Satala in Armenia

1. Location - population

Satala is known as one of the most important Roman fortresses on the eastern border of the Roman Empire. It was located on the border between Lesser and Greater Armenia, between the sources of the rivers Lycus and Akampsis. It was an important traffic hub at the crossroads of major Roman roads.1 It is a place name of Asia Minor origin, appearing in Roman and Byzantine written sources either in neutral plural2 or feminine singular.3

The modern-day Turkish name of the region is Sadak, reminding of the ancient place-name. The ruins of the ancient city have been spotted on the site, 100 km south of Trebizond and 25 km north of the river Euphrates (from Erzincan).

Satala was built at an altitude of 1,650 m, on an isolated plateau surrounded by high hills to the north, east and south.4 Although the site is not naturally fortified, it was selected for other, mainly strategic, reasons. Built at the crossroads of natural and historical roads leading from Persia to the Aegean Sea and from the Black Sea to the valley of the river Euphrates in Syria, Satala controlled the northern routes of a potential invasion in Asia Minor from the East. The site had other advantages as well: an abundant water supply from a nearby spring, food from the surrounding fertile valleys and a tributary of the river Lycus, the river Sadak Çay, which flows just 1 km to the east.

The fortress in Satala was built at the end of the 1st cent. AD and abandoned after the 7th cent. AD. At the end of the 2nd or the beginning of the 3rd cent. AD, an important Roman city was born around the fortress. The city continued to flourish until the age of Justinian (6th cent. AD). Apart from a Greek tombstone inscription dating back to the 2nd cent. AD, there are no other Greek documents dating back to the first centuries of the city. It seems that only the Latin language was used in Satala, both for military and political purposes. The existence, however, of a Greek-speaking population during Christian years is proven by the plethora of Greek tombstone inscriptions found in the region.

2. History

Satala was a temporary camp during the Roman general Corbulo’s operations in Armenia (52-63 AD), whereas the advantages of the site had probably been known since Claudius’s operations (41-54 AD) in the region. The first fortress is believed to have been built immediately after the reorganization of the eastern provinces of the Roman Empire in 71/72 AD by emperor Vespasian. It was initially the headquarters of legion XVI Flavia Firma, which arrived in Satala just after 75 AD.5 During Trajan’s Parthian campaign (114-117 AD), legion XVI Flavia Firma moved to Samosata and was replaced by legion XV Apollinaris, which was tranferred from the river Danube after 117 AD, remained in Satala until the beginning of the 5th cent. AD at least and constituted the core of Arrian’s operations against the invasions of the Alans.6 In 114 AD, Trajan gathered in Satala military forces from different parts of the empire for the campaign against Armenia and received honours and promises of peace from the majority of tribal chiefs from the regions of the Caucasus, the Black Sea and the Caspian Sea.7

A few years later, in 131 AD, Hadrian inspected the eastern border of the empire and received in Satala subordinate tribal chiefs from beyond the river Euphrates.8 In 134/135 AD, the Alans who had invaded Armenia and Cappadocia were repelled in Satala by Arrian, commander of Cappadocia at the time.9 In 256 AD, Satala was seized by Sapor I during his second campaign,10 but in 262 AD, it passed under Roman control again. A votive inscription of Gallienus, dating back to 262 AD, and a votive inscription of Aurelian, dating to 272/275 AD, prove that a series of repairs took place in the fortress after its sack by Sapor I and in view of the danger posed by the Goths and the Sassanids.11 In 385 AD, Satala was included in First Armenia and played a crucial role during the subsequent Persian wars of 421/422 and 441 AD. In 528 AD, it was incorporated in the newly-established province of Armenia Minor or First Armenia.12 In 530 AD, the Persians were defeated in front of the walls of Satala by Sitta, Justinian’s general. It was one of the most significant battles of Justinian’s war against the Persians (502-532 AD).13

The great strategic location of the site led Justinian to fortify it again during the 1st half of the 6th cent. AD. More specifically, Justinian replaced the old walls with new and stronger ones.14 However, the new walls did not deter the Persians from seizing the city in 607/608 AD. The fortress of Satala was lastly used by Heraclius as his operational base against the Persians during the 620s AD. Before long, Satala was gradually abandoned. There is no reference to Satala in the Arab historians and geographers of the Middle Ages and there is no evidence of its destruction by the Muslims. Its presence on the isolated mountains of Lesser Armenia depended on the military needs of the border on the river Euphrates. As soon as the border was abandoned and the need for communication between the north and the south vanished, the fortress in Satala was no longer useful and was abandoned.15

With regard to the city which was born around the fortress, there is not enough evidence to track its course in time. It is believed that it had already been an important city since the 2nd or the 3rd cent. AD.16 However, the first sure indication of the existence of an organized city in Satala dates back to the end of the 4th cent. AD, when Saint Basil the Great visited the city in 372 AD and promised the citizens of Satala to appoint a bishop worthy of their city.17 In addition, it can be deduced from a reference in Theodosius’ Novellae18 that its territory during the 5th cent. AD was particularly extensive: it reached the river Euphrates and the border with Greater Armenia. After the 7th cent. AD, the city was gradually abandoned along with the fortress.

3. Religion

The head of a statue of Aphrodite attests to a cult of the goddess. Apart from it, there is no other evidence of cults in Satala prior to the advent of Christianity. Satala constituted a flourishing Christian community and an important diocese from an early age. The number of Christians had already been significant since Diocletian’s reign (285-305 AD),19 whereas Euenthius, the bishop of Satala, participated in the Council of Nicaea in 325 AD.20 There are also references to bishops Elfridius (360 AD), Poemius (circa 378 AD), Anatolius (451 AD), Epiphanius (458 AD), Gregory (692 AD), Philip (879 AD) and Cosmas (1256 AD).21 The fact, however, that there are references to the diocese of Satala until the 13th cent. AD does not prove the existence of an organized city in Satala until that period. They are rather the traditional titles of priests who had already lost their seats.22

4. Culture

There are significant works of art from Satala, attesting to the degree of the vanished city’s growth. Among the fragments of bronze statues dating back to the Late Hellenistic and Early Roman periods, the famous head of Aphrodite, exhibited in the British Museum nowadays, stands out.23 The head dates back to the 1st quarter of the 1st cent. BC and belongs to a colossal bronze statue, which was probably transported to Satala from a temple of Asia Minor during the 2nd or the 3rd cent. AD. Pieces of Roman stone sculptures,24 pedestals, fragments of mosaics, Late Roman and Byzantine tomb monuments, Roman sarcophagi and numerous architectural parts were also found. A significant number of Roman and Byzantine coins,25 as well as Latin and Greek Christian inscriptions26 also come from Satala. The pottery from Satala dates back to the Late Roman period and is of good quality. Several intact Roman clay oil lamps and parts of Roman glass vessels were also found.


5. Topography and buildings

Satala was excavated by the British vice-consul in Trebizond, Alfred Biliotti, in September 1874.27 At the beginning of the 20th century, the Belgian archaeologists, Franz and Eugene Cumont, also visited the site.28 In recent years, research has been conducted by the British Archaeological Institute of Ancara.

The site can be divided into the following sectors: a) the fortress, b) the north settlement, c) the early Christian basilica to the northeast of the fortress, built over the ruins of a Roman aqueduct, d) constructions surrounding the spring to the west of the modern-day village and e) some remote constructions and fortifications.


6. The fortress

The ancient fortress extended over a relatively flat space between two streams and some higher ground to the north. It had a rectangular shape with rounded edges and occupied an area of 15.7 hectares. Nothing survives of the fortress dating to the age of the Flavians. On the contrary, the ruins of the wall dating back to the age of Justinian (6th cent. AD) are still visible. The wall was 4 m thick. Its two external sides were built with squared stones, whereas the internal sides were filled with irregular small stones and connective material. The east and north sides, as well as parts of the west side, survive in good condition. Gates are visible on the north and east sides. Similar gates must have stood on the other two sides as well. It was strengthened by square towers with sides ranging from 7 to 9 m. Parts of a tower reaching a height of 6 m survive on the northeast corner. According to the sources, there was also a hexagonal tower with a diameter of 8 m on the southeast corner of the fortress. No sign of the rampart mentioned by the historian Procopius has been spotted yet.29 In times of war, the rampart was used by peasants as a shelter. Nothing survives inside the fortress.

7. The city

Although no buildings have been spotted on the site, it can be deduced from surface finds that parts of the ancient city of Satala extended to the north of the fortress, from the northwestern corner and along the south side of the northern hill to the valley of the river Sadak Çay to the east.30 The ruined arches which stand 500 m far from the modern-day village,31 near the river Sadak Çay, belong to an Early Christian three-aisled basilica, probably a martyrium, dedicated to the patron saint of Satala, Saint Eugene. Several Early Christian tomb inscriptions come from there.32 It was probably the site of the Byzantine cemetery of Satala.33

On a hillside to the west of the fortress, at an altitude of 500 m, a 15 x 11 m reservoir, dating to the 2nd cent. AD, was found. The rainwater was collected in the reservoir and then transported to the fortress via clay ducts. Around the spring, a large yard of rectangular shape was also found. It was mainly used as a fortification, but had religious uses as well. Below the spring, the ruins of a building, which was probably part of a Roman bath, were also found.34 On a low hillside which stands 500 m far from the modern-day village, part of the cemetery of the legionnaires, dating back to the 2nd cent. AD, was found.35

Satala was surrounded by some remote small Roman fortresses and towers-observation posts. Such a tower stood on a low hill, not far from and to the north of the settlement, above the valley of the river Sadak Çay. Another fortified position was spotted on a low rise, 5 km east of Sadak. A Roman outpost or small fortress with two rows of rooms around a central yard and a second external yard was also found on the south side of a hill, not far from and to the north of Satala. According to Procopius, it could be a small fortress dating back to the 6th cent. AD, built by Justinian near Satala.36 Near the top of a hill to the south of Satala stand the ruins of a small church. Finally, there was a distinct settlement to the north of the modern-day village, dating back to the Early Bronze Age (3rd millennium BC).




1. Ptol., Geogr. 5.6.20, 1.15.10, 8.17.4; It. Ant. 183.5, 207.9, 207.10, 216.3, 217.4; Tab. Peut. 10.5; Miller, K., Itineraria Romana (Stuttgart 1916), pp. 676, 682.

2. Dio C. 68.18.2; Anderson, J.G.C. – Cumont, F. – Gregoire, H., Recueil des inscriptions grecques et latines du Pont et de l’Armenie (Studia Pontica 3.1, Brussels 1910), p. 48, no. 34 (εν Σατάλοις); Proc., Build. 3.4.2-5 (Σάταλα πόλις...Σατάλων); Not. episc. 1.231, 3.168, 8.283 (Σατγάλων), 9.192, 10.289, 13.148; Steph. Byz. See entry “Σάταλα” (Σάταλα, πόλις Αρμενίας).

3. It. Ant. 181.8; Ptol., Geogr. 1.15.10; Justin., Novell. 31.

4. See Proc., Build 3.4.2-5; Ptol., Geogr. 1.15.10.

5. It has been suggested that legion XVI Flavia Firma never camped in Satala. Therefore, the first legion that camped in Satala either is unknown or was legion XII Fulminata. However, the presence of legion XVI Flavia Firma in Satala is proven by two inscriptions on the tombstones of legionnaires. See Mitford, T.B., “The Inscriptions of Satala (Armenia Minor)”, ZPE 115 (1997), pp. 140-142, no. 4-5. On the relevant discussion, see Mitford, T.B., “Further Inscriptions from the Cappadocian Limes”, ZPE 71 (1988), p. 168, note 7.

6. Dio C. 55.23.5; It. Ant. 183.5; Not. Dign., Or. 38.5 and 13. The presence of legion XV Apollinaris in Satala is proven by seals on tiles, inscriptions on the tombstones of legionnaires and coins. See Dizionario epigraphico di Antihita Romane 4.42 (1983), p. 1315, see entry “Satala” (M. Malavolta); French, D.H. – Summerly, J.R., “Four latin inscriptions from Satala”, AS 37 (1987), pp. 17-22; Mitford, T.B., “The Inscriptions of Satala (Armenia Minor)”, ZPE 115 (1997), pp. 142-147, no. 6-11.

7. Dio C. 68.18.2, 19.2; Ptol., Geogr. 1.15.10; Magie, D., Roman rule in Asia Minor: to the end of the third century after Christ 1 (Princeton 1950), p. 607; Sartre, M., L’Orient Romain. Provinces et sociétés provinciales en Mediterranée orientale d'Auguste aux Sévères (31 avant J.-C. – 235 après J.-C. (Paris 1991), p. 47; Wylie, G., “How did Trajan Succeed in Subduing Parthia where Mark Antony Failed?”, The Ancient History Bulletin 4.2 (1990), p. 38.

8. Mitford, T.B., “Cappadocia and Armenia Minor. Historical setting of the ‘limes’”, ANRW 2.7.2 (1980), p. 1201.

9. Arr., Exp. 5.15 and 24.

10. Honigmann, E. – Maricq, A., Recherches sur les Res Gestae Divi Saporis (Brussels 1952), paragr. 18; Maricq, A., “Res Gestae Divi Saporis”, Syria 35 (1958), p. 295, row 18.

11. Mitford, T.B., “The Inscriptions of Satala (Armenia Minor)”, ZPE 115 (1997), pp. 138-140, no. 1-2; Similar repairs also took place in the fortress some time later. CIL 3 Suppl. 2.13630. Dizionario epigraphico di Antihita Romane 4.42 (1983), p. 1315, see entry “Satala” (M. Malavolta).

12. Cod. Just. 1.29.5; Justin., Novell. 8, notitia 23, Novell. 31.1.

13. Proc., Pers. 1.15.10; Greatrex, G., Rome and Persia at War 502-532 (Leeds 1998).

14. Proc. Build. 3.4.2-5.

15. Cumont, F. – Cumont, E., Voyage d’éxploration archéologique dans le Pont et la Petite Armenie (Studia Pontica 2, Brussels 1906), p. 346; Mitford, T.B., “Cappadocia and Armenia Minor. Historical setting of the ‘limes’”, ANRW 2.7.2 (1980), p. 1210-1211.

16. The view that a colony of veteran legionnaires was established in Satala at the end of the 2nd or the beginning of the 3rd cent. AD is not accurate. See Sartre, M., L’Orient Romain. Provinces et sociétés provinciales en Mediterranée orientale d'Auguste aux Sévères (31 avant J.-C. – 235 après J.-C. (Paris 1991), p. 268; Mitford, T.B., “The Inscriptions of Satala (Armenia Minor)”, ZPE 115 (1997), p. 149, no. 14.

17. Bas., Epistl. 99.3-4, 102-103 (Migne, PG 32, column 508). See Jones, A.H.M., The Cities of the Eastern Roman provinces2 (Oxford 1971), pp. 171, 428, note 46; Cumont, E., Voyage d’éxploration archéologique dans le Pont et la Petite Armenie (Studia Pontica 2, Brussels 1906), p. 345.

18. Theod., Nov. 5.3.

19. Life of St. Eustathius (Migne, PG 116, columns 470B, 483C).

20. Nomina partum Nicaen (Gelzer-Cunz, p. 26).

21. Hierocl., Synecd. 703.4; Not. episc. 1.231, 3.168, 8.283 (Σατγάλων), 9.192, 10.289, 13.148.

22. Cumont, F. – Cumont, E., Voyage d’éxploration archéologique dans le Pont et la Petite Armenie (Studia Pontica 2, Brussels 1906), p. 345, note 4.

23. BMCatalogue of Bronzes (1899), no. 266, photo 3; Burn, L., The British Museum Book of Greek and Roman Art (London 1991), pp. 128-129; Wilkinson, R.D., “The sculpture of ancient Armenia”, Journal of the Society for Armenian Studies 4 (1989), pp. 230-231, picture 12.

24. A significant group of tomb stelae with relief busts dating back to the 2nd cent. AD stand out among them; Mitford, T.B., “Further Inscriptions from the Cappadocian Limes”, ZPE 71 (1988), pp. 167, 171-173.

25. Lightfoot, C.S., “The coins from Satala’, in Ashton, R. (ed.), Studies in Ancient Coinage from Turkey (BIAA Monograph 12, London 1996), pp. 147-150, table 68, no. 4.20.

26. Mitford, T.B., “The Inscriptions of Satala (Armenia Minor)”, ZPE 115 (1997), pp. 137-167.

27. Mitford, T.B., “Biliotti’s Excavations at Satala”, AS 24 (1974), pp. 221-224.

28. Cumont, F. – Cumont, E., Voyage d’éxploration archéologique dans le Pont et la Petite Armenie (Studia Pontica 2, Brussels 1906), pp. 343-351.

29. Proc., Build. 3.4.2-5.

30. Some researchers, however, believe that it was the site of the ancient cemetery of Satala due to the tomb stelae dating back to the Mid-Imperial period found there. See French, D.H. – Summerly, J.R., “Four latin inscriptions from Satala”, AS 37 (1987), p. 17.

31. There were 7 arches in 1866, whereas only 4 survived until 1906. Only 3 arches survive nowadays. During the 19th century, the peasants who built the neighbouring armenian village removed all the chiseled stones, which covered the core of the walls, made of raw stone.

32. Mitford, T.B., “The Inscriptions of Satala (Armenia Minor)”, ZPE 115 (1997), pp. 156-167, no. 28-57.

33. On the proto-christian basilica and the byzantine cemetery, see Mitford, T.B., “Biliotti’s Excavations at Satala”, AS 24 (1974), pp. 233-235.

34. An early-ottoman sweating bath was spotted near the northwest corner of the fortress. Sinclair, T.A., Eastern Turkey: An Architectural and Archaeological Survey 2 (London 1989), pp. 330-331.

35. Mitford, T.B., “Further Inscriptions from the Cappadocian Limes”, ZPE 71 (1988), p. 167.

36. For the basilica and the byzantine cemetery, see Mitford, T.B., “Biliotti’s Excavations at Satala”, AS 24 (1974), σελ. 233-235.