Ancient sources reveal the extent of Roman plans to add Parthia to their empire during the first decades of Augustan rule (27 BC–AD 14). The preparations involved intelligence gathering and the creation of itineraries to map possible invasion routes through Iran. These accounts confirm the veracity of Roman ambitions and verify the scope of military preparations being considered by the Emperor Augustus.
During the first decade of Augustan rule there was a resurgence of Roman interest in eastern campaigns and Latin poets took the subject as inspiration to introduce dramatic situations into their narratives. For example, Propertius explored ideas of distant military service and the feelings of a Roman wife separated from her soldier-husband who was fighting in Bactria. The verse takes the form of a letter with the wife appealing to her husband not to be reckless in the pursuit of glory when Roman forces lay siege to Bactrian cities and take silks as plunder from the steppe hordes. She writes, ‘I beg you not to set so much glory in scaling Bactrian walls, or seizing fine fabrics from their perfumed chieftain, especially when the enemy launch the lead shot from their slings and fire their bows with such cunning from their wheeling horses.’ Propertius imagines that the soldier-husband would return when ‘the lands of the Parthian hordes are overcome’ and the Oxus River was established as a new imperial boundary. However, he hints at even more distant locations. The Roman wife expects her husband will be seen amid ‘the dark-skinned Indians who are pounded by the eastern waves’.
These ideas could have been prompted by the arrival of Indian envoys in the Roman Empire who might have offered the prospect of military alliances (26–20 BC). In another work, Propertius addresses a lover with the possible scenario, ‘What if I were a soldier detained in far-off India, or my ship was stationed on the Ocean?’ In this period it must have seemed possible that well-led Roman armies could exceed the eastern conquests of Alexander.
There are also some indications that the scenarios suggested by Propertius could be based on genuine military planning. Propertius mentions charts being circulated that mapped Parthian territory and provided details concerning enemy logistics. The wife reveals that she ‘studies the course of the Oxus River which is soon to be conquered and learns how many miles a Parthian horse can travel without water’. She also ‘examines the world depicted on a map and the position of lands set out by the gods to be sluggish with frost, or brittle with heat’. These details suggest that imperial authorities were gathering geographical and logistical information about the east in order to determine the practical prospects for conquest.
The Parthian Stations
The Romans knew the size and geography of Persia from Greek histories, including accounts written by authorities who followed Alexander. From these historical descriptions Roman commanders reconstructed hypothetical invasion routes as ‘itineraries’ listing the directions and distances between strategic sites. These itineraries might have been represented pictorially on charts containing geographic detail including mountains, lakes and rivers. But other documents were descriptive texts that explained the character of strategic locations and itemised possible invasion routes.
The Romans had supporters in the main Greek cities of Babylonia and these communities were in continual contact with Mediterranean merchants who travelled back and forth across the Euphrates frontiers. In particular, the Romans had a network of collaborators in the city-port of Spasinu Charax near the head of the Persian Gulf. Spasinu Charax was originally a military outpost established by the Seleucids, taking its name from the Greek word ‘Charax’ meaning ‘palisaded fort’. This fortified Hellenic town developed into a commercial city that was well-protected from siege or cavalry attack by the floodplains of the Tigris River. When the Parthians conquered the eastern half of the Seleucid realm, the local Greek commander Hyspaosines took the title of king and founded a new Hellenic dynasty at Charax to govern a region called Characene (127 BC). The new kings of Characene accepted Parthian suzerainty, but with a well-defended capital they could assert their independence and challenge outside interests.
Spasinu Charax was positioned at the junction between riverine and maritime travel. The city received traffic coming down the Tigris River from the heartlands of Babylonia, but it was also a staging post for maritime voyages into the Persian Gulf. Spasinu Charax therefore received trade goods from Arabia and India and served as a meeting place where Persian and Greek traders could engage with eastern merchants from distant lands. Consequently, the city was an ideal location to gather intelligence about political developments in the distant east and use as a base from which to reconnoitre possible invasion routes into foreign territories.
Pliny reveals how Greek operatives from Spasinu Charax provided Roman authorities with accounts of eastern geography and politics in preparation for a planned military action against the Parthians. He explains that ‘the most recent writer to have dealt with the geography of the world is Dionysius who was born in Charax and sent to the East by the Emperor Augustus to write a full account of this region.’ Pliny explains that Dionysus was given this responsibility sometime before 2 BC and ‘shortly before Gaius Caesar travelled into Armenia to take command against the Parthians’. The work has not survived, but it probably included the type of information suggested by Propertius when he described Roman charts recording the distance between Parthian watering-stations and the condition of the surrounding landscapes.
An ancient account known as the Parthian Stations could also be a product of these early intelligence gathering operations. The Parthian Stations was written by a Greek author called Isidore who also came from Spasinu Charax. Sometime before 10 BC Isidore charted a route through Parthia for the benefit of Roman authorities. If Augustus had ordered the conquest of the Parthian Empire, the route was intended for Roman forces to follow during their military campaign. Isidore also described valuable resources produced in Parthian territory, including pearls that contributed large revenues to eastern treasuries.
The Invasion Route through Parthia
The Parthian Stations gives an itinerary of ancient sites leading across ancient Persia from the frontiers of Roman Syria to the eastern edge of Iran. Isidore mentions distances between strategic locations and provides information about the character of prominent settlements. He maps out a possible route for Roman legions that follows the Euphrates River into Babylonia and then out across the Iranian Plateau to reach the eastern frontiers of Parthian jurisdiction in Arachosia (southern Afghanistan). Isidore mentions which settlements are fortified and notes which districts have access to the main wells. He records sites that were founded by the Macedonian regime and on several occasions includes details as to whether an urban population could be considered ‘Greek’ and thereby, by implication, pro-Roman.
Isidore suggested the invasion route should begin at the Syrian town of Zeugma on the Euphrates frontier. Zeugma controlled a large bridge that spanned the Euphrates, but he advised the main Roman force keep to the western bank so that any intercepting Parthian cavalry would need to cross the river to launch an attack. From Zeugma, the Roman army would march south through a line of fortified Greek towns and walled villages that had been founded by Alexander or the Seleucid kings. Isidore also notes the site of several ‘Royal Stations’ in this region that had been established by the Persian King Darius as part of an ancient Royal Road that connected his domains in the fifth century BC. It is possible that the Romans planned to ship supplies and personnel down the Euphrates on river craft and Isidore therefore notes any sailing hazards. At one point he warns, ‘Here the flow is dammed with rocks in order that the water may overflow the fields, but in summer this same barrier will wreck the boats.’
A village named Phaliga on the Euphrates occupied a strategic position in the invasion plans. Isidore records that the settlement lay almost halfway between the Syrian capital Antioch and the main city of Seleucia in central Babylonia. Downstream from Phaliga a tributary river flowed into the Euphrates near a walled village called Nabagath. At this point Isidore recommends, ‘Here the Legions cross over to territory beyond the river.’ This was the site where the Romans expected to bridge the Euphrates in order to advance down the east bank of the river.
There were Parthian garrisons guarding river outposts on the east banks of the mid-Euphrates and the Romans needed to occupy these locations in order to control this part of Mesopotamia (northern Iraq). Isidore mentions two Euphrates islands that the Parthians used as secure bases to store treasury funds. When a renegade Parthian Prince named Tiridates II temporarily took control of Babylonia in 26 BC, he was able to take these sites from King Phraates IV. Isidore noted that Phraates ordered his men to ‘cut the throats of the concubines’ when the exiled Tiridates had surrounded the loyalist outpost. This incident reminded the Romans that the Parthians would kill hostages and destroy property if they believed defeat was imminent.
Beyond the treasury island of Thilabus there was another island midstream in the Euphrates where the city of Izan was located. The Romans required river transports to seize these sites and Isidore mentions that the nearby city of Aipolis had bitumen springs used to waterproof ship-hulls. This material could repair any Roman transports damaged by sailing downstream, or perhaps the imperial invaders planned to construct new vessels at this site. Meanwhile, Roman land forces could advance towards the city of Besechana which had a prominent temple dedicated to the Syrian goddess Atargatis. Beyond this city the course of the Euphrates came close to the Tigris with a short canal connecting the two rivers. After capturing the Hellenic city of Neapolis, the legions would follow the path of this canal east to Seleucia on the banks of the Tigris River. The vast city of Seleucia was heavily fortified, but the Romans could expect support and assistance from its largely Greek population.
The twin capitals of Seleucia and Ctesiphon were positioned on opposite sides of the Tigris River, so the Romans needed to commandeer or build river craft to make a crossing to this monumental Parthian city. The Romans probably surmised that once Ctesiphon was captured, the Parthians would relinquish control over Babylonia, including Spasinu Charax at the head of the Persian Gulf. Isidore therefore suggests that the next stage in the Roman campaign was the invasion of Iran and the capture of Ecbatana, the second royal city of the Parthian Empire.
Babylonia was densely populated with wide well-irrigated field systems and relatively short distances between the leading urban sites. But the cities of Iran were separated by arid and mountainous tracks of land and this created difficulties for the invading force. Isidore uses a new terminology to describe sites on the route through Iran, including positions that he calls stathmoi or ‘stations’. These sites might have been caravan supply-stations (caravanserai), military installations, or communication posts used by Parthian administrators to relay government orders. It is also possible that many of these locations fulfilled multiple roles for travellers and because the Parthian regime depended on cavalry, these outposts were crucial in maintaining cohesion across their empire. As the ruling Parthian court wintered at Ctesiphon and spent the summer months in Ecbatana, the thoroughfare between these two cities was well maintained for official travellers. It therefore provided the Parthian rulers with a fast and effective escape route if the Romans sized Babylonia. Isidore describes Ecbatana as a metropolis that housed the Parthian treasury and reports that it also had a major temple dedicated to the Iranian goddess Anaitis.
The route from Ctesiphon to Ecbatana headed northeast to the Iranian Plateau. Leaving the banks of the Tigris, the Roman force would pass a Greek city named Artemita on the edge of Babylonia. From this point onwards the legions had to cross open terrain through a series of rural villages equipped with caravan stations. On route to the Zagros Mountain range they would pass through another Greek city named Chala before crossing into Media.
The legions needed to travel through ten villages in Median territory, each equipped with stations (stathmoi) for travellers. After these positions were secured, the Romans would reach a mountain city named Bagistana that controlled passage to the city of Concobar with its famous temple to the Greek hunting-goddess Artemis. By then the Roman army were close to the centre of Parthian rule in Media and if they marched further east they were advised to capture a custom station known as Bazigraban which controlled caravan traffic moving between Babylonia and Iran. Close by there was a royal summer palace named Adrapana which had surrounding parklands for the Parthian nobility to hunt game and engage in other equestrian sports.
From Adrapana it was suggested that the legions marched onward to capture the nearby capital Ecbatana. Ecbatana was crucial to the conquest of Media and after this city had been captured, Isidore recommended a further route through the region to seize three important caravan stations, ten villages at strategic locations and five additional cities. This part of the campaign route ended near a city called Rhaga which had a population larger than Ecbatana. Nearby was the city of Media-Charax which had developed around a fortified installation where the Parthians had settled some of their steppe allies known as the Mardi (176–171 BC). The city of Media-Charax was positioned beneath a mountain called Caspius and it controlled the main approach to the southern Caspian Gates. This location marked the edge of Media, so the capture of the city would have brought the western realms of the Parthian Empire fully under Roman dominion. According to Pliny, Ecbatana was 750 miles from Seleucia and 20 miles from a strategic pass known as the Caspian Gates.
Hostile deserts of salt-encrusted sediment filled large stretches of eastern Iran. These were the remains of ancient prehistoric seas that had entirely evaporated to leave broad wastelands between the mountain ranges that encircled the country. In order to capture the remaining Parthian territories, Roman forces needed to follow a course across Hyrcania, a fertile region that stretched around the southern shores of the Caspian Sea. This open coastline included grasslands, but inland there were deciduous broadleaved forests and upland alpine meadows that provided a habitat for the now extinct Caspian tiger. Leopards, lynx, brown bears, wild boars and wolves were hunted in these forests and its grassland peripheries.
However, to enter Hyrcania, the Romans had to pass through a narrow gorge known as the Southern Caspian Gates that cut through the Alburz Mountains. The legionaries taken prisoner at Carrhae had been marched through this bleak mountain pass, so the Roman authorities already had harrowing eyewitness accounts of the region. Pliny describes how the pass had been cut through the rock by Persian engineers, but the 8-mile roadway was scarcely broad enough for a single line of wagon traffic. He reports that the gorge was ‘overhung on either side by crags that look as if they had been burnt by fire and the narrow passage through the gorge is only interrupted by a salt-water stream’. Roman reports suggested that the surrounding country was almost entirely waterless for a range of 28 miles. The permanent mountain streams were saline and fresh water could only be obtained with the melt of winter snows. Consequently, the region would present serious challenges to any infantry-based Roman army trying to capture this strategic position from Parthian cavalry forces. Pliny concludes, ‘The Parthian kingdom is effectively shut off by passes.’
Pliny had read Roman itineraries that used information from Alexander’s campaigns to chart invasion routes through Iran. The Southern Caspian Gates was a central strategic point in these studies since it was estimated to be almost 600 miles from the River Jaxartes (the Syr Darya) where Alexander fixed the northern limits of his conquests in Sogdia. The gorge was also calculated to be about 450 miles from the Bactrian capital Balkh and 2,000 miles from the northern frontiers of ancient India.
A modern survey of these routes has confirmed the accuracy of these ancient figures. Pliny reports that the distance from the Caspian Gate to the Parthian capital Hecatompylos was 133 Roman miles (122 modern miles). The distance measured using modern techniques is close to 125 miles along a course that probably deviates only slightly from the ancient pathways.
Isidore outlined a route into the fertile lands of Hyrcania for any Roman forces that captured and held the Caspian Gates. Beyond the Gates, the legionaries would arrive at a narrow valley that led to the Iranian city of Apamia. From there, the invasion course had to turn east and occupy another line of villages equipped with caravan stations that probably operated as Parthian military outposts. There were no cities in this region and the Romans would travel through thirty-five villages with stathmoi (stations) on their route through Hyrcania. Only then would they reach the frontiers of the region known as Parthia and the original homelands of their enemy.
An Iranian city called Asaac (Arsak) was on the western frontier of Parthia. It was here on the southeast shores of the Caspian Sea that the founder of the Parthian regime, Arsaces I had been proclaimed king by his steppe followers after they had settled the region (250–211 BC). Isidore reports that Asaac was an important centre for an ancient Iranian religion known as Zoroastrianism and a sacred everlasting fire was maintained in the city temples.
Near Asaac was the fortified city of Nisa (Parthaunisa) which was the location of ancient royal tombs belonging to the earliest Parthian rulers. Excavations at this site, near Ashgabat in Turkmenistan, recovered carved ivory drinking cups or ceremonial libation cones known as rytons. Other finds from the city include thousands of fragmentary administrative records from the Parthian regime written on clay tablets in Persian script. These texts document deliveries of wine and other produce to the Parthian administrators at Nisa. They also record military titles including Border-Wardens and Fortress-Commanders who oversaw the conveyance of cash crops to the royal centre.
There were no travel-stations in this part of Parthia because the region already had sufficient cities to accommodate caravans and facilitate the movement of mounted armies. North of Parthia was the Eurasian steppe, but the lands to the south were covered by desert. This meant that any Roman invasion route had to pass directly through the region. Isidore lists a series of Parthian cities that would have to be captured on any campaign through this territory, including Gathar, Siroc, Apauarctica and Ragau. This would complete the anticipated Roman conquest of Parthia, but further east there were other territories subject to Parthian rule that might also be claimed for the Roman Empire.
Isidore outlined a route from Parthia east into Margiana that would have allowed Roman forces to take possession of the oasis site of Merv. The territory around Merv was almost entirely devoid of any settlements, but there were two Parthian villages on route to the oasis. Here Roman commanders expected to find the captive legionaries who had served under Crassus (53 BC) and Mark Antony (36 BC). By the 20s BC many of these prisoners would have spent most of their adult lives under Parthian governance.
The Romans received reports that Merv was enclosed by mountains that formed a 187-mile circuit around the oasis. Beyond the mountains was a large expanse of desert that extended for at least 120 miles to the east. The oasis at Merv received water from the Murghab River which flowed more than 500 miles from mountains on the edge of northwest Afghanistan into the Karakum Desert in Turkmenistan. Curtius records that Alexander the Great established six Hellenic towns on hill sites near Merv, ‘spaced only a short distance apart so that they could seek mutual aid from one another’. Pliny claims that Alexander also established a city near the river, but the settlement was abandoned or destroyed by enemy forces. Antiochus I Soter reclaimed the oasis by founding a walled Hellenic city called Antiochia Margiana close to the river (281–261 BC). He enclosed the countryside surrounding this city with a wall measuring almost 8 miles in circumference.
Archaeological remains and records from later eras suggest the appearance of this ancient territory. A Chinese soldier visited the city of Merv (Mulu) in the eighth century AD after he had been held captive by Iranian forces. He saw a caravan city surrounded by walls that were 3 miles in circumference and had iron gates. When he returned to China he reported that the ‘walls of the city are high and thick and the streets and markets are tidy and well-arranged’. The remains of ancient clay-wall barriers have been found stretching across certain northern districts of Margiana. These defences were probably built to protect the territory from mounted raiders, but the remains cannot be securely dated. They could be Hellenic, Parthian or perhaps Sassanid defences (AD 224–651) built or repaired by native peoples, or perhaps foreign prisoners of war.
Pliny describes how Margiana was ‘famous for its sunny climate’ and received recognition as one of the few territories in Parthia where grape vines were cultivated. Strabo emphasises the wine production at this site and describes fertile soil suitable for viticulture. He reports that ‘vine stocks are found that require two men to girth [10 feet circumference] and bunches of gapes grow to 2 cubits [3 feet]. This suggests that viticulture might have been well established when the first Roman captives were brought to Merv in 53 BC.
Many of the Roman captives were probably settled as agricultural labourers in towns near the city of Antiochia Margiana. Some of these Italian captives might have had pre-war experience in viticulture which was a valuable skill. One of the Parthian clay tablets recovered from the royal city of Nisa records wine deliveries from the oasis (before 40 BC). The delivery was arranged and overseen by two ‘Tagmadars’, a Greek title that designated unit officers. These men had the Parthian names Frabaxtak and Frafarn, but they could have commanded labour teams of Roman workers assigned to royal vineyards. A further possibility is that some Roman captives adopted Parthian culture and received the titles and responsibilities of their new regime. Horace asked his reader to imagine their fate: ‘Are the soldiers of Crassus, men of Marsi and Apulia, living under Median rule, joined in shameful marriage to foreign wives?’