Published online by Cambridge University Press: 08 May 2015
The extant Notitia Dignitatum describes itself as a list of the highest civilian and military officials of the Roman state and their administrations, or rather as two such lists, one for the parts of the empire in the East and one for the parts in the West. Each begins with a list of officials (the so-called index), followed by separate entries for these officials (the so-called chapters). These modern labels are dangerously familiarising devices which set up anachronistic preconceptions about the nature of the text and its parts. I shall refer to the two lists as the initial list and the sectional lists. The latter all have a similar structure, being divided into three parts: (i) an illustration representing the jurisdiction of the official in a primarily visual form; (ii) a list of the functions and/or other officials within his jurisdiction; (iii) a list of his bureaucratic staff. The text first surfaces in fifteenth century copies of a probably Carolingian manuscript in the Codex Spirensis.
1 Brennen, P., in Les littératures techniques dans l’antiquité romaine [Fondation Hardt, Entretiens sur l’antiquité classique 42] (1996) 147–178Google Scholar, for a more detailed analysis of the general nature and history of the text; 166 ff. for its ideological context.
2 Kelly, C., CAH 2 13 (1998) 138–183Google Scholar at 163 ff. for an excellent, recent review of the state of scholarship.
3 Bury, J.B., JRS 10 (1920) 131–154Google Scholar at 131-133; Clemente, G., La “Notitia Dignitatum” (Cagliari 1968) 366; Brennan (n.1) 150 f.Google Scholar
4 Claudian, , Epithalamium dictum Palladio et Celerinae 85Google Scholar: ‘cunctorum tabulas adsignat honorum’; ND Or. 45Google Scholar.
5 Berger, P., The Insignia of the Notitia Dignitatum (London 1981) 142 ff.; Brennan (n.l) 159 with references.Google Scholar
6 The critique of Seeck’s reconstruction of the text by Maier, I., Studies in the Textual Transmission of the Notitia Dignitatum, unpublished PhD Dissertation (University of Melbourne 1975) reveals some of the hazards of working from hypothetical archetypes.Google Scholar
7 Brennan (n.l) 159 ff. with references to examples and the modern debate on the relation between shield illustrations and military units in the text.
8 Auden, W.H., Nones (London 1952) 28.Google Scholar
9 Jones, A.H.M., The Later Roman Empire, 284-602 (Oxford 1966) 3.347.Google Scholar
10 Blockley, R.C., East Roman Foreign Policy. Formation and Conduct from Diocletian to Anastasius (Leeds 1992)Google Scholar; Braund, D., Georgia in Antiquity: a History of Colchis and Transcaucasian Iberia, 550 BC-AD 562 (Oxford 1994)Google Scholar; Zuckerman, C., ‘The early Byzantine strongholds in eastern Pontus’, Travaux et Mémoires 11 (1991)527-53.Google Scholar
11 Jones (n.9) 1.44 for similar commands in the West.
12 Cod, .Theod. 13.11.2.Google Scholar
13 Arrian, , Alan. 1-14, 18Google Scholar; see Roxan, M. in Goodburn, R. and Bartholomew, P. (ed.), Aspects of the Notitia Dignitatum (Oxford 1976) 59-79, 73Google Scholar; Spaul, J., Ala. The auxiliary cavalry units of the pre-Diocletianic imperial Roman army 2 (Andover 1994)Google Scholar; Mitford, T.B., ANRW 2.7.2: 1169-1228, 1188 ff.Google Scholar
14 Miller, K.O., Itineraria Romana (Stuttgart 1916) 676–684Google Scholar; Adontz, N., Armenia in the Period of Justinian, translated and revised by Garsoian, N. (Lisbon 1970) 62–66Google Scholar; Bryer, A. and Winfield, D., The Byzantine Monuments and Topography of the Pontus (Washington 1985) 1.25; Mitford (n.13) 1188 ff.Google Scholar
15 Bryer and Winfield (n.14) 1.325; ibid., 33 ff. for all routes east from Salala; Tab.Peut. 11.1 has a site, Aegea, on this route—see Miller (n.14) 676—but the general assumption that Aeliana is a corruption of a known site with toponymie similarities may be wrong; an unattested Aeliana is quite plausible.
16 For the unit, Spaul (n. 13) 187; Speidel, M., Roman Army Studies (Amsterdam 1984) 1.288Google Scholar shows that the identification with the Ιταλοί in Arrian’s army is unlikely and Roxan (n.13) omits the Notitia unit from her list of pre-Severan auxilia.
17 ND Or. 28.20-21; 34.37; 35.34.
18 Roxan (n.13) 74-75.
19 Zuckerman (n. 10) 530.
20 Procop, . Bell. 8.4.4–5Google Scholar; Aed. 3.7.8.Google Scholar
21 See Böcking, E., Nolitia Dignitatum et Administrationum omnium tam civilium quam militarium in Partibus Orientis et Occidentis (Bonn 1839) 1.438-9Google Scholar for these alternative identifications in earlier scholarship.
22 Berchem, D. van, L’armée de Dioclétien et la réforme constantinienne (Paris 1952)31–32Google Scholar, who also rejects Colchian Pityus for similar reasons; his localisation would still fit the Notitia pattern, but both localisations are effectively rejected by Zuckerman (n.10) 531 n.16.
23 Itineraria Antonini Augusti (hereafter IAA) 205, 214Google Scholar in Cuntz, O., Itineraria Romana (Leipzig 1929) vol. 1; Zuckerman (n.10) 531-3Google Scholar (and 538 for recognition of Notitia limitations).
24 Nyss, Greg.. V.Macrinae 36.Google Scholar
25 Jones, A.H.M., The Cities of the Eastern Roman Provinces 2 (Oxford 1971) 182Google Scholar with n.17; 539 (appendix IV); the changing use and boundaries of Pontus as a provincial and regional name in the Roman period are not easy to unravel—the Colchian town was always in one of the Pontus provinces, the Cappadocian one sometimes.
26 Chrys, Joh.. Epp. 13;14.Google Scholar
27 Zos. 2.33.1; see D. Braund (n.10) 265, who notes the weakness of the argument from silence and maximises the counter argument from vestigial traces.
28 Speidel, M., Roman Army Studies (Stuttgart 1992) 2.209–211Google Scholar for a praetensio which we happen to know about in Arabia; Colchian sites might have been similarly manned from the legion at Trapezus.
29 Reddé, M., Mare Nostrum (Rome 1986) 642Google Scholar f.; the Notitia records no eastern fleets, which is not to say that none existed.
30 E.g. the units named after Abasgi, , Hiberi, , Tzanni, , and perhaps Suani and Sannigae, if the obscure Suentium and Saginarum of the Notitia entries conceal confused references to the Colchian tribes with similar names (ND Or. 31.46, 55, 62, 65, 66)Google Scholar. Procop, . Bell. 8.4.3Google Scholar also uses Saginae for Sannigae.
31 Anon.Val. 35.
32 Zuckerman (n.10) 529.
33 Zuckerman (n. 10) 534; Bryer and Winfield (n.14) 326; both support the Lazic site.
34 Procop, . Bell. 2.29.18; 8.14.45–46Google Scholar. Bryer and Winfield (n.14) 304 supports the Lazic site; Zuckerman (n.10) 530-531 argues against it, noting errors in Bryer and Winfield and van Berchem (n.22) 32, but the case for the Lazic site remains viable despite their errors and his own methodologically flawed argument that such a site would be too isolated. We cannot know that the coastal Phasis site had no military force (note Zos. 2.33.1)—it could even be the unlocated Valentia. His own case for the site near Trapezus is subject to the same reservations he used to reject the nearby Zigana as an inhospitable permanent garrison site. Braund (n.10) 265-266, though implicitly rejecting the Lazic site, offers no alternative.
35 For Lazica: van Berchem (n.22) 32; Bryer and Winfield (n.14) 304; 326. For near Trapezus; Adontz (n.14) 81; Dodgeon, M. and Lieu, S., The Roman Frontier and the Persian Wars AD 226–363Google Scholar. A Documentary History (London 1991) 348Google Scholar, though confusing this Mochora with Mogaro(IAA 205.1)Google Scholar, far to the west, probably in Galatia. Zuckerman (n.10) 531-532, 534-535 alone sets the sites in different regions, placing one in Lazica, one near Trapezus.
36 Winfield and Bryer (n.14) 326; Zuckerman (n.10) 534; the same arguments they use against siting Ziganne here apply also to Mochora.
37 As postulated by Zuckerman (n.10) 535-536, but his belief that Mochora was part of a different Diocletianic disposition made it unavailable to him for his Valentian disposition; see also Zos. 2.33.1.
38 Procop, . Bell. 8.4.4–5Google Scholar; Aed. 3.7.8Google Scholar; Theodoret, HE 5.34Google Scholar; Jerome, , Ep. 60.16Google Scholar for another exile there in 396. Zuckerman (n.10) 533-534 makes an excellent case for this site and against alternatives.
39 Van Berchem (n.22) 31; earlier, Böcking (n.21) 434.
40 IAA 217; Adontz (n.14) 82. 41
41 As in Dodgeon and Lieu (n.35) 400, wrongly turning the IAA site into the far-off Black Sea port.
42 Although several cities in the west earlier were called Valentia, the three known eastern ones are likely to be associated with the emperor; see RE 2.7A.2148 ff.; for Valentia in Phrygia and Nea Valentia in Osrhoene, Jones (n.25) 72, 222.
43 Braund (n.10) 265; Bryer and Winfield (n.14) 346; both cite (different) Russian works unavailable to me. Braund, rightly, minimises the value for military sites of the Tale of St Orientius and his Brothers.
44 Zuckerman (n.10) 534-535. 45
45 Zosimus 2.33.1.
46 Miller (n.14) 675; Procop, . Aed. 3.4.10Google Scholar (Lutararizon).
47 For the other four; ND. Or. 38.20–22; 31.41. 48Google Scholar
48 Jones, (n.9) 3.360 on ND Or. 31.41 & 55Google Scholar, but this unit of Abasgi is not Theodosian, as Jones thought; it was already in the Thebaid in the early fourth century—Thomas, J.D., Yale Classical Studies 28 (1985) 115-25Google Scholar, 115 ff. The placing of this unit on both registers remains a testament to transience and muddle in bureaucratic processes. It thus seems coincidental that the three alae without dynastic names on the laterculum maius (ala Rizena, ala I Abasgorum, ala II Armeniorum) are all associated with places or peoples in the Pontic/Armenian region.
49 Cod, .Theod. 1.8.1–3Google Scholar; see Jones (n.9) 3.641.
50 See Zuckerman (n.10) 529 n.10 for an excellent analysis of these entries and identifications.
51 Adontz (n.14) 408 n.15. 52
52 Hoffmann, D., Das spätrömische Bewegungsheer und die Notitia Dignitatum (Düsseldorf 1969) 1.420Google Scholar ff.; Jones (n.9) 2.609.