Southport : Original Sources in Exploration

Archaeology of the Acropolis in Athens

Wilhelm Dorpfeld


The Age of the Old Parthenon

(Article originally published in 1902 in the journal
Communications of the German Royal Archaeological Institute, vol. 27 [Ath.Mitt.XXVII] , pp. 379-xxx)

Ten years ago I published in this journal (1892, p.158) an article on the "Older Parthenon," the temple discovered under the Parthenon of Pericles by L. Ross during his excavations on the Acropolis in 1835. Among other things, I tried to prove that this temple was not the old Hekatompedos destroyed by the Persians, as was previously generally believed, but a much larger temple that had only been built after the Persian wars. The older view had formed at a time when the "older Athena temple" further north was not yet known and therefore the "older Parthenon" had to be regarded as the only larger pre-Persian temple. The message from Hesych about an older temple burned by the Persians (s. v. "hekatompedos neos") could only be related to the remains lying under the Parthenon. Thinking of an older Erechtheum was out of the question because of the large size of 100 feet

But two facts remained unexplained in the older view. First, according to Hesych, the burnt temple was supposed to be fifty feet shorter than the Periclesian structure; but in fact the older Parthenon surpassed the younger temple by several yards. Secondly, the column drums built into the north wall of the Acropolis, which were attributed to ancient Hekatompedos, were made of marble and were unfinished, while the pieces of entablature built further west, which were also attributed to him, belonged to a finished and already painted poros building.

An attempt was made to remove the first objection by adding a smaller temple corresponding to Hesych's dimensions (p.380) to the large substructure with excellent foundations; so lastly F. C. Penrose in the Journal of Hellenic Studies 1891, 275. However, in that article I could assert against this that the protruding part of the substructure, if it had not had anything to support it, could not possibly have been founded to a depth of 10 m. The excellent foundation, made of regular blocks, must have supported the columns or walls of the temple itself, as the surviving remains of the steps directly prove. Also the proposal by L. Ross [Arch. Essays I 138), explaining the eastern part of the substructure for a later addition and thus raising the difficulty is untenable. The whole substructure is actually built in one go.

The second objection, relating to the difference in material and work, was attempted to be countered by the assumption that the temple provided with poros steps had a poros entablature, but exceptionally had columns of marble, and that, moreover, only the columns and steps were still unfinished. But even these assumptions could be described as very questionable.

It was only when a second ancient temple was discovered between the Parthenon and the Erechtheion in 1885 that both concerns were eliminated. This temple, which certainly dates from pre-Persian times, was apparently the Hekatompedos burned by the Persians. First of all, it corresponded to Hesych's indication of a size difference of 50 feet between the older and younger Hekatompedos. Without its ring hall, the "old Athena temple" was a hekatompedon: pronaos, naos and opisthodome together formed a 100-foot structure, a "hieron hekatombpedon". In the younger Parthenon, on the other hand, the east cella alone had a length of 100 feet and was therefore a "neos hekatompedos". But a west cella of about 50 feet was added to her. The Temple of Pericles could therefore be rightly said to be 50 feet larger than the pre-Persian 100-foot structure. Second, the dimensions of the finished Porous entablatures made them an excellent match for the foundations of the ancient Temple of Athena, while the unfinished marble columns (p.381) probably belonged to the unfinished older Parthenon. Both attributions were fully confirmed when the still missing column drums and capitals made of this material came to light in the northern castle wall below the entablature of Pores (cf. Middleton Plans and Drawings of Athenian Buildings, Pl. 6), and when also below the marble drums the same steps of limestone found on the older Parthenon have been identified.

This established that before the time of Pericles there had been two great temples of Athena on the Acropolis: the "older Parthenon" on the site of the present Parthenon and the "old Athena-temple" south of the Erechtheion. But were they both pre-Persian?

Since Hesych only speaks of a single pre-Persian temple of Athena, the Hekatompedos burned by the Persians, since our second temple, the older Parthenon, was demonstrably never completed and never even got beyond the substructure and the lowest column drums, everyone decided who have been writing about the temples of the Acropolis since 1885, for assuming that the older Parthenon was only begun after the Persian wars and, after a shorter or longer interruption, was only completed by Pericles. As far as I can see, only F.C. Penrose, in the essay cited above, maintained that the older Parthenon was pre-Persian. However, its dating, about a century before the Persian Wars (Journal of Hellenic Studies 1891, 295), could be described as completely impossible.

It is therefore now accepted that the older Parthenon was built between 479 and 447 BC. Opinions only differ about the exact time when construction began. Having myself espoused the older time of Kimon [Ath. Mitt. 1892, 188), A. Furtwangler suggested Themistokles as the builder (Meisterwerke p. 164), while F. Kopp (Yearbook of the Inst. 1891, 270) of the more recent time of Kimon and finally B. Keil (Anonymous Argent. p. 98) prior to the Eurymedon battle.

However, recent studies of the older Parthenon itself and of the (p.382) retaining walls erected on its south side have convinced me that all these dates are incorrect. The older Parthenon is pre-Persian. When the Persians destroyed the castle, it was in the middle of its construction. Its scaffolding was set on fire by the barbarians. From 479 to 447 BC, i.e. for a generation, the burned substructure of the temple that had been started could be seen next to the temporarily restored "old Athena temple". Only Pericles rebuilt the "great temple" according to a slightly different plan.

I.
Before presenting the evidence for this thesis, I must make some corrections and additions to my earlier essay about the ruin itself.

I had previously left it undecided whether the base of the temple had two or three steps (Athens Mitt. 1892, 187). There must have been three. In addition to the two lower steps that are still in place, several step stones have been found under the pillars of the northern Acropolis wall, which have similar dimensions and the same characteristic profile as the middle step of the substructure. Its material (Kara limestone) is harder than the limestone of the two lower stages. A fragment of the same harder stage is built into the S.W. corner of the Younger Parthenon as an underlay of the southern stylobate (see fig.1 below, where the profile behind Marble Stage 3 is indicated).



Fig.1: Cross-section of the step building of the older and the younger Parthenon.

The steps of hard stone formed the stylobate of the ring hall. The allocation is confirmed by the stone lengths: while the lower tier consists of stretchers of 1.70-1.80 m, the second tier is formed by trusses that are only about half the width (0.90-1.00 m) and have a very great, but different depth (in the east 1.59-2.32 m). Because of the latter fact alone, a stylobate of stones of even depth must have lain above it, the length of which in turn has to correspond to the runners of the lower level. In fact, the existing hard limestones not only have these (p.383) dimensions, but also agree with the previously inexplicable marks found on the substructure, from which I already suspected in my earlier essay [Ath. Mitt. 1892, 186]. Stones of this form had closed. We can now even determine the exact location of the joints of the stylobate on the south side of the temple from the marks, although very few of the stones themselves remain.

The shape of the step structure as it now appears is shown in fig.1 in section. The in situ stones of the older temple are double hatched, the added steps of the same hatched. The stones of the younger temple have received an even lighter hatching. The older stages are denoted by the numbers I. II. III, the younger ones by 1. 2. 3. The level originally designated as the lower level of the old temple, which later became Euthynteria, has received the designation IIIa.

Accordingly, if the temple had three steps, the length of the stylobate is calculated as 75.06 m and its width as (p.384) 29.60 m. Since the corresponding dimensions of the younger Parthenon are 69.51 m and 30.86 m , the older temple in the stylobate was 5.55 m longer but 1.26 m narrower than the younger building. The number of columns, as I had previously ascertained, was 8 on the short sides and 19 on the long sides. The axis width of the columns is calculated to be equal for both sides at about 4.12 m, a coincidence that can be regarded as valuable confirmation of the correctness of the entire calculation. A difference in the size of the ax widths on the short and long sides, as it occurs in several older temples and thus also in the ring hall of the "old Athenian temple", does not exist in the older Parthenon.

The ground plan of the inner temple can be drawn in the way I have explained earlier [op.cit. p.177), with the younger walls and rows of columns, taking into account the displacement of the temple axis, not exactly, but in general. As a result of the addition of the third step, the older foundations now apply a little better than was the case with my earlier drawing [op. cit. p. 177), together with the younger walls and rows of columns. The floor plan that was added hereafter is shown in fig. 6 (see below). The thickness of the cella wall is assumed to be about 1 m according to several marble blocks built together with the column drums. With regard to the design of the pronaos and the back hall, one can hesitate as to whether 6 Doric or 4 Ionic columns are to be assumed. In the ground plan drawing, I opted for the former because the column drums with a diameter of 1.90 m also have several smaller Doric drums with a diameter of 1.72 m Diuxhmcsser that only cover the top or bottom stone of the Column shaft may have formed. Since a lower diameter of 1.90 m has an upper diameter of less than 1.72 m, the smaller drums must be assigned to the anteroom and rear hall. They are too big for the inner columns of the cella. Ad. Michaelis [Arx Athens.Plate. VIII) only drew (p.385) 4 columns in his addition to the temple floor plan in the vestibule and thus decided in favor of the second possibility. Since there are objections to both solutions, a third possibility must be considered: the porches could contain only 5 pillars; the column drums with a diameter of 1.72 m would then fit better because of their strength. But such an unusual arrangement seems to me very little to recommend. A 1.95 m drum built into the north wall must be counted among the corner pillars of the ring hall, since the corner pillars on the younger Parthenon are also somewhat stronger than the remaining pillars.

II.
The terraces heaped up to the south of the temple, with their retaining walls and masses of rubble, are of greater importance to the history of the Parthenon than has been attributed to them up to now. Usually only two walls of this kind are mentioned, a polygonal lining wall of irregular limestone, running almost parallel to the south side of the temple at a distance of 10-13 m, and the Cimonian castle wall, which is considerably stronger and higher and about twice the distance from the Temple has (both walls at Middleton op.cit. pl.2). In reality, however, there is a third retaining wall that lies between the two in terms of time and space. Opposite the two corners of the temple it is built of regular blocks of porosity. No new wall was built between them, but the old Pelasgian castle wall was used as a lining wall for the masses of rubble and was probably raised a little. In a report published during the excavations (Ath.Mitt. 1888 p. 434), my then relative dating of these periods was correct, but the absolute one, as will be shown later, was wrong.

In order to understand the various epochs of temple construction and their dating, a precise consideration of the individual supporting walls and their backfillings is urgently required (p.386); it is still possible on the basis of the numerous photographs which were taken during the excavations for the collection of the German Institute and copies of which can be obtained from the Institute. (The catalog is published in Arch. Anzeiger from 1891 p. 75 and 1895 p. 55

In these images, which are unmistakable witnesses to the stratification and composition of the rubble, the various retaining walls and rubble masses that gradually formed south of the Parthenon can be examined and identified in detail. One can clearly see how behind the various walls several layers of dark earth alternate with those of broken building elements and sculptures, and how lighter stripes stand out in between, consisting of small stone chips (rubble) of limestone or marble.

Plate 9: Foundation of the Parthenon, revealed by excavation along the south side.

One of those pictures that I have previously published  (Ath.Mitt. 892 Taf. IX), may first be described here (plate 9). It shows the layers of earth between the Parthenon and the polygonal retaining wall. Above the bedrock we can see the foundation of the Parthenon on the right, and various superimposed layers of earth and stones on the left. The old layer of humus, which is supported by the Pelasgian castle wall further to the left and which is not visible in the picture, reaches up to the 3rd block layer of the foundation and up to the head of the man standing there. It has covered the rock since very ancient times and its surface indicates the height of the terrain before the beginning of the temple construction. All items it contained (including some red-figure sherds) must predate the temple.

At the beginning of construction, a pit was cut into this layer of humus, reaching down to the rock, which was filled with rubble again after laying the lower layers of blocks. The pit appears in several photographs (e.g. in the picture Acropolis No. 74) as a lighter triangle next to the temple foundation. In our picture, the line of demarcation is somewhat obscured by the man. A thin layer of small boulders follows above the humus layer, the surface of which stands out as a lighter line (p.387) and extends to the middle of the 4th cuboid layer. The following layer corresponds to the height of two cuboids and contains many fragments of broken poros buildings. Its surface, in turn formed by lighter building splinters, stands out strongly. In the vicinity of the foundation, instead of the larger pieces of foros, there are heavier masses of building rubble. Above this follows a layer corresponding to three layers of cuboids, which in its lowest part contains some lumps of porosity and then masses of earth mixed with small stones. Its upper limit is formed by a double light line at the level of the 9th cuboid layer. Two stone layers higher, you can see another light layer of building rubble in the darker masses of earth. Traces of another bright layer can be seen at the level of the 13th block layer.

L. Ross and others have already correctly recognized that these light-colored layers of building rubble arose from the fact that the stone splinters that fell off when processing the blocks were spread out on the respective terrace to secure the piled-up masses of earth (cf. also Ath.Mitt. 1892 p .162). In addition to the long, light lines running through the entire terrace, one notices in the picture near the base of the temple still shorter lines and thicker layers of light-colored rubble. Their origin is due to the fact that the entire surface of the substructure was cleaned of building splinters after the completion of each block layer. After they had been swept down from the wall, the latter formed larger or smaller heaps, some of which remained lying next to the substructure and others were spread out over the terrace. Only after the completion of each 2 or 3 layers of ashlars was the entire terrace and its supporting wall leveled out evenly.

Another important fact can be seen in our picture (plate 9). While the masses of rubble, reaching up to 9 blocks, are held together by a polygonal retaining wall, the remains of which lie to the left under the rubble, without being visible in the picture, next to the higher horizontal layers of earth, reaching up to 14 blocks, we see several obliquely sloping layers of Earth and (p.388) small stones, also crossed by a few thin light lines. These sloping strata extend beyond the polygonal retaining wall and appear to belong to a period when the backfill already reached beyond this retaining wall to a second wall. The old Pelasgian castle wall opposite the center of the temple served as such, which was probably raised a little at that time. Opposite the two ends of the temple, however, are the already mentioned walls of porosity blocks, which we shall later learn more about.

The mound around the temple does not appear to have reached higher than the 15th block layer, because firstly, the above-mentioned marks are located on the 16th layer, which were made when the stylobates were made, and secondly, the ergasterion (building hut ), of which a stone is just visible in our picture above left. An even higher deposit around the temple was only possible when a stronger and higher retaining wall for the masses of earth was created by the Cimonian castle wall. These upper layers of earth, which are missing in our picture, were already removed by L. Ross during the excavations of the 1830's. The description by L. Ross [Arch. Onto. I p. 104) is sufficient. In it he found splinters of porous and marble (i.e. building rubble, "latupe") mixed with head-sized stones blown off the living rock of the Acropolis. It is obviously rubble that was spread out at the time of the construction of the Parthenon in Pericles. When Ross adds that those head-sized pieces of rock were blasted off the Acropolis rock when the temple foundations were laid, he is wrong, because the pieces of rock would then have to be found in the lowest layers of rubble. Rather, they were only blasted off the rock when, after the completion of the Periclean Parthenon, the plateau east and north-east of it was created by working down the rock. The fact that a 1.15 m and originally even 1.70 m (p.389) lower level around the temple was planned for the older Parthenon and therefore the rock was to be worked down even deeper earlier leaves no doubt about this that the blasting off can only have taken place under Pericles. So it was only then that the last heaping up of earth took place on the south side of the Parthenon. The upper part of the southern castle wall may have been built at the same time as the last elevation, while Kimon certainly built the lower part. I am therefore referring to the lower, larger part of the south wall in the passage in Aeschylus' pleading for protection (v. 134), to which Bücheler (Rhein. puree XL 629) pointed out.

Plate 13: Excavations on south side of Parthenon.

A second photograph reproduced in the accompanying Plate XIII (Acropolis No. 81) gives a more comprehensive view of the different periods of the deposits. On the right we can again see the substructure of the Parthenon, the lower blocks of which are still covered by the earth. Next to it lie the almost horizontal layers of porosity fragments and earth that we are already familiar with, again interspersed with thin and thick layers of light-colored rubble. Here, too, the gradual growth of the terrace can be followed step by step. In the two lower visible layers, many fragments of poros buildings, broken into manageable pieces, are used for filling, in the upper layers only earth and small stone chips. From the polygonal retaining wall, which holds these masses of rubble together, only a few upper stones can be seen in the middle of the picture, their lower part is still in the ground. Above this wall and further to the right we note the sloping strata of earth formed as the upper plateau widened. They are not as strong here as in other photographs, and moreover seem to be somewhat disturbed in their upper part by the east wall of the ergasterion, which is founded on column drums and ashlars. This building hut will only date from the time of Pericles, because it requires the presence of the Cimonian wall visible on the left edge of the picture and the Periclean wall in its upper part. The horizontal layers below the (p.390) column drums, criss-crossed by lines of light-colored building rubble, might initially be taken for the backfilling of the Cimonian wall and therefore dated to the time when this wall was built, but a closer study of the layers and the ground plan soon convinces us that the right half of this rubble, composed of horizontal strata (from the polygonal wall to the first two workers), is still of the third period, and was once supported by the now absent structure of the Pelasgian wall. The appearance of the layers of rubble corresponds to that of the higher layers next to the temple. On the other hand, only the masses of rubble lying further to the left, which clearly show a different composition in the photograph, belong to the construction of the Kimon wall. Only this last rubble is, as we shall see later, definitely the real "Persian rubble"; all other layers of our picture date from pre-Persian times.

Plate 14: Excavations on south side of Parthenon.

We publish a third picture on Plate XIV (Acropolis No. 91). It shows a profile of the strata slightly to the west of the previous photograph, adjacent to one of the two inner walls of the Ergasterion. Little can be seen of the Parthenon's substructure. The adjoining layers of earth of the same age as it, on the other hand, are clearly visible. The obliquely sloping layers, which were only small in the previous picture, are particularly strong here. They can be traced to the lower edge of the picture and apparently already extend beyond the line of the polygonal retaining wall, which is not visible. Above them lies a layer of ashlars from the foundation of the inner wall of the ergasterion. On the left edge below this wall, the remains of layers of rubble can still be seen, which do not yet belong to the Kimonian wall but, like those sloping layers of rubble, belong to the second construction period of the temple. Younger layers of earth and debris masses are not shown in the picture.

On the basis of these various photographs and sketches that Georg Kawerau made during the excavations and kindly made available to me, I have compiled the two sections shown in figs.2 and 3. They are intended to show the facts next to the southeast corner (p.391) of the temple and adjacent to its western half. In order to better illustrate the gradual formation and composition of the various terraces, I had to piece together the facts of several photographs taken at different locations.



Fig.2: Section t

In Fig. 2 on the right the substructure of the older Parthenon (2) with the steps of the younger building (5) is drawn in section. The steps of the older temple, insofar as they were later removed, and two unfinished pillar drums are indicated by dotted lines. Follow the rock on which the temple rests to the left under the Kimonian castle wall (4). Below the temple, the rock line rises to the right, only below the temple wall will it be worked horizontally a bit. Since the strength of this wall is not known, I drew the line of the rock completely horizontally. Before the construction of the temple began, only the Pelasgian wall 1 and the associated layer of earth I lay above the rock. The building pit described above, which was cut into this oldest layer of humus, bears the letter b. The accumulation of layers of earth and the associated polygonal retaining wall 2 kept step with the growth of the temple substructure. When the substructure rose above the 12-layer course of ashlars, it was found that the retaining wall was not sufficient, and so the sloping masses of rubble soon began to fall over the wall, especially as it could not be raised because of its thinness. The heaps of rubble gradually covered the polygonal wall 2 and in part reached as far as the Pelasgian castle wall 1. Opposite the center of the temple, this probably received a superstructure made of ashlars (3), of which, however, nothing has survived. Opposite the corners of the temple, however, special ashlar walls were erected from Pores. Its location, which can be seen in the ground plan in Fig. 5, and its shape can only be discussed after this ground plan. How the structure of the Pelasgian wall was designed cannot be said with certainty, because not only its later structure but also its old top is so badly destroyed that we can no longer even determine the original shape and height (p.392) of the wall know. That the structure, as shown in the drawing, consisted of regular ashlars, I infer from the circumstance that the two walls opposite the corners of the temple are built in this way. The backfilling of these ashlar walls was again done in horizontal layers of stones and earth, between which lighter lines of rubble can be seen. The sloping heaps reaching over the polygonal wall and those horizontal layers together form the second building period. On average they are marked III.

But even by this measure the terrace could scarcely be brought to the height intended for the older temple, which is marked in our section by two dotted lines. The fact that it actually did not reach this height at that time can be seen from the level of the ergasterion, whose floor can probably be assumed to be at the height of the second layer of ashlars of its foundation. In order to make the terrace even higher, a new strong retaining wall had to be built further south. The southern castle wall (4) served as such, which according to reliable literary tradition was built by Kimon (Paus. I 28, 3; Plut. Kimon 13). Simultaneously with it, the horizontal strata of rubble rose, which are marked IV on our section and consist of real "Persian rubble," i. H. of rubble containing numerous structures and sculptures burned and smashed by the Persians.

Somewhat later, when Pericles raised the level around the temple when building the temple again, more rubble had to be spread around the temple and the Kimonian wall had to be raised even more. The upper part of the wall erected at that time, which is significantly wider than the Kimonian wall, I have designated 5 and the terrace behind it V. As the rubble of the latter was removed by Ross as early as 1835 and drawings or photographs of them do not exist, I have left the upper part of the terrace white. The drawing published by Ross [Arch. Essays I, Taf. V) gives a (p.395) further eastern section and is therefore used as valuable material in the other profile.


Fig.3: Section t

The second section (fig.3) lies in the extension of the eastern front of the temple and is intended to show not only the retaining walls shown in fig.2 but also the retaining wall made of porous blocks (3) belonging to the second period of temple construction. I owe the height figures of this drawing to Georg Kawerau. I took the shape of the earth masses from various institute photographs (e.g. Acropolis Nos. 74, 85, 86, 87, 114, 116, 119, 121) and the sketches by Kawerau and Ross. The substructure of the Parthenon is not drawn in section, but viewed from the east (cf. the photographs of the Acropolis, Nos. 4-9). The teaching edge at the corner only starts at the 9th layer from the bottom (at c). The originally intended height of the floor next to the temple is indicated by a single line, as in Fig. 2, the height of the floor intended later when the steps of the older Parthenon were made by a double dotted line, the height actually achieved after the completion of the younger Parthenon by two full ones lines indicated. The various retaining walls have been given the same numbers as on the other intersections. The remains of an ancient dwelling house la can be seen within the oldest layer of humus, the floor and remains of the wall of which had already been buried when the temple was built. The triangular building pit b for the temple has been cut into the same layer of humus. The polygonal wall (2) is lower than in fig. 2 because the terrace it supports slopes towards the east (cf. the photograph of the Acropolis, no. 87, which shows the facade and the foundations of the wall). Sloping layers of debris that had fallen over them are here of very little thickness.

The retaining wall of the second construction period (3), made of porous ashlars, is only secured by one stone in the line of our average. However, since it is better preserved farther east (cf. the photographs of the Acropolis, Nos. 65-67), it was allowed to be added to the drawing. We do not know its former height. Only a few stones (p.396) are preserved here from the Pelasgian wall (1); it was partially broken off during the construction of the strong Cimonian wall. In the absence of drawings, the backfilling of the latter could only be indicated schematically. For the uppermost layers of earth of our section, which may be called the backfill of the Perikleisian wall 5, I have used the drawing by L. Ross (Arch. Essays I Taf. V) as a basis. Particularly characteristic of the rubble masses of this youngest part of the terrace is the stripe of white marble splinters left light in the drawing, which arose during the processing of the marble blocks of the temple.








[Return to table of contents]



Southport main page         Main index of Athena Review

Copyright  ©  2023    Rust Family Foundation.  (All Rights Reserved).

.