These last few months have been spent in a form of devilish torment. Our museum conservators, helped and hindered by my input, have been piecing together a silver jigsaw puzzle – pieces of Roman vessels, many only millimetres across. They have come together to reveal a remarkable find from Fife, on the east coast of Scotland: a hoard of Roman hacksilver, the earliest known beyond the Empire’s edge, which casts fresh light on Roman frontier politics in the later 3rd century AD.
What is hacksilver? It is a term used for items, mostly silver vessels, which have been deliberately cut, chopped, and crushed into fragments – converted from beautiful works of art into a weight of silver. Although this habit has a long pedigree in the Old World, our main interest is with the Roman period. Hacksilver has long been the ugly sister of Roman silver research – scholars have preferred hoards of intact vessels, such as Mildenhall (see CA 229), to battered fragments. If they were considered at all, they were usually seen as the fault of raiding barbarians who lacked the refinement to appreciate such fine Classical art.
Yet the picture is more complex. Over the last few years, a research project based at National Museums Scotland has been looking at this hacksilver in the context of the great Traprain Treasure, the biggest and best of these hacksilver hoards (CA 238). The work of the late Kenneth Painter in particular has shown that this was no barbarian habit: hacksilver is found on both sides of the Roman frontier, especially in the north-western provinces, from the 2nd century AD to the 6th, though with a strong focus in the 4th and 5th centuries. Kenneth also showed that this silver was treated as bullion: much of it clearly correlated to multiples or fractions of Roman pounds and ounces, indicating the use of Roman weight standards. Either our barbarians were terribly careful with their brutal hacking, or there was a Roman economic motive behind this process – hacksilver was a process of turning objects into bullion at times of need.
One of those needs was dealing with barbarians. Hacksilver was sent north, whether as diplomatic gifts or payments for mercenary services, to Ireland, Denmark, the Netherlands, as far as Ukraine – and to Scotland. For a long time, this ‘barbarian’ hacksilver was seen as a phenomenon of the late 4th and 5th centuries AD. The new Fife find changes this. For the first time, we can see this policy of bullion payments starting earlier, in the late 3rd century.
Detecting and digging
As with so many finds of valuable metals, this one started with a metal-detecting discovery. The field in eastern Fife had no known archaeological significance, but it was good farming land and, critically, it was harvested and available for a Detecting Scotland rally. Over 100 detectorists were searching in the field, but it was schoolboy David Hall, then 14 years old, who found the first fragments. He quickly realised their significance, and over 200 pieces of silver were recovered that first day. Most were tiny scraps, for this hoard had been hacked twice – first by the Romans, and then by the plough, which had scattered the fragments widely. Only one piece of silver remained in situ.
The hoard was quickly reported to Scotland’s Treasure Trove Unit, who got me involved, knowing my taste for both Roman silver and muddy fields. Working with a group of detectorists, including the original finder and his dad, we scanned the area methodically, plotting further scraps, and then carefully machine-stripped spits of soil from a large area around the main findspot, discovering and plotting many more silver sherds as we went. Then, at the base of the ploughsoil, traces of features began to appear. Careful cleaning showed that there was more here than the post-medieval field drains that showed at first sight. The hoard had been placed between a small peat bog on one side and a cluster of features on the other. The peat bog is in an odd setting, perched on the edge of a terrace, and may arise from a former spring – such wet places were often used for offerings.
The group of features was equally intriguing, for two of them contained stumps of standing stones, one in association with probably Bronze Age pottery. This hoard had been buried in a memorable, perhaps even a sacred site, between an intriguing wet spot on one side and, on the other, two standing stones, which were already ancient by the time the silver was buried. Such burial of valuables in association with prehistoric monuments can be readily paralleled elsewhere. It suggests this silver was placed under the care of the gods, probably as a sacrifice rather than a burial for safekeeping. Geophysical survey showed no trace of any occupation in the field, and we remain in the dark over where the nearest settlements were – this is a target for future work.
Pieces of silver
The 408 fragments of silver that we recovered came from only four vessels in total, and the one least plough-scarred was the most puzzling of the lot: a silver cylinder of rather thick metal, which preserves a rounded lip on part of its edge. The surface is irregular, and this may be a flawed casting of a blank intended for sheet-working, or a partly recycled piece. As mentioned before, only one piece of the hoard had remained in situ, though unfortunately the detectorists lifted it themselves, so we couldn’t sample the fill of the shallow pit it lay in. This was a folded quarter of a large flat dish, about 360mm in diameter, with a beaded rim and an ornate niello-inlaid medallion at its centre. Fragments of a second quarter, badly damaged by ploughing, were also recovered – when whole, this would have been an impressive sight.
The third item was part of a fluted bowl, originally used for washing one’s hands at the dinner table. This had initially been cut in half, and each half then cut again, and two pieces, each around a third of the vessel, had been folded into packages and buried with the hoard. Both show traces of attached handles, which had been removed before it was chopped up.
The final vessel is the most remarkable, but also the most difficult to reconstruct. It was made from thin, highly fragile silver, and seems to have been carefully placed, intact and upside down, to act as a cap to cover the rest of the silver. Unfortunately, this arrangement left it extremely vulnerable to the plough, and the base has been almost totally lost, but most of the rim survives.
We have been able to reconstruct its decoration: at the top is a wreath of olive leaves, and below this a series of slightly bossed roundels alternate with vases or archways. The vases are piled with grapes, while vines sprout from their rims. This type of object is rare but well known in the north-western provinces, and was probably made in this area as a drinking vessel.
So, from the hundreds of fragments, we can identify one complete vessel and five packages of silver from three more. They are too damaged to discern their original weights, but the care involved in hacking them fits the wider patterns noted above. They are distinctive in both form and decoration, allowing us to date them to the later 3rd century AD, fully 100 years before the next known hacksilver hoard from barbaricum. This silver hoard was something special.
Buffering the frontier
Why then did the hoard end up in Fife? To understand this we need to take a wider view. The Romans had been seeking either to conquer what is now Scotland or to create stable neighbours for their empire for over 200 years. Armies had come and gone; forts had been built and abandoned. Now silver was being used as a tool to buy peace on the frontier. The Romans had form here: a wealth of silver coin hoards is known from east-central Scotland in the late 2nd century, marking a diplomatic onslaught in the decades before the Emperor Severus led his legions back into the area on a punitive campaign in AD 209-210 (see CA 181).
Hacksilver was the start of a similar policy in a different medium, and with very different effects. The earlier coins had been useless as we would understand money, since they came to an area with no interest in coinage. They were also shunned as a raw material – analysis of crucibles of the period shows no sign of silver-working in Scotland until around AD 400. Instead, these coins were used as prestige goods, marking the owner’s favoured links with the Roman world.
Hacksilver was different – created as bullion, it was made to be recycled. Local craftworkers started melting it down to create a range of local products drawing on wider styles, from fine pins and rings to massive silver chains. The surviving hacksilver hoards, and the distribution of silver finds of the 4th-5th centuries, shows a strong east-coast bias, straddling the area of north-east Scotland where it seems the troublesome Picts emerged. This suggests deliberate diplomatic targeting of groups either side of the Picts to create friendly ‘buffers’.
Whether the Roman political aims were successful or not is debatable. What is certain, though, is that this silver acted as a prestige good and valued raw material for the next 500 years. It seems the long-term legacy of Rome can be found in a precious metal – silver – which acted to show power long after Rome was just a memory.
Fraser Hunter is the Iron Age and Roman curator at National Museums Scotland.