BMCR 2022.06.15

The Temple of Artemis at Sardis

, The Temple of Artemis at Sardis. Archaeological exploration of Sardis report, 7. Cambridge: Harvard University Press, 2020. Pp. 336, 24 p. of plates. ISBN 9780674248564 $150.00.

Open Access[1]

This monumental book is the final publication of more than a century of research and excavations of one of the largest and most important temples of Asia Minor. The excavations of Howard Crosby Butler from 1910 to 1914, published in 1922 and 1925, were continued by George M. A. Hanfmann from 1960 to 1970. Further research and publications include those by Gottfried Gruben in 1961, Wolfram Hoepfner in 1990, and Thomas Howe in the early 1980s (with publications into the late 1990s). In 1987, Crawford H. Greenewalt invited Fikret Yegül to study the temple. The result, after 33 years of work, is a magnificent opus magnum.

The difficulties of the study of the temple of Artemis at Sardis derive from two circumstances: the temple has more than one building phase, and it was never finished. It has been debated over decades whether the temple had two or three building phases, and how they should be dated. Yegül has convincingly solved these problems through meticulous observations, especially of the construction technique, and shows that there were only two building phases. A Hellenistic temple, dated to the years of Seleukid rule after 280 BC, was probably planned as a dipteros but only the cella was finished. In a Hadrianic phase the cella was divided into two cellae, and a peristasis was added, of which only the east front was finished.

After the elaborate prefaces by the editor Nicholas Cahill and the author follows — at an unusual, but useful place within the book — the English and Turkish summary. The book is divided into four chapters starting with the introduction. It comprises an overview of the history of the temple and its investigations, a summary description of its architecture — the latter repeating some of the information in the general summary — and descriptions by early travelers who saw the temple in a better condition than it was at the time of Butler’s excavations. The early remains of the sanctuary dating to the late Lydian period of the 6th century BC are noteworthy. Important for the understanding of the temple are the 6thcentury “purple sandstone base” in the east cella and the 6th or 5th century BC Lydian Altar LA 1 which was framed and enlarged, but not replaced by the Hellenistic Altar LA 2. The latter was right in front of the temple and may have been physically connected to it. The early structures were so important that they have been preserved within and in front of the new temple, and obviously were formative for the general layout of the architecture. This chapter also describes the cleaning of the ruins between 2014 and 2018; the marble walls and columns now appear bright white, and almost all photos in the volume show this new appearance.

The second and largest chapter is the very detailed description of the building. It is certainly not a chapter that will be read as a whole, but it is absolutely necessary as a reliable reference for anyone who approaches the temple with new questions, or reaches different conclusions on certain aspects than the author. The value of this chapter lies in the observation of all the details which are crucial for the overall interpretation of the temple and the identification of its phases. This becomes clear by the sub-chapter on the construction technique where Yegül — building on the work of Gruben and others — clearly defines the two phases in the lewis, clamp and dowel holes: the Hellenistic phase uses the so-called ‘Carian’ lewis system, bar clamps, and edge dowels, while the Hadrianic phase prefers the standard lewis, dovetail clamps in the foundations and bar clamps above, and square dowels with pour channels. There are also blocks showing both techniques, proving that a Hellenistic block has been re-used or re-located in the Hadrianic phase. The chapter includes a stylistic assessment of the column bases, capitals and other decorated architectural elements. The quality of the architectural decoration is indeed outstanding, and the Hellenistic features have been copied in the Roman phase so well that it is mainly due to technical differences that they can be assigned to one or the other phase. Exceptional are Hellenistic capital C with refinements like helices and palmettes on the eggs of the echinus, or Roman column base 6 with a torus covered with oak leaves with barely visible scorpions, lizards, and snails. From the entablature, only few architraves were found; obviously the temple construction had stopped at that point.

The third chapter, “Building History and Chronology”, starts with a re-assessment of the earlier studies on the Artemis temple since Butler’s excavations. Yegül then evaluates the historical, epigraphical, and numismatic evidence. A historically quite plausible argument for a Seleukid engagement in the temple building comes from the Seleukid residence in Sardis which included a royal archive, and the presence of queen Stratonike in Sardis which is attested by her dedication of a marble ball. A total of 126 Hellenistic coins were found within the Archaic purple sandstone base stuck into the joints between the blocks — a parallel therefore to be added would be the temple of Athena Polias at Priene. In the following sections, many aspects are discussed in detail; worth mentioning here is the sounding between the northeastern anta and column 16, a fill with ceramics of mostly the first half of the 1st century AD. In his preface, Nicholas Cahill, expressing his disagreement with the author, dates the column foundation to the mid-first century AD, while Yegül accepts this date only as a terminus post quem. Yegül’s careful argument against the scenario of a Julio-Claudian pseudodipteros is indeed convincing. The Hadrianic eastern front of the early Hellenistic Apollon temple at Didyma may be added as a possible parallel: its bases were stylistically dated to the Julio-Claudian period by Gliwitzky,[2] so that there could have been a first attempt in the 1st century AD which was then finished by Hadrian.

The dating of the Roman phase is also addressed in this chapter. Among other considerations, the base of the ‘talking column’ and its inscription is discussed in detail, including the parallel of its corona-shaped torus to the column of Trajan in Rome, and the archaising inscription recalling the archaic dedication of the colossus of the Naxians at Delos with the formula “made from one stone”. This indeed fits very well the spirit of the Second Sophistic which is closely connected to the emperor Hadrian.

Another issue is the partition of the cella in the Roman phase that has been explained with the introduction of the imperial cult, i.e. the second neokoria of Sardis. This previously has been connected to Antoninus Pius, while Yegül convincingly refers to recent arguments for a Hadrianic date which may be connected to Hadrian’s visit of Sardis in AD 124. A large collection of nine colossal heads, some of them rather fragmented, belonged to acrolithic statues that must have stood in the temple; out of them, Antoninus Pius, Marcus Aurelius, Commodus and Faustina Maior were identified with some certainty, and the reconstruction of a gallery of ten pairs, starting with Hadrian and Sabina and going down to Commodus and Crispina, is plausible. Yegül places them in the new eastern cella, accessible through the new colossal eastern door, whose huge size fits well the colossal emperors. Hadrian and Sabina would have stood on the base, while the others would have stood between the interior columns of the cella. Yegül assumes the location of the imperial cult in the eastern cella without any discussion or argument. This matter should have received more attention because this would mean that Artemis of Sardis would have been moved away from her sacred Archaic base which then would have been occupied by Hadrian.

A decades-long debate has been waged over the issue of the moving columns. The Hellenistic temple had a cella oriented towards the west, with a double row of interior columns, plus six slightly larger columns in the western pronaos and two in antis at the opisthodomos. In the Roman phase, the cella was cut into two; at the same time the entrance of the western front was re-located further west, thus providing space for a larger western cella; and in front of both fronts, deep porches of six columns each were erected from re-used Hellenistic columns. Their lateral, larger columns then would have used the former eight pronaos and in antis-columns, while the two smaller middle columns on each side which were put on top of pedestals originated from the interior of the western cella. Cahill, in a further disagreement mentioned in his preface, argues that the columns in antis were not removed before the 4th or 5th century AD; but then four columns for the Roman porches would have been missing, which makes Yegül’s reconstruction convincing. In any case, the western cella had to be re-designed because the columns — formerly in- and outside of the Hellenistic cella — were not in a line, nor was the floor level. This made a new roof construction necessary, possibly carried by two pillars or columns upon the “rough piers”; this is confirmed by evidence for a Roman phase of the marble roof. On the other side, the eastern cella most probably remained almost unchanged.

The fourth and final chapter provides an architectural analysis. The first phase of the temple must have been planned as a dipteros, similar to those at Ephesos and Didyma, although only the cella was finished. The plan was restricted by the two venerable elements of the Archaic altar and base, which is the reason why the altar stands so close to the temple and the base stands in the middle of the cella instead of at its rear. The relation to Ephesos is close since the Sardian Artemis is affiliated to the Ephesian one — yet a closer comparison to Didyma and the Athenian Olympieion, both being Seleukid colossal temples, would be desirable.

The Roman phase encircles the naos with the peristasis of a pseudodipteros, but in a strange and irregular form, not found in other pseudodipteroi: the wide pteroma appears only on the long sides of the temple, while the two temple fronts were supplied with a deep porch. While the pseudodipteros plan connects the temple with the architecture of Hermogenes and the regional architecture, the deep porches come from the Italian-Roman temple architecture which otherwise remained foreign to Asia Minor. The Archaic element of gradually increasing intercolumnia towards the center of the east front sets it apart from any other Hellenistic or Roman pseudodipteros. Yegül connects that feature to the past of Sardis, the golden age of Kroisos, and to Ephesos, in the spirit of the Second Sophistic, while adding elements of Italian temple architecture. Yegül rightly emphasizes the genius of the architect, but it should not be overlooked that this eclectic arrangement is a typical element of imperial cult temples in Asia Minor.

The book is very carefully edited and copiously illustrated. A fundamental feature is the 24 fold-out plates in a separate box with the excellent plans and drawings of Yegül, all produced in traditional technique. They provide the documentation of all the observations on which the interpretations rely. Irritating is only the wrong designation of the peristasis as “peristyle” throughout the book.

To sum up, Yegül’s book is a worthy final publication of this exceptional and colossal temple, and it succeeds in solving most of the problems that have been discussed for decades. It is a milestone in the archaeology of temples in Asia Minor.


[1] This is a review of the print edition; the online open access edition contains the same content in a slightly different form.

[2] C. Gliwitzky, “Hadrianisch oder caliguläisch? Zur kaiserzeitlichen Bauphase am Apollontempel von Didyma”, in: Otium. Festschrift für V. M. Strocka (Remshalden 2005) 97–106.