Palaces in Ancient Egypt: Cities for Kings and Gods

Illustration of the ancient palace of Malkata

The grandeur that early European explorers had come to expect in royal building programs seems to have been reserved for sacred space and funerary complexes.

By Dr. Steven Snape
Reader in Egyptian Archaeology
University of Liverpool


For early European explorers in Egypt, it was inconceivable that the massive monumental structures they saw were not obvious evidence of ‘oriental despotism’ at work creating ego-affirming monuments designed to liken the king with the gods. In particular – and perhaps with the elaborate residences of Ottoman Istanbul in mind – it was easy to think of Egyptian kings creating huge and impressive royal palaces, and in the seminal publication produced after Napoleon’s expedition to the country, Description de L’Egypte, each of the New Kingdom Theban complexes at Karnak, Luxor and Medinet Habu was described as a ‘palais’.

The ‘North Palace’ at Amarna has been the subject of many years of excavation. The model shows how it may have appeared when complete. Model and photo by Eastwood Cook; concept by Mallinson Architects and Kate Spence.

Since the decipherment of the hieroglyphs that adorn their walls, and the archaeological investigations within their precincts, we now know that the major monumental architectural features that so impressed these early visitors were in fact massive temple and mortuary complexes. The urban dwellings of king and subjects were and are harder to identify because of continued building work on top of ancient sites (sometimes through to the present day) and because, contrary to expectation, the highly stratified hierarchy of ancient Egyptian society was often not evident in its urban landscape: the grandeur Europeans had come to expect in royal building programmes seems to have been reserved for sacred space and funerary complexes.

However, these early visitors were to a very limited extent right in identifying major temples as palaces: Karnak and Medinet Habu did indeed contain royal residences, but only as a comparatively tiny adjunct to what was, overwhelmingly, a residence for gods. The great royal palace, built on the same scale of temples and tombs, was simply not to be seen. There was no direct ancient equivalent of Versailles, Schönbrunn or Buckingham Palace – i.e. a great architectural confection built in or near the capital city of a centralized state, where external appearance reflects the power and dignity of the person or office it houses. The pattern of royal residence in dynastic Egypt was quite different to that of the 18th- or 19th-century European monarch.

The ‘North Palace’ at Amarna has been the subject of many years of excavation. Courtesy Barry Kemp/The Amarna Trust.

Perhaps the best way of thinking about palaces is to consider their different potential functions: as a private dwelling for the ruler and the ruler’s family, a place for royal administration and the bureaucracy that supports it, as a venue for ceremonial activity (both public and private). In the Egyptian context (as with others) a fourth major function can be added: the location for production of rare or important goods of intrinsic value or requiring specialist skills.

The Peripatetic Pharaoh

Part of the reason that palaces were so much more than just a residence was that the king was an active, mobile ruler, not just on foreign campaigns (although for a significant number of rulers they seem to have been a regular event), but throughout Egypt. Egyptian sources refer to the ‘Mooring Places of Pharaoh’, giving the impression of riverside stopping-places for a king who travelled – unsurprisingly – by river.

In this model of royal residence the major requirement is a relatively large number of relatively small residences that could be equipped at short notice. An example of the preparations required for the imminent arrival of a Ramesside king is found in Papyrus Anastasi IV, which demands, among many other things, nearly 30,000 loaves of bread, 60 sacks of pomegranates, 50 bowls of honey and 100 stands for floral bouquets. It should be noted, however, that these provisions were not just for the king but also for the ‘army and chariotry’ who were with him. A variation on the movable royal residence is to be found in the fact that towns attached to individual royal pyramids of the Old Kingdom served as multiple important centres of royal administration.

[LEFT]: The palace of King Merenptah at Memphis, which appears to have essentially been a building designed for royal audience, with relatively modest living quarters attached. Steven Snape.
[RIGHT]: This small palace (marked by the box) built as part of the mortuary temple complex of Ramesses III at Medinet Habu may or may not have acted as a real royal residence during the king’s lifetime, but it was certainly used as such in the Third Intermediate Period. Steven Snape.

The only significant exception to this model of constant progress by the king (at least while he was young and vigorous enough to travel) is found at Amarna, where the stated intention of Akhenaten never to leave his new capital finds its archaeological reflection in the extensive palace complexes at the city.

Because the Egyptians did not adopt the model of having a small number of very large multi-functional palaces in a few major centres, individual ‘palaces’ could have their own distinct character based on its specific, limited functions. The palace of Merenptah at Memphis is a good example of this, essentially consisting of a large audience/reception hall, with a very modest set of personal apartments for the king at the rear.

A ‘Window of Appearance’ was designed to provide an appropriate setting for the royal reward of officials. This scene, on the walls of Meryra II’s tomb at Amarna, shows Meryra being thrown gifts by Akhenaten and Nefertiti. Otto Georgi.

The palace of Ramesses III at Medinet Habu was also primarily an ‘audience palace’, but with the additional twist that its connection to an open court of the adjacent mortuary temple provided a very suitable location for more public royal display and reward in the architectural setting of a ‘Window of Appearance’ where the king presented himself in a manner reminiscent of the balconies favoured by 20th-century totalitarian leaders to appear before a gathered public. Indeed the main way palaces appear in visual art from dynastic Egypt is in tomb scenes showing the rewarding of the tombowner by the king at a ‘Window of Appearance’. However, even here the identification of a modest royal residence is by no means straightforward, as some scholars have identified these ‘palaces’ attached to New Kingdom mortuary temples as structures for the use of the dead, rather than the living, king.

Building the Palace

Given the requirement that palaces could be built very quickly (in the event of a particular royal progress, or because one king wished suddenly to have a palace in a location not used by his predecessors), but might be used once only, or on a very temporary basis, it is hardly surprising that these buildings were regarded as essentially disposable. Exceptions might be the palaces attached to temples that had a particular long-term connection with the king, meaning that their royal palaces were reused more frequently. This applies particularly to the mortuary temples at Thebes, where both the Ramesseum and Medinet Habu had small palaces as part of their grand scheme – but here, as we have seen, the house of the living king was significantly smaller than the space provided for the god Amun. Owing to the modest nature of royal palaces, it is sometimes difficult to differentiate archaeologically between residences constructed for kings in important provincial centres and ‘palaces’ for regional administrators (see Bubastis in the Eastern Delta and Ayn Asil in the Dakhla Oasis).

In most known cases the material of choice for the royal palace, as for the houses of the most humble of the king’s subjects, was mudbrick. Just as for ordinary dwellings, the survival of royal palaces in the archaeological record is a matter of accident and, one suspects, atypical examples. However, the ability to construct a palace in a very short time is best attested in Amenhotep III’s ‘festival palace’ complex at Malkata, on the West Bank at Thebes, which was built, used and then abandoned within a relatively short period.

The Palace Harem and the Harem Palace

Western conceptions of the harem as an institution tend to be fixated on the erotic possibilities of the creation of an assemblage of sexually available women for the sole access of one male individual. The eastern harem was a regular subject in 19th-century genre painting, and although the harem concerned was usually that of Ottoman Turkey, the range was occasionally extended to include the imagined harem of Pharaoh, portrayed with no less an atmosphere of languorous eroticism.

It is likely that the reality was somewhat different and that the Harem Palace was developed in Egypt as a means of coping with the sheer number of women who were part of the royal household. To take the best-known example, when Princess Gilukhepa came from Mitanni to marry Amenhotep III as a diplomatic bride, the commemorative scarab issued by the king to mark this event also notes that she brought with her 317 ‘chief women of the harem’. The ancient Egyptian royal harem was not so much a sexual supermarket for the king, but a community that chiefly comprised female and infant members of the king’s extended family. A related element of the palace complex was the Kap, a royal nursery or school that educated both royal children and those of favoured members of the court. To have been a ‘Child of the Kap’ was an important claim for members of the court in the 18th Dynasty, as it indicated closeness to the king from an early age.

Surviving architectural remains make it difficult to envisage the internal appearance of royal palaces in ancient Egypt. The ‘princesses fresco’ from the King’s House at Amarna perhaps comes closest – it depicts an informal scene of royal family life, including two daughters of Akhenaten and Nefertiti perched on a cushion. Ivy Close Images/Alamy.

The ancient Egyptian harem differed from the Ottoman variant in an important way. Although it was designed to accommodate significant numbers of royal women, it was not designed to segregate them from the outside world, although the most common word for harem, ḫnr(khener), seems to be derived from the verb ‘to restrain’. Instead the harem was a viable institution in its own right, with its own economic assets, income-generating activities and administrative officials. The most important of these activities indicates how these women passed their time: the first mention of what seems to be the earliest Egyptian word for harem, Ipet, on a sealing of the First Dynasty (reign of King Semerkhet) refers to a ‘weaving workshop of the Ipet’.

The production of high-quality linen seems to have been the most important activity at the Harem Palace at Medinet el-Gurob. This site, in the Faiyum, is unusual as a harem because it was located well away from main centres of power. Other buildings known, or suspected, to have housed royal women have been found at Thebes, Memphis and Amarna, but always in association with larger palace complexes. The reasons why Medinet el-Gurob was established a significant distance from the main centres of political power may be to do with a desire to remove potentially competing royal wives and mothers away from those arenas. If this is the case, then the involvement later in the New Kingdom of harem-women in the attempted, and possibly successful, assassination attempt on Ramesses III shows the wisdom of such a policy, particularly when multiple royal wives and numerous royal children had the potential to make the succession to the throne extremely contentious.

Fieldwork carried out at Medinet el-Gurob since 2005 by British Egyptologist Ian Shaw for the University of Liverpool supports the idea of the Harem Palace there as a substantial town with its own production facilities and substantial residential blocks, which prospered from its foundation in the reign of Thutmose III to its eventual abandonment, probably during the reign of Ramesses V.

The Palace City

Although it now survives as a rather desolate collection of spoil-mounds amid the traces of mud-brick buildings, the palace complex of Malkata built by Amenhotep III is the most extensive royal residence known from ancient Egypt. Steven Snape.

Many New Kingdom urban centres – including Thebes, Memphis, Pr-Ramesses and Amarna – can be referred to as ‘Royal Cities’ because of the way in which the desires of a king, or a series of kings, shaped their appearance. An important subset of the royal city is a settlement composed of an extended royal palace with its ancillary buildings. Malkata, which dates to the New Kingdom, is the best surviving example of such a palace-town. The full extent of Malkata is not yet known, but it covered an area of at least 35 hectares (86½ acres) on the West Bank at Thebes. Its centrepiece was the palace itself, 125 by 50 metres (410 by 164 ft), with a series of internal rooms that have been interpreted as a central columned audience hall. At one end of the hall was a small throne room and behind this were the royal apartments (including a bedroom and bathroom). Flanking the central hall were a series of smaller suites, which may have been used to accommodate other important members of the royal household. This palace opened up onto a series of open courtyards. Other structures within this vast complex included further, smaller palaces, an area for the celebration of the king’s jubilee-festival, a series of storerooms, workshops, kitchens and bakeries, and houses, the last presumably to be occupied by members of the court of Amenhotep III. Perhaps most striking was the construction of a huge T-shaped artificial lake (at c. 2 by 1 km, 1¼ by ⅔ miles, almost six times the area of the city), now known as the Birket Habu.

Another type of palace-town was built at Deir el-Ballas, located 45 km (28 miles) north of Luxor on the west bank of the Nile and chiefly excavated between 1980 and 1986 by American Egyptologist Peter Lacovara. The site consists of a series of related groups of buildings spread across some small hills and wadis (dried valley beds) over an area about 2 km (1¼ miles) long. The core structure is the ‘North Palace’ within an enclosure of about 300 by 150 metres (984 by 492 ft). Other parts of the town include large residences, observation towers and what appears to be a workmen’s village on the model later followed by Deir el-Medina. Like Malkata, Deir el-Ballas had a short period of occupation, but its purpose seems to have been very different: it was a primarily military staging-post in the Theban wars against the Hyksos in the late 17th and early 18th Dynasties.

From The Complete Cities of Ancient Egypt, by Steven Snape (Thames & Hudson, 09.16.2014), published by Erenow, public open access.