Hostname: page-component-586b7cd67f-t7czq Total loading time: 0 Render date: 2024-11-23T22:44:07.153Z Has data issue: false hasContentIssue false

Fortifications in the North-Western Roman Empire from the Third to the Fifth Centuries A.D.

Published online by Cambridge University Press:  24 September 2012

Harald von Petrikovits
Affiliation:
Rheinisches Landesmuseum, Bonn

Extract

Roman methods of fortification in the north-western portions of the Empire change significantly during the second half of the third century, the difference from the Principate being more apparent in military building than in civil. We may accept the universal view that these changes were due to increasing insecurity in Dacia, on the Rhine and Danube frontiers, and along the coast. From its beginnings in the first half of the century, the threat to the north-west by Germans and tribes from the steppes reached such a pitch in and after the 250's that it seriously endangered Roman rule in Europe. The Goths broke through the Danube frontier into Moesia several times from 238 onwards, and Roman morale was gravely affected when they killed Decius and his son in the disastrous battle of Abrittus (251). The northern barbarians fell upon towns in Greece and Asia Minor, and plundered them; only in the years following 268 did some emperors succeed in mastering the danger.

Type
Research Article
Copyright
Copyright ©Harald von Petrikovits 1971. Exclusive Licence to Publish: The Society for the Promotion of Roman Studies

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)

References

1 Basic: Unverzagt, W., Terra sigillata mit Rädchenverzierung (Frankfurt/M. 1919)Google Scholar; Chenet, G., La Céramique gallo-romaine d"Argonne du 4e siècle, etc. (Mâcon 1941)Google Scholar. On questions of dating, see Hübener, W., BJb. 168, 1968, 241 ff.Google Scholar and the literature there cited; idem in; Jahresber. Gesellschaft pro Vindonissa 1968 (Brugg 1969), 7 ff.Google Scholar

2 There is no need to cite the literature on British late Roman pottery to readers of the JRS. The literature on Pannonian late Roman pottery is to be found in Mócsy, Pannonia 681.

3 Examples of period 1: Loeschcke, S., BJb 127, 1922, 320 ff.Google Scholar, Taf. 10; idem, Trierer Jahresber., N. F. 13, 1923, 103 ff., Taf. 11 f.; Petrikovits, H. v., BJb. 142, 1937, 325 ff.Google Scholar (1st stratum); the period has close ties with the Niederbieber horizon.

Period 2: jars and jugs with continuous red stripes, and the associated pottery. Characteristics: special forms of ‘heart-shaped’ lip profile. See Petrikovits, H. v., ‘Landschaft und Geschichte’, Festschrift f. F. Petri (Bonn 1970), 401Google Scholar, n. 67; Binsfeld, W., Kölner Jahrb. 6, 1962/1963, 93 ff.Google Scholar

Period 3: finds from Cologne: Binsfeld, W., Kölner Jahrb. 5, 1960/1961, 73 ff.Google Scholar; idem, Kölner Jahrb. 6, 1962/63, 89 ff. Other finds from St. Gereon in Cologne (unpublished) and from Boppard (unpublished). Characteristics: a combination of ‘heart’ - and ‘crescent-shaped’ lip profiles.

Period 4: most of the types found at Alzey, the rest may go back to period 3: Unverzagt, W., Die Keramik des Kastells Alzei (Frankfurt 1916)Google Scholar; idem, Germania 13, 1929, 177; idem, Ber. RGK 49, 1968, 74 ff.; H. v. Petrikovits s.v. Period 1 (stratum 2 f); Stehlin—von Gonzenbach, 119 ff. Cf. H. Cüppers in H. Hinz, Kreis Bergheim (= Archäologische Funde und Denkmäler des Rheinlandes 2, Düsseldorf 1969) 120, n. 291. Characteristics: ‘crescent-shaped’ lip profile.

Period 5: Loeschcke, S., BJb. 126, 1921, 56 ff.Google Scholar Taf. 4; Hussong, L., Trierer Zeitschr. II, 1936, 76 ff.Google Scholar Represented mainly by the pottery from the Runder Berg near Urach (unpublished). Characteristics: degraded types of period 4.

4 Especially in the published material from Schaan (8, 82), Moosberg (1, 86), Lorenzberg (2, 87) and Auf Krüppel (1, 83).

5 Typology of the ‘cross-bow’ brooch: literature in Heurgon, J., Le trésor de Ténès (Paris 1958), 22 ff.Google Scholar; Garbsch, Moosberg 65; Keller, E., Die spätrömischen Grabfunde in Südbayern (= Münchner Beiträge zur Vor- und Frühgeschichte 14, München 1971)Google Scholar. On chip-carved belt-fittings, see Bullinger, H., Spätantike Gürtelbeschläge (= Diss. Gandenses 12, Brugge 1969)Google Scholar.

6 The work done on late Roman fortifications varies from country to country. None has been adequately investigated in the Spanish diocese. See Richmond, I. A., JRS 21, 1931, 86 ff.Google Scholar; Balil, A., in Legio VII gemina (León 1970), 608 ff.Google Scholar

6a Forts are accordingly distinguished as frontierforts and road-forts. Smaller forts holding up to c. 30 men are described as fortlets.

7 Hist. Aug., Tyr. Trig. 5,4.

8 Kellner, H.-J., in Limes-Studien (Basel 1959), 56 f.Google Scholar; Sch. p. 179; Garbsch, Donau-Iller-Rhein-Limes 7.

9 Wagner, F., Ber. RGK 37/38, 1956/1957, 224Google Scholar No. 30; Kellner, H.-J., in Limes-Studien (Basel 1959), 56Google Scholar; Sch. p. 179.

10 Zosim. 2, 34, 1; Paneg. Lat. 9, 18, 4.

11 The principal argument for Diocletianic date, the presence of brick-stamps of Legio I Martia, no longer holds good: see below, p. 185. We can only argue from the likelihood that Kaiseraugst replaced Colonia Augusta Raurica, and from the termini ad quem of the church, the mid-fourth-century silver hoard, and CIL 13, 5270.

12 H. Nesselhauf, Abh. Preuss. Akad. Wiss. 1938, Ph.-hist. Kl. 2, 51 ff. Oudenburg: Sch. 15.

13 Mócsy, Pannonia 636, and below, p. 184.

14 For late Roman brick-stamps of the 8th and 22nd Legions: CIL 13/6, p. 23 type 89 and p. 56, cap. 8. Baatz, D., Mogontiacum (= Limesforschungen 4, Berlin 1962), 52Google Scholar, No. 27 and p. 79. Hoffmann 2, 147 f., n. 289. Further examples are cited under: 5, 46; 6, 45, 56, 58, 63.

15 Sági, K., Acta Arch. Acad. Scient. Hung. I, 1951, 87 ff.Google Scholar; Szilágyi, J., in Intercisa I (Budapest 1954), 47 ff.Google Scholar, Mócsy (see n. 13). The late Roman fort of Piro torto-Zwentendorf in Noricum Ripense also has fan-shaped angle towers: Hampl, F. und Stiglitz, H., Die Ausgrabungen in Zwentendorf (Vienna 1961), 4 f.Google Scholar; plan in Vetters, H., Gymnasium 76, 1969Google Scholar, Taf. 13, 1 at p. 495. Professor S. S. Frere drew my attention to a possible fan-shaped tower at Causennae-Ancaster, Lines.: JRS 55, 1965, 205Google Scholar and fig. 12.

16 Petrikovits, H. v., BJb. 161, 1961, 475 ff.Google Scholar

17 Sch. p. 182; Garbsch, J., Bayer. Vorgeschichtsbl. 32, 1967, 73 ff.Google Scholar

18 Mócsy, A., Folia Archaeol. 10, 1958, 99 ff.Google Scholar; idem, Pannonia 629 and 631 ff. Bricks of Frigeridus dux, Legio X Gemina, and others, were found in the burgus of Visegrád (7, 98) which is dated by an inscription to 372: Soproni, S., in Limes Romanus, Konferenz Nitra (Bratislava 1959), 140Google Scholar.

19 For military brick-stamps in Noricum and Pannonia during Valentinian's reign, see: Alföldi, A., Der Untergang der Römerherrschaft in Pannonien 1 (Berlin-Leipzig 1924), 85Google Scholar; Szilágyi, J., Inscriptiones tegularum Pannonicarum (= Diss. Pann. 2/1, Budapest 1933), 94 ffGoogle Scholar; R. Egger, Anz. Österr. Akad. Wiss., ph.-h.KL 1954, 101 ff. = idem, Römische Antike und frühes Christentum 2 (Klagenfurt 1963), 180 ff.; Soproni, S., Arch. Ért. 85, 1958, 52 ff.Google Scholar; Mócsy, Pannonia 631 f. For the dating of Legio I Martia brick-stamps, see Staehelin, Schweiz 279; Hoffmann I, 348. The brick-stamps cannot be used in dating Argentovaria-Horbourg, Argentorate-Strasbourg, Epamanduodurum–Mandeure, Altenburg near Bragg (2, 76) and the landing-place of Wyhlen opposite Kaiseraugst.

20 Mócsy, Pannonia 642 f.

21 Bogaers, J. E., Ber. ROB 18, 1968, 156 f.Google Scholar

22 Roblin, M., RÉA 67, 1965, 368 ff.Google Scholar For eastern examples, see MacMullen, R., Soldier and Civilian in the Later Roman Empire (Cambridge, Mass. 1963), 77 ff.Google Scholar

23 cf. Taberna, which came under the Dux Mogontiacensis (ND Occ. 41, 4 and 16). See Hoffmann 2, 146, n. 277. Also Foetibus and Teriolis, which came under the Dux Raetiae (ND Occ. 35; 10, 11, 21, 22 and 31). The emendation ‘Fano Martis’ for ‘Marcis’ (Occ. 38, 7) is uncertain. For the arms-factories controlled by the magister officiorum, see ND Occ. 9, 16 ff.

24 Late Roman limitanei: see now Hoffmann, passim, whose index of regiments (2, 271 ff.) satisfies a long-felt need. Size of fortifications: see J. Garbsch, Donau–Iller–Rhein–Limes, 14 f. For calculation of the size of the Bürgle's (5, 91) garrison from its living-quarters, see Bersu, G., Die spätrömische Befestigung ‘Bürgle’ bei Gundremmingen (Munich 1964), 46 ff.Google Scholar

25 General accounts: Blanchet, A., Les enceintes romaines de la Gaule (Paris 1907)Google Scholar; Grenier, , Manuel I, 403 ff.Google Scholar; Lot, F., Recherches sur la population et la superficie des cités remontant à la période gallo-romaine, 3 vols. (Paris 19461953)Google Scholar (incomplete); Butler, R. M., The Arch, Journ. 116, 1959, 25 ff.Google Scholar For individual areas, see the various local surveys: Richmond, I. A., JRS 21, 1931, 86 ff.Google Scholar; Frere, Britannia; Vercauteren, F., Étude sur les civitates de la Belgique seconde (Brüssel 1934)Google Scholar; Byvanck, A. W., Nederland in den romeinschen Tijd, 2 vols. (Leiden 1943)Google Scholar; V. Petrikovits, RR; Staehelin, Schweiz. For Aurelian's walls at Rome, see Richmond, I. A., The City Wall of Imperial Rome (Oxford 1930)Google Scholar.

26 Egger, R., Teurnia5 (Klagenfurt 1963), 27Google Scholar.

27 Marasović, J. and T., Der Diokletianspalast (Zagreb 1968)Google Scholar. For the Landmauer near Trier, see Steinhausen, J., Trierer Zeitschr. 6, 1931, 41 ff.Google Scholar; idem, Archäologische Siedlungskunde des Trierer Landes (Trier 1936), index s.v.; Wightman, E. M., Roman Trier and the Treveri (London 1970), 170 f.Google Scholar; Hoffmann 2, 152 f., n. 332; Wild, J. P., Textile Manufacture in the Northern Roman Provinces (Cambridge 1970), 9Google Scholar.

28 H. Cüppers, in Kempf, Th. K. and Reusch, W. (edd.), Frühchristliche Zeugnisse im Einzugsgebiet von Rhein und Mosel (Trier 1965), 152 ff.Google Scholar; Petrikovits, H. v., BJb, 169, 1969, 579 f.Google Scholar

29 Dyggve, E. and Vetters, H., Mogorjelo (Vienna etc. 1966)Google Scholar.

30 For a general account: Schleiermacher, W., Ber. RGK 33, 19431950, 176 f.Google Scholar

There are many examples of late Roman hilltop refuges.

Belgium: Mertens, J., Pays gaumais 15, 1954Google Scholar, Nos. 1-2; idem, Annales de l'Inst. Archéol. du Luxembourg 92, 1961, 73 ff.; idem, Archaeologia Belgica 63, 1962 and 76, 1964.

Germany: Uslar, R. v., BJb. 153, 1953, 136 and 138 f.Google Scholar; Schindler, R., Studien zum vorgeschichtlichen Siedlungs- und Befestigungswesen des Saarlandes (Trier 1968), 159 f.Google Scholar; Sprater, F., Die Pfalz unter den Römern I (Speyer 1929), 58 ff.Google Scholar; Schleiermacher, W., BJb. 162, 1962, 173Google Scholar; Eckrich, L. and Kaiser, Kw., Mitt. Hist. Ver. d. Pfalz 68, 1970, 101 f.Google Scholar No. 328; Moosberg (1, 86).

France: Sch. 44.

Switzerland and Liechtenstein: Wittnauer Horn (1, 69), Auf Krüppel (1, 83).

Austria: R. Egger (see n. 44). Dolenz, H. and Görlich, W., Carinthia I 125, 1935, 133 ff.Google Scholar; Jantsch, F., Mitt. d. Anthropolog. Ges. Wien 68, 1938, 337 f.Google Scholar, ibid. 73–77, 1947, 168 ff.; A. Hild, Jahrb. Vorarlberger Museumsverein 1941, 5 ff. and 11 ff.; Kohla, F. X., Carinthia I 132, 1942, 67 ff.Google Scholar; Pohl, G. & Stiglitz, Pro Austria Romana 17, 1967, 14 ff.Google Scholar; Vetters, H., Gymnasium 76, 1969, 500 ff.Google Scholar Further literature cited by Petrikovits, H. v., Trierer Zeitschrift 19, 1950, 81Google Scholar, n. 21 and Saria, B., Historia I, 1950, 484 f.Google Scholar

Hungary: Mócsy, Pannonia 637 ff.

Yugoslavia: B. Saria, Ant. Inschr. Jugosl. p. 3, 15 ff., 104 and 109; idem, Carinthia I 132, 1942, 102 ff.; Klemenc, J., Ptujski grad v kasni antiki (Ljubljana 1950)Google Scholar.

Spain: Balil, A., in Legio VII Gemina (León 1970), 611Google Scholar.

31 Miltner, F., ÖJh. 40, 1953Google Scholar, Bbl. 81 ff. and 41, 1954, Bbl. 82 ff.; Vetters, H., Anz. Österr. Akad. Wiss., Ph.-h.Kl. 106, 1969, 75 ff.Google Scholar

32 Tacitus, , Hist. 4, 23, 3Google Scholar.

33 Auxiliary forts were sometimes reconstructed in the late Roman period to meet new military requirements, as, for example, Remagen (1, 40) in Germania II; Schlögen in Noricum Ripense (Eckhart, L., Das römische Donaukastell Schlogen, etc. ( = RLiÖ 25, 1969), 53 f.Google Scholar); the forts on Hadrian's Wall; and Brough-on-Humber (Wacher, J. S., Excavations at Brough-on-Humber (Leeds 1969), 34 ff.Google Scholar).

34 Garbsch, J., Bayer. Vorgeschichtsbl. 32, 1967, 62 ff.Google Scholar; idem, Donau-Iller-Rhein-Limes 15 and fig. 22 f. For centenaria in Raetia, see Schleiermacher, W., Aus Bayerns Frühzeit (Fr. Wagner-Festschrift, ed. Werner, J., Munich 1962), 195 ff.Google Scholar

35 Examples: Hornsby, W. and Laverick, J. D., Arch. Journ. 89, 1932, 203 ff.Google Scholar; Barb, A. A., ÖJh 37, 1948, Bbl. 263 ff.Google Scholar; Stehlin-von Gonzenbach; Garbsch, J., Bayer. Vorgeschichtsbl. 32, 1967, 51 ff.Google Scholar; Mócsy, Pannonia 639. If only its ground-plan survived, the Heidentor at Carnuntum would have been interpreted as a burgus with central pillar; but the vaulting of its ground-storey rules out this interpretation. Swoboda, E., Carnuntum4 (Graz-Cologne 1964), 171 ff.Google Scholar and 289. On the development of road burgi from beneficiarii posts, see v. Petrikovits, RR 75 f. Baisweil (1, 88). Binding, G., Rheinische Ausgrabungen 3 (Düsseldorf 1968), 121 ff.Google Scholar Watch-towers on the frontier, and presumably on roads in the hinterland too, had their names: summa rapida (‘highest rapids’) (7, 73) and commercium (7, 94).

36 As H. Eiden (Koblenz) kindly informs me. The strength of late Roman walls: J. Garbsch, Moosberg 58.

37 Petrikovits, H. v., BJb. 161, 1961, 477Google Scholar, n. 43. Half-projecting rectangular towers had a long history in Roman town walls. Lissus is an example: Wilkes, Dalmatia 363.

38 Mócsy, Pannonia 637.

39 Mócsy, Pannonia 637. On fan-shaped angle-towers, see n. 15 above.

40 Amphitheatres incorporated into enceintes: examples at Tours, Périgueux, Amiens, Trier and Salonae. J. Heurgon, Bull. Soc. Nat. Ant. de France 1952/53, 149 ff. Wilkes, Dalmatia 360 and fig. 16. The amphitheatre at Trier was the inhabitants' last refuge: Fredegar, Chron. 2, 60 (ann. 407). Even the legionary fortress of Aquincum had its amphitheatre reconstructed as a stronghold: Szilágyi, J., in Limes-Studien (Basel 1959), 170Google Scholar. Similarly the theatres at Madaurus in the sixth century and at Miletus in c. eighth century: Müller-Wiener, W., Istanbuler Mitt. 17, 1967, 279 ff.Google Scholar

41 Goodchild, R. G. and Perkins, J. B. Ward, JRS 39, 1949, 90Google Scholar.

42 Neynaber, A., Die Wehrbauten des Irak (Berlin 1920), 49 ff.Google Scholar

43 Stein, G., Saalburg-Jahrb. 19, 1961, 8 ff.Google Scholar

44 Duel near Paternion-Feistritz, on the northern side of the upper Drave valley, Carinthia, was excavated by R. Egger and G. Bersu from 1928 to 1931. Egger, R., ÖJh 25, 1929, Bbl. 159 ff.Google Scholar; Vetters, H., Gymnasium 76, 1969, 505Google Scholar.

45 e.g. the forts built by the magister militum Solomon at Thamugadi-Timgad (Numidia) and probably Ammaedara-Haïdra (Byzacena). Diehl, Ch., L'Afrique Byzantine (Paris 1896)Google Scholar, index s.v. Ammaedara and Thamugadi, plans 1, 195 and 200 (without internal buildings), and Haïdra restored, p. 164. Cagnat, R., Carthage, Timgad, Tébessa3 (Paris 1927), 140Google Scholar (a reconstruction of Haïdra). Romanelli, P., ‘La riconquista Africana di Giustiniano’ in Africa Romana, Scritti di A. G. Amatucci, F. Arnaldi, etc. (Milan, 1935), 123 ff.Google Scholar, giving a new plan of Haïdra without the via sagularis; idem, Topografia e archeologia dell'Africa Romana ( = Enciclopedia classica 3/10/7, Turin 1970), 398 ff., esp. 405 and tav. 354a.