Today, Torksey is an ordinary Lincolnshire village that lies alongside the River Trent and astride the busy A156. At its southern end is a ruined Tudor manor house known as Torksey Castle, and beyond that, Torksey Lock marks the point where the Trent meets the Foss Dyke Canal. This channel leads to Lincoln, some nine miles to the south-east, and is thought to have been constructed by the Romans. The lock is a favoured mooring point for a flotilla of motor cruisers and narrow boats, but in AD 872 a rather different kind of fleet came to Torksey, rowed by what is described in a contemporary annal as the Viking Great Army. Scores of longships were drawn up on the floodplain north of the village, and a throng of several thousand warriors, traders, craftworkers, women, and children disembarked to camp on the higher ground east of the Trent. There they spent the winter months, processing the loot they had plundered during the previous year’s campaigning, turning Church gold and silver into ingots and jewellery, trading in slaves and precious metals, and gambling on the outcome of games of strategy and chance, which they played with distinctive lead gaming pieces. Over the last decade, we have been investigating the remains of their camp, which was effectively a town on the move (see CA 281). Recently, however, our attention has focused on what happened in this area after the Viking army moved upriver to Repton (CA 100 and 352) in AD 873. While the island that had hosted the Viking army camp was largely abandoned, reverting to shifting sand dunes until it was later levelled and taken over for farming, the higher ground to the south became an important town, a Saxon borough or burh that might be one of the seven boroughs referenced in the Anglo-Saxon Chronicle entry for 1015 (alongside the better documented ‘five boroughs’ of Derby, Nottingham, Leicester, Lincoln, and Stamford). The presence of a mint in Torksey by the beginning of the 11th century testifies to its importance, and the town was also a key place on long-distance routeways from London, with the Domesday Book recording that ‘if the king’s messengers should come [to Torksey] the men of the same town should conduct them to York with their ships and their means of navigation’.
The settlement was also a significant early industrial centre, producing the distinctive Torksey ware pottery, which was one of the first wheel-thrown industries to emerge in post-Roman England. This legacy is clear in the archaeological record: in a field adjacent to the castle, south of the modern village, we have found traces of more pottery kilns than in any other place in late Saxon England, as well as evidence for experiments in glazing. For us, it isn’t a coincidence that an industrial and urban revolution happened in the wake of the Viking visit – and so our research has sought to understand the connection, and to examine the wider legacy of the Viking invasion for Anglo-Saxon England.
While the advent of Torksey ware is a key clue to this Viking legacy, it is an interesting industry in its own right. The pottery, with its distinctive black, gritty fabric, appears on many sites in eastern and northern England from the late 9th century AD, and the way it was made is very different from earlier pottery in the region, which was hand built from individual coils of clay and fired in small batches on a bonfire. Torksey ware, by contrast, was thrown on a fast potter’s wheel – under a high-powered polarising microscope you can see elongated grains parallel to the edges of the pottery, which are formed as the vessel spins and the potter squeezes and stretches the clay walls upwards. Analysis of its colour, and the shape of the pores in the clay that are affected by heat and oxygen levels, also suggest it had been fired at very high temperatures, at least 800-850°C, requiring the use of kilns.
These kilns were made of the clay on which they were built, adjacent to the Trent – a sensible location, given the large amounts of clay and water needed for pottery production, and one that also gave the potters ready access to one of the major routeways into midland England; they could also use the Foss Dyke to take their products to Lincoln. The pottery itself, however, was made from a different band of clay, from an outcrop more than a kilometre east of the village. Our microscopic analysis revealed that this source contained naturally occurring sand inclusions, and so did not need additional tempering – indicating that the potters were experienced and knew just what they needed.
The form and fabric, and use of the fast wheel, have prompted suggestions that this revolution in production had continental origins. Similar industries are known from Northern France and the Low Countries, areas in which Viking armies were campaigning in the 860s before they came to England. Tellingly, similar transformations have been identified at other sites in eastern England, including at Stamford (Lincolnshire) and Thetford (Norfolk), and while it is not certain precisely when these new pottery industries emerged, they coincide with the areas of Scandinavian conquest and settlement.
Just as the Scandinavian leaders (such as Guthrum) who took control of territory in England relied upon continental moneyers to produce coins in the style of Anglo-Saxon kings and Carolingian emperors, so it now seems that the pottery industries of eastern England saw the arrival of another group of continental craftworkers. Perhaps they had been part of the Great Army, or were among the traders and craftworkers following in its wake, but, either way, they took advantage of the possibilities opened up by the army’s territorial conquests.
The Castle Field
The study of Torksey ware goes back to 1960, when Maurice Barley and students from Nottingham University’s Adult Education Department embarked on a series of summer excavations, beginning south of the village, adjacent to the castle. Their aim was to investigate the 16th-century antiquarian John Leland’s assertions that: ‘the old buildings of Torksey were on the south of the new town, but there now is little scene of old buildings, more than a chapel, where men say was the parish church of old Torksey.’
Although three years of excavation revealed few traces of the medieval town, Barley and his team did uncover two pottery kilns, as well as their products – bowls, cooking pots, and storage jars, some of them decorated with the thumb impressions of the potters. Five more kilns would emerge at the southern edges of the village during Barley’s work, and subsequent excavations carried out in advance of housing development nearby have added another eight to that total. When we set out to investigate the connection between the Viking overwintering and the development of the town, then, the site of Barley’s investigations was the obvious place to start. After Torksey had fallen into decline towards the end of the Middle Ages, as the Foss Dyke silted up and trade from Lincolnshire began to be focused on the east-coast port of Boston, this area had remained undisturbed by modern development, providing us with an important archaeological opportunity.
A combination of metal-detector survey, fieldwalking, magnetometer survey, and targeted excavation has provided clear evidence for the development of the Torksey pottery industry in the Castle Field – but, critically, also for some limited Viking activity, including a lead gaming piece, a clench nail from ship repair, and a bark stripper for preparing timber for woodworking. Unlike our metal-detector survey of the winter camp, however, there were also Roman and later medieval finds; and, whereas very little Torksey ware was found during fieldwalking of the camp, in the Castle Field we recovered 5,370 sherds, three-quarters of the total fieldwalked pottery.
Interestingly, there were also chronological variations seen in the distribution of different concentrations of pottery and kiln waste. Later 9th-century sherds with roulette decoration and simple everted rims were focused in one part of the field, whereas those from another area were primarily mid-10th-century to mid-11th-century in date. We also recovered two sherds of glazed Torksey ware, both with roulette decoration dating to the later 9th century – a significant find, as they are the first evidence for glaze being used on this type of pottery. The introduction of glazing is normally associated with another new industry based in Stamford, but this discovery suggests that the Torksey potters also experimented with it.
Magnetometry held the key to explaining the existence of these groupings. One anomaly clearly coincided with the location of a kiln excavated by Barley, while others were tied to clusters of Torksey ware, presumably indicating wasters from the firing process and pointing to at least five additional kilns. Scatters of other magnetic anomalies, 2-5m across, in the immediate area may represent associated structures, such as workshops or accommodation, and clay pits. The industrial area was even more expansive, however; a second magnetometer survey recently undertaken by Headland Archaeology on behalf of the Environment Agency in advance of planned flood defence works on the western and southern edges of the site has revealed another ten features, which are interpreted as yet more kilns.
Clues from kilns
As for excavation, our first trench revealed the remains of an updraught kiln, the 16th to be unearthed in Torksey. It was surprisingly close to the modern ground surface, and although truncated by plough damage, what survives is probably the fire pit that had originally been below ground. In constructing the kiln, a circular bowl-like depression had been created in the sandy subsoil and slabs of clay had been pressed into it – we could still see the handprints of the people who had made it. There was a shallow flue pit for stoking on the north-eastern side, and although it is not possible to say exactly how much the kiln would have extended above the surviving level, it was c.1m in diameter, and we could see that the walls appeared to turn inwards just at the top of the remains. Meanwhile, the fact that large numbers of moderately intact fire bars were found in the fire pit suggests that they had radiated from their central pedestal at a point just above the preserved portion.
In the base of the kiln was a substantial number of Torksey ware sherds, and these represented a variety of vessel forms, including bowls and jars, confirming that multiple types of pottery were manufactured in one kiln, and even during one firing. Unfortunately, archaeomagnetic dating was only able to confirm a broad 9th- to 11th-century date range for the final firing of the kiln, but the pottery being produced there provided some helpful clues. The large number of bowls with in-turned rims, combined with a marked absence of rouletting common from the later 9th to the early/mid-10th century, and a lack of thumb-impressed decoration that is typical of the later 10th century, indicates that our kiln is likely to have been in operation in the middle decades of the 10th century.
The form of the kiln itself also offers interesting insights into how this industry developed at Torksey. Our example, with its central pedestal and radiating fire bars, is typical of the larger kilns previously excavated on the site – yet, at just over 1m in diameter, it is much smaller than these, more like two earlier kilns, neither of which possess a central pedestal. This, along with the mid-10th-century date of the pottery, suggests that Kiln 16 is a transitional type between the two principal forms of kiln: a ‘missing link’ that will be extremely important for our understanding of the evolution of kiln structures at Torksey.
We were also able to investigate the products of neighbouring kilns, thanks to the large quantities of Torksey ware fragments recovered from the plough soil excavated from our trench. Their forms indicate that the kilns in this part of the site were generally late 9th- to mid-10th-century in date, and we can deduce other details from traces on the sherds’ surfaces: 16 bear traces of a dark green glaze, while sooting on some of the fragments hints at domestic use of the pottery, indicating that the potters were living adjacent to their work.
A town of cemeteries
While the Castle Field investigations have yielded plentiful evidence for the activities of the living, excavations within Torksey village itself have identified little evidence for early medieval occupation, and our main insights into the development of the borough come from the funerary record. Later Anglo-Saxon and medieval burials have been excavated at four locations – including next to the Castle Field kilns, where Barley uncovered the remains of around 30 adults, who had been interred in an orderly fashion, although without evidence for coffins or grave goods. A few of these graves had been cut into an undated pit in which lime had been burned – Barley interpreted them as 13th-century, but we now believe that burial commenced much earlier.
At the southern end of the present village, ten more adult burials were excavated at Castle Farm in the 1990s. No coffins or grave goods were noted, although a terminus ante quem was provided by a pit which disturbed one of the burials and contained ten coins dating to the late 11th century. Again, these individuals had been interred adjacent to a cluster of kilns, with the presence of a ditch dividing the two spaces suggesting that the cemetery and industrial features had been contemporary.
A third group of burials was excavated in 2002, to the east of Main Street. There, seven graves were found to contain eight individuals, including three children and an infant. As four of the burials were contained in stone cists, they were thought to date to the later Anglo-Saxon period – and they were not alone; the disarticulated remains of a further five individuals were recovered during these investigations, and five more graves were noted but not excavated. In 2005, two more burials of similar date were excavated close by, one of which contained traces of gold thread deriving from high-status ecclesiastical or secular clothing. This latter pair have been radiocarbon dated to the 11th century, and antiquarian accounts suggest that they may lie in the vicinity of the long-demolished medieval parish church of St Mary.
Traces of a fourth cemetery emerged in 2007, during an archaeological evaluation to the north of the village that revealed the graves of 19 individuals and the disarticulated remains of a further 11. There was an intriguing similarity to the Castle Farm cemetery, in that the graves lay close to a pottery kiln, on the other side of a ditch, but this was the least orderly cemetery found to-date, with intercutting graves suggesting that it had been used for more than a single generation. Its burials were an eclectic group, arranged on varying alignments, with two containing multiple individuals, and some people laid to rest on their sides. It is unclear whether it was associated with a church, although an adjacent stretch of wall suggests it may have been. The locations of two of Torksey’s parish churches, St Peter’s and the demolished St Mary’s, are known, but the third, All Saints, remains lost. Might this cemetery, or those at Castle Farm or Castle Field, provide a clue? At any rate, it is apparent that Torksey had more later Anglo-Saxon and medieval cemeteries than recorded parish churches, suggesting either there were further undocumented religious buildings that have since disappeared, or that some burial grounds were not associated with a church at all.
The Castle Field cemetery enclosure
Returning to our investigations in Castle Field, one of the more unexpected discoveries during fieldwalking in this area was 852 fragments of human bone, including multiple pieces from skulls. These were probably ploughed up from the cemetery that Barley had encountered in this area, and further remains have been disturbed by badger setts on the western side of the site. Taken together, a total of 2,140 human bones or bone fragments have been discovered in this field, representing the heavily fragmented and commingled remains of a minimum of 22 adults and 16 juveniles.
After plotting the distribution of fragments found during fieldwalking, we could see a clear concentration on the higher ground that runs along the western edge of the centre of the field. There, our magnetometer survey had identified a D-shaped enclosure, with its straight side formed by the edge of the bank above the floodplain, and an apparent entrance in its southern side. Given that nearly all the human remains have been found within this area, or eroding from the bank below it, it seems likely that the enclosure marked a cemetery boundary. Our survey also demonstrated that the distribution of kilns respects the D-shaped enclosure, something reinforced by pottery and metalwork finds, which were largely located outside this space. This strongly suggests that the kilns and burials were contemporary, and that industrial activity was taking place almost alongside human burial. It also adds to an interesting pattern emerging in Torksey’s archaeology: as already noted, there is evidence for at least three other cemeteries, also in proximity to kilns, and apparently demarcated by ditched enclosures.
We opened a second trench over the south-east corner of the ditched enclosure, in the hope of retrieving some dating evidence. We were not able to establish when the ditch had been first cut, but we can tell (from the presence of an abraded sherd of Torksey ware) that its final backfilling was no earlier than the late 9th or 10th century. This excavation also highlighted a marked absence of burials in the southern part of the enclosure – something that might be explained by the location of a church or chapel, though no trace of a stone structure was identified in either geophysical survey or fieldwalking. Barley did excavate traces of a possibly later 10th-century building c.50m north of Kiln 1, though. This was represented by a wall slot and a posthole, with a row of three further postholes c.1.7m to the south, and we now know it would have stood immediately inside the enclosure and adjacent to the break in the ditch. Such a construction seems far too insubstantial to be a candidate for a church or chapel, but might it have acted as some form of small gatehouse?
As for the burials within the enclosure, radiocarbon dating indicates that they ranged from the middle to late Saxon periods and into the 13th century. There is a concentration in the 9th and 10th centuries, although there is a possibility that the earliest date from the 8th century. These results confirm that the main period of use of the cemetery was contemporaneous with the Torksey pottery industry operating outside the enclosure, suggesting that the potters and their families were amongst those being buried there. It also appears that the cemetery largely went out of use at the same time as the industry declined, from the 11th century onwards, with continued (or later) usage on a small scale in the 13th century. There may also have been newcomers to eastern England among the cemetery population; we have carried out stable isotope analysis on eight individuals, and while this work is still in its early stages, preliminary results indicate that at least four were non-local individuals.
The seeds of a town
In summary, our work in the Castle Field leads us to some important conclusions about what happened in Torksey after the Vikings left, and their wider legacy for Anglo-Saxon England. When the army moved on, they left the seeds of a town behind. We believe that some of the craftworkers who had travelled with them from continental Europe stayed behind, or possibly came to Torksey to supply a new demand. What became one of the most important pottery industries in late Saxon England began here, and a large permanent community grew up, many of them being buried in enclosed cemeteries dispersed amongst the industrial areas.
A very similar picture emerges from Thetford, although excavation there has been constrained by modern development. The Viking Great Army overwintered in Thetford in 869-870, and while the location of the camp has not yet been securely identified, like Torksey, Thetford was located at a strategically important place, at a fording point over two rivers and where they were crossed by the Roman Icknield Way. Also like Torksey, it was during the late 9th and 10th centuries that Thetford grew into a sizeable settlement, and a major pottery industry emerged, with kilns located adjacent to contemporary cemeteries.
We believe this pattern was probably repeated throughout eastern England, where migrant potters began related wheel-thrown industries in other sites with Viking connections.
There is nothing to suggest that places like Torksey, Stamford, Thetford, and Lincoln were already urban settlements when the Great Army arrived. We have already suggested that the winter camps were akin to towns on the move.
Now we can see that they were also catalysts for urban and industrial development throughout eastern England.
Further reading Dawn Hadley and Julian Richards, The Viking Great Army and the Making of England, Thames and Hudson, £25, ISBN 978-0500022016
Acknowledgements We are grateful to Tim Allen of Historic England for facilitating Scheduled Monument Consent; to Ian George, Historic Places Manager at Lincolnshire County Council, for his support; and to Steve Dean of the Environment Agency for sharing reports. Excavations were by Allen Archaeology on behalf of the University of York. Magnetometer survey was by Hannah Brown; metal detector survey by Dave and Pete Stanley, and Neil Parker. Thanks too to landowners Edward and Kit Dickinson, and Dick Denby. Funding for the Tents to Towns project has been provided by the British Academy and Society of Antiquaries of London. Plans and distribution maps were drawn by Helen Goodchild.
All images: Tents to Towns Project, unless otherwise stated.