Hillforts after 300 BC
Text by Barry Cunliffe
The story of the late phase of Danebury Ring, though obscure in points of detail, is well established in outline and can act as a guide to the interpretation of other sites in the region. In the immediate neighbourhood it seems that the forts at Figsbury, Quarley and possibly Bury Hill were abandoned sometime around 300 BC, leaving Danebury as the only fortified enclosure on the downland block between the Test and the Bourne. Woolbury, too, seems to have remained largely unoccupied, though a few pits were dug in the interior.
A similar picture can be traced elsewhere in southern Britain and is particularly clear along the downs east of the Itchen where each area of chalkland, isolated by river valleys, seems to have been dominated at this time by a single hillfort: St Catherine’s Hill, OldWinchester Hill, Torberry, Trundle, Cissbury, Devil’s Dyke and Caburn. A similar situation can be traced over much of Wessex.
Wherever there has been sufficient excavation it can usually be shown that the hillforts that survived – known as ‘developed hillforts’ – grew out of existing settlements or enclosures. This rise to dominance of selected sites must reflect a significant, and apparently quite rapid, change in the socio-political structure – it looks very much as though there was a sudden coalescence of power in the hands of a smaller number of communities.
At Danebury Ring, as already noted, the new defensive line followed exactly that of the original circuit and this probably happened in the majority of cases, but some forts were greatly extended at this time. A prime example is Maiden Castle where the fort was more than doubled in area when the fortifications were thrust eastwards to embrace the crest of an adjacent hill. At Hambledon Hill, overlooking the Stour Valley, at least two successive enlargements can be traced by a careful study of the existing earthworks, even though the site has not been excavated, while at Yarnbury (Wiltshire), excavation has shown that the original circuit was totally abandoned and a new system of defences built outside it to enclose a much larger area.
Wherever it is possible to check the date of these reorganisations they seem to coincide with a development in pottery styles equivalent to the transition between the middle and late periods at Danebury Ring. They are likely, therefore, to be of approximately the same date but it is impossible to say how sudden this change was and we must allow that the evolution from densely scattered early hillforts to more widely spaced, developed hillforts may have taken anything up to a century to complete. The archaeological data, at present available, do not allow greater precision.
The century following 300 BC was, then, a period of dramatic change in Wessex when centres of power were crystallising out, but there is no reason to suppose that they were all of equivalent status, or that the status of each remained static during the next 200 years. A more likely model anticipates a complex situation with some forts pre-eminent and commanding the allegiance of lesser forts but with the patterns of allegiance changing as the fortunes of the major centres rose and fell. It is entirely beyond the limits of the present evidence to discern even the broadest direction of such a dynamic situation but the individual forts which have been excavated show a wide variety of patterns.
At Winklebury (northern Hampshire), although the early fortifications were rebuilt, the community within appears to have remained small and lacked any sign of careful organisation. Danebury, on the other hand, was densely occupied throughout and developed grandiose entrances late in its life, while Maiden Castle grew to colossal proportions with a vastly complex system of multivallate defences and with most unusual double gateways. The three examples are sufficient to demonstrate a range of variation reflecting the very different political positions occupied by these sites in the contemporary settlement hierarchy.
The recent excavations at Bury Hill, only 6km (4 miles) to the north of Danebury have shown that sometime towards the end of the late period of Danebury, probably late in the second century BC or early in the first, the old, long-abandoned fort was suddenly brought back into active use when a smaller circuit was defended by a massive ditch flanked inside and out by substantial ramparts. The style of defence differs from that of traditional hillforts and looks forward to the latest Iron Age fortifications characterised by Suddern Farm.
Inside, occupation was quite intensive and the finds, which include large quantities of horse harness and chariot fittings and an unusually large number of horse bones, strongly suggest an aristocratic warrior settlement. Suddenly then, Danebury Ring, after more than two centuries of dominance, now had to face competition. Who the builders of Bury Hill were and the nature of their relationship with Danebury will never be known, but soon after, the long occupation of Danebury came to an end.