Bryn Mawr Classical Review

The purpose of Zahariade's work is to collect and analyse the evidence for Thracians serving in the auxilia of the Roman army during the first to third centuries A.D. Zahariade succeeds admirably in the first of these aims, presenting the reader with extensive catalogues of soldiers who served in the alae and cohortes Thracum, ranging from the equestrian commanders to the equites and milites themselves. However, the book does not always succeed in exploiting this material to its full advantage and some of the conclusions are problematic.

Chapters one ('The Background: Thrace and the Thracians') and two ('Rome and Thrace. A Political and Military History of the Roman conquest of Thrace') are largely preliminary to the main body of the work, as Zahariade discusses the origins and disposition of the Thracian tribes and their incorporation into the Roman empire. In chapter three, 'The Recruitment and the Setting of alae and cohortes Thracum', the author examines the employment of Thracians in the Roman auxilia, beginning with the six units styled Augusta Thracum (three alae, three cohortes), raised in the first century A.D. Using military diplomas and other epigraphic evidence, Zahariade proceeds to analyse the establishment of other units bearing the epithet Thracum, emphasising the extent to which the Roman empire relied on manpower from the Thracian tribes.

In the next four chapters, four to seven, Zahariade examines the history of the twelve alae and twenty-six cohortes Thracum. Chapter four, 'Auxilia Thracum: Identification, Structure, and Number' provides a brief overview of the units, which are then individually analysed in chapters five ('Alae Thracum') and six ('Cohortes Thracum'). Each unit is assigned its own number, making it easy to follow cross-references throughout the text. Chapter seven, 'The Dynamics of the Thracian Regiments', contains a particularly useful table charting the movement of the auxiliary units over the course of the first three centuries A.D. There follows two largely descriptive chapters, eight ('Auxilia Thracum at War') and nine ('Auxilia Thracum at Work'), which assemble the epigraphic evidence for the involvement of these units in military campaigns and in building activities. A brief conclusion summarises the major themes of the work.

The book contains full appendices which will be useful to scholars working on the history of the Romanauxilia, although they often repeat the same information in different formats. Appendix I contains lists of the men who served in the auxilia Thracum, from the equestrian commanders to the rank and file soldiers, with those of specifically Thracian origin marked with an asterisk. Zahariade also collects the evidence for Thracians attested in other auxiliary units. Appendix II.1 lists all the Thracians who served in the auxilia, separating them into groups based on their origo. In Appendix II.2, Zahariade collects and reproduces, sometimes in abbreviated format, all the inscriptions for Thracian auxiliaries. These lists are divided by unit for the auxilia Thracum, but not for the other regiments. The work concludes with forty pages of maps and illustrations of inscriptions (pp. 359-89).

Zahariade has collected and synthesised a vast amount of evidence, mainly epigraphic, which will be useful to scholars working on a variety of aspects on the Roman army, from recruitment patterns and troop movements to concepts of ethnicity and identity. However, there is some inconsistency in Zahariade's arguments. It is well established that from the Flavian period onwards, the majority of auxiliary units gradually lost their original ethnic make-up, as they were replenished by recruits from other areas of the empire, particularly the regions in which they were stationed.¹ Thracians constituted one of the largest ethnic groups in the auxilia, but they were dispersed throughout the empire, rather than being restricted to units which bore the epithet Thracum.² While Zahariade is clearly aware of this (see his comments on pp. 69, 119), he sometimes assumes that Thracians were specifically recruited to replenish the auxilia Thracum. This conclusion is often based on insufficient evidence. Regarding the ala II Augusta Thracum, stationed in Mauretania Caesariensis from the late first century A.D., he writes that "Thracians continue ... to supply the regiment in Africa, in the 2nd century", although this verdict is supported by only one fragmentary inscription (p. 61). On p. 311, the author states that the presence of one third-century Thracian veteran in the ala I Thracum "indicates a constant supply of the unit with recruits from the Thracian stock late in the 3rd century", a claim which extends far beyond the evidence. Other explanations, such as the transfer of troops between units, are equally plausible.

Some of the statistical conclusions are susceptible to similar criticisms. For example, the table on p. 290 shows 18 Thracians attested as equites in the alae Thracum, out of a total of 80 recorded soldiers in these units. The equivalent figure for Thracians as equites in the cohortes Thracum is 12 out of 24 examples. Zahariade states that that the percentage of "equites of Thracian origin in Thracian regiments reaches 40.21% from the total of 92 subjects". This cannot be true: his own figures show 30 Thracians out of 104 examples, which is 28.85 per cent, although it is misleading to use such precise percentages given the deficiencies of the evidence.³ Nor does it justify the conclusion that there was a "constant process" by which the administration specifically directed Thracian recruits to the auxilia Thracum during the first to third centuries A.D., as Zahariade claims (p. 290). Throughout the work, Zahariade repeatedly cites an inscription recording the dispatch of 1,000 Thracian troops to Mauretania Tingitana, probably dated to the Severan period (CIL VIII 9381 = ILS 2763, mentioned on pp. 66, 72, 77, 94, 116, 292, 312).⁴ Since no auxilia Thracum were stationed in this province,⁵ the inscription eloquently demonstrates that Thracians were sent wherever they were needed in the empire.

There is consequently a problematic thread which runs throughout the work, namely that auxilia Thracum is not always synonymous with 'Thracians in the auxilia'. The issue is particularly glaring in chapters eight and nine, which chronicle the employment of the auxilia Thracum in warfare and in building works, despite the fact that in the second and third centuries A.D. Thracians were not the largest ethnic group in these regiments. Zahariade collects the inscriptional evidence for the involvement of these units in military campaigns or for their participation in the construction of roads, bridges, fortifications, bath-houses and temples. However, there is nothing specifically Thracian about the military expeditions or the construction undertaken, apart from the names of the regiments themselves. It will hardly come as a surprise to scholars that auxiliary troops were involved in building works.

There are other issues which would have benefited from more searching analysis. The discussion of Thracian communities in the provinces (pp. 90-5) goes little further than stating that there were such groups. Descriptions of ethnicity are often unhelpful: the soldier Aurelius Tatius is described as having a name "of Illyrian but also Thracian tinge" (p. 101).

Rather than assuming that certain auxilia Thracum remained Thracian based on the somewhat limited evidence, it would have been better to use the epigraphic material to track changes in the ethnic composition of these units over the course of the first to third centuries (the material for such an examination is available, since Zahariade lists non-Thracians in the auxilia and cohortes Thracum on pp. 261-304). The treatment of theoretical concepts such as 'Romanization' is cursory, showing little awareness of current debates in this area (p. 91).⁶

A random check of references shows some minor slips, but these are to be expected in a work which contains so much epigraphic material.⁷ The standard of English expression is not always adequate, and sometimes obscures the author's meaning.⁸ The quality of binding was disappointing, as the front cover detached from the spine on opening the work, and the back cover during the first reading.

In summary, this is a very good collection of evidence for the presence of Thracians in the Roman auxilia and a history of the units bearing the epithet Thracum. However, readers will need to carefully examine the appendices to determine whether all the interpretations are supported by the evidence available.

1.   P. A. Holder, Studies in the Auxilia of the Roman Army from Augustus to Trajan, BAR International Series 70 (Oxford, 1980), 118.
2.   Holder (n. 1), 123; M. P. Speidel, 'A Thousand Thracian Recruits for Mauretania Tingitana', Antiquités Africanes 11 (1977), 167-72, at 168; and M. M. Roxan, Roman Military Diplomas 1985-199 3 (London, 1994), 255.
3.   There is also an error in column two of the table, labelled 'Thracians in alae non-Thracum' which gives the total number of subjects as 76 and the number of Thracians as 67. Both these figures should be 67.
4.   The author repeatedly dates the inscription to A.D. 200, but it cannot be more closely dated than the early third century A.D. See Speidel (n. 2).
5.   M. M. Roxan, 'The auxilia of Mauretania Tingitana', Latomus 32 (1973), 838-55.
6.   See for example: D. J. Mattingly (ed.), Dialogues in Roman Imperialism: Power, Discourse and Discrepant Experience in the Roman Empire, JRA Supplementary Series 23 (Portsmouth, RI, 1997) and J. Webster, 'Creolizing the Roman Provinces', AJA 105 (2001), 209-25. Note also A. Wallace-Hadrill, Rome's Cultural Revolution (Cambridge, 2008), although this was probably published too late to be included.
7.   Vitellius' general who attacked the Helvetii was Caecina Alienus, not Caecina Severus (p. 172); AE 1933, 209 (Au[---]) should read 1933, 213 (p. 261); ILS 110 (Catonius Vindex) is actually 1107 (p. 262); CIL III 4857 (Ulpius Moderatus) should be 4851 (p. 274); Sex. Iulius Possessor should read Sex. Iulius Iulianus (p. 312). Other minor quibbles: the title defensor provinciae suae recorded on CIL VIII 9045 (= ILS 2766) surely does not refer to an official post (p. 191); it is doubtful whether the testimony of the Historia Augusta (Sev. Alex. 58.1) constitutes sufficient evidence for a war on the Danube in A.D. 228 (p. 192).
8.   For example, "pedestrians" is used to mean infantry (p. 96), likewise "setting" for establishment (p. 213). The definite article is frequently used with names, "the Niger" (p. 138), "the Trajan" (p. 165). Some sentences are very difficult to understand, such as the one beginning "The eleven Thracian soldiers..." on p. 103. There are some typographical errors, such as "Maggie" for the scholar D. Magie (p. 86) and "Judeea" for Judaea (p. 172).