Location: South of the Panathenaic Way, below the NE corner of the Mint. Next to no. 28 in the Agora plan of the Guide: Μc Camp II, J., The Athenian Agora, A Short Guide to the Excavations, Excavations of the Athenian Agora, Picture Book no 16, American School of Classical Studies, Princeton 2003, pp. 2 and 24-25.

The SE Temple, which was discovered in 1959, lies in close proximity to the Eleusinium, on the west side of the Panathenaic Way, on the spot were until the 1st cent. a Mint existed (more specifically at the NE corner of this large enclosure). It belongs to a large group of monuments relocated here from other parts of Attica (SW Temple, the so-called Temple of Ares, Altar of Zeus Agoraios). 

The building’s direction is W-N, in an area dominated by the SE side of the Agora until the middle of the 2nd cent. AD, when in that area the Nymphaeum was erected. The building is of rather modest dimensions: its total length comes to 20.60m, while its width measured 12.10 (pronaos) and 11.20 (the cella). The cella’s dimensions (on the inside of the walls) were 12.50 Χ 9.56m. The thickness of the walls is approximately 0.90m. They do not have any special foundations. Remnants of the lower layers of the cella’s stone walls survive, which are formed by limestone held together by grey mortar, in secondary use. On the inside of the walls traces of turquoise coating decorated with red dots survive, which at a later phase was covered by a thick white coat of inferior quality. No traces of an altar were found, but at the centre of the cella a large pedestal (4.40 Χ 6.70m) was unearthed, on the inside of which roughly arranged building material at second hand use was discovered. The pedestal supported a colossal statue, two fragments of which were found in close proximity to the building; it was sculpted out of Pentelic marble and depicted a standing female figure wearing a cloak. Today it is widely accepted that these fragments come from a statue of Demeter in the so-called Capitol Museum type, dating to around 420-410 BC. 

In the rear part of the cella, the floor was made of compacted clay (height from sea-level 68.57m), while its front part, in front of the statue, apparently had a marble floor. Large rectangular slabs of marble had been incorporated and were later excavated in a nearby tower of the Late Roman wall. The pronaos possibly had the same type of flooring, though no substantial traces of it survive. 

The pronaos, although belonging to the same period as the cella, is completely different, in terms of building material as well as in terms of the superior quality of its foundations’ construction, which feature large slabs of limestone or pebbles (cobblestones?) (measuring 1.7 Χ 1.4 Χ 1.35m) indicating the existence of square slabs in the superstructure. The elevation on the uppermost layer of the foundation is 67.47m, that is 1.10m below the level of the ground in the cella.

Suppositions about the superstructure

The material found incorporated in the Late Roman fortification at a distance of about 3m from the SE Temple comes from a Doric building and consists of columns and capitals; these are made of marble which comes not from Hemyttus nor from Penteli, but, according to Thompson, belongs to a type characteristic of the monuments in the deme of Thoricus. It is also likely that the 5th cent. material was taken from a Doric temple of 7 Χ 14 columns excavated in 1812 by the Society of the Dilettanti, a temple which was left uncompleted. The idea is that in the 1st cent. AD, when Thoricus had already been abandoned, the unfinished temple was dismantled and its architectural members were moved to a new temple in the Athenian Agora. It is also possible that the material in question was taken not from a temple but from a stoa.

This identification is connected with that of another ‘wandering’ temple in the South square of the Agora, the so-called SW Temple, to which Thompson attributes the Ionic material from the Temple of Athena at Sunium, which Dinsmoor attributes to the SE Temple. Correspondingly, Dinsmoor considers the SW a Doric temple. The papers published after Dinsmoor adopt his identification, more importantly Mc Camp in the 4th edition of the Guide to the Agora and in his book. A recent study by Masimo Osanna seems to confirm Thompson’s view, relying mostly on the fact that one of the Doric capitals was unearthed very close to the SE temple.

More specifically, the Doric architectural members in question are the following: (Α) one segment of the Doric epistyle (Α 2983), which was discovered in the foundations of the tower of the Late Roman wall. This segment preserves the fillet (taenia), as well as the guttae. Its dimensions are 0.76 in height 2.115m in its current condition, 0.466m. (Β) Three fragments from the rear section of a Doric epistyle, which do not fit together. (C) Two metopes of Pentelic marble, discovered in the foundations of the wall’s tower. (D) Eight fragments of triglyphs, also from the foundations of the tower. These come from five different buildings. Two belong to the same building (Α 2975 and 2976), one belongs to the building the rear segments of the Doric epistyle come from (Α 2977). Three more belong to a separate monument (Α 2979, 2980, 2981), while two more from a fourth and fifth building, respectively. All these were cut to the same height (0.695-0.70m) to be installed in the same building. (Ε) Fragments of the volutes from four columns (Α 3008-3011) and one capital (Α 2987) discovered inside the Late Roman wall. One more capital (Α 2988) was discovered in the Panathenaic Way, at a distance of 14.5m north of the SE Temple. Lastly, two more capitals, severely damaged (Α 3356a-m), as well as the upper part of a volute from a column, were found incorporated in the tower of the Late Roman wall. These are the architectural members taken from Thoricus. (F) The Doric capital of a post, made of Pentelic marble and measuring 0.48 in height,  1.002 and 0.953  in diameter, 0.82 and 0.74m at the top (Α 2989). It had been used as a cover of a duct in the area of the tower of the Late Roman wall. It is thought that it belongs to the same building, as its lower diameter corresponds exactly to the lower diameter of the aforementioned capitals. It dates to the Roman period, but it imitates classical models of the 5th century. (G) Eight slabs in good condition (and two more in a poor condition), found incorporated in the Late Roman wall; these are made of Thoricus marble and preserve signs of anathyrosis. These do not come from the same building as the columns, but possibly from another structure, which Dinsmoor tentatively identifies as the Temple of Dionysus in the theatre of Thoricus. Judging by their thickness, the walls of the building must have been approximately 0.625m thick.

This material, taken from a number of buildings, and not only from the temple at Thoricus, was re-worked and adapted to the dimensions of building of the Roman Period. It is noteworthy that the columns of the temple or stoa in Thoricus had not received fluting, only their lower parts were fluted. This work was completed in the Agora after the columns had been relocated there.
 According to Thompson, this material must be tentatively reconstructed as belonging to the SE Temple. Based on the fragment of the epistyle (Α), we should reconstruct a Doric frieze with 12 triglyphs and eleven metopes, 0.444 and 0.548m in width respectively and 11.36m in total length.  That is, it must have been a building of 6 Χ 2 columns. In this case, the central intercolumnal distance would have been significantly larger, encompassing two intercolumnal triglyphs. The other distances would have been smaller, with one intercolumnal triglyph corresponding to each.

This is the point where Dinsmoor raises his objections, judging that these members could not have come from this temple. The central interaxial distance must have been 2.976m (this is the conjectured length of the surviving epistyle), the transitory 1.984 and the corner 1.76m. Because, however, the lower diameter of the columns made of Thoricus marble has a diameter of 0.984-1.001m, necessarily the intercolumniation would have been extremely small (1.00 in the centre and just 0.76m in the corners). It is likewise difficult to explain why only four were taken from Thoricus, if we reckon that the construction of the building necessitated the use of 8 columns (accepting, in addition, the optimistic hypothesis that we have discovered all of the columns). Thus, it is rather safe to suppose that the building was not a hexastyle prostyle, but a tetrastyle in antis. The size of the post (F) which was discovered does not support this view. Additional, minor difficulties led Dinsmoor to the conclusion that the Doric material is not compatible with the SE Temple, unless one were to think that the fragments of the Doric epistyle belong to another building, while the epistyle of this particular temple has not been discovered yet.  One final argument has an aesthetic basis: a temple that was so makeshift and composite, composed of material gleaned from at least five different buildings had no place in such a prominent area, in the immediate vicinity of the raised part of the Panathenaic Way.

So Dinsmoor supposes that the SE Temple must be supplemented with the Ionic material coming from roughly the same part of the Late Roman wall, and which was almost certainly produced in an Attic workshop of the late 5th cent. BC.  This particular material comes from the Temple of Athena in Sunium.

This series consists of twelve tambours from Ionic columns, a complete capital and three more fragments, as well as a fragment from the base of a column, which was discovered very close to the SE Temple.  There is also a small fragment from the epistyle, also unearthed very close to the SE Temple.  To these members we must add three more volutes, eight fragments of capitals, a part of the cornice and one more part (epicran), all found incorporated in other buildings of the Byzantine and Ottoman periods. The volutes come from eight out of the total twenty seven columns that existed in Sunium (most probably none was left behind).  The capitals and the fragments represent thirteen columns in total.  This number may be even smaller (12 or 11). According to Dinsmoor’s reconstruction, the temple was an octastyle prostyle, with a depth of three columns.  The interaxial distance would have been 1.531m in the façade, and 1.667m on the flanks, that is, equal to the interaxial distances in the façade and the sole colonnade of the wing in the temple of Athena at Sunium, respectively. 12 Ionic columns are needed in total.

Thompson dates the relocation of the Doric building from Thoricus and the construction of the SE Temple in the 1st cent. BC, during the period of the reign of Augustus, a view to which other scholars subscribe as well (Osanna) – this was a period during which extensive works are being carried out in the Agora and the face of the square is radically transformed.  Dinsmoor, on the contrary, relying on the study of the pottery unearthed during the excavation of the building, favours a date close to the beginning of the 2nd cent. AD.  One of his main arguments for this dating is the fact that the upper part of the euthenteria lies 0.15m below the surface of the Panathenaic Way, which is probably earlier, late 1st cent. AD).  The temple must have been earlier than the Nymphaeum and Pausanias’ visit to Attica, for he only saw one temple at Sunium, the one dedicated to Poseidon, although he incorrectly identified it as the Temple of Athena.

The figure of Demeter renders almost certain the identification of the building with one of the two temples of Demeter and Kore witnessed by Pausanias in the Agora, as he approached the Eleusinium.  The second temple, in which the existence of a statue of Triptolemos was reported, might be the temple-like structure discovered very close to the Eleusinium.  There is also the view that the area of the SE Temple can be identified as the location of a temple devoted to Kore, the so-called Pherphateion.  According to another opinion (Massimo Osanna), the temple should not be connected with the Eleusinian hypostasis of the divinities, but rather with the Thesmophorion.  If we accept Thompson’s addition of material from the Stoa or from the temple of Demeter and Kore in Thoricus, there is a further piece of evidence pointing towards the Eleusinian cult, for, according to legend, the goddess disembarked at Thoricus, when she first came to Attica.  Certain dating complications arise, however, in this identification.  The statue dates to the 5ο cent. BC, while the temple is Roman.  The monuments Pausanias describes also belong to the same period. We would have to suppose then that for 500 years before its relocation, the statue stood in another location in the Agora.
    

DINSMOOR, W.R. Jr., “Anchoring Two Floating Temples”, Hesperia 51 (1982), 410-451, table 95-96.
HARRISON, E.B., “New Sculpture from the Athenian Agora, 1959”, Hesperia 29 (1960), pp. 368-392, table 81-86 (esp. pp. 371-373, on the identification of the cultic statue).
Mc CAMP II, J., The Athenian Agora: A Guide to the Excavation and Museum, 4th ed., Athens 1990.
Mc CAMP II, J., The Athenian Agora: Excavations in the Heart of the Classical City², Cambridge University Press 2001), pp. 222-224. 
OSANNA, M., “Thesmophorion ed Eleusinion ad Atene: problemi topografici e culturali”, Ostraka IV.1 (1995), pp. 103-118.
THOMPSON, H.A., “Activities in the Athenian Agora: 1959”, Hesperia 29 (1960), pp. 327-368, table 73-80 (esp. pp. 339-343).
THOMPSON, H.A., WYCHERLEY, R., The Agora of Athens. The American Excavations in the Athenian Agora, vol. XIV, American School of Classical Studies at Athens, Princeton 1972, pp. 167-168.



SE-Temple, Representation in VR environment