This atlas presents an updated set of maps and plans for the island of Delos, continuing a long tradition of cartographic excellence maintained by French scholars since the beginning of systematic excavations by the École française d’Athènes in 1873. The authors aim at providing “une nouvelle étape dans l’histoire des relevés de l’île et des fouilles” (7). This goal is certainly fully accomplished, and the authors deserve the highest admiration and praise for providing this monumental masterpiece, which, for the first time, makes the achievements of the long-lasting excavations in Delos available in one single volume, in a standardized format, and with superb quality. If the following brief assessment includes some critical remarks, they are by no means meant to diminish the unquestionably paramount value of this atlas, whose price—if quite an investment for individual scholars—is fully justified.
The foreword lists three reasons for publishing this atlas (7): 1. some of the previously published maps of the islands and plans of the ancient remains are outdated, incomplete, and faulty; 2. so far, maps and plans were not systematically produced and provided digitally; 3. topographical instruments and computers are now available that allow such a big project to be brought to fruition in less than ten years. Indeed, the project was started in 2002 and completed in 2011 (according to the date of the foreword), even if its results were only published in 2015. While the first two reasons are excellent arguments, ideally they should have been supported by the provision of digital versions of the plans and maps, in addition to the printed results.
The printed version consists of a large box that includes a thin booklet with an introductory text (39 pages), and an extra hardcover box with 51 foldout plates of different sizes, ranging from 30.5×57 cm to 117×62 cm. The plates include large overview maps (pls. 3-9) of the entire island in the scale of 1/5000 and of the excavated remains in scales of 1/5000, 1/2000, and 1/1000; 36 detailed plans of the archaeological remains (scale 1/200; pls. 10-46); and five sections of the different quarters of the city (scale 1/200; pls. 47-51). The atlas is incorporated into the prestigious excavation series of Delos ( Exploration archéologique de Délos / EAD), as volume XLIII, a choice that put certain constraints on the format of the maps and plans. Between 1909 and 1995, volumes in the series had a uniform format of about 28×35 cm, but their size was diminished from EAD XXXVII onwards. Currently they measure 23.5x 30.5cm, dimensions to which the foldout plates had to be adapted (20).
The introductory booklet includes three parts. After a succinct, very useful history of the cartography of Delos since 1873 (9-18), the aims of the atlas as well as the techniques and methods of surveying and documentation are discussed (19-25), followed by an outline of graphic conventions (27-32). The text is complemented by a list of the plates (33-34) and an index of the monuments (35-39).
Since the introductory text is focused entirely on Delos, one would have appreciated a brief contextualization of the project within the existing scholarship, reflecting upon possible models and comparable endeavors. For example, the term atlas is used in research for a broad variety of projects that have little in common with EAD XLIII. Large-scale excavation projects of ancient cities such as Corinth, Cyrene, Ephesos, Kassope, Miletus, Morgantina, Ostia, Paestum, Pergamon, Pompeii, Priene, Solunto, Thera, and Timgad are not (yet) documented with comparable collections of plans.1 Works that come close to EAD XLIII are the documentation of Olynthus, the recent publication of the Punic settlement on the Acropolis of Selinunte, and the atlas of Megara Hyblaea.2 In its systematic coverage of an entire island and ancient site and its holistic approach, however, the Delian atlas project seems unique, a distinction that further underlines its groundbreaking significance.
All maps and plans use a clearly readable color coding system: brown for contour lines; red for elevations; white, black, and different gray shades for different types of walls; and blue for water on pls. 3-9. Contour lines and elevations are consistently, and in case of the latter abundantly, indicated on all plans, a scheme which is a major asset of the atlas. All plates are labeled on the back (number, scale, title; in addition, on pls. 5-51 thumbnails of the plan are present), but only pls. 3-9 include a north arrow and scale bar on the plan itself. It would have been helpful to add these also to the plans (pls. 10-46) for easier use and reference.
The overlap of the plans is in general sufficient (cf. pl. 2), but sometimes (too) thin (e.g., pls. 37 and 38). Care was taken to show large buildings, such as the Agora of the Italians, in their entirety on at least one of the plans. Overlapping plans, however, are not always fully congruent (cf., e.g., pls. 28-29, 33; pls. 36-38). While the differences may be minimal and unavoidable, given the size of the foldout plans, they are still noticeable and impede an easy assemblage of the plans at the 1/200 scale (one must go back to the 1/1000 scale maps, pls. 6-9). All foldout plans can be handled without difficulty, but they are still too large for convenient scanning, and thus use in teaching and research. Once again, providing digital versions of the plans would have helped enormously.
When dealing with an island that extends for ca. 350 ha and with excavated remains that cover a surface area of about 25-30 ha, clear decisions must be made about what to record and how. Photogrammetry was used to generate the large overview maps (pls. 3-9) from aerial photos that the Greek army had made in 1982 (17, 22-23). For the detailed plans (pls. 10-46), the original intention was to digitize and update the existing drawings, but they proved to be too faulty. Therefore, all buildings were newly surveyed with a total station and differential GPS technology, plans were then drawn with Adobe Illustrator, and finally controlled twice on site. This new documentation includes a significant number of structures that had not appeared on any earlier plan, e.g. in the Quarter of Skardhana, to the east of the Agora of the Italians, in the north-east area of the Quarter of the Theater, as well as the southern part of Warehouse/Magasin γ, and the House of Fourni. Finally, the five sections (pls. 47-51), for which no models were available, are a welcome novelty in Delian cartography.
While the newly drawn plans (pls. 10-46) are not stone-by-stone representations, but rectified and hence somewhat abstract ones (28), they still show an impressive amount of structural detail, using a differentiated repertoire of marks and colors (white, black, gray shades) for walls and materials. Some of the marks are too small and too lightly printed, however, and not suited for indicating the exact state of preservation; this affects particularly pavements, whose type and extension are hard to distinguish clearly on the plans.
Walls and structures are visible on the surface in many areas of the island, and it is not always easy to determine their date (ancient or modern). The atlas project opted for a reasonable compromise in determining what to show. For the detailed plans (1/200, for size and location cf. pl. 2), all structures, ancient and modern, excavated or just visible on the surface, were surveyed and drawn. A similarly comprehensive approach could not be adopted for the terrain beyond these areas, but pl. 4 at least shows areas marked in grey where structures abound on surface. Alongside the atlas project, Philippe Fraisse had begun to record all visible walls in the entirety of Delos; since this project has not yet been completed, its results could not be incorporated into the atlas (21).
For the new plans, only structures visible today were recorded, omitting those that had been destroyed or reburied after excavation (19). This choice is convincingly explained with the need to deal with a major remodeling of the site, especially in the Sanctuary of Apollo and around the Sacred Lake, which was underway when the atlas project team was working in Delos. For areas less affected by the rearrangement of the site, this approach seems, at times, somewhat too conservative, rendering the new plans in some aspects less informative and detailed than the older ones, which were made shortly after excavation or at least before large-scale restoration with modern cement floors took place. For example, several drains that appear on the earlier plans of the Quarter of the Theater and of the Stadium Quarter are missing on the new plans. As a result, in the Stadium Quarter, room ε and room c of House I B/Perfumery are not clearly identifiable as latrines because the channel that served them is not shown, or rather breaks off incomprehensibly where it is no longer visible today.3 One wonders whether at least some of the achievements of earlier documentation projects could not have been included into the new plans, maybe clearly marking structures as “no longer visible” or “reburied”; this would have resulted in even more comprehensive and more easily readable plans.
The decision that this reviewer regrets most, however, is mentioned in a single sentence without any further explanation on p. 28: “Les coups de sabre ne sont pas notés, sauf pour les rues, les portes et les fenêtres bouchées.” This is a major drawback for those who want to study the history and development of buildings and insulae. For example, for the Quarter of the Theater with its dense occupation and complex history, the new plans suggest a homogeneous construction in a single phase. Since the plans do not show the remains cut at a uniform level (28), but switch levels in order to provide maximum detail and information (e.g., niches and windows), one may also have expected the marking of wall joints. In this regard, the gray walls of the new simplified plans, which would have easily allowed the graphical indication of joints, seem much more deceptive than the black walls used in earlier plans.4
Minor disagreements notwithstanding, this is an excellent volume that should serve as a model for the publication of other sites and excavation projects, in the traditional printed version and ideally in a future open access online format. While the plans (gladly!) do not excuse those interested in studying Delian archaeology and urban development from thorough examinations on site, they constitute a prime example of basic research and a superb, intellectually stimulating working tool for future investigation.
Notes
1. This list is by no means complete, and Athens and Rome with their specific tradition of dictionaries and atlas-editions are deliberately omitted here.
2. Olynthus: The comprehensive documentation of all remains is spread over several volumes of the excavations series and has not been systematically updated since then: D. M. Robinson, Architecture and Sculpture. Excavations at Olynthus II (Baltimore 1930); D. M. Robinson and J. W. Graham, The Hellenic House. Excavations at Olynthus VIII (Baltimore 1938); D. M. Robinson, Domestic and Public Architecture. Excavations at Olynthus XII (Baltimore 1946). Selinunte: S. Helas, Selinus II. Die punische Stadt auf der Akropolis (Wiesbaden 2011). Megara Hyblaea: G. Vallet , F. Villard, and P. Auberson, Mégara Hyblaea 1. Le Quartier de l’Agora archaïque (Rome 1976). The text is complemented by a box with three “portefeuilles”, among them an atlas including 51 foldout plans (45×28 cm) that document in the scale of 1/100 the excavated area within the Hellenistic fortification; building phases are distinguished by different colors and signatures.
3. Cf. here pls. 10, 28 with A. Plassart, Quarter d’habitations privées à l’Est du Stade, Bulletin de correspondance hellénique 40, 1916, 144-256, pls. V-VII, and J. Chamonard, Le Quartier du théâtre. Exploration archéologique de Délos VIII (Paris 1922-24) pls. III-VI. Drains or parts of drains are missing in a significant number of buildings on the new plans, probably because they were overgrown, covered by earth or by modern cement.
4. EAD VIII, pls. III-VI (op. cit. n. 3); these plans even include some joints, marked white.