Wadi Faynan was one of the biggest copper mines in the Roman Empire. It lies south of the Dead Sea on the east side of the Wadi Arabah which separates Israel from Jordan. The first intensive modern research was conducted by the German Mining Museum at Bochum from 1983 under the direction of Andreas Hauptmann and Gerd Weisgerber. They noted that the production of copper from Faynan was in quantities ‘unparalleled in the southeastern Mediterranean, with the possible exception of Cyprus’, with up to 200,000 tons of slag left behind. They identified mining and smelting dating from the Chalcolithic period (4500 – 3100 BC) and running fairly continuously until the 4th century AD. A final phase of activity occurred during the Mamluk period (1250 – 1516 AD), although it has been suggested that this involved reprocessing some of the ancient slags.
The ore deposits are part of the same geological formation as those at Timna on the western (Israeli) side of the Wadi Arabah, which has been better documented, and where more mines have been identified. However the size of the Faynan mines suggests a more intensive exploitation than at Timna. Recent investigations conducted by the Jordanian Natural Resource Agency have shown that Roman mining largely exhausted the ores. In the face of recent proposals for a major dam project, Hauptmann was able to argue that while other copper mines may have been more important in the past, only in Faynan is such an extensive suite of remains preserved. The Jordanian Department of Antiquities is considering whether to apply for World Heritage Status for the industrial landscape.
In 1994 I was commissioned to conduct an archaeological survey for the RSCN, and shortly afterwards became involved in the BIAAH project being established by the new Director, Alison McQuitty (continuing from 1998 as the Council for British Research in the Levant, CBRL). This was designed to be a ten year research programme under the patronage of Princess Sumaya bint Hassan.
The main Roman and Byzantine centre at Faynan is a series of ruins known as Khirbet Faynan (Khirbet means ruins) where two of the major streams that rush down from the mountains in the winter join together to form the broad Wadi Faynan. It is a large mound covering an area of about 70,000m², and what survives on the surface are some surprisingly well preserved Byzantine remains. From the Roman period there is also a large reservoir, an aqueduct and a water mill. It was probably the site of the Roman settlement mentioned by Eusebius known as Phaino, and possibly the site known in the Bible as Punon (Numbers 33, 42-3) and documentary evidence shows that the town was the centre of a bishopric in the 5th and 6th centuries.
One of the big surprises was the discovery and excavation of two Neolithic villages, only 100 metres apart, up one of the side valleys down which a stream flowed down to the plain. Both belonged to the Pre-pottery Neolithic, one to the Pre-pottery Neolithic A (PPNA), the other Pre-pottery Neolithic B (PPNB). The PPNA site of Wadi Faynan 16 which I have been excavating with Steven Mithen has some very early radiocarbon dates starting c.10,900 BC calibrated.
Architecturally it was a surprisingly complex site which lay well beyond the Mediterranean woodland zone seen as the core area for this period. Even at this early date people had begun to exploit the local copper ores, using them in their native state for beads and possibly for pigment. The PPNB site of Ghuwayr 1 and the pottery Neolithic site of Tell Wadi Faynan was excavated by Mohammad Najjar of the Department of Antiquities, initially with the German team, and later with a team from the University of Nevada. The dates from the PPNA and PPNB sites overlap, and Faynan is very unusual in the southern Levant for preserving the period of transition.
Around 4,000 BC the environment became more arid, and settlement expanded out into the main wadi, and in the Early Bronze Age (around 3,500 BC) more structured systems of floodwater farming were developed. The field systems still visible maintain some elements of the earliest systems, indeed it appears that some of the walls have a long history, being repaired and refurbished as sediment accumulated around them – making them much deeper and more substantial than is seen on the surface. As mining and ore processing became more intense in the Iron Age both farming practices and smelting became more sophisticated under the Nabatean kingdom. When the Romans annexed the Nabateans in AD 106, this led to a huge increase in activity. In particular the organisation appears to have been centralised, and the water management system became a single integrated system. Pairs of parallel walls, once interpreted as trackways, are now seen as a series of water conduits, and have been excavated to reveal well made clay linings, with Roman pottery dating their construction. Not surprisingly the major industrial intensification noted by the Germans had to be matched by an agricultural basis that could feed the workforce.
Located on the south bank of the Wadi Ghuweir, opposite Khirbat Faynan, are the substantial remains of a water management and storage system. The main components of the system are an open channel, an aqueduct across Wadi Sheger and a large sunken reservoir. From the reservoir, water was possibly channelled to a watermill and possibly on to the extensive field system beyond. Further west there are the remains of an aqueduct that originally spanned the Wadi Sheger over a distance of approximately 120m. A series of arches would have been used to span the wadi. The aqueduct would have carried water to an open channel that is currently buried under large slag heaps. The team from the Bochum Mining Museum has dated these slags to the Roman period, between the 2nd and 5th centuries AD, which helps to confirm the date of the rest of the system. The channel emerges further west, leading right into the settlement tank of the reservoir.
The water mill is a raised tower or arubah penstock mill. These are fairly common in the Near East. The Wadi Faynan mill is an example of a single tower mill similar to others in the wadis flowing into the Jordan Valley. Remains of the upper channel that brought water to the mill are still visible for more than 11m. It is one course high and plastered on the inside. This channel leads to an open chamber at the centre of which is the circular opening of the vertical shaft lined with plaster. The opening is 0.45m in diameter and at least 3.3m deep. The raised tower is well preserved, standing to around 4m above ground on the outside. A chamber abutting the foot of the tower is practically demolished, its remaining walls being preserved to a maximum height of only three courses. Recent restoration work has revealed the wheel chamber still surviving below this mill room. It would be logical to assume that the mill is part of the same system as the aqueduct and water channels to the east and is therefore also Roman in origin. However, penstock mills in Jordan are generally of Islamic date and if the Faynan mill proves to be Roman it will be unique in Jordan.
The mining came at a cost: the area became heavily polluted, and Graeme Barker’s team from Leicester revealed the extent of the pollution. Their geochemical analyses have revealed a record of changing levels of pollution with heavy metals trapped in the sediments. The pollution can be traced back to the Early Bronze Age, with marked increases in the Iron Age and the Nabatean period, and then an enormous increase associated with the Roman arrival. By the late Roman/Byzantine period pollution has declined, with a later peak (with a different suite of pollutants) apparently matching the evidence for Mamluk activity recorded by the German team.
Barker’s team have also examined the pollen record: this and other palaeoecological studies indicate that the landscape continued to degrade during the Bronze Age, possibly due to the activities of farmers, but possibly also the result of continuing climate change. The landscape was still further degraded by the Roman period. This is more obviously the result of intensive farming and the presumed enormous use of wood for fuel in smelting. Charcoal analysis from the smelting sites suggests that by this period, timber for the furnaces had to be brought down from the plateau because none was available locally. Barker’s team suspect that the extensive sherd scatter throughout the field system may be the result of manuring, where the pottery is thrown onto the fields along with all the organic waste. They argue that increasing pollution will have diminished crop production, making manuring even more important for soil fertility.
Who worked the mines? The sample of skeletons excavated in the late Roman to early Byzantine cemetery has revealed that part of the population suffered from severe osteoarthritis. This is significantly higher in males (33.3%) than females (23%). Although osteoarthritis is present in most populations, and indicative of ageing, the pathology increases in severity as a result of heavy physical work. Apart from this however the skeletal evidence shows that the population was generally healthy. However, analysis of the metal content of skeletons has shown that many (44%) had absorbed concentrations of copper and lead at levels indicating industrial exposure. The pathologies showing evidence of manual labour have been used to argue that most of those with normal metal contamination (i.e. not exposed to the polluting air of Faynan for any length of time) probably died soon after they arrived in the area – though this is asomewhat far-flung speculation.
The main historical evidence comes from Eusebius, a somewhat unreliable source as far as Christian persecution is concerned. He claimed that during the early 4th century persecutions by the Roman authorities, Phaeno was used as a place of exile for Christians mainly from Gaza and Egypt. Indeed he reports that so many Christians were present in the early 4th century that Silvanus, a convict from Gaza, served as Bishop until he was executed by the authorities (Eusebius, Bishop of Caesarea). Bishops were recorded at Phaeno from 431 until 587 when Theodore is mentioned as Bishop on a building inscription from Khirbat Faynan.
Eusebius indicated that life expectancy in the mines of Phaeno was short. Grattan and his co-workers have suggested that the greatest metal poisoning may therefore have affected the overseers and specialists who directed the penal slaves. They may have escaped the brutalizing treatment handed out to the slaves, but have suffered the serious health problems associated with copper and lead poisoning and dust inhalation. Nevertheless, the skeletal evidence suggested that the population was generally healthy.
Sadly, some of the effects of the mining pollution remain present today. Continued aridity and grazing pressure has resulted in the disappearance of some plant species and possibly associated herbivores. These effects are even more pronounced on the metalliferous spoil tips, which remain highly toxic. Radon gas, which rapidly decays in the atmosphere to isotopes which are radioactive and dangerous and may promote cancers of the lung is present in the mines at elevated levels.
This pollution may affect the current longterm residents of the area. Without the sustained interest in copper it has become the home to nomadic people. Ethnoarchaeological fieldwork was conducted to look at the recent and contemporary use of the landscape by Bedouin, and to provide information on the potential signature of Bedouin camp sites for the survey. The most common form of Bedouin tent, beit sha’ar, has three centre poles allowing the tent to be divided into up to four sections. In winter, a common arrangement is to divide the tent in three and shelter goats overnight in one half of the tent, dividing the other half into men’s and women’s sections. The men’s section is where guests are received and its main feature is an excavated central hearth, which may be stone-lined. There is another hearth in the women’s section, where food, particularly bread, is cooked. Mattresses are often stored on a raised platform made from large stones and some groups define a sleeping area with stones. Goats create thick layers of dung in the winter which requires tents to be moved. When the weather becomes hotter in the summer people change their goat hair tents for cooler, sacking tents and move from sheltered wadi-edge terraces to cooler, breezier, ridge-top locations. Summer camp sites are more ephemeral in their nature than winter camps. Other families move up to the plateau for cooler conditions, although increasingly school has disrupted traditional mobility. Now there are small settled villages developing and there is a greater need for studies to assess the potential harm from the ancient pollution.
The research in Wadi Faynan is informing us of ancient human settlement, the very early beginnings of sedentism and farming, the development of pastoralism, and important stages in copper mining and working. At the same time it provides a fascinating insight into how people can develop different strategies to live in a changing environment.
This article is an extract from the full article published in World Archaeology Issue 13. Click here to subscribe