[image ALT: Much of my site will be useless to you if you've got the images turned off!]
Bill Thayer

[image ALT: Cliccare qui per una pagina di aiuto in Italiano.]

[Link to a series of help pages]
[Link to the next level up]
[Link to my homepage]

This webpage reproduces an article in
American Journal of Archaeology
Vol. 38, No. 1 (Jan.‑Mar. 1934), pp93‑106.

The text is in the public domain.

This page has been carefully proofread
and I believe it to be free of errors.
If you find a mistake though,
please let me know!

 p93  The Repair of the Athena Parthenos:
A Story of Five Dowels

In 1910, while drawing a plan of the Parthenon in connection with Dr. Hill's impending publication of the Older Parthenon,1 I was obliged to pay due attention to the pedestal of the gold-and‑ivory Athena Parthenos, the masterpiece of Pheidias dedicated at the Panathenaic festival of 438 B.C. Before 1910, just as in more recent years, there have been several discussions of the pedestal;2 but, since no thorough examination of its site has yet been made, it seems desirable to retrieve some of the observations buried in my old notebooks.

The well-known traces of the statue pedestal consist, primarily, of a portion of the floor area (about 2.62 × 6.53 m)3 wherein, for reasons of economy, thirty blocks of poros limestone were substituted for eight marble pavement slabs. At the exact center of the poros area is a rectangular hole (0.542 × 0.755 m and 0.372 m deep) for the central mast of the statue's skeleton.4 And all around the poros area are dimly perceptible traces, sometimes engraved lines, and sometimes merely variations of weathering, outlining an area 4.096 × 8.043 m.5 On the marble surface, between the engraved outline and the poros area, and also within the poros area itself, are dowel holes and pry cuttings marking joints of the lowest course of the pedestal, showing that the blocks were uniformly about 1.03 × 1.30 m. Six of these blocks, of marble 0.295 m high, have recently been identified by means of their dimensions and by the fact that one of them reproduces the socket for the central mast; they have been replaced, therefore, approximately in their original positions (Fig. 1), but, unfortunately, all have been so mutilated that none retains any trace of the original external finish of the pedestal.

[image ALT: A large ruined stone room open to the sky, with a single monumental door in the background. It is a view of the cella of the Parthenon in Athens.]
Fig. 1. — Cella of the Parthenon with Pedestal of Athena in Foreground

With regard to the jointing of the pedestal, Lehmann-Hartleben argues that the lowest course was composed of twenty-four blocks, each 1.03 × 1.31 m, and thus obtains his total dimensions of 4.12 × 7.86 m. These dimensions conflict with the actual outline of the pedestal engraved on the marble floor (4.096 × 8.043 m), and also with the jointing. It is clear, from seventeen pry cuttings and two dowel holes (one of the latter a T‑dowel) on the poros area, that the core of the structure was  p94 composed of eight slabs in two rows of four. The periphery, instead of containing sixteen blocks, had only eleven. This is shown by twenty-four pry cuttings and eleven dowels (four of them T‑dowels), of undoubted original workmanship, on the surrounding strip of marble. While the rear or west edge of the pedestal contained six blocks, the front or east edge had only half as many, three blocks of double length; similarly on the north and south edges there were only three blocks (counting the angle ones twice) instead of four, the central ones being of double length. Thus there were five joints on the back, two on the front, and two on each end, corresponding to the number of original dowels on the marble strip.6

The unnoticed fact which I wish to stress at this point is the existence, on the same marble strip, of five large dowel holes of late type, in shape utterly different from the fifth century dowels of the original pedestal, and accompanied, furthermore, by pour  p95 channels (Fig. 2). Four of these dowels, on the east face, are 0.06 m‑0.08 m long and 0.04‑0.065 m wide; they have short pour channels (0.06‑0.08 m in length) leading from a point 0.18 m behind the east face of the original pedestal. Similarly, a single dowel at the northwest corner, 0.03 by 0.06 m, has a long pour channel (0.18 m) leading from a point 0.18 m behind the west face of the original pedestal. These totally unexpected dowel holes incite us to look for additional traces of reconstruction. And they are not lacking. Thus we find two pry cuttings, one at the north and the other at the south, 0.18 m behind the north and south faces of the original pedestal, but in no possible relationship to the original joints. In the middle of the east face the marble pavement shows considerable signs of wear for about 0.19 m within the area covered by the original pedestal. And faint weathered traces appear on the north about 0.555 m beyond the poros area. All these bear incontrovertible testimony to a reconstruction of the pedestal with its face 0.188 m within that of the original, covering an area of 3.72 × 7.667 m.7

[image ALT: A plan of the pedestal to the monumental statue of Athena in the Parthenon in Athens.]
Fig. 2. — Plan of Pedestal of Athena Parthenos

Thus we are confronted by a problem hitherto unsuspected. The gold-and‑ivory statue apparently stood on its pedestal from 438 B.C. (the dedication) until 375 A.D. (Zosimus, VI.18), or even until about 435 A.D. (Marinus, Proclus, 30). Yet at some intervening period the pedestal was reconstructed, beginning with the bottom  p96 course. Was this reconstruction limited merely to the facing of the pedestal? Or was the statue itself rebuilt?

*   *   *   *   *

The removable gold plates of the statue,8 a total weight of 44 talents or 1151 kilograms (2,538 lbs.),9 and about ¾ mm in thickness,10 must have formed a constant temptation to the lawless. We hear that they were actually removed under official supervision in 433 B.C., in order to prove that Pheidias was innocent of embezzlement.11 During the disturbances of the Peloponnesian War a certain Phileas or Philourgos, according to the orators Aeschines, Andocides, and Isocrates, succeeded in stealing the gold-plated Gorgoneion from the shield;12 but it was recovered at least as early as 398 B.C.13 For such reasons, after any political or administrative disruption, the golden statue received a careful inspection. Thus, while the annual inventories of the cella of the Parthenon (the Hekatompedos Neos) ordinarily make no reference to the statue, we find it occupying the first place on the list in 384, 376, 343, 320, and 316 B.C.14 I have referred elsewhere15 to the fact that these occasions always (with the possible exception of 343 B.C.) coincide with the first year following some disturbance.16 In the winter of 304/3 B.C. we hear that Demetrius Poliorcetes profaned the Parthenon by taking up his quarters and staging his orgies within it.17 A fragmentary decree passed on Poseidon 23 = December 28, 304 B.C., beginning πειδὴ . . 17 . . τοῦ ἀγάλματος . . 15 . . ἐν τῶι ἑκατομπέδῶι, may have referred to the condition of the statue at this time.18 Yet we cannot affirm that anything of serious nature happened to the statue before 300 B.C.

With the end of the fourth century our great series of official inventories of the various chambers of the Parthenon (IG I2, 232‑309; II2, 1370‑1492) comes to a sudden termination. The reason for this, obviously, was the disappearance of the precious offerings recorded in the inventories, because of the depredations of Lachares. By means of a coup d'état in 300 B.C. he established himself as "tyrant" at  p97 Athens,19 and there maintained himself against Demetrius for nearly five years. For the payment of his mercenaries a precedent had been set during the Sacred War (356‑346), when the Phocians paid their mercenaries by melting gold offerings at Delphi, to the value of 4000 talents, and silver offerings of more than 6000 talents.20 So Lachares, according to four passages which have come down to us, melted down the golden shields dedicated on the Acropolis, the golden Nikes, and even the removable gold from the image of Athena herself; as it was laconicallyº expressed, "Lachares left Athena nude."21 With such explicit testimony, this removal of the gold of the statue by Lachares, who doubted by some, appears to be fully authenticated. We have, furthermore, a series of gold coins (Fig. 3), such as were struck by Athens only in times of great stress, which are stylistically intermediate between the other known gold issues of 406 and 88 B.C., and therefore can only belong to Lachares.22 The gold of the Athena Parthenos could have furnished 132,000 of these staters.

[image ALT: A photograph of the obverse and reverse of a small gold coin.]
Fig. 3. — Gold
Perhaps from the Parthenos
(Coin of Lachares, ca. 297 B.C.)
If we now consider, on grounds of probability, the epoch at which the gold might logically have been restored, we find ourselves in difficulties. The brief period of the second domination of Demetrius Poliorcetes (295‑288) was quite unfavorable: Athens was free but poor during 288‑262 B.C., and the last years were occupied by the checkered course of the Chremonidean War.23 After a brief period of suppression (262‑256), Athens enjoyed a more liberal treatment under Antigonus Gonatas and Demetrius II between 256 and about 230 B.C., though little seems to have been done for the embellishment of the city. Another period of poverty-stricken independence filled the last third of the century. In fact, for at least a century after the removal of the gold, it is impossible to suggest a moment at which it could have been restored.

During this long interval the statue may have existed with its ivory portions more or less intact, and likewise its helmet, shield, and other accessories. But are we to assume that the wooden (or, more likely, plaster) backing of the drapery was left exposed to view? It is more probable that the missing gold would have been replaced by gold leaf, which could have been done with only one fiftieth of a talent instead of 44 talents weight of gold.24 Such a repair would undoubtedly have been made not long after 295 B.C. But there would have been no reason for carrying this  p98 work down through the pedestal; and the shape of our five dowel holes seems to belong to a much later period. Are we to suppose, furthermore, that Pliny, Plutarch, Pausanias, and all later writers were deceived by the gold leaf?25

*   *   *   *   *

At this point we may recapitulate a series of more or less related facts pertaining to late repairs of the architecture of the Parthenon. While these have never yet been considered in their entirety, or even as the results of a single cataclysm, yet I believe that careful examination will show that they are all repairs made after a great conflagration.

1. The marble roof tiles of the Parthenon were evidently replaced in antiquity. Those which actually survive were executed in a technique quite foreign to the fifth century, the planes of the prismatic cover tiles, in particular, being merely cut with a saw instead of being worked with a chisel and rubbed smooth. The marble, too, is of a very inferior quality, filled with narrow parallel pink veins.26 Such a repair might, of course, have been necessitated by decay of the supporting timbers; but it would seem more plausible to regard it as the result of a fire.

2. Together with these repairs of the roof we may associate traces of alteration in the east pediment. Some of these alterations were noted by Sauer in 1890; I had taken additional notes in 1911; but the most detailed record is that published by Carpenter in 1933.27 The latter says, "An interesting architectural study might be made to center upon block 19 (of the east cornice)." As Carpenter shows, it is apparent that four of the pedimental figures were removed, a standing figure (probably Hermes) and K, L, M (the so‑called "Three Fates"); and four cornice blocks (19‑22) were lifted from below them. To facilitate their removal and replacement, the original tong holes of block 21 (the original last-laid block) were employed, but new cuttings of tremendous depth and careless workmanship were made on blocks 20 and 22. And block 19, the sixth north of the center, was so damaged that a substitute was carved for its place, very roughly tooled on the top,28 and hoisted by means of a lewis hole.29 No clamps were employed for block 19, though the half T‑clamps in the adjoining blocks 18 and 20 show that the original block 19 had been securely fastened between them. After resetting the cornice, the "Three Fates" were restored to their places and fastened by dowels of a form not used elsewhere in the pediments; but, in the absence of any fragments or of any traces on the new cornice block 19, there is no evidence that the fourth statue was ever replaced, so that it may have  p99 been broken and abandoned like the cornice block 19 below it. Carpenter suggests that an earthquake was responsible for the damage; but it should also be observed that this new cornice block lies directly below the middle purlin of the north half of the roof, and it is conceivable that a fire and the collapse of the roof construction might have developed a thrust which could have overturned the tympanum block and the statue before it.

3. Any fire that could have affected the roof must also have destroyed the wooden ceiling below; and the chances are that such a fire would have been carried by draft out through the great east and west doorways of the cella building, and would have badly injured the door enframements and particularly the lintels (as was the case, on two successive occasions, in the north doorway of the Erechtheum). It so happens that the west door lintel, which alone survived (until 1926), was very badly calcined by fire, a condition which led to its support by extra jamb linings in Byzantine times (exactly as in the Erechtheum), to the insertion of a segmental arch in 1872, and to the final removal of the lintel in 1926. But, just as in the Erechtheum the Byzantine repair of the north doorway was preceded by repairs of the reign of Augustus after an earlier fire, so in the Parthenon the Byzantine jamb linings were merely the successors of earlier repairs. The marble enframements of the original doorways were replaced by later substitutes, for which additional cuttings were prepared in the jambs, exactly as in the Propylaea.

4. A fire darting out through the great doorways would undoubtedly have injured the inner faces of the Doric capitals and architraves of the pronaos and opisthodomos porticoes. And this is exactly what happened, as we learn from a detail which was first observed by Bötticher. "The echinus of the column capital is partly studded with groups of iron pegs . . . set in carefully bored holes. . . . These cylindrical pegs occur on the inner face of the echinus toward the opisthodomos." He speaks also of "their recurrence in the echinus of the pronaos columns, as in the capital of the still existing southeast column, and in the two capitals from pronaos columns now lying in the cella."30 But Bötticher was deceived by their neat workmanship, and advanced the erroneous theory that they belonged to the original construction of the temple, forming the attachments of the intercolumnar grilles! Also Durm regarded them as part of the original, but pointed out that Bötticher had been in error in claiming that they did not occur on abacus or epistyle. In Durm's opinion, they were repairs, for the attachments of patches, the capitals (and epistyles) having been injured during erection and therefore turned with their patched faces inward.31 Penrose gave the correct interpretation: "There are numerous iron pins fastened on to the echinus of some of the capitals on the side towards the temple as well as to some other parts. These have evidently been intended as keys for plastering, for the purpose of making good the surface where injured by fire."32 At present these key-holes can be seen on the calcined and flaked inner faces of the first, second and third capitals from the north, in the opisthodomos, on the abacus as well  p100 as on the echinus, and likewise on the southeast pronaos angle capital in situ, and on two pronaos capitals lying outside the southeast corner and one lying inside the northeast corner of the Parthenon. The same holes occur near the bottoms of the inner faces of the second and third epistyles from the north, in the opisthodomos.33 With this interpretation in mind, however, Penrose inferred that they were repairs made after the fire which according to Fanelli broke out after the explosion of 1687, "executed in stucco, because, as is well known, the Turks used this material."34 It is clear that Penrose, while correctly interpreting these pegs as "merely intended for the sake of subsequent repairs," was even more incorrect than Bötticher as to their date.35 The inner face of the southernmost epistyle of the opisthodomos is badly calcined and its bottom is gone, and the southern return is likewise in a bad state; these two blocks, built into the campanile-minaret of the fourteenth century, prove that the fire occurred before that date. And since it would be inconceivable that any medieval builder would have restored these Doric profiles, we must regard these repairs as ancient, a solution which is indicated likewise by their technical characteristics.36

5. It is conceivable that such a fire would have injured the Doric internal colonnades of the eastern or Hekatompedon chamber. Here, as it happens, we know that even the visitors who arrived before the explosion of 1687 did not see the original twenty-three Doric columns in each tier.37 And the fact that no remains of these original colonnades have been identified would suggest that they were demolished in ancient times.38 On the other hand, what the early modern travellers saw was  p101 two tiers of twenty-three columns each, the central one in the lower tier having been removed for the sake of the church and its entablature blocks replaced by an arch.39 The lower story, at least, was of the Doric order;40 the upper may have been "Aeolic."41 Their positions are now marked by circles 0.65 m in diameter, placed nearer the edge of the inner stylobate than the original columns (1.117 m in diameter), and more closely spaced (2.49 m instead of 2.603 m).42 At their centers are large dowel holes, with long pour channels, such as are totally absent from the original construction. Fragments of the columns of the lower tier have been noted on several occasions. Wilkins in 1802 discovered that "the lower portion of one whose diameter corresponded with this dimension was lying within the cella. The method observed in fluting proves this column to have been of the Doric order, but it is to be observed that the lower part of the shaft was planed and not fluted."43 Dodwell also noted "fragments of some small Doric columns," and "some soffits of corresponding dimensions."44 And I have myself at various times observed pieces of such shafts, with their lower portions divided into twenty facets, in or around the Parthenon, and had long ago come to the conclusion that they belonged to the internal colonnade.45 The corresponding pieces of entablature, mentioned by Dodwell, were carefully drawn by Penrose: "the length 8.21 (= 2.50 m) conforms to the reduced range" and "agrees with the traces left on the pavement of the later columns." The architrave is 1.47′ (= 0.448 m), the triglyph frieze 1.70′ (= 0.518 m), the cornice 0.83′ (= 0.253 m), giving a total height of 4.00′ (= 1.219 m). Penrose notes that "the workmanship of the entablature seems to be all of one period," and it was certainly made for this place.46

 p102  As for the date of this internal colonnade, one is astonished to find that it is universally attributed to the time of the transformation into a Byzantine church. Wilkins quotes Fauvel as giving "a positive assertion that the interior columns of the Parthenon, of which there were then remaining several portions, were of the lower Greek Empire."47 Those who did not recognize this Doric order as belonging to the Parthenon (including Michaelis), attributed the new colonnades to the building of the church.48 And, influenced by such statements, even Penrose asserted that "the smaller columns had been substituted for those of the original structure when the temple was converted into a church," and fancied that "the feeble character of the detail assigns it to a late period, which, however, still retained something of the classical tradition; in fact somewhere between the time of Constantine and Justinian, at which period we may presume the Parthenon was adapted for Christian worship and the previous solid columns removed."49 But such a date is absurd; the Byzantine constructors would have been quite unable to design or execute such an order. The details of the order, which was clearly designed for this place, and the technical characteristics, including the fastening to the stylobate, recall rather the Pergamene work of the second century B.C.50

Here, at last, we have evidence for a reconstruction of which the date would agree with the form of the five dowel holes in the pedestal, and a reasonable explanation of the complete reconstruction of the pedestal. But it also suggests a more weighty consequence. If the interior of the temple was gutted by fire at the middle of the second century B.C., the statue itself could hardly have escaped. That we hear nothing of such a catastrophe in literature may be due, in part, to the gaps in our knowledge of Hellenistic Athens; but it may also be due to the fact that, the gold plates not having been restored, it did not mark a great economic disaster. From the artistic viewpoint, however, it was a notable disaster: it is difficult to escape the conclusion that most of our numerous copies of the Athena  p103 Parthenos reproduce, not the statue by Pheidias, but a Hellenistic and therefore more or less inaccurate replica.

*   *   *   *   *

We turn, therefore, to the copies, from which we must seek the answer to two questions. (A) If the original statue was destroyed in the second century, on what material did the maker of the replica base his reproduction? (B) Do we find, among the series of copies of the Athena Parthenos, a moment at which it acquired sudden popularity, which might be suggested as the epoch of its reconstruction?

In seeking the answer to the first question, we are immediately struck by the enormous number of free copies and reflections of the statue in minor works of art, particularly during the eighty years after the original dedication, and so before the influence of Scopas and Praxiteles had supplanted that of Pheidias. For instance, we have about two dozen Athenian marble reliefs well distributed over the first half of the fourth century, some of them very close copies, one even representing the column supporting the right hand.51 Again, we find numerous representations dating from this period in jewelry (particularly the two gold medallions from Kul Oba near Kertch in the Crimea, Attic workmanship of about 400 B.C.), terracotta disks and moulds, and in relief ceramics. As for the coins, it seems evident that the Athena Parthenos was also the direct inspiration of the heads on the coins of the Athenian colony of Thurioi, toward the end of the fifth century, and from this center the type spread over numerous cities of the west.52 It was not until a late period, in the so‑called "New Series" (229‑30 B.C.), that the Athena Parthenos appeared on coins of the home city;53 at the time of the origin of the type the statue still retained the original ivory face and bronze helmet.

Even more important, however, are the actual copies in the form of statues, dating from this early period. Of these, we may cite at least three:

[image ALT: A photograph of one side of an ancient coin, depicting a standuy draped woman, holding a shield against the floor with her left hand; her right, apparently supported on a small tree, holds a winged being with a plam or a wreath. It is a coin of Aphrodisias, representing a cult statue of Athena, possibly based on the now vanished statue of Athena Parthenos in the Parthenon.]
Fig. 4. — Contemporary Replica of the Parthenos
(Coin of Aphrodisias, ca. 375 B.C.)
(1) Marble copy in Acropolis Museum, Athens (Casson, Catalogue, II, no. 1362), probably of latter part of fifth century — though not of the best execution and nearly one quarter of the actual size. Collignon regarded this as the best copy.

(2) Cult statue of the fifth century in the sanctuary of Athena at Side in Pamphylia; the sanctuary is mentioned by Strabo (XIV, p667); the complete statue, no longer extant, was represented on coins beginning at about 400 B.C. and persisting for more than a century. In the later examples the figure stands on a pedestal and so clearly is a statue.54

(3) Replica, probably a cult statue, at Aphrodisias in Cilicia, represented on coins about 375 B.C., showing the Parthenos with a Nike on her extended right hand, supported on a tree-trunk, a substitute for the column of the original, and therefore clearly a statue (Fig. 4).55

It is evident, from these few surviving examples, that there was no dearth of material to which the rebuilders of the statue could have referred.

 p104 We now turn to the second question, the moment of the revival of popular interest in the type. And here we meet a series of strange coincidences. After a period of approximately two centuries, during which interest in the creation of Pheidias had waned, we find a sudden revival toward the middle of the second century B.C. as evidenced by half a dozen examples:

(1) Marble copy in Berlin, found at Pergamon, one third the size of the original. It is a free, rather than an accurate copy, and in style resembles the work of Eumenes II (197‑159) on the Great Altar. The statue was set up, furthermore, in the Library attached to the stoa surrounding the temple of Athena Nikephoros; both the stoa and the Library were likewise erected by Eumenes II.56

(2) Marble cult statue at Notion, within the Pergamene zone of influence, about two fifths of the size of the original. The discoverers note that "it presents very close analogies to the colossal statue of Athena coming from the Library of Pergamon."57 Likewise the surrounding stoa resembles closely the Pergamene architecture of the second century B.C., though the temple itself is a later reconstruction, apparently of the period of Hadrian.58

(3) Gold-and‑marble cult statue in the temple of Athena at Priene, celebrated by Pausanias (VII.5.5), and shown to be a copy of the Athena Parthenos by coins of the Roman imperial age, ranging from Vespasian to Valerian.59 Fragments of the actual statue (front part of the colossal left foot, part of left hand, and left upper arm, of marble: gilt bronze wings of Nike, with charred wood of body) are in the British Museum. The missing drapery was undoubtedly of wood, covered with gold leaf or gold plates, for which holes of attachment remain. The dimensions shows that it was half of the size of the Athena Parthenos.60 The date is given by seven tetradrachms of the pretended son of Ariarathes IV, King Orophernes (159‑157) of Cappadocia,61 who is known to have deposited 600,000 tetradrachms within this temple; the seven discovered formed part of a foundation deposit, in special cavities provided in the statue pedestal, which is clearly later than the temple.62

 p105  (4) Reproductions of the entire statue on Cappadocian silver coins from Ariarathes IV (220‑163) to Ariarathes X (42‑36 B.C.).63 Whether this represented an actual statue somewhere in Cappadocia, or was adopted as a symbol for some other reason, remains to be seen.

(5) Reproductions of the entire statue on Syrian coins, from Alexander I Bala (150‑145) to Antiochus XI (92 B.C.), as well as on coins of Cyrrhus (where a statue was perhaps erected) issued under Alexander I Bala.64

(6) Reproduction of the entire statue on a tetradrachm from Gortyna in Crete.65

From all this evidence of a sudden revival of interest in the Athena Parthenos at about the middle of the second century B.C., I might attempt to reconstruct its history as follows. Our reproductions of this period all come from the area dominated by the kings of Pergamon, Cappadocia, and Syria.66 We are at once struck by the coincidence that, not only were these the three kingdoms which Eumenes strove to unite in a triple alliance, but they were also the three kingdoms whose rulers were honorary citizens of Athens. As Ferguson says, "Carneades, the head and third founder of the Academy . . . had the honor of being the teacher of two princes, who were severed by a difference in age of twenty years, but united by a mutual affection for Stratonice, queen of Pergamum (the one was her brother, the other her brother-in‑law and lover), and by common college days in Athens — Attalus, subsequently the second king of that name, and Ariarathes who, later, as king of Cappadocia acclimated Greek culture, and especially Greek philosophy in his boorish and ill-reputed country. They joined in erecting a statue of their brilliant teacher, and, upon receiving Athenian citizenship, had themselves enrolled in the upper Cephisus deme Sypalettus. This was perhaps in 178 B.C., and the connection thus established was maintained both with one another and with Athens for their respective lifetimes."67 Two years later came Antiochus, soon to become Antiochus IV of Syria; he likewise became an honorary citizen of Athens and served as one of its mint magistrates for the year 176/5 B.C., when the Athenian coins bore his symbol, the elephant.68 He owed his throne to the friendship established with the Pergamene princes during the sojourn at Athens. Ariarathes, too, may have served as one of the mint magistrates.69

In return, these Asiatic potentates were lavish in their display of affection for Athens. Antiochus IV, upon becoming king, began the enormous Olympieion at Athens,70 and erected a huge gold-plated Gorgoneion against the south wall of the Acropolis.71 In this friendly rivalry Eumenes II retorted by erecting the Stoa of Eumenes along the south slope of the Acropolis.72 Attalus II followed with the  p106 Stoa of Attalus in the Agora.73 Just what form the benefactions of the Cappadocian monarchs took we do not know; we hear of gifts to the club of Dionysiac artists at Athens,74 but there were clearly others to the State as well, if we may judge from the statues erected in honor of Ariarathes V and his Queen Nysa. One or more of these benefactors may have been responsible for the reconstruction of the interior of the Parthenon and of its gold-and‑ivory statue, perhaps accidentally destroyed by fire shortly before 165 B.C. The expense of such an undertaking would have been as nothing to Antiochus. Possibly he collaborated with his friendly brother-in‑law, Ariarathes IV.75 In this way Cappadocian kings would have secured the right to employ the statue as an emblem on their coins. Antiochus, too, might well have employed it on his coins; but his main interest lay rather in Zeus Olympius, as evidenced by his coins, and by the hangings in the temple at Olympia, the new temples at Athens and Lebadeia, the erection of a full sized replica of the Olympian Zeus of Pheidias, likewise of gold-and‑ivory,76 and by the lining of the walls of the temple of Zeus at Antioch with plates of gold.77 That he did not forget Athena, however, is demonstrated by the huge gold-plate Gorgoneion. And some color is given to this theory by the fact that, when the Pergamene-Cappadocian-Syrian alliance was renewed in 150 B.C. by the accession of Alexander I Bala at Antioch, the Athena Parthenos at once appeared on the Syrian coins, and stayed there.

Thus restored, the Athena Parthenos served as the model for several important statues of the period. At Pergamon, Eumenes II, whose wife Stratonice was the daughter of Ariarathes IV and niece of Antiochus IV, installed a copy in his Library before 159 B.C. At Notion, which belonged to Pergamon, a copy served as the cult statue of the temple of Athena. And at Priene, when the original cult statue was injured in 157 B.C., during the siege of the city by the brothers-in‑law, Ariarathes V and Attalus II, in connection with the Orophernes affair, it was rebuilt probably under the patronage of Ariarathes.78

To recapitulate, we have found that the pedestal of the Athena Parthenos was reconstructed in late Greek times. As for the statue, while stripped of its gold before 295 B.C., conditions would not have been propitious for a complete restoration in the original material until more than a century had elapsed. Before this could be done, what remained of the original statue was apparently destroyed by a fire, as shown by the architectural evidence, in the second century B.C. And the rebuilding of the interior of the temple, and the erection of a new statue (from which most of our copies are derived), apparently occurred just before the creation of numerous copies dating from about 160‑150 B.C.

William Bell Dinsmoor

Columbia University

The Author's Notes:

1 Hill, A.J.A., 1912, pp535‑558.

2 For the older discussions, see Penrose, Athenian Architecture (2nd ed.), pl. 4; Dörpfeld, Ath. Mitt., 1881, pp293‑297, pl. XII; Puchstein, Jb. Arch. I., 1890, pp111‑117; Reisch, Eranos Vindobinensis,º 1893, pp4‑5; Perry, A.J.A., 1896, pp335‑346; Reisch, Jh. Oest. Arch. I., 1906, p221; Winter, Jb. Arch. I., 1907, pp55‑70. For discussions since 1910, see Winter, Jh. Oest. Arch. I., 1915, pp1‑16; Vannoy, Univ. Iowa Humanistic Studies, I, 5 (1917), pp13‑20; Dinsmoor, A.J.A., 1921, p128; Lehmann-Hartleben, Jb. Arch. I., 1932, pp17‑46.

3 The measurements vary at different points since the poros blocks advance or recede to fit the marble pavement. Thus Lehmann-Hartleben gives 2.63 × 6.52 m.

4 Schrader (Phidias, p36) gives 0.54 × 0.80 m and 0.40 m deep after Magne (Collignon, Parthénon, pl. 4); Lehmann-Hartleben gives 0.52 × 0.78 × 0.39 m.

5 Penrose gives the length as 32.233 - 5.798 = 26.435 ft. = 8.057 m; Dörpfeld measured 4.09 × 8.04 m; Reisch 4.09 × 8.08 m (misprint?); Wiegand gives 3.50 × 7.60 m. (Priene, p110 n.), a curiously deficient estimate; and Lehmann-Hartleben calculates 4.12 × 7.86 m by the erroneous process of multiplying the extant blocks of 1.03 × 1.31 m by 4 and 6 respectively.

6 It is clear from the relationships of the pry cuttings to the dowel holes, and of the latter to the axis of symmetry, that of the eleven blocks forming the periphery the southeast corner block (1.21 × 2.72 m) was laid first, then the central blocks on east (2.60 m) and south (1.676 m), next the northeast (1.21 × 2.72 m) and southwest (1.21 × 1.42 m), and so on, the west block adjoining that at the northwest corner being the last laid and undowelled.

7 Only three of the four corner blocks were dowelled in this late repair.

8 Removable according to Thucydides (II.13). Diodorus (XII.40), Plutarch (Pericles, 31), and Pausanias (I.25.7).

9 The weight given as 40 talents by Thucydides (II.13), from whom Plutarch quotes (de vitando aere alieno, 2), and as 50 talents by Diodorus (XII.40); but Philochorus (in Schol. Aristophanes, Peace, 605), who gives 44 talents and seems to have quoted directly from the inscriptions and archives on the Acropolis, is to be preferred.

10 I estimate 0.77 mm, without including the helmet, shield, serpent, pedestal reliefs, etc., which were undoubtedly of other materials. Earlier estimates were those of Quatremère de Quincy (Le Jupiter Olympien), 1.13 mm; Albizzati (J.H.S., 1916, p399), 1 mm; Vannoy (op. cit. pp36‑39), 0.6 mm.

11 Plutarch, Pericles, 31.

12 Isocrates, 18.57 (oration delivered 403‑400 B.C.); Synesius, de calvitie, 19, p83 A; Suidas, s.v. Φιλέας, Φιλοῦργος; Photius, s.v. Φιλοῦργος; Eustathius, ad Od. XI.633; Bekker, Anecd. gr., p315, 20. Cf. Schriftquellen, 681‑684; Michaelis, Parthenon, p270, 43‑44; Arx Athenarum, p99, n42.

13 Mentioned in inventories of 398/7 to 368/7 B.C. as stored in Hekatompedon awaiting restoration (IG II2. 1388, 1393, 1400, 1401, 1415, 1421, 1425); compare also another piece from the shield (IG II2.1423, 1425, 1428, 1429).

14 IG II2.1407, 1410, 1443, 1468, 1477.

15 A.J.A., 1932, p163, n7.

16 Another inscription no longer extant, but said by Pittakis to have been found near the Parthenon (IG II2.1513) bears the letters καὶ τὸ χρυσίον ὃ . . . . τὸ ἄγαλμα τῆς θεοῦ . . . . μὴ ἔλαττον, σταθμόν . . . Unfortunately the date and meaning of the inscription are as uncertain as its present location.

17 Ferguson, Hellenistic Athens, pp118‑119; Dinsmoor, Archons, pp14, 37, 64, 383.

18 IG II2.482; Dinsmoor, Archons, p37, n2.

19 It had formerly been considered that this occurred in 295 B.C., which was really the date of his fall. For the true date, see Ferguson, Cl. Phil., 1929, pp14‑15; Proc. Mass. Hist. Soc. LXIV, 1930/32, p363; cf. Dinsmoor, Archons, pp. xiv, 64, 389‑390.

20 Diodorus, XVI.56.

21 Pausanias, I.25.7; Athenaeus, IX.70, p405F; Plutarch, Isis and Osiris, 71 (cf. Schriftquellen, 685‑687; Michaelis, Parthenon, p268, 21; Arx Athenarum, p63, n44). See now also the new Oxyrhynchus papyrus XVII.2082 (Ferguson, Cl. Phil., 1929, pp1‑20).

22 For the coins, see Svoronos, Monnaies d'Athènes, pl. 21; cf. Mrs. Shear, Hesperia, II, 1933, pp248‑249. The specimen illustrated is from the collection of Mr. E. T. Newell.

23 For the dates, see Dinsmoor, Archons, pp. xiv-xvi, 56‑88.

24 Pliny (H. N. XXXIII.61) says that one ounce of gold (27.29 grams) made 750 leaves, each 4 dactyls (0.074 m) square. Thus an area 1 m square would require 183 leaves, and the whole Athena and Nike (77.90 sq. m) would have required 14,255 leaves, equivalent to 19 Roman ounces (0.518 kilogram), or 1/50 Attic talent.

25 Pausanias is sufficiently specific with regard to statues that were merely of wood covered with gold leaf (I.42.4; II.2.6; VII.26.4; IX.4.1) and not of actual gold.

26 These are evidently the tiles which Dörpfeld (Ath. Mitt., 1889, p332) regarded as pre-Persian, and so assumed that the process of sawing tiles was invented by Byzes of Naxos. Wiegand (Poros-Architektur, p180) excluded them from the pre-Persian period, and regarded them as post-Persian. Lepsius (Gr. Marmorstudien, p123) and Durm (Baukunst d. Griechen, p206, n2) speak of them as if they were Periclean (cf. Frazer, Pausanias, III, p503). But the original fifth century tiles are of the best Pentelic marble, and show no trace of the saw; these are unquestionably due to late repairs.

27 Sauer, Antike Denkmäler, I, p51; Ath. Mitt., 1891, p70; Carpenter, Hesperia, II, 1933, pp8‑9, 11‑12, pl. I‑II.

28 Carpenter gives a photograph on his pl. I, and a plan on his pl. II.

29 During this resetting process the upper left edge of metope XI, directly below, seems to have been injured; at least, there exists here a repair patch of Parian marble (Praschniker, Parthenonstudien, p173).

30 Bötticher, Untersuchungen auf der Akropolis zu Athen, p148, fig. 20 (cf. p158 for the pronaos).

31 Durm, Zeitschrift für Bauwesen, 1871, p483, fig. 16‑18; Centralblatt für Bauverwaltung, 1895, pp222‑223, fig. 10, 11, 13, 14; Ἀρχ. Ἐφ. 1895, pp22‑24, fig. 6, 7, 9; Baukunst der Griechen, p396 (cf. p171, fig. 144).

32 Penrose, Athenian Architecture, p21.

33 Apart from the sketches by Durm, cited above, see the photograph in Collignon, Parthenon, pl. 72, upper left illustration.

34 Penrose, Πρακτικά, 1895, pp194‑195.

35 The very fact that they appear in the pronaos capitals, overthrown in 1687, shows that they cannot be later.

36 After the manuscript of this article had been written, I received (January 5, 1934) an article from Professor W. Kolbe (Forschungen und Fortschritte, IX, 1933, Dec. 10/30, pp497‑498), in which he also calls attention to these traces in the opisthodomos of the Parthenon and observes that they must be due to ancient repairs after a fire. He mentions another fact which I had not observed, namely, that part of the fluting of the column built into the minaret is restored in stucco; and he properly interprets this as proof that the fire was earlier than that of 1687, and consequently that the repairs occurred in antiquity. Kolbe utilizes these traces as evidence against my argument (wherein I follow Dörpfeld) that the Opisthodomos par excellence (the state treasury) was not in the Parthenon at all, but in the Poros Temple of Athena (A.J.A., 1932, pp143‑172, 307‑326). For, he says, the damage to the Parthenon resulted from the fire in Opisthodomos par excellence, of which we learn from Demosthenes (XXIV.136), and which, I believe, occurred in 377/6 B.C. A vital objection to Kolbe's theory, and consequently to this early date of the repairs, is the fact that the traces of fire are just as prominent in the pronaos as in the opisthodomos, so that it would hardly have been merely a fire in the Opisthodomos. Again, a fire sweeping from end to end of the building would certainly have attacked the Athena Parthenos: yet, as we have seen, the statue was certified as intact in 384 B.C. (before the fire of Demosthenes) and also in 376, 343, 320 and 316 B.C. (after the fire of Demosthenes).

37 Beulé (Acropole, II, pp37‑40) claims that they did see the original Doric columns, and that the repairs were executed after 1687. This is impossible, since the temple formed an open ruin after 1687, with a small new mosque in the middle; no reconstruction of the cella colonnades took place at that time. Beulé was misled by the apparent necessity of reconciling Magni's account of the Doric columns (shortly before 1687) with the existence of Ionic bases within the cella (see below).

38 Penrose found a fragment of a fluted shaft which seemed to him to fit the original traces on the stylobate; but I have not seen this fragment and cannot verify the identification. Penrose also assigned to the original upper colonnade a Doric capital which we now know to belong to the monument of Nicias (A.J.A., 1910, pp470‑471, 483‑484; Ath. Mitt., 1911, p62). Penrose suggests that certain recut beams formerly lying in the Parthenon were the original upper entablature, shortened to fit the closer intercolumniations of the alteration.

39 Wheler, Journey into Greece, p363: "On both sides, and towards the Door, is a kind of Gallery, made with two Ranks of Pillars, Twenty two below, and Twenty three above; the odd Pillar is over the Arch of the Entrance, which was left for the Passage." So also Spon, Voyage, p155.

40 Cornelio Magni, Relazione della Città d' Atene, p502: "Dividesi il Tempio interiore in trè Navi, . . . spartita da Colonne . . . Gli architravi e capitelli risaltano in ordine Dorico."

41 Vienna Anonymus (ca. 1460; cf. Michaelis, Parthenon, p353): "Toward the southwest, northwest, and (west) there is the (double) row of columns, which reach far up into the heights, the capitals of the columns are hewn palmiform through working with iron."

42 While the number of late columns was the same as in the original scheme, they were all thrust westward in order to equalize the east and west end intercolumniations, which had originally been very unequal.

43 Wilkins, Atheniensia, p100 n.

44 Dodwell, Classical Tour in Greece, I, p330; cf. Michaelis, Parthenon, p48.

45 Certain Ionic bases found inside the Parthenon have sometimes been attributed to this later colonnade, with the result that Beulé (Acropole, II, pp37‑40) claimed that they belonged to a Turkish repair after 1687. Ussing (Griechische Reisen, pp180‑182) pointed out that there could have been no Turkish reconstruction of the whole interior after 1687, and argued that they belonged to the colonnade of the Byzantine church, ignoring Magni's description of the Doric columns. In this he was followed by Michaelis (Parthenon, p48), who thought that the colonnade was Corinthian. All disregarded the late Doric columns mentioned above, which, with their perfectly fitting entablature, sufficiently disprove the theory. The Ionic bases are of Asiatic form, much earlier than the Parthenon; I have seen four of them, and they may have been brought into the Parthenon to support the four red porphyry columns of the baldachino, mentioned by Spon and Wheler, of which Dodwell saw fragments east of the Parthenon.

46 Penrose, Athenian Architecture, p20, pl. 9A. He also cites recut marble blocks of approximately the same length, but yielding greater lengths if we restore the curtailed ends (ibid., pp20‑21, pl. 9A). These may well have been taken from an early building, or even, as Penrose suggests, from the upper entablature of the Parthenon cella, and, forming an inconspicuous portion of the repair, have been thus barbarously treated. It may be noted that this entablature is employed by E. Flagg in his analysis called The Parthenon Naos (New York, 1928). He admits that it is late, to be sure, but infers that it was an accurate copy of the original entablature of the upper interior order, with some alteration of the triglyph spacing.

47 Wilkins, Atheniensia, p101 n.; according to Legrand (R. Arch. 18972, p97), no such statement is to be found among Fauvel's manuscripts.

48 Michaelis, Parthenon, p48.

49 Penrose, Athenian Architecture, pp10, 20.

50 I had been inclined to assign to this reconstruction (cf. A.J.A. 1932, p160, n8, pp170‑171) an inscription from the Athenian Acropolis, referring to "the men elected to supervise the preparation of the temple of Athena" (IG II2.1023). For this, according to Kirchner, dates from the second half or latter part of the second century. I am indebted to Mr. Sterling Dow, at Athens, for more exact information about this inscription: "IG II2.1023 is cut in large deep letters with great flaring apices. The letters are widely spaced, and blank spaces are used freely, for punctuation. The wreath at the bottom is cut in relief." He finds the same characteristics in IG II2.1028 (archon Medeios, of 101/0 B.C.). "In the case of lettering of so pronounced a style, great confidence is justified; in the case of such perfect similarities, doubt simply disappears. 1023 and 1028 are by the same hand; there are no other decrees by this mason." In view of this evidence that IG II2.1023 dates from about 101/0 B.C., we must either infer that the repair at this later date was a minor one, or, more probably (since the word is κατασκευήν and not ἐπισκευήν), that it was the temporary decoration of the temple on some special occasion.

51 Lack of space forbids listing all these minor works. A list that is comprehensive up to 1911 is given by Robinson, A.J.A., 1911, pp500‑503, C-D; see also Robinson, Olynthus: IVterracottas, no. 358, pl. 37. The relief with the column is in Berlin (Sammlung antiker Skulpturen, III, K 104, pl. 83).

52 These types are discussed by Lermann, Athenatypen auf griechischer Münzen pp59‑68.

53 See, for example, Svoronos, Monnaies d'Athènes, pls. 33‑81.

54 Lermann, op. cit. pp75‑77, pl. II, 8; British Museum, Coins of Lycia and Pamphylia, pl. XXVI, 4‑11.

55 Lermann, op. cit. pp77‑78, pl. II, 6; Imhoof-Blumer and Gardner, Numismatic Commentary on Pausanias, pl. Y, 22 (= J.H.S., 1887, pl. LXXIV, 22). The example illustrated is in the collection of Mr. E. T. Newell.

56 Winter, Jb. Arch. I., 1907, pp55‑70; Altertümer von Pergamon, VII, pp33‑46, pl. VIII; Collignon and Pontremoli, Pergame, pp143, 145; Lehmann-Hartleben, Jb. Arch. I., 1932, pp21‑27.

57 Demangel and Laumonier, B.C.H., 1925, pp322‑323.

58 B.C.H., 1923, pp362‑373; 1925, pl. XIII-XIV.

59 British Museum, Coins of Ionia, pl. XXIV, 13; Dressel, Sitzb. Berl. Akad. 1905, pp467‑476; Regling, Münzen von Priene, nos. 191‑197, 209, 211, 220‑221; Lehmann-Hartleben, Jb. Arch. I., 1932, p27, fig. 10.

60 Antiquities of Ionia, IV (1881), pp26, 29, 31; Rayet and Thomas, Milet et le Golfe Latmique, pl. 15, fig. 19; Smith, British Museum Catalogue of Sculpture, II, no. 1150, 1‑4 and pieces in magazine: Walters, British Museum Catalogue of Bronzes, no. 1728; Wiegand, Priene, pp110‑111; Lehmann-Hartleben, Jb. Arch. I., 1932, pp27‑31.

61 British Museum, Coins of Galatia and Cappadocia, pl. VI, 5; Hill, Historical Greek Coins, p145, pl. XI, 86; Regling, Münzen von Priene, pp44‑45.

62 Antiquities of Ionia, IV, pp25‑27; Hicks, J.H.S., 1885, pp261‑274; Greek Inscriptions in the British Museum, III, no. 424; Wiegand, Priene, pp84, 111; Smith, British Museum Catalogue of Sculpture, II, pp146‑147; British Museum, Coins of Galatia and Cappadocia, pp. xxviii-xxix; Regling, Münzen von Priene, pp8‑10, 44‑45.

63 Lermann, op. cit. p79; British Museum, Coins of Galatia and Cappadocia, pl. VI, 2‑4, 6‑9; pl. VII.2‑11.

64 Lermann, op. cit. p79; British Museum, Coins of the Seleucid Kings, pl. XV, 5; XVI, 12, 15; XVII, 10; XVIII, 12; XX, 6; XXII, 4; XXIV, 2, 9; XXV, 12; XXVI, 8.

65 Lermann, op. cit. p79; Head, Historia Numorum, p467.

66 Cretan mercenaries at this time in the employ of the Syrian kings, and even, after 164 B.C., assuming control of the country, might explain the appearance of the type at Gortyna.

67 Ferguson, Hellenistic Athens, p300.

68 Ferguson, op. cit. pp302‑303; for the coins, Svoronos, Monnaies d'Athènes, pl. 44.

69 The theory that Ariarathes V was one of the mint magistrates is dismissed by Ferguson (op. cit. p301, n1), who argues that it was a later Ariarathes. But Svoronos places the coins of Ariarathes far earlier: see Svoronos, Monnaies d'Athènes, pl. 53.

70 Ferguson, op. cit. pp303‑306.

71 Pausanias, I.21.3; V.12.4.

72 Ferguson, op. cit. pp299, 342.

73 Ferguson, op. cit. pp300‑301, 342, 367, 442.

74 IG II2.1330; Ferguson, op. cit. pp301, 370.

75 I might call attention to the curious analogy in that a later Cappadocian king, Ariobarzanes II (62‑52), rebuilt the Periclean Odeum which had been burnt during Sulla's siege in 86 B.C.

76 Ammianus Marcellinus, XXII.13.1; Granius Licinianus, XXVII; the Nike on the hand was also of gold (Justin, XXXIX.2.5).

77 Livy, XLI.20; Granius Licinianus, XXVII.

78 The mere fact that coins of Orophernes were found under the statue need not imply that Orophernes rebuilt the statue. The cost, to be sure, may have been deducted from the 400 talents which were to be restored to him. The same would apply to the "sacred stoa" along the north side of the Agora, which is usually attributed to Orophernes because of the fragmentary inscription Βασιλέως ἀριαράθου on its architrave (Wiegand, Priene, pp214‑215; Inschriften von Priene, no. 173); this construction might with more propriety be attributed to Ariarathes V, the true son of Ariarathes IV.

[image ALT: Valid HTML 4.01.]

Page updated: 18 Sep 12