The Roman Frontier with Persia in North-Eastern Mesopotamia : Fortresses and Roads Around Singara

The Roman frontier with Persia in north-eastern Mesopotamia investigates the Roman city of Singara and the fortifications and roads in the surrounding area. The physical frontier between Rome and Persia has been little studied, in part because of the difficulty of access for scholars. In comparison with other parts of the Roman ‘limes’, this frontier was of great importance because it separated the two major civilisations of the early first millennium CE. Although the frontier stretched north to Armenia and the Black Sea, north-east Mesopotamia was for long periods the major area of confrontation. After a brief review of the history of north-east Mesopotamia and its role as the setting for repeated clashes between the two empires, the book focuses on Singara, its fortifications and the surrounding frontier zone. This town was one of the strongpoints on the Roman frontier as it existed up to 363 CE. The volume then addresses the ancient road network around Singara and the links to Nisibis and to the Khabur valley to the west. It makes use of old aerial photographs and satellite imagery to illustrate fortifications, roads and associated sites, in particular those mentioned in the Peutinger Table. A final chapter addresses the nature of the frontier in this region. Anthony Comfort is an independent researcher and an associate member of the Centre for the study of Greek and Roman antiquities at Corpus Christi College, Oxford. He has an MA in archaeology from the University of Leicester and a PhD in archaeology and ancient history from the University of Exeter. Previously he worked as an official of the European Parliament’s secretariat, principally in the Research Department. He is a specialist in the use of satellite imagery for archaeology in the Middle East but has also conducted research on ancient roads in Aquitaine and in Romania. With Michal Marcia he is co-author of How did the Persian king of kings get his wine? The upper Tigris in antiquity (Archaeopress 2018). Cover 1 Title Page 3 Copyright Page 4 Contents Page 7 List of Figures and Tables 8 Map 1. Setting, showing area of concern 14 Map 2. Fortresses and roads 20 Figure 1. Copied from Oates, 1968 Fig 8. Also in this publication are plans of the Roman south gate and Bastions 5, 6 and 7 33 Figure 2. Singara Corona 11 Dec 1967 34 Figure 3. Singara Google Earth 13 March 2021. Rampart of Roman town indicated in red 35 Figures 3a and b. Ain Sinu/Zagurae. a) Plan copied from Oates 1968 p83. b) Corona 11 December 1967.36°27'07"N, 42°10'43"E 37 Figure 4. Majnuniye. Corona 11 Dec 1967. with Tel Hayal circled in yellow 38 Figure 5. Sikeniye. Corona 11 Dec 1967. Abandoned town circled in yellow 40 Figure 6. Fort at foot of Bara pass. CAST and Google Earth 5 Sept 2020 41 Figure 7. Poidebard’s photograph of ‘Thebeta’. La Trace de Rome dans le désert’, Pl.CLX.3 43 Figure 8. Thebeta, according to Poidebard. 36°49’25’’N, 41°41’43’’E. Corona 1967 44 Figure 9. Qohbol/Gohbal. Corona 11 Dec 1967. 36°30’52”N, 41°57’24”E 45 Figure 10. Al- Han – situation to the east. Corona 5 Nov 1968 48 Figure 11. Fort at al-Han. Google Earth 23 November 2020. A modern structure has been super-imposed. 48 Figure 12. Fort at al-Han. Corona 5 Nov 1968. 36°17’36”N, 41°04’14”E 49 Figure 13. Poidebard’s map of the Khabur bend (Pl CXL, ‘Coude du Habour’ Large rectangles indicate fortified towns; small ones = Roman posts (castellum/castrum). 50 Figure 14. Forts North of Al-Hol. CAST accessed 22.01.2022. A is 94x84m, B is about 294x219m, C is 160x160m. A tower is visible in the SW corner of C in the Corona image.below. 51 Figure 15 a and b. The fort north of al-Hol. a) Zoom Earth (accessed 26/01/2022) 36°25’52”N, 41°07’56”E. b) Extract from Poidebard Pl. CLI - Fort to east not visible 52 Figure 16. The forts at al-Hol. Corona 5 Nov 1968 53 Figure 17. Batitas Corona 11 Dec 1968 54 Figure 18. Batitas Google Earth 23 Nov 2020 Extract of mosaic (2 images) for part of the town 36°17’20”N, 41°14’00”E 54 Figure 19. Batitas CAST accessed 26 Jan 2022 55 Figure 20. Oumtriye/Tell Mashnaqa/Amostae? An extract from Poidebard’s plate CXIV. 56 Figure 21. Oumtariye/Tell Mashnaqa/Amostae. Corona 11 Dec 1967 36°17’49”N, 40°46’39”E (approx.) 56 Figure 22. Tell Taban. CAST - accessed 26 Jan 2020 57 Figure 23. Tell Taban. Google Earth 5 Jan 2019. 36°20’08”N, 40°47’16’=3E 58 Figure 24. Tell Tuneynir (Thannouris) Zoom Earth, accessed 28/01/2022. 36°24’N, 40°53’S 59 Figure 25. Thannouris. Plan by Kennedy traced from relevant portions of Poidebard, 1934: Pl CXVII. Note north to bottom. 60 Figure 26. Touloul Mougayir from Poidebard’s Pl CL 61 Figure 27. Touloul Mougayir a) Google Earth 20/08/2020 and b) Corona 5 Nov 1968 36°24’30”N, 41°01’27”E. 62 Figure 28. Hirbet Hassan Aga. Corona 5 No=v 1968 Camp (above – 36°22’48N, 40°59’51”E) and small fort (bottom left circled on yellow). 63 Figure 29. a) Enlargement from Figure 28 above showing the camp (Corona 1968) and b) Situation of fort guarding spring - CAST accessed 27/012022 64 Figure 30. Hirbet Hassan Aga. Extract from Poidebard’s Pl CXLIII. 65 Figure 31. Extract from Poidebard’s Plate CXLVII, oriented to the north and showing the course of the wall on the Jebel Cembe. The dashed yellow line is placed just to the south of the wall. The structure called by Poidebard ‘ouvrage de défense indigène’ 66 Figure 32. Circesium. Figure 78 from Sarre and Herzfeld Vol 1, p172 68 Figure 33. Circesium. Figure 79 from Sarre and Herzfeld Vol 1, p 173 69 Figure 34. Osrhoene (NW Mesopotamia) from Dillemann (1962: fig. XI) with additions showing sites along the Khabur river (‘Aborras’) 70 Figure 35. Tell Araban a) Corona 5 Nov 1968 b) Google Earth 27 Feb 2021 72 Figure 36. Tell Ajaja. Herzfeld 1911 Vol1: 185 73 Figure 37. Tell Arban Poidebard Pl.CXIII. 36°05’26”N, 40°42’15”E 74 Figure 38. Marqada. Corona 5 Nov 1968. 35°44’38‘’ N, 40°46’00’’E 75 Figure 39. Bridge at Soufeiyé Poidebard Pl. CXIX.3 76 Figure 40. Tell Bizari. a) Corona 5 Nov 1968. b) Google Earth 24 Oct 2020 36°36’02”N, 40°47’09”E 77 Figure 41. Poidebard’s post at Tell Zenbil (north bank). a) Pls. CXXI and b) CAST accessed 05/02/2022 36°38’18”N, 40°58’53”E 78 Figure 42. The post Tell el-Bab (south bank) Poidebard Pl.CXX 79 Figure 43. Reproduction of Dillemann 1962 Fig XXVI 81 Figure 44. Tell Awan Poidebard Pl. I.1 82 Figure 45. Tell Awan Triple enceinte. Poidebard Pl. III 83 Figure 46. Tell Awan Corona 5 Nov 1968. 36°42’46”N, 41°04’59”E 83 Figure 47. Tell Awan. Extract from previous figure 84 Figure 49a. Castellum of Tell Brak – west side and plan. Poidebard Plate CXXV 85 Figure 49b. Castellum of Tell Brak – east. Poidebard Plate CXXVI 85 Figure 50. Tell Brak Poidebard Plate CXXIII. 86 Figure 51. Tell Brak showing Poidebard’s castellum upper right. Corona 5 November 1968 87 Figure 52. Tell Brak – castellum seen from above Poidebard Pl CXXIX 88 Figure 53. Brak: Photo of gate. Poidebard Pl. CXXVII.2 89 Figure 54. Tell Bati. Poidebard’s photograph (Pl. CXXXIX) re-oriented to show North at top and ‘fort’ circled in yellow. 90 Figure 55. Abra Extract from Poidebard’s Plate CLIX 94 Figure 56. Abra Corona 2 Aug 1969. 37°01’40”N, 41°47’15”E. Poidebard’s ‘castellum’ circled in yellow. 95 Figure 57. Qubur al-Bid. Copied from Wood (2004: 400). 37°24’07”N, 42°04’00”E. 95 Figure 58. Seh Qubba. Corona 11 Dec 1967 36°51’32”N, 42°29’30”E 96 Figure 59. Seh Qubba Ball 2003. 64 96 Figure 60. Babil. Extract from Poidebard’s Pl. CLIX 97 Figure 61. Babil/Kebeli Corona 11 Dec 1967. 37°11’12”N, 42°01’41”E 97 Figure 62. Tell Lelan. Poidebard Pl. CIX.1 ‘ 98 Figure 63. Tell Lelan. Corona 11 Dec 1967. 36°57’26”N, 41°30’18”E 98 Figure 64. Wilkinson and Tucker’s site 54 (1995). The adjacent site 48 is a prominent tell with occupation from the Assyrian to the Parthian periods but not Sasanian. 101 Figure 65. Site 54 of Wilkinson and Tucker (near Gar Sur village). Corona 11 Dec 1967 36°43’03”N, 42°25’17”E 101 Figure 66. Oates 1 “Syria and Parthia c A.D. 200” 104 Figure 67. Oates 2 “Northern Iraq in the Roman period” 104 Figure 68. Extract from Poidebard’s map entitled ‘Le Limes romain dans le désert de Syrie : observations aériennes et vérifications au sol (1925-1932) (map drawn at scale 1/1 000 000 by ’la Compagnie Aérienne Française’, Suresnes 1933). Large rectangles i 105 Figure 69. a) and b) The probable site of the ‘Roman’ bridge at Kisik 36°28’6”N, 42°40’19”E Corona December 1967 and Zoom Earth January 2014 108 Figure 70. Possible route across middle of Mount Sinjar from Karsi. Extract from Google Earth 110 Figure 71. The road 14km east of Sinjar. Corona 11 Dec 1967 111 Figure 72. The road 28km east of Sinjar. Note crossroad and forts. Corona 11 Dec 1967 112 Figure 73. Extract from US army map of 1942 (Tell Afar) showing position of Gonaisiya; 1 inch to 4 miles 112 Figure 74. The course of the ancient road 8km west of the current Syria/Iraq frontier. Corona 11 Dec 1967 (see also figure 71 above) 113 Figure 75. Extract from Peutinger Table, Segment XI: Singara and Sirgora circles in yellow. 114 Figure 76. Haste/Muezzar Corona 1968. 36°15’26”N, 40°19’49”E (Poidebard ‘La Trace’ Pls.CXXXIII-IV) 117 Figure 77. Site 11.5 km SW of Tell Muezzar at 36°10’02”N, 40°16’02”E - Manajid Fawqani? 118 Figure 78. Site on road from Haste to Tell Ajaja – possibly ‘Tell Touwenan’ of Poidebard Magrus? Corona 1967 36°11’26”N, 40°27’30”E 118 Figure 79. Possibly, Hirbet al-Han with a Roman (?) fort (36°03’36”N, 41°15’19”E) – 40km NE of Semsaniye, near current Syria/Iraq frontier Zoom Earth © ESRI, HERE, Garmin 120 Figure 80. The crossing of the Khabur at Ash Shaddadi (‘Sheddadi’ - Corona 1968) – 36°04’00”N, 40°29’20”E 122 Figure 81. Sheddadi Zoom Earth. The mound is 3km south of the crossing. 122 Figure 82. Sheddadi. Sarre and Herzfeld (1911: 183) 123 Figure 83. The road near ‘Qseybe’ (route 3 - see also fig 11 above) 36°15’58”N, 41°15’3”E. Zoom Earth August 2012 124 Figure 84a. Tell al-Hadhail (Vicat according to Hauser). Corona 11 Dec 1967. 36°08'56"N, 42°03'16"E 125 Figure 84b. Ibrat as-Sagira (Vicat according to Oates). Corona 11 Dec 1967. 36°22’28”N, 42°19’’53”E Corona 11 December 1967. 36°22’28”N, 42°19’53”E 126 Figure 85. Uğur. Corona 11 Dec 1967. 37°10’59”N, 41°57’00”E 127 Figure 86. Extract from the Peutinger Table (see also figure 75 above) showing the road from Amouda (Thamaudi) and Nisibis to the Tigris 128 Figure 87. The road west of Cizre (Bazeft on the horizon). Poidebard PL CLVIII.3 128 Introduction 13 Aerial photography, satellite imagery and maps 18 Geographical and historical background 19 Forts and fortresses 28 Dating the fortifications 29 The city of Singara 31 The forts at Ain Sinu 33 Alaina 36 Thebeta 41 Qohbol/Ghobal 44 Towards the Khabur 47 al-Han 49 al-Hol 51 Amostae 53 Batitas 53 Thannouris (Tell Tuneynir) 57 Touloul Mougayir and Hirbet Hassan Aga 60 The wall of Jebel Cembe 65 The Khabur valley 67 North of Circesium 69 Tell Ajaja/Arban/Horoba 71 Tell Brak and the Jaghjagh 74 Tell Brak 80 Castra Maurorum 91 Cizre and Bezabde 91 Eski Mosul and the Tigris 93 The roads 102 Roads north of Singara 106 Roads east of Singara 110 Roads to the west of Singara and to the Khabur valley 113 Roads south-east of Singara 124 The northern road from Nisibis east to the Tigris 126 The frontier in north-eastern Mesopotamia 130 Conclusions 137 Bibliography 138 Placename index 144 Roman,Frontiers,LIMES,Persia,Mesopotamia,Forts,Fortifications,Roads,Transport,Singara The Roman frontier with Persia in north-eastern Mesopotamia investigates the Roman city of Singara and the fortifications and roads in the surrounding area. The physical frontier between Rome and Persia has been little studied, in part because of the difficulty of access for scholars. In comparison with other parts of the Roman ?limes?, this frontier was of great importance because it separated the two major civilisations of the early first millennium CE. Although the frontier stretched north to Armenia and the Black Sea, north-east Mesopotamia was for long periods the major area of confrontation.0After a brief review of the history of north-east Mesopotamia and its role as the setting for repeated clashes between the two empires, the book focuses on Singara, its fortifications and the surrounding frontier zone. This town was one of the strongpoints on the Roman frontier as it existed up to 363 CE. The volume then addresses the ancient road network around Singara and the links to Nisibis and to the Khabur valley to the west. It makes use of old aerial photographs and satellite imagery to illustrate fortifications, roads and associated sites, in particular those mentioned in the Peutinger Table. A final chapter addresses the nature of the frontier in this region