Re at Heliop- olis was a family cult, involving a genealogically- related ennead; and the king's connection with Re also had a genealogical basis- he was Fig. 11. A fragmentfrom the templeofDjoser at Heliopolis, Re's son. The of the of Re now in the Egyptian Museum, Turin. This drawing is growth importance based on a slide taken the author. and his cult seems to have brought about a new by stress on family, children, and posterity. (The

63 W. S. Smith, A History of Egyptian Sculptureand Painting older dynastic deities, Horus and Seth, in con- in the Old Kingdom(Boston, 1946) 113, fig. 48 right. The trast, had no spouses or children to speak of.) A names, but not the figures, of Djoser's wife and daughter similar increase in on the wife and also on 79 re-used steles and markersat his emphasis appear mortuary children of the took at the end of complex (Firth and Quibell, TheStep Pyramid, 119, pls. 86- king place 87). A collection of four statuesof different sizes (ibid., 114, the Eighteenth Dynasty along with the rise of pl. 63 bottom), of which only the feet are preserved,might another solar cult. suggest that they were depicted as statues in the complex; but it is equally possible that these statues represent deities. 65 Alreadyon one of the Djoserfragments from Heliopo- W[ilfried] S[eipel], "Konigin,"LA III, col. 465, notes lis Geb, Shu and Seth seem to be represented. W. S. Smith, that the queen began to outrank the king's mother and The Art and Architectureof Ancient Egypt, 2nd ed., revised by took on more important roles in royal iconographybegin- W. K. Simpson (Harmondsworth,1981), 64. ning in the reign of Amenhotep III. SOCIAL CHANGE IN THE FOURTH DYNASTY 55

The ultimate origin of the new forms adopted mental characteristics of the Egyptians' belief for royal tomb complexes in the early Fourth system: their expectations about life after death Dynasty, by this analysis, resembles that of ear- and their relationship to their king. In both of lier analyses. The changes derive from the new these areas, the change represented a departure association between the king and the sun god, from the backward-looking views of the first an association which, it should perhaps be three dynasties towards a forward-looking de- noted, neither my analysis nor the traditional in- pendence on posterity, a posterity that was pro- terpretations explain. The value of the applica- duced by the family relationships that the sun tion of spatial analysis here lies in its elucidation cult stressed. Ironically, the endowments and of the intermediate effects of this association, perpetual mortuary service that this new view re- and their wider consequences. quired resulted in a proliferation of ancestor This broader view of the architectural changes cults, which came to dominate Egypt's society of the early Fourth Dynasty reveals that the new and economy, and ultimately shackled to the form of the royal tombs was not an isolated phe- past the very posterity upon which they de- nomenon, resulting from an esoteric philosoph- pended. In its initial effects, however, the burst ical and religious emphasis on the sun god that of pyramid building that the new solar ideology was limited to the king himself. Instead, the produced at the beginning of the Fourth Dy- changes occurred in all levels of elite mortuary nasty seems to have been one of many "ratchets" architecture, and represented the culmination that propelled a basically backwards-looking of a larger, slower, and more far-reaching shift culture into the future. in the focus of the Egyptians, from the past to the future. This shift radically altered two funda- Philadelphia, PA