The study of the city in the ancient, late antique and medieval world involves complex questions that concern several fields of research. The challenge of combining archaeology, geography, landscape archaeology, literature and historiography by focusing attention on a specific place and in a more or less limited period is a noteworthy ambition; even more so if different studies are inserted inside a single framework. For this reason, the experimental work carried out by the editors of this volume is most welcome.

The book gathers the talks at the conference ‘The Art of Siege Warfare and Military Architecture from the Classical World to the Middle Ages: an International Conference for the Study of Poliorcetics, Military Historiography and the Archaeology of Battlefields’, organised in Haifa, Israel, in February 2017. It is a good example of how different scholars and experiences can be brought together, with the aim of providing scholars with a broad chronological framework (early fifth century bc–sixteenth century). The geographical scope is also wide, from Larissa Phrikonis to Sicily, from Smolino to Acri.

The collection is part of a trend that is producing interesting results (i.e. the research project ‘Prokop und die Sprache der Bauten’ of the Johannes Gutenberg-Universität of Mainz) and is experiencing great publishing success, including companions and miscellaneous books (e.g. J. Armstrong and M. Trundle [edd.], Brill's Companion to Sieges in the Ancient Mediterranean [2019]) as well as specialist studies. It is certain that the book will occupy a prominent position in this field, since it offers the reader a concise, clear, well-informed collection of contributions, accompanied by useful illustrations.

The focus of the work is essentially archaeological, but this does not mean that the essays are exempt from an interdisciplinary approach. Archaeology and architecture, archaeology and history, archaeology and military strategy communicate and understand each other. Scholars who make a link between archaeology and geography sometimes fall into an almost inevitable trap of those who focus on a specific geographical area but neglect, or rather imply, the theoretical aspects of geography, while elaborating models of knowledge similar to those of anthropic geography, which nevertheless are not always well known by a non-specialised public.

The relationship between archaeology and literary sources is more problematic and deserves some reflection. In this review I will not dwell upon the content of each chapter, dealing with specific cases in the manner cited above, but I will instead add some observations regarding the dialogue between the two disciplines.

I would like to begin with the editors’ introduction: ‘This idea was born during a spontaneous debate that we had about terminology’ (p. 1). The idea is original: a collection of archaeological studies is born from a question of lexicon. Unfortunately, for a non-expert public, this starting point has not been successfully transformed into a dynamic structure around which the different essays are built.

Therefore, there is no explanation of the possible diachronic evolution of terms such as toichobates, phrourion, emplekton etc., although their contextualisation in a specific context is no small feat. For example, diateichisma usually defines a cross-wall, and it is given this meaning in the book (Chapters 1, 6, 9). However, this term acquires a series of semantic nuances: ‘dividing wall’, ‘fortification wall’ (Polyb. 8.32.2; 34.9; Joseph. BJ 5.199). Above all, it takes the meaning of ‘cross-wall’ from the cross-wall at Athens, and it is used several times in Procopius (Aed. 2.7.15: city of Edessa; 4.2.22: Heraclea and Mycopolis; 4.3.21: Potidaea / Cassandrea; 4.11.12: Anastasiopolis; 4.11.7: Toperus / Rhousion). In short, diateichisma may have been a lexical key on which some topics could have focused, both to illustrate its applications at a local level and above all to carry out a diachronic and diatopic investigation in relation to city architecture and the organisation of urban space. We should point out that this was research already carried out by A. Sokolicek (Diateichismata [2009]), in a volume of which some of these essays could have been excellent case studies with in-depth analysis.

The case of emplekton is even more interesting. This term is used in ancient military architecture to indicate a particular type of masonry, built with heterogeneous materials, earth and different-shaped stones, with which the space between the two vertical parameters of the wall is filled. This meaning is never given in Greek (see LSJ, s.v.); instead, the noun is used in a technical sense thanks to Vitr. 2.8.7, which clearly illustrates its meaning in relation to different types of construction, as a clear sign that it was used by experts, but was not entirely clear to others and therefore needed explanation.

These linguistic notes introduce the more general problem of the relationship between archaeological evidence and literary sources, in our case, between the technical texts that have been transmitted and the archaeological evidence on sieges and city buildings. The problem is extremely complex, and the volume deals with it only marginally, when B. Jansen (pp. 76–8) highlights that Aeneas Tacticus (32.12) recommended building a second wall behind the first to prevent enemies from entering the city. I would add that a clear example of this building is found in the siege of Plataea in 429 bc (Thuc. 2.6.3), but also in Perinth (Diod. Sic. 16.74.3); at Tyre the citizens built a second wall at a distance of five cubits from the first and filled the gap with stones and earth (Diod. Sic. 17.43.3), while in Pelusium in 525 bc the garrison was able to quickly build a second wall when a large portion of the first was falling, to prevent the enemy's advance (Diod. Sic. 16.49.1). Therefore, I would also add that Philo of Byzantium gives us more details on this structure and recommends building a triangular counter-wall, with the point facing inwards, through which many windows are to be made on both sides. In this case, if the curtain covering the wall should fall, the enemies advancing inwards will be wounded in the flanks by both the artillery and the troops, and also by the citizens (Philo 18 [p. 92.22–6 Thévenot]).

Could Philo of Byzantium's manual have been used by architects for the fortification of cities? Jansen rightly poses this question, which should be the prerequisite for the problem of terminology from which the whole volume originates. This question goes beyond a mere comparison between incongruous testimonies, which is generally a risky operation often producing questionable results. Indeed, we should consider the characteristics of these technical manuals, their relationship with the literary tradition on the one hand and with contemporary reality on the other, and finally the extent of the witnesses that have been transmitted. Studies on sieges and military architecture certainly cannot be separated either from the Poliorketika of Aeneas Tacticus or the Paraskeuastika and Poliorketika of Philo of Byzantium for the classical period. In the same way, the De obsidione toleranda should be a point of reference and comparison for those who deal with these issues for the Byzantine period (an excellent study for this period is provided by G. Ravegnani, Castelli e città fortificate nel VI secolo [1983], unfortunately an almost unknown volume to non-Italian speakers). An entirely neglected source is Syrianus Magister's De re strategica, which dedicates a specific section to poliorcetics and military architecture (chapters 9–13, pp. 28–44 Dennis).

In short, even if the military manuals did not give the architects and soldiers of their time or later ages advice on how to build and defend a city, they could certainly help scholars to interpret the archaeological data.

However, the opposite is also true. For instance, the text in Philo 44 (p. 83.7–10 Thévenot) ταύτῃ δὲ <ἡ> πριονωτὴ παραπλήσιος οὖσα τυγχάνει, ἣν Πολύειδόν φασιν εὑρεῖν τὸν μηχανοποιὸν ἐν τῇ με-<…κα>τά τινας τῶν ἐπικαίρων τόπων is corrupt. Scholars have proposed various hypotheses to amend and make it understandable. Among these, H. Diels and E. Schramm (Exzerpte aus Philonis Mechanik B. VII und VIII [1920], p. 28) proposed ἐν τῇ Με<γαλοπόλεως περιτειχίσει⋅ μάλιστα δὲ αὐτῇ χρῆσθαι δεῖ κα>τά. They believed that Philo was referring to Megalopolis, on the basis of the English excavation campaign underway in that period (E.A. Gardner, Excavations at Megalopolis, 1890–1891 [1892]). A.W. Lawrence (Greek Aims in Fortification [1979]) considered the conjecture by Diels and Schramm as arbitrary and inappropriate, above all because ‘Polyeidus is not known to have worked there’. Indeed, the Thessalian Polyeidos, whom Philo considers to be the heuretes of the πριονωτὴ τειχοποιία, is traditionally linked to the siege of Byzantium led by Philip II in 340 bc (Athen. 10.5–9; Vitr. 7, praef. 14; 10.13.3), when he constructed the helepolis, and was active in Rhodes (PBerol 13044 col. 8, ll. 6–7). The conclusions drawn by Diels and Schramm were erroneous, although their method was excellent.

We very much hope that this volume will provide useful food for thought regarding the study of literary sources on sieges and military architecture. Considering its wealth of ideas, we believe that this will be the case.