(Spasinou) in pre-Islamic times; Characene is the name Pliny gives for the later region of Mesene (called Mēšān or Mēšūn in Middle Persian, Maysān/Mayšān in Syriac, and Maysān in Arabic) in southernmost Mesopotamia, which formed a political district of that name in the Seleucid, Parthian, and Sasanian periods.


CHARACENE and CHARAX (Spasinou) in pre-Islamic times. Characene is the name Pliny gives for the later region of Mesene (called Mēšān or Mēšūn in Middle Persian, Maysān/Mayšān in Syriac, and Maysān in Arabic) in southernmost Mesopotamia, which formed a political district of that name in the Seleucid, Parthian, and Sasanian periods. The toponym Characene is derived from the Greek name for its chief city Charax, which was also called Mēšān in Middle Persian, but Karḵ Maysān in Arabic (see EI1 IV, p. 146, on the various forms of this word). In the surviving historical texts, the state founded by Hyspaosines in lower Mesopotamia (see below) was first called Mesene by Strabo during the late first century b.c.e. (Strabo, 16.4.1). The term is probably a Greek adaptation of the equivalent Aramaic or Persian toponyms. Pliny and Ptolemy call the region which surrounds Charax “Characene,” a term that appears to designate almost the identical extent of land as Mesene/Mēšān (Pliny, 6.31.138; Ptolemy, Geography 6.5). The name Chara­cene, which is derived from Charax, would appear to have originated as more of a political identification of a geographical region, whereas the older Mesene gave a more specific geographical meaning. The Aramaic­-Syriac form Maysān was later adapted by the Arab conquerors and so survived as the name of the whole of Southern Iraq until the late Middle Ages.

History. The city, founded by Alexander the Great as Alexandria on the Tigris, was constructed on an artificial mound to protect the site from the flood waters of the adjoining rivers. Alexander probably intended the new town to serve as a major commercial port for his eastern capital of Babylon, a port which would handle the rich sea trade both from newly conquered India, and from the Arabian peninsula. To provide a sufficient population Alexander directed that the city should be settled partly by invalid Macedonian soldiers drawn from the ranks of his returning armies and partly by a transfer of residents from the city of Durine nearby. The Macedonians were established in a quarter (dēmē) of the port called Pella after Alexander’s own town of birth (Pliny, 6.31.138).

During the subsequent years Alexandria did not wholly live up to its founder’s expectations. We know, for example, that by the latter half of the 3rd century b.c.e., the independent Arab city of Gerrha, situated on the lower Persian Gulf, had become the major trans­shipping port for goods from India and Arabia, to both the Seleucid territories and to Ptolemaic Egypt (Tarn, pp. 63, 367). Alexandria on the Tigris, meanwhile, was destroyed by floods (Pliny, 6.31.138). These established channels of commerce in the Persian Gulf thereafter remained largely unchanged until Antiochus IV (q.v.) acceded to the Seleucid throne. Antiochus proved to be an energetic and ambitious ruler, who was determined to remedy the faltering economy of his realm. To further this end he attempted to divert the Indian sea trade to his own domains. A key factor in the scheme had been the rebuilding of Alexandria as an Antiochia. At the completion of the restored port (in 166/65 b.c.e.) Antiochus appointed Hyspaosines, son of a certain Sardodonacus, as governor (eparch) of Anti­ochia and its surrounding district (Pliny, 6.31.139). Both of these names are Hellenized Persian forms, and we may deduce from this that the family of Hyspaosines was of Persian background.

Though Antiochia did enjoy a new measure of prosperity, the effect on the whole kingdom was to prove of little lasting value, for the early death of Antiochus in 163 b.c.e. marked the end of the king’s plans to revitalize the economy of Mesopotamia. Now too, moreover, centralized Seleucid control throughout the entire realm was greatly weakened, as rival claim­ants to the throne continued to wage civil strife in Syria.

As an eventual result of this period of internal instability, several small areas from within the empire cast off their political ties with the Seleucids and declared themselves to be independent states. Elymais, which comprised much of the territory of Ḵūzestān in south Iran, a region that lay to the east of Mesene, was a notable example. Hyspaosines, on the other hand, seems to have remained loyal to the Seleucid dynasty, though he too most probably found himself a virtually independent sovereign at Antiochia. His willingness to continue as governor only was possibly prompted by the knowledge that if his present position was main­tained, there would be no interruption in the lucrative trade which the refounding of Antiochus had caused to be channeled through that city to the empire capital of Seleucia on the Tigris. Other former vassals of the Seleucids proved more ambitious. By 141 b.c.e. Mithradates I of Parthia had defeated the Seleucid contender Demetrius II in Mesopotamia and shortly thereafter took Seleucia (for an account of these events and references see Debevoise, p. 25). Although De­metrius later regained a part of Mesopotamia, the threat and closeness of the Parthian menace apparently persuaded Hyspaosines to declare himself independent ruler in Antiochia. Nodelman suggests that this final break from Seleucid authority occurred sometime be­tween 141 and 139 b.c.e., when Demetrius was again defeated and further taken captive by Mithradates (Nodelman, p. 87).

During the latter part of his reign Hyspaosines struck silver tetradrachms patterned after the coins of contemporary Seleucid rulers. Their Greek legend is basileōs yspaosinoy “of the king Hyspaosines,” with dates calculated in the Seleucid era of ĒPR, QR, or AQR, 125-121 b.c.e. (Le Rider, p. 230).

Alone, Hyspaosines could not indefinitely forestall the renewed Parthian advance and in 121/20 b.c.e. we find bronze coins of his being overstruck with a type of Mithradates II. This occurrence would indicate final and complete conquest by the Parthians (Bellinger, pp. 60-61).

Though defeated Hyspaosines was allowed to retain his small kingdom at the head of the Persian Gulf. As a vassal to the Parthian throne his old authority, how­ever, was much weakened.

During the continuing reign of Hyspaosines the embankment walls of Antiochia were again de­stroyed by flood. Pliny notes that to protect his capital Hyspaosines caused new embankments to be raised which extended in length a distance of nearly 2 miles, or 3.2 km (Pliny, 6.31.138). Because of this impressive rebuilding Josephus and other writers of the late Classical period, refer to the city as Charax Spasinou, literally, Palisade of Spasines or Hyspaosine (Josephus, Antiquities 1.6.4). In the Palmyrene inscriptions the city was usually given the Aramaic form Kark Ispasina (Corpus Inscr. Semit. II, no. 3928). It was sometimes also called Karka de Maysān or fortress of Maysān (Starcky, pp. 13-14).

Dated and inscribed tetradrachms in Greek attest that Hyspaosines was followed as Parthian vassal in Mesene by Apodakos ca. 110 b.c.e.; Tiraios I (ca. 95-­88) and Tiraios II (ca. 78-44). A unique tetradrachm of the next succeeding ruling prince, Artabazos, is dated DXS (48-47 b.c.) and displays on the reverse an extended Greek inscription: basileōs artabazo theopatoros aytokratoros sōtēros philopatoros kai philellēnos “of the king Artabazes, of divine descent, ruler in his own right, the deliverer, who loves his father and the Greeks” (Le Rider, p. 246). The square arrangement of this epithet spaced around a typical Greek Herakles, is copied from the conventional style of contemporary Parthian coinage.

When the Roman emperor Trajan descended the Tigris in 116 c.e., he received the temporary sub­mission of Attambelos, the ruling prince in Mesene (Dio Cassius, 78.28). Apart from this occasion, however, the Parthian state presumably remained in at least nominal control of Mesene during the nearly 350 years they dominated Mesopotamia east of the Euphrates River.

Continued vassalage to Parthia did not interrupt the prosperity which the people of Mesene enjoyed as major agents of east-west commerce. By the first century of our era new and quite lucrative overland trade routes had been established with the Nabatean city of Petra in what is now southern Jordan.

For more than a century the Nabateans, through Charax, became major providers of a considerable eastern trade with the rich Roman west. Still other caravan routes joined Charax with the Syrian desert emporium of Palmyra. Following the assimilation of the Nabatean kingdom by Rome in 106 c.e., the Palmyrenes were to enjoy a practical monopoly of such trade.

The center of commerce within the kingdom of Mesene, meanwhile, appears gradually to have moved south. By the third quarter of the first century Pliny mentions the city of Forāt, which was subject to the Kings of Characene (Mesene) as being frequented by the people from Petra (Pliny, 6.32.145). Palmyrene inscriptions of the second century also testify to success­ful caravan journeys undertaken between that city and both Charax and Forāt (Starcky, nos. 81, 112, and 114). Forāt was located some 11 miles or 17.7 km below Charax on the Tigris. Nodelman (p. 113) suggests that Mithradates IV of Parthia (r. 128-47) moved the capital from Charax to Forāt.

A recently discovered bilingual inscription in Greek and Parthian infers that Mithridates, a son of the Parthian King Pacoros, ruled Masan independently for some years as a rival to the Parthian throne. The text relates that after a lengthy struggle in the whole of Mesopotamia Vologases III (r. ca. 148-92), the legiti­mate Parthian king, defeated Mithridates in Mesene in 150/51 (Turin, pp. 423-25). This Mithridates was the last prince to issue coins in Mesene which display comprehensible Greek inscriptions; they bear dates equivalent to 142/43 c.e. and legends which read “king Mihrdat, son of king Phoba” (Seyrig, p. 254). A few later tetradrachms with devolved, unreadable legends, are the last Greek issues of Mesene.

From the second half of the second century c.e. the coins of the rulers in Mesene were either uninscribed or in a variant form of Aramaic that probably represents the contemporary dialect spoken there. A bronze tetradrachm in this late series shows on the obverse a profile bust of a prince, with the Aramaic legend ybyngʾy mlkʾ “Abinergios (?) the king” (see Hill, nos. cciv-ccv).

The Aramaic legend on tetradrachms attributed to one of the last rulers in Mesene reads: mʾg zy ʾstʾbyʾz mlkʾ, possibly “Maga, (son) of king Attambelos” (Hill, nos. ccviii-ccix). The Aramaic coins of Mesene are undated.

Very little is known of the closing decades of the history of Mesene/Characene. Even the question of whether the kingdom, during this late phase, con­tinued as a dependent province under a waning Parthian suzerainty remains conjectural. What we do know is that in 221-22 c.e. Ardašēr (q.v.), satrap of Persis (Pārs, Fārs) to southeastern Persia, rose in revolt against his Parthian overlord Artabanus V. Ardašēr quickly gained the support of all Persis and shortly thereafter subdued the vassal state of Elymais. He next marched against Maysān and according to Ṭabarī (I, p. 818) killed its last ruler, Bandū, and built Karḵ Maysān, which was renamed Astarābād-Ardašīr (Ṭabarī, I, p. 820).

Textual references on the political history of Mēšān during the Sasanian period are rare. In a Manichean fragment Mihršāh, Lord of Mēšūn is mentioned as a brother of the Sasanian king Šāpūr I (r. ca. 242-72), but the historicity of the text is not certain (Sundermann, pp. 61-63). In the inscription of Narseh I (r. 293-303) at Paikuli Ādurfarnbay, King of Mēšān, is portrayed as one of the supporters of Warahrān (III), son of Warahrān II, but his official Status remains uncertain (ed. Humbach and Skjærvø, pars. 34-39, comm. III/2, pp. 70ff.).

By the 5th century, Forāt had apparently succeeded Karḵ Maysān as the provincial capital, for under its Syriac name Perāt de Maysān this place is listed (between 410 and 605) in Nestorian church annals as the chief city of Maysān (Chabot, pp. 272, 478).

Göbl, following Herzfeld, has suggested that the mint monogram PR which appears on numerous Sasanian coins from the reign of Kobād I (r. 488-531) to that of Ḵosrow II (r. 590-628), represents Perāt (Göbl, p. 91). “Maysān” and “Forāt” later appear as Omayyad mints; the former from 79/698-99 to 97/715-16) and the latter from 81/700-01 to 97/715-16). Because of persis­tent flooding, changing river courses and a lessening of trade with the west, the two cities seem to have been finally abandoned during the 3rd/9th century.

Archeology. A search for the sites of the major cities of Mesene, Charax and Forāt, was undertaken by the present writer in 1965 (Hansman, pp. 21-58). According to Pliny, as we have noted above, the name Charax, meaning “palisade,” identified the nearly two-mile long (3.2 km) embankment walls which protected the city from flooding. Pliny further states that Charax was located near the confluence of the Eulaeus river canal and the Tigris river (Pliny, 6.31.138).

On the evidence of Pliny and other classical writers I had supposed that the ancient Eulaeus river should be identified with the modern Karḵa river, which rises in Persia and flows into the swamps between Persia and Iraq (Hansman, pp. 27-36). Aerial photographs clearly showed a substantial, now abandoned water channel leading in a southeasterly direction from the present Eulaeus-Karḵa river to an equally substantial, now disused channel of the Tigris river. Near the point where these two abandoned water courses met, a distinc­tive trapezoidal feature appeared in the photograph. Upon visiting this site in 1965 very substantial remains of embankment walls were discovered, arranged in the form described. The northern and eastern embank­ments were largely intact, but those to the west and south had suffered from water erosion. The surviving walls, in fact, rise to a height of 4.6 m. Series of protrud­ing “bastions” are visible at intervals. The embankments consist almost entirely of earth, although cours­ing and loose scatterings of square, baked bricks are found at most of the bastions. Much of the ground within the site is covered with alluvial deposits laid down by the frequent flooding to which this area was subjected. Because of alluviation, surface sherds were found only on the embankments and at several high points within the site. Those that could be identified belonged to the Sasanian or early Islamic periods (Hansman, pp. 36-45).

The northern length of the walls measures approxi­mately 2.8 km, the western wall 1.5 km, the eastern 1.3 km, and the southern embankment 2.9 km. It was therefore possible to determine that Pliny’s measure­ment of the embankments of Charax, given as two milia passuum or 2.8 km, almost exactly equaled the length of the northern embankment of our site. The local name for these impressive remains is Naysān, a quite possible colloquial Arabic corruption of Maysān, a form of the toponym attested by Yāqūt (Boldān III, p. 860; IV, p. 207). Bearing in mind also that the site of “Naysān” is situated at a point near where former arms of the Tigris and Eulaeus rivers met (as described by Pliny), I believe that the embankments of Naysān are to be identified as those of the ancient city of Charax/Mēšān/Maysān.

In respect of the city of Forāt, Pliny states that “Pratta” was located on the “Pasitigris” 12 milia passuum below Charax (Pliny, 6.32.145). The name Pasitigris is usually associated with the Kārūn river in neighboring south Iran but is also identified by Strabo with the lower reaches of the Tigris into which the Kārūn now flows (Strabo, 15.3.4). Therefore, having accepted the remains of Naysān as marking the site of Charax/Mēšān, in order to find a hypothetical location of Pratta/Forāt, we measured downstream along the former course of the Tigris, already noted. Using known parallel distances measured in Roman miles, we cal­culated that 12 milia passuum equaled 17.4 km. Moving south of Naysān along the abandoned Tigris channel we arrived at a large archeological site, called Maḡlūb (Ar. “conquered”), almost exactly 17.4 km below Naysān. Maḡlūb measures approximately 1.8 x 1.3 km in area, and rises to an average height of two meters above the surrounding alluvial plain. The exposed surface sherds are to be associated with the Sasanian and early Islamic periods. On the basis of the relevant measurements, we identified the site of Maḡlūb as that of ancient Forāt. No other archeological sites of any size were located in the region. No excavations have been possible at either Naysān (Charax) or Maḡlūb (Forāt) because of the political troubles that persist in those areas.



A. R. Bellinger, “Hyspaosines of Charax” Yale Classical Studies 8, 1942, pp. 53-67.

J. Chabot, “Synodicon orientale,” Notices et extraits des mss. de la Bibliothèque Nationale, Paris, 1902, pp. 664-85.

N. Debevoise, A Political History of Parthia, Chicago, 1938.

R. Göbl, Ein Asiatischer Staat I, Wiesbaden, 1954.

G. F. Hill, Catalogue of Greek Coins of Arabia, Mesopotamia and Persia, British Museum, 1922.

J. Hansman, “Charax and the Karkheh,” Iranica Anti­qua 7, 1967, pp. 21-58.

H. Humbach and P. O. Skjærvø, The Sassanian inscription of Paikuli, 3 vols. in 4, Wiesbaden, 1978-83.

G. Le Rider, “Monnaies de Characène,” Syria 36, 1959, pp. 229-53.

S. A. Nodel­man, “A Preliminary History of Characene,” Berytus 13, 1960, pp. 83-121.

H. Seyrig, “Antiquités syriennes 38. Inscriptions grecques de l’agora de Palmyre,” Syria 22, 1941, pp. 225-70.

J. Starcky, Inventaire des inscriptions de Palmyre X, Damascus, 1949.

M. Streck, “Maisān” in EI1, pp. 146-55.

W. Sunder­mann, “Studien zur kirchengeschichtlichen Liter­ature der iranischen Manichäer III,” AoF 14/1, 1987, pp. 41-107.

W. Torn, The Greeks in Bactria and India, Cambridge, 1952.

Turin, Centro Ricerche Archeologiche e Scavi di Torino, The Land between two Rivers, Turin, 1986.

(John Hansman)

Originally Published: December 15, 1991

Last Updated: October 13, 2011

This article is available in print.
Vol. V, Fasc. 4, pp. 363-365