Apollodorus of Damascus, architect and master-builder associated with the emperors Trajan and Hadrian, is the purported author of this text, a manual for building siege devices. David Whitehead has done us all a service by producing both the first English translation and the first comprehensive commentary in any language. Greek text and English translation are on facing pages (pp. 36-67); suspect passages are given in reduced font, and page numbers are taken from Carl Wescher’s German edition.1 The commentary (pp. 69-136) is philological and not primarily historical, a point that is already grieving at least one historian, but which provides the foundation needed for future historical work.2 There is a two-page endnote on Trajan’s Column, six pages of illustrations, a short bibliography (32 items), and sixteen pages of indices of passages cited and of Greek and English terms (pp. 137-162).
Whitehead had a fruitful relationship with Henry Blyth, a retired schoolmaster trained in engineering, with whom Whitehead co-produced his earlier text and commentary on Athenaeus Mechanicus.3 Although not a co-editor in the Apollodorus volume, Whitehead is clearly indebted to Blyth, not only through the book’s dedication and because Blyth had published the only article on the work that Whitehead could find while reading at Oxford in 1999.4 Whitehead credits Blyth with the “brilliant initial realization that, as transmitted, ‘Apollodorus’ is heavily interpolated with later material.” This article should be read along with the book because it offers explanations, missing in the book, for why such material is identified as such. Blyth held strongly that all ancient military writings of a technical nature “had to be tested against the laws of physics and engineering” in order to distinguish “the practical and the impracticable, the real world and the world of fantasy.” (p. 10) Such testing, however, has no part in the book at hand.
The Introduction (pp. 15-34) deals with the text sources, the identity of the author, and the internal arrangement of the work. The text derives from an edition by Rudolf Schneider, who reused Wechser’s page and line numbers, maintained by Whitehead throughout.5 The author’s identity and dates remain elusive; Whitehead surveys Procopius, the SHA, and Dio, while holding that his putative identity remains that of PIRA 2 A.922, the Syrian-Greek architect and engineer.6 In considering the many later interpolations in the text, Whitehead thinks that the present version was enlarged from the original text, given shape prior to the tenth century, perhaps by a Byzantine military writer known as Syrianus Magister.
The questions of whom Apollodorus is addressing and the identity of the historical events connected to that unnamed despota are also vexatious ones. Whitehead takes the problem apart succinctly in the introduction, concluding that the present treatise was addressed to Trajan rather than Hadrian, and that the military campaign at issue was the first invasion of Dacia (AD 101-102), a conclusion that follows Blyth. The treatise notes that the author is not informed of the specifics of Dacia—especially its topography—and this best fits the first rather than second invasion. The tone of the salutation also suggests Trajan rather than Hadrian, with whom Apollodorus had strained relations and by whom he may have been murdered. This hypothesis is admittedly sketchy and does not always fit with other sources, but certainty is not possible here.
The internal arrangement of the work begins with an epistolary preface, leading to a rough “list of contents” that is followed by the main text, undivided in the manuscripts but now road-mapped with subheads provided by scribes, some competently and others not, and often in disagreement with each other. The text is marked with Wechser section numbers; line numbers must be counted. Whitehead divides the text as follows, showing the Wechser line numbers (pp. 25-26): Preface 137.1-138-17
List of Contents 138.18-139.8
Ch. 1: Protecting the attackers 139.9-143.5
Ch. 2: Excavation 143.6-147.6
Ch. 3: Brick walls 148.2-152.5
Ch. 4: (Firing) stone walls 152.6-153.7
Ch. 5: Rams 153.8-161.8
Ch.6: A device for reconnaissance 161.9-164.4
Ch. 7: Towers 164.6-174.7
Ch. 8: Ladders 175.2-188.9
Ch. 9: An assault raft for crossing rivers 189.1-193.5
The question of illustrations has several levels. Apollodorus mentions the hypodeigmata . . . diagrapsas (137.3-4) and schēmata … diegrapsa (137.7-8), which, Whitehead notes, refer to serious technical drawings and not models. What we have today are medieval drawings, with uncertain relationships both to the original manuscripts as well as to the many allusions to drawings in the text. Whitehead generally translates the common schēmata as “figures,” katagrapsai as “depictions,” and opseis as “views.” Following other language markers, including the nature of the text as “an exposition in continuous Greek prose,” Whitehead concludes that the text is not a series of comments made to the illustrations, but rather a text into which illustrations were inserted. (p. 27) The illustrations are not intended to be definitive or precise, but are rather “what I believe his designs would have looked like.” (p. 14)
The commentary provides a base-line for reading the text; it is clearly not intended to offer a comprehensive historical interpretation of this rather dry, technical, un-literary text. That this has become a pleasure to read is itself a testimony to the expert care taken in producing it here. Two samples of text, translation and commentary may be the best way to convey the flavor of the work. Near the start, Apollodorus sets out the obligatory plea to the emperor, along with a statement of his own limitations as an author (text lines 138.13-17):
Ἐάν δέ τι ἐν τοῖς ἐπὶ ἑκάστου συστήματος ἐπιλογισμοῖς ἀσαφῶς εἴπω, σύγγωθι, δέσποτα. Καὶ γὰρ τὰ ὀνόματα τῆς ἐπιστήμης ἀσυνήθη ἔσται τoῖς κοινοῖς λόγοις, καὶ ποικίλην θεωρίαν ἔχει τὸ ἔργον, καὶ αὐτὸς ἐγὼ εἰπεῖν τάχα ἀσθενέστερος · τάχα δὲ ἡ μεγαλο – φυἶα σου διορθοῦται, καὶ συγγινώσκει ἡ εὐμένεια.
If there is anything unclear in what I say in the descriptions applying to each apparatus, excuse me, master; the vocabulary of the science will be unfamiliar to everyday speech, the task involves complex theory, and I myself am perhaps rather weak with words. Perhaps, though, your natural genius puts this to rights, and your graciousness forgives it. (section 138)
Whitehead provides six commentary items on these lines (p. 73). For instance, regarding line 13, en tois . . . epilogismois, Whitehead notes that Liddell, Scott and Jones list this as the only instance of sustēma meaning “machine, apparatus,” and that mēchanēma or ergon would have sufficed. For line 14, ta onomata tēs epistēmēs as “the vocabulary of the science” has a precedent in [Xen.] Ath. 1.19, Athenian rowers and their slaves learn onomata . . . ta en tē nautikē. In line 16, Whitehead notes that the insertion of “perhaps” ( tacha), repeated in the next sentence, indicates that Apollodorus, rather than seeing his works as weak, “is quite pleased–and reasonably so–with his efforts.” Such comments serve a purpose different than that undertaken by Blyth, who was more concerned with differentiating the authentic from the spurious sections of the text, and with the practical aspects of the devices. Blyth in many ways offered fuller explanations for his editorial decisions.
Another passage can illustrate the technical terms and measures at play, from text lines 140.9-13:
Ἡ δὲ χελώνη ἑμβόλου σχῆμα ἔχουςα, ὑπὸ ὁπλιτῶν φερομένη, ἀνάγεται τετρα – γώνοις ξύλοις ποδιαίοις, λεία οὖσα τὸ σχῆμα, ἢ ἐκ τῆς ἕδρας τροχοῖς <ἥλους> σιδηροῦς ἔχουσα ἵνα, ὅταν τίθηται, ἐμπηγνύηται τῇ γῇ καὶ μὴ ὑπὸ τῆς συμβολῆς ἐπισύρηται.
The tortoise, having the shape of a wedge, carried by armed men, is brought up on one-foot (-thick) (29.57 cm.) square timbers – being smooth as to its surface – or ?[on] wheels (emanating) out of the base, and it has iron [nails]? So that, when set in place, it would be pegged to the ground and would not be dragged along by the collision.
This is from the section on the wedge tortoise, which is illustrated on p. 139, and which Whitehead connects to a depiction on Trajan’s column.7 Five comments are set forth. Whitehead criticizes his own translation by noting that hoplitōn does not mean “armed” men, but rather “armoured.” Whitehead draws on Schneider in placing “leia ousa de schēma” in reduced font as “textually suspect,” perhaps because schēma is used in a different sense than earlier in the same sentence, “surface,” versus “shape.”
As to the tortoise itself, Whitehead adopts a functional view of the device, noting that it must move uphill and withstand the impact of rolling objects. This leaves open whether wheels were used in such devices, whether those wheels were left in place after it was moved in, and whether (and how) the device was anchored to the ground. Blyth maintained that there were no wheels in any of the devices in the treatise, and used this criterion as a way to identify suspect text. One hopes that Whitehead, in identifying passage 140.11-12 (concerned with wheels) as original, is not being logically circular by using his own conclusion about wheels to guide his evaluation of the text. In any event, reconstruction of the tortoise must remain conjectural, given the uncertainties in the use of nails versus wheels, for instance, or whether the device would have been wheeled to the site and then pegged to the ground or rather dragged in.
Cast 308 of Trajan’s Column, discussed in the endnote on the column, shows three of what may be such devices, clearly mounted on wheels. Whitehead here rejects claims that these devices were either deployed by the Dacians (perhaps as scythed chariots, or as aids to the building of walls), or used by the Romans as mobile ramming devices. He rather sees them as used by the Romans to deflect rolling objects, a conclusion that fits with Blyth, who called them “triangular deflectors” rather than “tortoises.” This fits with both the chapter name, “Protecting the attackers,” and the subheading of the first section of the chapter, “The problem: avoiding objects propelled from elevated positions.” (p. 77)
The passages on ladders (sections 175-188, pp. 56-65) have substantial material which is indicated as later additions, and Whitehead is surely correct to note that ladders “in their simplest form(s) will generally have been beneath the notice of military engineers intent on grander designs” (p. 120). But their production and use is far from simple, starting with the types of wood needed, and including the means by which sections twelve feet long and probably pre-assembled before the campaign could be fitted together, and used both for scaling and for bridging across open space. The two illustrations of possible ways to connect the ladder sections (p. 144, one referring to Lendle) illuminate the nature of the problem, but offer no practical resolution to which alternative is preferable.
The commentary, to sum up, is a guide to the text, not a comprehensive historical or technical examination of the devices it depicts. Such investigations are left to others, as they should be. Whitehead has done all classicists, military historians, and students of ancient engineering a service in providing our first detailed look at an important, and heretofore neglected, treatise—and in stressing Blyth’s point that it is critical to evaluate the practical use of such designs.
1. C. Wescher, Poliorcétique des Grecs: traits théoriques, récits historiques (Paris, 1867), 137-193.
2. E. Wheeler, Review in Journal of Military History 75.2 (April, 2011): 619-621.
3. D. Whitehead and H. Blyth, eds. and trs., Athenaeus Mechanicus, On Machines ( Historia – Einzelschriften 182: Stuttgart, 2004).
4. H. Blyth, “Apollodorus of Damascus and the Poliorcetica,” GRBS 33 (1992): 127-158.
5. R. Schneider, Griechische Poliorketiker mit den handschriftlichen Bildern herausgeben und übersetz, I: Apollodorus, Belagerungskunst (Berlin, 1908).
6. E. Groag, et al., Prosopographia imperii Romani saec. II (Berlin, 1933-).
7. Referencing O. Lendle, Texte und Untersuchungen zum technischen Bereich der antiken Poliorketik ( Palingenesia 19: Weisbaden, 1983): 133-136, and 184-187 for the possible depiction on the column.