Uncovering magnificent mosaics in the heart of Roman Leicester

The largest excavation undertaken in Leicester for over a decade has shed vivid new light on the city’s early Roman history, as well as revealing evidence of luxurious dwellings, including one of the biggest fragments of mosaic floor found in the city in 150 years. Gavin Speed reports.


Until recently, the block of land in central Leicester known as ‘Stibbe’ was a factory-filled industrial area packed with premises for the manufacture of goods including hosiery, boots and shoes, and cigars, as well as engineering products. This phase of the site’s life was not to last, however, and its last industrial buildings (the Stibbe engineering works) were demolished at the start of this century. After this, the area would lie largely untouched for almost two decades – until, last year, landowners Charles Street Buildings Group decided the time was right to redevelop the site, sparking a project that would uncover an array of much earlier remains beneath the modern buildings.

Previous trial trenches investigated in 2001 by University of Leicester Archaeological Services (ULAS) had already demonstrated that, although the factory basements had taken huge bites out of the site’s archaeology, destroying medieval and later street frontages and leaving only small traces of their backyard plots to hint at what had existed before, pockets of well-preserved Roman features had survived.

left Found in the main reception room of an elegant townhouse in the heart of Roman Leicester, this is part of one of the finest mosaics to be excavated in the city in 150 years.
Found in the main reception room of an elegant townhouse in the heart of Roman Leicester, this is part of one of the finest mosaics to be excavated in the city in 150 years.

If the site was to be transformed as planned (into hotels, offices, and open public space covering an area of about 4,500m2), it was imperative that these islands of archaeology should be excavated and fully recorded – which is where ULAS came in. Such a complex urban site demanded an innovative approach, for which we turned to cutting-edge digital methods like GPS and photogrammetry rather than the usual hand-drawn plans to record our discoveries. This step allowed us to document the site with much greater precision, and more quickly, than if we had stuck to traditional techniques. As a result, we now have a rich and versatile archive that contains a wealth of information about Roman Leicester, allowing us to begin to answer some long-standing questions about the city’s past.

Streets ahead

The earliest days of Roman Leicester (or Ratae Corieltavorum), between the Claudian Conquest of AD 43 and the town’s formal establishment as the civitas capital of the Corieltavi tribe in c.AD 100, have long been shrouded in obscurity. We know that the town had taken on official status by the start of the 2nd century, thanks to the presence of regular grid patterns of streets and of spaces set aside for public buildings dating from this period that have been excavated elsewhere in the city during previous archaeological work – but precious little evidence had been identified to illuminate what the settlement had looked like in its infancy.

Our excavation has helped to start to fill in some of these gaps. The Stibbe site is located just to the north of a cluster of some of Ratae’s key buildings – including the town’s forum (beneath what is now Jubilee Square); macellum, or market hall (beneath a Travelodge); and public baths (whose remains now form part of the city’s Jewry Wall Museum) – and overlies three residential blocks: insulae numbers IXa, IXb, and XVI. Close to these, we discovered tantalising traces of earlier structures – groups of post-holes pointing to the presence of a number of timber buildings. Two of these yielded coins from the reign of Augustus (r.27 BC-AD 14), and we believe that the buildings date from the mid- to late 1st century AD – rare glimpses of the earliest incarnation of the Roman town.

right The Stibbe site lay in the north of the Roman town, close to the forum and public baths.
The Stibbe site lay in the north of the Roman town, close to the forum and public baths.

Interestingly, the area where these early structures once stood seems to have remained an important focus of the later settlement: the footprints of the timber structures had later been covered by major roads forming part of the formal grid pattern of the established town. Two streets were discovered during our work, one running east–west and the other north–south. The latter, which covered the full length of the site for 55m, is the longest stretch of Roman road yet revealed in a single archaeological project in the city.

Both roads were strikingly well preserved, surviving to thicknesses of over 1.5m in places, which allowed us to unpick a wealth of details about their construction. First, a metalled surface made up of compacted sand and fine gravel had been laid down as a base, which was then covered by layer after layer of tightly packed gravel to create a cambered surface 7-8m wide. These appear to have been important thoroughfares – the roads had been carefully maintained during their use, leaving very little evidence of erosion, damage from traffic, or accumulated detritus for us to see – and they were lined with closely packed wooden buildings whose outer walls can still be traced through the post-holes and beam-slots that they left behind.

Further pits and boundary ditches were set a little further back from the roads, painting a picture of a bustling and intensely occupied part of the town in the late 1st and early 2nd centuries – a picture that we hope full analysis of the pot fragments and other small finds discovered here will help to clarify further. Such remains provide intriguing hints of the activities of the town’s ordinary inhabitants – but nearby there were also echoes of more elite social strata.

ULAS’s excavations uncovered the remains of two Roman streets packed with the footprints of timber buildings.

Magnificent mosaics

The discovery of elegant Roman townhouses at Stibbe was not entirely surprising: it was already known that this area had been home to high-status residences, with one substantial structure excavated at Vine Street a decade ago, and another discovered during the investigation of an adjacent insula early last year (see CA 325). We have now doubled this total, though, uncovering the foundations of two more luxurious dwellings dating from the later Roman period (after AD 250). These were glamorous, well-appointed homes boasting desirable locations on one of the main streets through the town, and had been decorated with careful attention to detail and the fashions of the day. Fragments of painted wall plaster suggest that most of their rooms were adorned using a variety of colours and patterns, while both houses were also furnished with colourful mosaic floors. These ornate surfaces only survive in fragments today, but their remains show us quite how impressive these houses must once have been. The best preserved of these was truly outstanding, though: the largest and highest-quality mosaic to have been uncovered in Leicester in over 150 years.

Overlooking part of the largest and most intricately decorated mosaic that was discovered inside the townhouses.

Decorated with complex geometric patterns, this floor had been cut by medieval pits and a 1950s cellar, but enough survives for us to reconstruct a surface stretching 10m in length and 6m wide. Its intricate design included a border composed of four intertwined strands (known as a guilloche), surrounding shield (peltae) and Greek key (meander) patterns. Expert analysis suggests that this is one of the finest mosaic floors yet found in Leicester, showing quality and workmanship comparable to the Blackfriars and Peacock pavements that were found in the town in the 19th century and are now housed by the Jewry Wall Museum. Its doubtless proud owner, probably one of the town’s social elite, had chosen to show off their wealth by installing the impressive mosaic in the main reception room of their house, covering the hypocaust system that provided under- floor heating.

We were keen too to share this remarkable discovery with as many people as possible, and made the decision to open the site to the public – something that our client was very receptive to, even constructing a special elevated platform to provide a good (and safe!) vantage point from which to look out over the trenches. Ratae’s modern citizens responded with great enthusiasm: over 5,000 people visited our open weekends and lunchtime tours, as did more than 500 local schoolchildren.

left & above The fine mosaic’s imagery included shield, or peltae, patterns (the motifs seen left, and at the top of the image above).
The fine mosaic’s imagery included shield, or peltae, patterns (the motifs seen above, and at the top of the image below).

Once our public events had concluded, though, it was time to carefully lift the mosaic so that it could be conserved at Skillington Workshop Ltd. Efforts to preserve the pavement are nearing completion, and the mosaic should be ready for public display by this winter – initially at the University of Leicester, before taking up permanent residence at Jewry Wall Museum (which is currently undergoing a multimillion pound redevelopment and is due to reopen in 2019). In the meantime, you can see an interactive 3D model of the floor in situ at https://skfb.ly/6ps9y.

Step 1: Clean off loose soil and dust, and record the mosaic at 1:1 scale using Melinex and marker pens to complement the photographic and photogrammetery records.
Step 2: Each section of mosaic is then faced up with two layers of fine cotton gauze, fixed in place using conservation grade PVA emulsion adhesive.
Step 3: Next, the faced-up mosaic is lifted in section, complete with a solid coherent bed, by inserting rods underneath at regular intervals and lifting as a slab.
Step 4: The numerous sections are now at a workshop, where the existing backing is being removed, and the mosaics being prepared for rebacking, ready for public display.

Images from the arena?

As well as mosaic floors, the townhouses yielded a wealth of other clues to the occupants’ tastes and interests, from vast quantities of pottery and coins to personal items like brooches, beads, and hair pins, as well as objects such as gaming pieces that hint at how these well-off individuals liked to spend their leisure time. None of these are unexpected finds for a high-status Roman residence – but one object, recovered from between two floor surfaces, was more enigmatic.

Cast in copper alloy with a square iron tang running through it, the ornately decorated artefact is slightly curved and fits comfortably in the palm of the hand, although it is surprisingly heavy. We have interpreted the find as a handle for a knife or key, but its decoration is of an exceptionally high standard. The main element shows a male figure, possibly a Roman caricature of a Germanic ‘barbarian’, dressed in trousers with a wide waistband and his neck-length hair swept straight back. The details of his face are strikingly well preserved: a short beard covers his chin from ear to ear, and we can also see his broad nose and outsized eyes. These staring eyes could suggest fear – certainly the man’s position is perilous, as he is shown under attack from a male lion, which clings to the unfortunate individual’s waist with its jaw gaping wide and its tail curled around his left leg.

right This ornate handle, possibly for a knife or a key, was found in the townhouse with the fine mosaic. It depicts a man being attacked by a lion, possibly an arena scene.
This ornate handle, possibly for a key, was found in the townhouse with the fine mosaic. It depicts a man being attacked by a lion, possibly an arena scene.

This is not the full extent of the handle’s ornate artwork, though. The man is not depicted alone, but forms part of a complex arrangement of human bodies, standing on the heads of four naked, curly haired men. These are arranged in pairs, the outermost of each being larger and apparently older in appearance. Some are shown holding objects or animals. What could this tableau mean? The composition is surprisingly complex for such a small object (other known examples tend to have limited themselves to just one or two figures in their decoration), and it was clearly designed to exploit the handle’s three-dimensional possibilities, revealing different details as the artefact is turned. We have also captured these details in three dimensions for further study: an interactive digital model of the (partially cleaned) handle can be found at https://skfb.ly/68w99.

As for its subject matter, it has been suggested that the central figure might represent a damnatio ad bestias scene, depicting a condemned captive facing execution by being thrown to the wild beasts in the amphitheatre. Indeed, it would be particularly apt if the handle does represent some form of (albeit grisly) public entertainment, given that the townhouse in which it was found once stood just 50m from what may have been a Roman theatre.

The stage is set

The southern edge of our excavation had seen archaeological investigation before – by John Wacher in the 1950s, and ULAS in the early 2000s – revealing the remains of a townhouse that had been demolished in the early 3rd century to make way for a large, purpose-built market hall whose hefty foundations suggest that it could have stood as much as 11m high. Little was known about what lay just beyond these buildings, however – until now.

At the northern end of the macellum, under a series of Victorian cellars, we found two large, curving walls – part of a substantial structure, almost certainly a Roman public building, with foundations nearly 2m deep. The outer wall was entirely mortared, with its superstructure surviving over 1.5m high, while the narrower inner wall was made up of pitched granite blocks. In the curved gap between the two, a space 2.25-2.8m wide, there were numerous compacted earth floors and make-up layers. Interestingly, though, the walls did not lie on the alignment of the main street grid. What might they belong to?

There were a number of other clues, in the form of architectural fragments – a column, stone mouldings, painted wall plaster – hinting at a fairly grand public structure, and preliminary dating evidence placed the remains in the early 3rd century, making the building roughly contemporary with the nearby market hall. It could be that the curved walls might have formed part of the supports for a cavea – the curved, stepped seating of a Roman theatre. If this was a theatre, it would be one of the smallest examples yet found in Roman Britain (with a projected diameter of around 35m), but it is a tantalising possibility, not least because such a structure is hitherto unknown in Leicester.

left & Below The remains of a curved public building, accompanied by a small apsed structure, have been interpreted as possibly representing a Roman theatre the first to be found in Leicester.
The remains of a curved public building, accompanied by a small apsed structure, have been interpreted as possibly representing a Roman theatre the first to be found in Leicester.

Traces of other public entertainments enjoyed by Leicester’s Roman populace have already been found in earlier excavations in the city: a small, graffitied sherd of red-slipped pottery bears the intriguing inscription ‘Lucius the Gladiator’, while gladiatorial combat is depicted in a scene on a 1st-century glass cup. In addition, the Stibbe townhouse with the spectacular mosaic hints that its owners may have had a fondness for, or a connection to, the theatre – not only in the decorated handle that perhaps depicts an arena scene, but in the mosaic itself. One of its panels is thought to represent a possible ‘vela’ design, using motifs that show the awnings which covered a Roman auditorium.

There were further clues to come from the archaeology: as our excavations continued we found a small, late Roman building next to the curved wall. This structure also had a mosaic floor, although it was a much simpler surface, with grey tessellated patterns bounded by a red border, as well as a well built from stone and reused rooftile. It was the building’s shape that was most instructive, though. It was rectangular in form, with a curved apse at one end. Could this structure have had some kind of religious purpose? Theatres and temples are known to have been associated during the Roman period, with such pairings known from sites like Ostia and Leptis Magna. That said, large buildings with curved walls could perform a number of other public functions, from odea (concert halls for recitals or singing) to bouleuteria (council chambers for political meetings). Whatever the interpretation, it is an exciting new discovery for Roman Leicester, and one that we hope to explore further as post-excavation analysis progresses.

We thank Charles Street Buildings Group for funding this project, and for their enthusiasm and keen interest in the archaeological remains. Many thanks also to the thousands of members of the public who visited the excavation; the team enjoyed explaining the finite archaeological heritage and showing off the super Roman buildings and mosaics in situ.
All images: ULAS.