Mycenaean Residential Architecture: Palaces and Ordinary Housing

Megaron at Pylos / Photo by University of North Carolina

By Dr. Jeremy B. Rutter
Professor Emeritus of Classical Studies
Sherman Fairchild Professor Emeritus in the Humanities
Dartmouth College

Distribution and Dating of Mycenaean Palaces

Mycenaean palatial structures are now known at the following sites on the Greek Mainland:

Megaron ruins at Tiryns / Wikimedia Commons

(a) Argolid: (1) Mycenae (2) Tiryns [Palaces were probably also located at Midea and Argos but have either been eroded away by natural processes or destroyed by subsequent building operations.]

(b) Messenia: (1) Pylos (see featured image)

Menelaion mansion ruins / Photo by Dartmouth University

(c) Laconia: (1) Menelaion

(d) Attica: (1) Athens (probably)

Treasury of Minyas. Roman Altar of Imperial Cult at Center of Mycenaean Tomb Chamber at Orchomenos / Photo by Dartmouth University

(e) Boeotia: (1) Thebes (2) Gla (3) Orchomenos (probably)

(f) Thessaly: (1) Iolkos (possibly)

The best preserved palaces, fully cleared, are those at Pylos, Tiryns, and Gla. Those at Mycenae and the Menelaion are only partially preserved, while those at Thebes and Orchomenos have been only partially exposed. The palace at Athens has been almost totally destroyed, to the extent that we can say little more than that a palace almost certainly once existed on the Acropolis. A substantial building at Iolkos is claimed to be a palace by its excavator, but the only part of it to have been exposed does not prove it to have been a palace.

By far the earliest palatial building to have been excavated is that at the Menelaion of LH IIB-IIIA1 date (ca. 1450-1400 B.C.). The palaces of Mycenae, Tiryns, and Thebes need be no earlier than LH IIIA2 (ca. 1400-1340 B.C.), while those at Pylos and Gla were built no earlier than LH IIIB (ca. 1340-1200 B.C.). The construction dates of the palatial structures at Athens, Orchomenos, and Iolkos are either unknown or unpublished. The Menelaion building was destroyed in LH IIIA1 and not rebuilt on the same scale until late LH IIIB. The palatial building at Gla appears to have been destroyed early in LH IIIB. The palaces at Mycenae, Tiryns, Thebes, and Pylos, as well as the rebuilt Menelaion complex, were finally destroyed at the end of LH IIIB. No LH IIIC palaces are known, and this period (the 12th and earlier 11th centuries B.C.) appears to have been comparable to the Post-Palatial period on Crete in terms of the absence of large administrative complexes.

Features Common to More Than One of the Mycenaean Palaces

Central Megaron Unit

Throne room at the Pylos megaron ruins / Photo by David Gill, Wikimedia Commons

This appears at Pylos, Tiryns, Mycenae, the Menelaion, and probably at Orchomenos. Smaller, less elaborately furbished megaron units occur in non-central positions at Gla. Characteristic features of this unit are: (a) Tripartite division into porch, vestibule, and throne room; (b) Large circular hearth, centrally located in the throne room; (c) Four columns arranged in a square around the hearth in the throne room; (d) A throne against the middle of the right-hand wall in the throne room (only Pylos and Tiryns for certain, but probably also Mycenae); (e) Plastered floors decorated with painted patterns in the throne room, the vestibule, and the porch (Pylos, Mycenae, and Tiryns); (f) Access to the throne room only from the vestibule, through an axially placed doorway; (g) Two columns between antae in the porch; (h) Rich decoration of the walls throughout the unit by means of frescoes. The megaron at the Menelaion lacks a central hearth, columns, and most of the other features listed above but the palace building there is closely comparable in its overall design to the architectural layout of the palace at Pylos.

Court from Which the Megaron Unit Is Entered

Courtyard at the Tiryns megaron ruins / Wikimedia Commons

A large court lies directly in front of the megaron unit at Mycenae, Tiryns, and Pylos. This court is surrounded by colonnades on three-and-a-half sides at Tiryns, on two-and-a-half sides at Pylos, and probably on just one-and-a-half sides at Mycenae. The court is entered at both Tiryns and Pylos from a propylon placed slightly off the short axis of this rectangular court. At Mycenae the court is entered either by means of a corridor or from the top of the so-called “Grand Staircase”.

Secondary Throne Room (or Queen’s Megaron)

Throne room (Queen’s Megaron) at the Pylos megaron ruins / Wikimedia Commons

This feature is attested at Tiryns (in the form of a two-room megaron unit, the throne room having a central rectangular hearth while the porch in front of it opens onto a partially colonnaded court), at Pylos (a single elaborately decorated room on the east wing with a central circular hearth but no certain throne), and at Mycenae (a single room with a throne against its north wall, on the opposite side of the court from the main megaron unit).


This feature is attested only at Pylos and Tiryns. In neither case is the room in question linked closely to any megaron unit. Each has its own anteroom.

Concern for Views

The placement of the palaces at Mycenae, Tiryns, Pylos, the Menelaion, and Gla on the summits of hills or limestone outcrops could be taken as evidence that a view was a common concern of Mycenaean palatial architects, but it is far more likely that the occupants of these palaces simply wanted their residences to be located physically above any other structure within their capitals as symbols of their own elevated social status. That is, the Mycenaean palace dominates its immediate physical environment in much the same fashion as one imagines that a Mycenaean monarch dominated his social one.

Individual Features Peculiar to Specific Palaces


Storage area at the Palace of Nestor ruins in Pylos / Wikimedia Commons

In its last phase, this palace features extensive storage areas and workshops, both around the central megaron unit (but separated from it by corridors) and in independent buildings clumped around the central block of the palace. A banquet hall comparable to those in Minoan palaces and, like those, set off from other quarters of the palace is located in the west wing. A small one-room unit to the southeast of the main palace block within a complex of rooms identified as the palace armory is often identified as a possible shrine, but the evidence for such a function is far from compelling. The site as a whole was not fortified during the LH IIIB period, although it appears to have been in an earlier LH IIIA phase.


Corbelled east galleries corridor at the ruins of Pylos / Wikimedia Commons

At Tiryns, two additional large courtyards precede the court in front of the megaron unit, the three being linked by two large H-shaped propyla (gateways), each having a pair of facades with two columns between antae. The outermost courtyard, approached through a series of gates from the main entrance to the citadel, is flanked on the east by a colonnade fronting a series of square chambers, an architectural ensemble possibly modelled after the earlier LM IIIA “Agora” at Ayia Triadha on Crete. Below the square chambers is a series of corbel-vaulted subterranean chambers opening off a long corbel-vaulted corridor, the so-called “East Galleries”. Like the similar set of “South Galleries” at the southern end of the citadel, these underground chambers probably served as storage chambers, functionally comparable to the magazines in the west wings of Minoan palaces.


The most distinctive individual feature of the palace at Mycenae is the “Grand Staircase”, a monumental stair which in two flights provides access to the palace court from the terrace below to the south.


Ruins of the Palace of Gla / Wikimedia Commons

The so-called “palace” at Gla is altogether unlike the palaces of Mycenae, Tiryns, and Pylos in its design. It is L-shaped, has no dominating central megaron unit, and lacks a court which precedes the main megaron. It is, however, located at the highest point within a walled citadel and is thus comparable at least in terms of its location to the palaces at Tiryns and Mycenae. The two wings of the L-shaped palace are both flanked by a major corridor, and the individual rooms within one wing correspond quite closely in size and placement within the wing to the rooms of the other wing. This symmetrical effect is then repeated in the large complex of buildings (called “the Agora” for no very good reason) located to the south of the palace within its own perimeter wall. The architectural peculiarities at Gla suggest that the “palace” here had quite a different function from that of the Argive and Messenian palaces from which most of our evidence for Mycenaean palatial architecture comes.

Mycenaean and Minoan Palatial Architecture Compared

Grand staircase of the palace ruins at Tiryns / Wikimedia Commons

The location of the central megaron unit within each of the three best preserved Mycenaean palaces clearly indicates that it was the architectural focus of the entire palatial structure. A visitor to the palace is inevitably led, indeed steered, directly toward this megaron. From the fact that the megaron contained a throne against the right-hand wall of its principal room (compare the position of the throne in the earlier Throne Room at Knossos, but also note the presence of a second throne in the “Queen’s Megaron” at Tiryns and in the so-called “Throne Room” at Mycenae), as well as from the extraordinary decorative embellishment of the megaron complex, this unit seems likely to have been the place where the administrative authority resident in the palace held court. In this sense, Mycenaean palatial architecture may be characterized as centripetal, in contrast to the centrifugal nature of Minoan palatial architecture. Note that the megaron block in a Mycenaean palace is isolated and not an integral part of a more complex unit, as all large or elaborately decorated spaces within a Minoan palace appear to be.

Mycenaean palaces reflect much more individuality in their design than do Minoan palaces in the sense that there are far fewer recurrent features in the former than in the latter. Individual features within specific Mycenaean palaces appear to have been adopted almost haphazardly (at the whim of a particular ruler?) from Crete: the banquet hall at Pylos; the “Grand Staircase” at Mycenae; the wall with multiple apertures connecting porch and vestibule in the Tirynthian megaron, loosely comparable to a Minoan pier-and-door partition; the lavish use of colonnades at Tiryns. The Mycenaean heritage from Crete in terms of palatial architecture appears to have consisted primarily of the concept of the palace itself as an administrative center. The heart of a Mycenaean palace, the main throne room with its large central hearth, is simply a monumental version of the normal private house of the EH III and MH periods and has no connection with Minoan architectural forms. Individual Minoan architectural features are, to be sure, taken over by the Mycenaeans, but these are selectively rather than universally adopted, except perhaps in the case of the extensive uses of frescoes, cut stone masonry, and downward tapering columns. These last, however, are architectural attributes of arguably lesser overall significance than the designs of architectural spaces. As in the cases of funerary architectural forms, most notably the tholos, the Mycenaeans have adopted something Minoan but have altered, even perverted, it to suit their own societal needs. With regard to both their tholoi and their palaces, it is the Mycenaean ruling class rather than the Mycenaean population at large which has absorbed Minoan forms into Mainland Greek building practices. With the downfall of this ruling class as a dominant political force ca. 1200 B.C., the Minoan forms represented by Mycenaean palatial architecture, the Linear B writing system, and even tholoi in areas other than Messenia and Thessaly disappear from Mainland Greece.

Whereas the Minoan palace qualifies as an artifactual type because it consists of a combination of constantly recurring units, the Mycenaean palace is not immediately recognizable as a type because the units of which it consists are so variable. Indeed, it is perhaps legitimate to view the Mycenaean palace simply as a highly elaborate Middle Helladic megaron with the nature and degree of the elaboration left up to the individual dynast. In this connection, it is worth emphasizing again that Minoan features adopted in Mycenaean palaces consist of relatively simple forms and never of a complex of two or more major architectural spaces. The emphasis on circulation of light and air in Minoan architecture is altogether disregarded by the closed and somewhat stuffy nature of Mycenaean palatial design. Imagine what the interior of a Mycenaean throne room must have been like with a fire always burning on the large central hearth! There is nothing in Mycenaean domestic architecture comparable to the villas and mansions of Neopalatial Crete. “Lower towns” existed below the hilltops at Mycenae, Pylos, and Tiryns, but little of these has as yet been cleared and it is therefore still impossible to compare a Mycenaean town with such Minoan equivalents as Gournia or Palaikastro, notwithstanding the attempt by the Minnesota Messenia Expedition in the late 1960’s and early 1970’s to expose the plan of an entire Mycenaean settlement at Nichoria. The hilltop of a Mycenaean palatial center was always reserved for the residence of the monarch and its associated outbuildings, and frequently this distinction between royal domain and that of the common people was emphasized by the construction of massive fortification walls around the former. Such a distinction is of course lacking on Crete, not only because Minoan palaces were not always placed on hilltops (Mallia, Zakro) but also because Minoan Neopalatial fortifications as such are to all intents and purposes non-existent.

It is important to make the distinction between “palace” (a term identifying the residential architecture reserved for royalty) and “citadel” (a term which marks the site so identified as fortified but not necessarily the seat of a royal residence). Pylos in the LH IIIB period is a palace but not a citadel; Mycenae, Tiryns, and Athens are both palaces and citadels; Gla is a citadel but arguably not a palace; Thebes is a palace but perhaps not a citadel.

Along the lines of the palatial relative chronology commonly used for the Minoan Bronze Age of Crete, one could outline the following Mycenaean palatial chronology for the Greek Mainland during the Late Bronze Age only:

LH I-IIA Early Mycenaean (or ca. 1680-1500 B.C.


LH IIB-IIIA1 Protopalatial ca. 1500-1400 B.C.

LH IIIA2-IIIB Neopalatial ca. 1400-1200 B.C.

LH IIIC Post-Palatial ca. 1200-1050 B.C.

It is tempting to associate the beginning of the above Protopalatial period with the LM IB destruction horizon which marks the end of the Neopalatial period on Crete. Likewise, the dividing line between the above Protopalatial and Neopalatial periods is perhaps to be connected with the destruction of the palace at Knossos early in the LM/LH IIIA2 period, ca. 1385/1375 B.C. Thus far, the only building identifiable as a “Protopalatial” Mycenaean palace is the Menelaion of Periods 1 and 2 in Laconia, although the excavator of Tiryns has claimed that the earliest palace there under the LH IIIB megaron dates from the LH IIIA1 period.