Subscribe now for full access and no adverts
Housesteads. Birdoswald. Carvoran. Today these words are familiar to us as the names of forts along Hadrian’s Wall – but if you said them to a Roman soldier, you would be met with blank incomprehension. We have few contemporary written sources to tell us what these fortifications were called by the troops who garrisoned them, as most inscriptions from such sites name the resident unit and/or their commanding officer rather than their location. There are some rare exceptions, however. Thanks to the unusual wording of sandstone altars from Birdoswald and Housesteads (RIB 1905 and RIB 1594), we can deduce that they were respectively called Banna and Vercovicium. For the names of many others, though, we must turn to late Roman or even early medieval sources: the Notitia Dignitatum (a 4th-century administrative document), and the Ravenna Cosmography (a list of places compiled c.AD 700, based on a now-lost map). The date of these writings, however, and the fact that they have come down to us as copies preserved in late medieval manuscripts, mean that the nomenclature that they supply should be used with careful caveats.
There is another, rather more contemporary body of evidence to help confirm some of these names, however: a handful of small, bronze vessels dating to the 2nd century. Found scattered across England, France, and Spain, they are variously described as ‘cups’, ‘pans’, and ‘skillets’, or using the Latin terms trullae and paterae. Each has vibrant enamel decorations, most of which depict a stylised image of a turreted wall, and some of them also bear the names of forts featured in the manuscripts described above (albeit with some variations in spelling). With the discovery of a sixth member of this enigmatic group recently announced in the journal Britannia (see ‘Further reading’ below), this is a good opportunity to explore what they can tell us about Roman perspectives on the Empire’s northern frontier.
The fragments making up the remains of the Berlanga Cup (top pair of images) and (below) a digital reconstruction of the artefact. Recently found in Spain, it is the latest addition to the ‘Hadrian’s Wall vessels’.Image: 3D Stoa. Arqueología y Patrimonio
A colourful collection
The first artefact in this group is the Rudge Cup, which was found around 400km (250 miles) from Hadrian’s Wall in Rudge Coppice, near Froxfield in Wiltshire. It was discovered in 1725, during the excavation of a presumed Roman villa (the investigations uncovered the remains of a building with a fine mosaic floor), not within the structure itself, but in a nearby well whose unusual contents included coins, animal bones, and a number of human skeletons, perhaps speaking of a ‘special’ deposit. Now housed at Alnwick Castle in Northumberland, the cup measures around 4.6cm (1.8in) tall and around 9.4cm (3.7in) in diameter. Traces of red, green, and blue enamel survive on its surface, including a regular, geometric pattern depicting a crenellated wall, outlined in red and filled with small squares evoking regular blocks of stone. Most significantly, it also has the names of five forts from the western end of Hadrian’s Wall moulded in relief just below its rim: Mais (Bowness), Aballava (Burgh by Sands), Uxelodunum (Stanwix), Camboglanna (Castlesteads), and Banna (Birdoswald).
For more than 200 years, the Rudge Cup remained an intriguing anomaly – but in 1949 a strikingly similar artefact emerged from the soil, this time in Amiens, northern France. It was associated with a Gallo-Roman house, and was found in a room with a hypocaust, alongside a pipeclay goddess figurine. This artefact is slightly larger than the Rudge Cup, and as it has a handle (albeit found broken off); it is known as the Amiens Skillet. Despite the distance between their findspots, both are strikingly similar in design. The Amiens artefact (now in the collections of the Musée de Picardie) also has a red enamelled line tracing the same mural motif, with a chequerboard of ‘masonry’ squares picked out in green and blue underneath, and both vessels use the same image of back-to-back crescents and dots to fill spaces between the crenellations. Again, we see moulded text beneath the vessel’s rim, preserving the same five fort names as appear on the Rudge Cup – but here there is a sixth, too: Aesica (Great Chesters).
Scholars would not have to wait so long for a third example to be found. Numerous artefacts have been recovered from the sacred thermal spring that rises beside the courtyard of the Temple of Sulis Minerva in Bath and feeds the site’s famous Roman Baths – and in 1980 these included another vessel with a mural design. The Bath Pan is similar in size to the Rudge Cup, with a (broken) handle like the Amiens Skillet, and turreted decorations in common with both objects (albeit inverted). There is no inscription, though lettering had been added to the handle at a later date, picked out in tiny punched holes. The surviving text suggests that it was a dedication to Sulis Minerva, the site’s presiding syncretic deity, who combines the local goddess Sulis and the Roman goddess Minerva.
A blending of British and Roman traditions is extravagantly represented on the fourth vessel within this set, too. In 2003, a metal-detectorist discovered a particularly vibrant artefact while searching near the Staffordshire village of Ilam (see CA 188). Known variously as the Staffordshire Moorlands Pan and the Ilam Pan, its dimensions are again similar to the Rudge Cup, and, while it has no surviving handle, a patch of solder hints that one may have originally been present. This artefact bears the names of four frontier forts, and while some – Mais, Uxelodunum, and Camboglanna – are familiar from the other vessels, here a new name, Coggabata (Drumburgh), appears second, in place of Aballava.
Significantly, the inscription includes four additional words: Rigore Valli Aeli Draconis. The meaning of the first two is clear enough: vallum is used to refer to Hadrian’s Wall in the Notitia Dignitatum, which lists the frontier forts under the heading per lineam valli, ‘along the line of the frontier’. Rigore valli is thought to express the same idea – and it is possible that the pan also preserves the Roman name for Hadrian’s Wall itself. Aelius was the Emperor Hadrian’s family name (in full, he was Publius Aelius Hadrianus), and Valli Aeli could refer to ‘the Aelian frontier’, in the same way that we have Pons Aelius, ‘the Aelian bridge’, at Newcastle-upon-Tyne. Draconis, meanwhile, probably indicates that the pan was made or owned by a man called Draco (or Aelius Draco, if following another possible reading of the text).
Another, more obvious difference is found in the pan’s decoration. Unlike the vessels described so far, this one (whose ownership is shared between the British Museum, Tullie in Carlisle, and the Potteries Museum and Art Gallery in Stoke-on-Trent) does not depict the Wall. In stark contrast to the sharply geometric designs of the others, the Ilam Pan is adorned with elaborate, swirling, Celtic-style patterns in red, blue and green enamel, creating an intriguing mix of distinctively local artistic ideas and text that evokes sites associated with imperial authority. Its exuberant design fits well into the tradition of highly skilled enamellers known to have been operating in Roman Britain (CA 222). Excavated finds show that colourful patterns were being applied to an eclectic array of objects, from flasks and jugs to dragonesque brooches (CA 311) and cockerel figurines (CA 381). This industry is thought to have been particularly associated with workshops in northern England, and Carlisle has been suggested as a possible production centre – one that, of course, lies on the line of Hadrian’s Wall.
Perhaps the most important difference between the Ilam Pan and its neighbours, though, is in the way that its inscription was created. Whereas the text on the Rudge Cup and Amiens Skillet was cast as part of the vessels’ original design, the fort names on the Ilam Pan had been incised later, with each letter cut deeply into the metal before being filled with enamel. This suggests that, while the first two objects had been deliberately designed to commemorate the Wall in some way, the Ilam Pan was a general-use (if rather flamboyant) object that was later customised to serve the same purpose.
Fragmentary finds
Alongside these largely intact examples, there are two fragments that are thought to come from similar objects. The most recent such discovery is the Basildon Fragment, which was found by a metal-detectorist in Essex in 2017 (and, the Portable Antiquities Scheme database notes, was ultimately returned to its finder). Measuring just 4.1cm by 3.2cm (1.6in by 1.3in) the slightly curved piece has lost its enamel and too little survives to reconstruct the vessel that it came from, but it does preserve a single occurrence of the now-familiar turreted design, next to a ‘saltire’ motif formed from four pointed ovals, which seems to have filled the spaces between crenellations like the crescents on some of the other finds. Immediately above this imagery, on the very edge of the fragment, a series of four letters and the edge of a fifth represent tantalising traces of an inscription. The surviving letters are thought to read ‘…NUM CA…’; could this be the start and end of ‘Uxelodunum, Camboglanna’, which appear together on the Rudge Cup, Amiens Skillet, and Ilam Pan?
The Basildon Fragment, found in Essex in 2017. Image: Colchester and Ipswich Museum Service
Completing this set, the Hildburgh Fragment is the only one not named after its findspot: instead, it takes its name from the collector Dr L W Hildburgh, who purchased the 8.5cm by 12.2cm (3.3in by 4.8in) piece in the early 20th century and ultimately donated it to the Victoria and Albert Museum in London, where it resides today. Discovered between Zamora and León in north-west Spain, its decorations of turrets and small squares are particularly close to those of the Rudge Cup and Amiens Skillet – though it uses the pointed-oval ‘saltire’ motif seen on the Basildon Fragment, too. Sadly, this piece appears to have come from a lower portion of the vessel, with none of the rim surviving, meaning that any trace of an inscription has been lost; however, the colours of its enamel squares are strikingly similar to those used on the other examples.
Together, these artefacts form an intriguing collection, clearly closely associated with Hadrian’s Wall but all found far from these fortifications. The use of frontier-fort names suggests local knowledge, and while none of the vessels are identical in design or in the wording of their inscriptions, they draw from a common repertoire of artistic ideas and share enough details that it has been suggested that they could have been made in the same area, possibly even the same workshop (though probably not by the same person) close to the Wall. What, though, were they for? The very specific inscriptions on these objects – all slightly different – might hint at a personal meaning rather than these being general ‘tourist’ keepsakes purchased by casual visitors to the frontier. The fact that they resemble trullae, the shallow, horizontal-handled cups that formed a standard part of Roman military kit, adds a martial flavour that complements the decorations seen on many of the vessels. In this light, and as all of the artefacts were found a long way from the place that they commemorate, it has been suggested that they could be souvenirs carried home by veterans of one of the auxiliary units drawn from across the Roman Empire that are known to have served on the northern frontier. The quality of the cups’ materials and manufacture might indicate that they were prestige items rather than something readily available to the common soldier, and it is possible that, as an alternative to a simple memento, they were presented as an officially sanctioned retirement gift or award of distinction to high-ranking individuals.
All of the inscriptions described above list their respective forts from west to east, as if viewing them from the Roman side of the frontier, and it is also notable that the sites named on these vessels all lie within the western and central sectors of the Wall. Now, almost exactly 300 years after the discovery of the Rudge Cup first brought such artefacts to archaeological attention, a sixth vessel has just been discovered – and it adds an exciting new dimension to this picture, as the four forts inscribed on its surface lie far to the east.
A veteran from Celtiberia?
Like most of its predecessors, the Berlanga Cup is named after its findspot: Berlanga de Duero, a village in the Soria province of north-central Spain that is traditionally associated with a Roman settlement. Recovered from farmland, it was fractured and incomplete when found, but around 80-90% of the vessel still survives, allowing researchers at the Consejo Superior de Investigaciones Científicas and the Museo Arqueológico Nacional to create a 3D digital reconstruction documenting its decorations in detail, and to carry out scientific analysis that has revealed a wealth of information about the object’s likely origins.
The reconstructed cup measures 11.34cm (4.5in) across the top and 7.89cm (3.1in) tall, making it the largest Hadrian’s Wall vessel yet found. Its decorations, however, are largely in keeping with its predecessors, comprising three horizontal friezes adorned with red, green, and blue enamel. The uppermost features a red line depicting a turreted wall, interspersed with pointed-oval saltires like those on the Basildon and Hildburgh Fragments. Below this is a series of small, colourful squares like those on the Hildburgh, Amiens, and Rudge finds, while the lowest level is more innovative, formed from scale-like patterns of colourful enamel. Around the very top runs an inscription that gives the names of four forts from the eastern sector of the Wall, again listed from west to east: Cilurnum (Chesters), Onno (Halton Chesters), Vindobala (Rudchester), and Condercom (Benwell). Significantly, this text had been incised after the cup was cast, like on the Ilam Pan, rather than moulded like the Rudge and Amiens artefacts – but, unlike the Ilam Pan’s swirling decorations, the Berlanga Cup’s mural motifs suggest that its link to Hadrian’s Wall had always been intended, with its rim left blank for later personalisation.
A drawing of the Berlanga Cup’s decorations and inscription. Image: Francisco Tapias
As well as exploring the cup’s imagery, the Britannia paper describes analysis of its materials, which for the first time offers scientific confirmation that such objects could have been made in Britain. Working with Durham University, the researchers used portable X-ray fluorescence spectrometry to establish that the Berlanga find is made of a zinc-lead-bronze alloy that is fairly typical of metal artefacts from 2nd-century Britannia, seen in around 30-40% of analysed objects of this date. The clearest insights, however, came from lead-isotope analysis, which revealed that this metal had most likely been mined in northern England (specifically the North Pennines or Durham) or Wales.
By combining this data with historical information about when the named forts had been built and were in active service, the team suggest that the Berlanga Cup was created between AD 124 and 150. This is a useful addition to the long-standing debate about when the Hadrian’s Wall vessels were made: it is usually suggested that they reflect a period when interest and pride in the fortifications would have been particularly high, with the most likely candidates being the 130s, when the Wall was new and exciting, and the 160s, after the Antonine Wall had been abandoned and the northern frontier re-established at Hadrian’s Wall.
Turning from the artefact’s origins to the end of its use-life, the Berlanga Cup had travelled some 2,000km (1,200 miles) to where it was eventually found. If it was a souvenir carried by a veteran returning home, we might be able to glean clues about the unit to which he belonged, as Hispanic auxiliary units are known to have been based at some of the forts named on the Cup. Surviving inscriptions and the Notitia Dignitatum link the Ala I Asturum Hispanorum to Benwell/Condercom (albeit in the 3rd century), while the Ala II Asturum left numerous inscriptions at Chesters/Cilurnum to attest to their presence from the late 2nd century. These dates do not match the suggested age of the Berlanga Cup – but, perhaps tellingly, it was found within the traditional homelands of another unit: the Cohors I Celtiberorum. These troops are rather more mysterious, as they are otherwise represented in Britain only by a series of 2nd-century discharge diplomas granting citizenship to their veterans. We know that they served here during the early to mid-2nd century, but precisely where they had been posted has long remained obscure. Might the names of the forts that were chosen to be added to the Berlanga Cup shed light on this question? As for the Cup’s final destination, a ground-penetrating radar survey of the site where it was found identified a small group of buildings including a rectangular structure measuring c.17m by 14m (56ft by 46ft), which was divided into several small rooms. The investigation also recovered quantities of tiles, metalworking slag, and glass, and it is thought that the remains collectively represent part of a larger villa-type complex.
The recently published research adds illuminating new details to an ever-evolving picture. Hadrian’s Wall vessels have been found on both rural and urban sites, sometimes associated with high-status residences, sometimes with ‘special’ deposits in watery locations. Carried far from the frontier that they commemorate, these finely worked artefacts were clearly both prestigious and precious to the people who owned them. With the Berlanga Cup now radically expanding the scope of the forts known to have been named on these vessels, this discovery raises exciting questions about what other inscriptions may have existed, and what future finds might yet add to our understanding.
R De Pablo Martínez, S De Luis Mariño, J Garcia Sanchez, I Montero Ruiz, and P Aparicio Resco (2026) ‘The Berlanga Cup: new evidence of Hadrian’s Wall pans found in Hispania Citerior (Spain)’, Britannia, https://doi.org/10.1017/S0068113X26100701 (open access).
Acknowledgements:
Grateful thanks to Professor David Breeze for providing additional information and for his kind assistance in sourcing images.
To read more about the other vessels, see D J Breeze, The First Souvenirs, Enamelled Vessels from Hadrian’s Wall (Cumberland & Westmorland Antiquarian & Archaeological Society, 2012).
