Wales at the end of the Bronze Age appears to have experienced a marked deterioration in the climate. The emerging competition for food and conflict between neighbouring peoples was one of the characteristics of the Iron Age, beginning in about 700 BC, expressed most vividly in the proliferation of defended settlements. With the Roman occupation, from about AD 47 to about AD 410, enormous changes were visible in the landscape, but there was much continuity in the character of everyday life for most people.
There have been considerable advances in understanding of the Iron Age – the crucial centuries prior to the Roman invasion. Traditionally, it has been identified with the arrival of iron tools, the emergence of new types of defended enclosures, and the absence of the ritual monuments dominant in earlier periods. This seemed to signal greater territoriality and aggression as a response to the climatic deterioration. However, the complex changes in the landscape suggest that there was not a sudden break at the beginning of the Iron Age. This was a time of developing complexity in society, of enormous endeavour in the construction of communal settlements, and of achievements in metal-working and art. Long-distance trade brought overseas contacts in Europe, including the expanding Roman empire. The term ‘Celtic’ is a useful shorthand, but while Celtic traditions influenced artistic expression, language and perhaps social structures and methods of inheritance across much of Europe, there was not a uniform culture. Local and regional leaders presided over well-defined territories that in some parts of Wales may have been as densely settled and farmed as they were in later times. Settlements continue to be discovered today, and those in the lowlands particularly contribute to understanding of possible population densities and territorial divisions.
Hillforts have long dominated research on the Iron Age. However these complex structures are just the most visible of the variety of defended enclosures that date from this period, the size and character of which varied from ring-forts and raths in lowland Pembrokeshire to the large earthen hillforts of central Wales and the border and the stone-built fortresses of north Wales and the Ll?n peninsula. A. H. A. Hogg of the Royal Commission, who spent much of his life advancing the study of British hillforts, wrote in 1975, ‘No archaeologist is satisfied with the term… the enclosures may have corresponded to anything from a cattle kraal to a small town, but were seldom exclusively military.’ In Wales they were usually heavily-defended hilltop enclosures with encircling earth or stone ramparts topped with timber palisades and entered through well-defended gateways. Inside there were thatched round houses, raised granaries, smithies and industrial areas.
The study of Welsh hillforts began in the later nineteenth century, with work on the Clwydian hillforts by Wynne Ffoulkes and S. Baring Gould’s study of Pembrokeshire hillforts. The Royal Commission’s early [5-5-1]inventories described hillforts, but the monuments were still poorly understood in archaeological terms. Before the Second World War it was still assumed that they were built by invading immigrants who arrived in southern England and spread to the rest of England and Wales. It is now clear, on the contrary, that hillforts were indigenous creations, though they were influenced by developments in mainland Europe. While some have been shown to have originated in the later Bronze Age, the majority were probably built in the later centuries of the Iron Age as wealth from the land increased and powerful leaders consolidated their territory. Many continued in occupation and flourished in Romano-British times.
Impressive entrance defences and gateways at even small hillforts served as deterrents to would-be attackers and as statements of power and status. When freshly built, the ramparts and entrances would have been even more striking than they are today. The greatest of them were vast in scale and unprecedented in the effort invested to create them. Additional protection might be provided by chevaux de frise, upright stones or timber stakes outside the ramparts. These are rare in Wales except for a handful of examples in Gwynedd and Pembrokeshire, though more may await discovery – timber examples in particular will have left few traces. However, it is still debatable whether hillforts should be viewed as military strongholds in a dangerous world or, rather, as farms and villages designed to ward off aggressors through the appearance of strength. Excavated finds and chance discoveries indicate that many hillforts were armed with the most basic – yet effective – of ammunition, the sling-stone. In experienced hands these were accurate and deadly, with a range of 60 to 100 metres. Roman histories of the conquest of Britain, as well as archaeological evidence, indicate that Roman troops clashed with Iron Age tribes.
A useful analogy when considering the role of hillfort defences may be the post-medieval Scottish chiefdoms of the highlands and islands, for which feuding and feasting were the principal forms of display by chiefs: aggression took the form not of sustained warfare but of periodic raids on rival clans, destroying crops and stealing cattle. The existence of this kind of localised, intermittent, and unpredictable conflict in the Iron Age is supported by archaeological finds of ostentatious weaponry, chariot fittings and horse harnesses. The culture required leaders to assert their political status through investment in elaborate gates and walls, and reasonable defences for settlements and fields, but it is likely that many hillforts never experienced attacks. Iron Age leaders would doubtless have worked to retain their positions by forging alliances with rival communities, helped by impressive hillfort architecture that demonstrated their power.
Hillforts were only one form of settlement, but they have survived conspicuously by virtue of their massive engineering and their siting on marginal land. On surrounding slopes and along valley floors were a plethora of lowland forts, small defended farms and concentric corrals – specialist sites consisting of a larger outer enclosure and an inner yard for settlement. In many parts of Wales, Iron Age farmers sited their settlements on the most productive agricultural land, where farmers from medieval times to the present day have continued to cultivate. Decades or centuries of ploughing have gradually worn away defences and other traces. However, aerial photography has rediscovered hundreds of these settlements, the ghosts of which are visible in patterns of differential crop growth. Many parts of pre-Roman Wales would have been busy, industrious, and cosmopolitan places. Great field systems extended across parts of the country, some first laid out in the Late Bronze Age. Goats and sheep provided meat, dairy products, leather, and wool, and helped to maintain pasture against the return of forests. Woodland management, hunting and fishing supplemented the agricultural economy and diet. Cattle were highly prized in Iron Age societies as sources of wealth and for their additional use in traction – they were celebrated as iconic beasts in art and metalwork.
Networks of trackways (precursors of historic droving routes) connected communities across the country. Finds of salt brought from the Cheshire plain, fine ceramics, and imported goods from more distant parts indicate a lively economy. Although coastal and river transport would have been important for many purposes, there is evidence that overland routes were used. It is significant how quickly the Roman armies established bases and supply depots in the Welsh heartland and penetrated the hills through mountain passes, seemingly drawing on local knowledge of existing routes.
The advent of the Romans led to the death or enslavement of many of the native peoples. Subjugation of the tribes took about thirty years after AD 47, in the face of stubborn resistance organised by leaders such as Caratacus. However, Agricola’s defeat of the Ordovician rebellion of AD 77 brought an end to large-scale campaigning. From then until Roman rule collapsed in the fourth century, tight control was exercised from strategic fortresses at Chester and Caerleon and an infrastructure of lesser installations that maintained law and order. The most intense period of garrisoning followed the final conquest. Within a few years military units were being moved to the north of Britain and by the mid-second century only a few forts in mid-Wales and Segontium on the north coast were maintained to control the nearby populations. The latter part of the third century saw the coast defended from seaborne marauders with a fort at Cardiff and refurbishment elsewhere. The garrison at Caerleon was probably withdrawn by the end of the century, though it may have had a skeleton force as late as the mid-fourth century. The military dispositions in the fourth-century, at Cardiff, Caernarfon, Holyhead and Caerhun seem to have been concerned with incursions by Irish bands. All recognisably military posts seem to have been abandoned in about AD 393 when soldiers were needed to counter rebellion in Gaul, though there is some evidence that troops were left to guard the towns of Carmarthen and Caerwent even into the fifth century.
The Roman armies on campaign were lodged in temporary marching camps. Once an area was pacified the troops were billeted in forts of various sizes from which they could patrol and collect levies. The headquarters of the legions responsible for maintaining the ‘Roman Peace’ (Pax Romana) were the fortresses of Chester and Caerleon. These controlled the land in their vicinity: it is thought that the second Augustan legion drained the Gwent levels to supply pasture for horses. Inevitably, military bases attracted settlements of merchants, craftsmen, entertainers, and also the troops’ unofficial spouses. Such settlements were called canabae outside legionary fortresses or vici outside auxiliary forts. The system of roads was constructed under army direction in the decades after the Flavian conquest. The two settlements that developed into urban centres (civitates) for the tribes of the Silures and Demetae were Venta Silurum and Moridunum, respectively Caerwent and Carmarthen. Caerwent had many of the features of Roman cities, such as a forum, temples and well-appointed houses. Carmarthen’s best-known feature is the amphitheatre outside the north-east gate. Besides acting as relatively secure homes for those of the tribal aristocracy who accepted imperial rule, the cities were centres of trade, manufacture, and tax collection.
Roadside settlements, or small towns, were a common feature of Roman Britain, but in Wales they were mainly confined to the south-east, usually developing at former army sites. Usk, Monmouth, Cardiff, Abergavenny, Cowbridge and probably Chepstow are examples. A feature common to most was the production of iron objects. There may have been two similar settlements near Ruthin, and at Ffrith in Flintshire a small settlement produced lead and silver.
Villas – well-appointed houses with associated agricultural buildings, showing signs of acceptance of Roman values and lifestyles ? are found on fertile lands and within reach of urban centres. The distribution of known examples reinforces the idea that the lands of the Silures were much more integrated than the territory of the Ordovices – the settlement forms in north-west Wales appear not to have been influenced significantly by Roman styles, whether through cultural resistance or limitations of investment. Equally, in mid-Wales and the territory of the Deceangli in the north-east there is little evidence of Roman influence on rural settlement, except at Plas Coch, Wrexham. By contrast, economic conditions and the disposition of local rulers in the lands of the Demetae in the south-west seem to have favoured the building of small villas, such as Trelissey in Pembrokeshire and Cwmbrwyn in Carmarthenshire.
The Silures, who offered the greatest initial resistance to the Romans, inflicting a humiliating reversal on a campaigning legion in AD 51, were the people that most readily embraced aspects of Roman culture once they had been defeated. The presence of the second Augustan legion at Caerleon would have been a major influence, but there must have been a conscious adjustment by the ruling classes from the end of the first century. The proliferation of villas in Silurian territory reflects the wealth accumulated by a class of landowners who had thrown in their lot with Rome. Some might have been able to retain their ancestral homes and develop them in the Roman style, as at Whitton in Glamorgan. A villa estate was a world away from the constructions of pre-Roman times – with masonry buildings, imposing architectural facades, tile or stone roofs, a bath house, a hypocaust heating system, mosaics and wall paintings.
Local dignitaries would have been introduced to imported goods such as oil, wine, pottery, glass and ceramic lamps when guests of senior army officers. The introduction of coinage to Wales signalled a new system of exchange that lasted until peace and stability were undermined. Coinage was essential to pay taxes, but judging by the paucity of coin finds, non-monetary exchange persisted alongside the cash economy throughout the Roman era. The continuing presence of military and imperial officials who required supplies, the development of a money economy and the growth of new markets around towns in the south encouraged agricultural production. In upland areas pastoral agriculture was the norm, but the balance between cattle and sheep might vary from place to place and there was localised crop growing. In the lowland valleys and coastal zones arable farming usually predominated.
Extractive and manufacturing industries operated at various times and places, encouraged by improvements in sea and road transport that enabled exports. The imperial government was interested in the extraction of precious metals, if only to offset the costs of maintaining a large army. Gold was probably mined at Dolaucothi in Carmarthenshire, though the total haul from Wales must have been disappointing. More lucrative was the extraction of silver for coinage: ingots, mines and settlements have been found at Prestatyn, Ffrith and Pentre in Flintshire. Copper mining is attested by ingots from Anglesey and Caernarfonshire and adits at Llanymynech on the Montgomeryshire-Shropshire border. However, disarming the population resulted in a decline in the production of iron until settled conditions developed.
Several manufacturing industries were stimulated by the needs of the military and the growing taste for Roman goods, particularly in the south-east. The army needed tiles, and kilns were built near bases, for example at Holt, to supply the twentieth legion at Chester. As garrisons became settled, itinerant potters set up kilns to supply army needs, while the taste of senior ranks for better quality tableware was satisfied by Rhenish and Gallic imports. Kitchenware manufactured in the west of England dominated the market during the second century and found its way to a wide range of settlements. After the mid-second century there is little sign of pottery being produced in Wales.
Classical-style art could be seen in military centres and the Romanised south-east. Stone sculpture was particularly associated with the army, especially at Caerleon, where it was in demand for burial monuments and religious objects. Fragments of statuary of imperial subjects are known from the towns, mosaics adorned villas and urban properties in Caerwent, and wall painters decorated country houses, military buildings, and private and municipal edifices in towns.
The native inhabitants possessed greater sophistication of ideology and religion than Roman propagandists admitted. Many of the deities of native religion continued to be revered, even if sometimes associated with gods of the Roman pantheon. The army introduced the principal Roman gods, headed by Jupiter; later, eastern cults popular with the soldiers gained a footing, such as that of Mithras at Caerleon and Segontium. Other cults popular in Romanised areas included those of Mercury, Fortuna and Mars. Romano-Celtic temples were built at Caerwent and Carmarthen, and a circular stone temple at Gwehelog in Monmouthshire. Isolated finds of votive objects suggest the widespread presence of small shrines, such as Llys Awel at Abergele where a statuette of Mercury and three models of dogs have been found. The evidence for Roman Christianity is not strong in Wales, but its presence was felt: there are the possible remains of a church at Caerwent. Cremation was the norm until the mid-third century, ashes being deposited in cists or pottery vessels, sometimes accompanied by personal objects; inhumation prevailed later, the bodies of wealthier individuals being placed in stone coffins. Long-term cemeteries like those near Caerleon were marked by gravestones for both soldiers and civilians, and some mausoleums. A few barrow cemeteries are known, for example at the military complex of Tomen-y-Mûr in Merioneth.
Wales shared in the Roman empire’s political, economic and social tensions as it struggled against fragmentation in the fourth century. Despite worsening conditions, urban life continued at the end of fourth century at Caerwent and Carmarthen. However, the threats to stability are shown starkly by defensive installations at Cardiff, Caer Gybi and Holyhead Mountain and the maintenance of garrisons at sites like Segontium, almost certainly to deter and combat raiding by Irish bands. Peace and stability were essential to the limited market economy. As these were threatened and coin ceased to circulate, the Roman economic system probably broke down rapidly.