The voice of archaeology in Britain and beyond

Cover of British Archaeology 88

Issue 88

May/June 2006



Meadowsweet flowers in prehistoric graves

Strange fish

"Find of several lifetimes" – cathedral archaeologist

Oldest houses in Scotland

Archaeologists mourn loss of two popular colleagues

In Brief


End of the line: St Pancras Station
How many graves would you have saved? Phil Emery rescues French revolution refugee history.

The floors that Rome built
Stephen R Cosh and David S Neal are recording every Roman mosaic.

Protests at Bling King's grave
What can the Prittlewell protesters hope to achieve?

on the web

Recommended websites


Views and responses

CBA news

Headlines from the CBA office.


ISSN 1357-4442

Editor Mike Pitts


The floors that Rome built

For a few centuries Roman mosaics paved the floors of wealthier Britons with bright, ornate designs, sometimes telling stories and often expressing exceptional craft and artistry. In a unique project, Stephen R Cosh and David S Neal are illustrating every one. Part two of their corpus is just published.

One of the most exciting finds for an archaeologist, particularly for us as specialists, is a Roman mosaic. After it has been uncovered and recorded, the inevitable question is: "What do we do with it?" Where the mosaic is about to be destroyed by development, lifting is often the solution; but when such a threat does not exist, usually the best strategy is to rebury it deeply. Lifted mosaics take up considerable space that museums seldom have, and the expense of having the pavement properly backed for display is usually prohibitive.

But mosaics are not merely ancient works of art, worthy of display. They are important archaeological tools. Wealthy Romano-Britons' lives can be gauged from the scant remains unearthed over the years by archaeologists and enthusiasts. Time, nature, stone robbers and the plough have normally reduced Roman edifices to floor level (and below): the grand dwellings are known only as ground plans, and relatively little of their decoration survives.

Mosaics are a notable exception, not only in their continued existence, but also from what can be learnt from their quality, style and content. As floors, they are made from durable materials - they were, in effect, paintings in stone. Fortunately the tesserae (the small cubes of stone and tile from which mosaics were made) could not be recycled or adapted for other purposes, and, therefore, mosaics were thankfully spared by people from late antiquity to Victorian times searching for building material.

Although not all survived ploughing and development, some 2,000 Romano-British mosaics are known. We are cataloguing and illustrating them for a comprehensive corpus to equal, or surpass, other such studies from elsewhere in the Roman empire. Volume 1 of Roman Mosaics of Britain (Northern Britain, the Midlands and East Anglia) was published in 2002, and the second of what will be four volumes - south-west Britain - was launched in January.

Many of the hundreds of coloured illustrations in these books are our own tessera-by-tessera paintings. We took a conscious decision that all mosaics capable of illustration would be shown in plan form. This is how early illustrators worked, including the great Samuel Lysons (1763–1819: several of his and others' engravings are featured in the new book), but since the introduction of photography it has rarely been the case.

Most were recorded in monochrome photographs, taken obliquely from the side of the trench, seldom of the whole pavement, often with parts obscured by an excavator, shadow, or glare which can have the misleading effect of making dark, shiny tesserae appear light and white matt tesserae appear dark. Only recently have scaffolding towers and cherry-pickers enabled anything close to a vertical shot. More recently still computer technology is having an impact on the recording of mosaics, especially in photogrammetric surveys in which a number of vertical shots are set together like a patchwork. However, even these can produce disappointing results, for it is difficult to clean a mosaic adequately. For these reasons we have produced a series of coloured tenth-scale paintings, which if not drawn from the actual mosaic (as many are), are based on photographs with a grid imposed, corrected for perspective. We are often aided by an excavator's plan, description or samples of the tesserae and occasionally rubbings or tracings.

British conviviality

Our second book describes 444 mosaics from about 120 sites in Bristol, Cornwall, Devon, Dorset, Somerset and Wiltshire. It is a particularly interesting region. Perhaps the earliest datable fragment in Britain came from the legionary baths at Exeter, showing the emblem of Legio II: a globe flanked by Pegasus and Capricornus. But in a civilian context no first and few second century mosaics are known, which is markedly different from the situation in south-east Britain.

Although early mosaics are conspicuously rare in the area, there are a surprisingly high number which are datable to the second half of the fourth century. Even in the Roman town of Dorchester, with over 60 known mosaics, only one can be dated (stylistically) to the second century with any degree of certainty; by contrast several can be dated externally (normally by sealed coins or pottery) to no earlier than the mid fourth century. By this time there were relatively dense clusters of villas, particularly in the vicinity of Bath and Ilchester, Somerset, the latter having become an important centre by the mid fourth century.

The mosaics vary in quality from simple pavements in coarse red and grey tesserae, to complex geometric designs in fine tesserae of several colours. Perhaps the most expensive and prestigious mosaics contained figured work, and this region has some of the best fourth century examples in Britain. Around Ilchester, there are several pavements which include mythological scenes (indeed, half the British examples of such scenes - as opposed to depictions of gods and goddesses - occur within 30km of the town). In other words, the pictures are telling stories.

From the villa at Low Ham, Somerset comes the famous mosaic that illustrates scenes from Virgil's Aeneid, perhaps based on the owner's illustrated manuscript. Other mosaics were perhaps inspired by literary works such as Ovid's Metamorphoses. In addition to the several hunting scenes, these pavements proclaim the villa owner's wealth and sophistication - and in the large mosaic at Hinton St Mary, Dorset, the owner's Christianity. More usually a pagan god presides over the scenes featured. In the Low Ham mosaic, Venus appropriately looks from the central compartment upon the tragic love of Dido for Aeneas depicted in four panels around her.

The best mosaics often adorn dining rooms, many of which are of the large bipartite type popular in the fourth century. The centres of several pavements have been lost, but despite this, perhaps as many as five in the area have Bacchus, the god of wine, as a focal point, and 25 others have a prominent wine-cup (cantharus), twice flanked by leopards, an animal associated with Bacchus. This god occurs more frequently on mosaics in Britain than elsewhere in the empire. From this we can perhaps construe that for Romano-British villa owners and their dinner guests there was an emphasis on conviviality. In the north of the region, in the former tribal area of the Dobunni, Orpheus is the preferred god, normally surrounded by animals which he is supposedly pacifying by the sound of his lyre. Five certain, or probable, examples are pictured in volume 2 of our corpus.

Literary prowess

Another notable feature of south-west Britain is the opulence of the villa bathsuite, a standard rarely seen elsewhere in Britain. This is normally the result of enlargement and refurbishment of baths in the third quarter of the fourth century, particularly in Dorset and Somerset. At Lufton, in south Somerset, a large indoor octagonal pool was added to the frigidarium (the cold room of the baths) and a mosaic was laid all around it, including a frieze of fish, of which 29 survive. Equally spectacular was the octagonal pool at Dewlish, Dorset, although its surrounding pavement was not so well preserved; it included an exceptionally fine panel of mythical sea-beasts. Aquatic scenes are highly appropriate for baths, but the aforementioned Aeneid mosaic at Low Ham adorning the frigidarium demonstrates that the baths were another opportunity to show off literary prowess. A large L-shaped mosaic, each arm some 12m long and 4.5m wide, from Halstock, north Dorset, has one arm running alongside a large rectangular pool.

With so many villas being built or refurbished around Ilchester and Bath, it is not surprising to find the same craftsmen working on mosaics at different sites, recognisable by the same schemes and motifs they used. Research for volume 2 has resulted in the recognition of new groups of mosaics. This, of course, has important implications for dating. Several mosaics form the Lindinis Group, named after the Roman name for Ilchester, its notional base. These craftsmen specialised in pairs of interlaced squares, and a fairly worn coin of Valentinian I (emperor AD364-375) discovered below one of their pavements at Ilchester Mead shows that they operated here, and presumably at other sites with their distinctive work, after AD364, and perhaps some time after. Such late dating should not surprise us. Avery similar mosaic from Cirencester has been dated to after AD388, and some mosaics from Ilchester seal Valentinianic coins.

The demand in the region was so great that we suspect craftsmen from Cirencester moved to Ilchester, and laid large mosaics in the town and nearby villas, notably at Halstock, north Dorset, and Lopen, Somerset, at some time after AD350. The more impoverished situation in the southeast at this time meant there were not enough commissions to sustain professional mosaicists. The large mosaic from Old Broad Street, London was almost certainly laid by those "Corinian" mosaicists from Ilchester. We have also identified fourth century groups around Bath, one of which, the South-Western Group, laid some of the mosaics in the grand villa at Keynsham, resplendent with structures of hexagonal plan at either end, probably towering above the rest of the building. Several other villas, and a town-house in Bath, feature their work, including the recently discovered pavement at Bradford-on-Avon, Wiltshire.

South-west Britain certainly has a spectacular collection of mosaics, several lifted and displayed in Dorset County Museum, Dorchester, and the Somerset County Museum, Taunton. Some of the most famous mosaics occur in the region. The wonderful series of mosaics from Frampton villa, Dorset, contain a bewildering display of mythological tales, but also, enigmatically at the margin of one, a chi rho, the Christian symbol formed by the first two letters of Christ written in Greek. These pavements were excavated by Samuel Lysons in 1796–97, and are known only from his engravings.

Very similar to the pavement with the chi rho at Frampton, the most famous pavement of all was found in 1963 at Hinton St Mary in the same county; this time the chi rho is at the centre partially behind a bust, generally (though not universally) accepted as being Christ. With the chi rho in such a dominant position, it is probable that the remainder of the pavement has a Christian connotation. The accompanying panel with Bellerophon mounted on Pegasus, spearing the monster, the Chimaera, is probably an allegory for good overcoming evil, and is surely the precursor of the Christian St George and the dragon. Figures in the corners, apparently adapted from Winds, are perhaps the four evangelists, and the scenes of hounds chasing deer may represent the Christian's struggle for salvation. It is to be hoped sincerely that the British Museum, who own the lifted mosaic, will soon put the entire pavement on display again, rather than just the bust.

Now that volume 2 of the corpus has been published, we can concentrate our efforts on the rest of the series. Volume 3 is almost ready: it covers south-east Britain, with many mosaics from the townhouses of Canterbury, Chichester, Colchester, London, Silchester, Verulamium and Winchester. It also includes many of the early mosaics of Roman Britain, but relatively few fourth century examples. Volume 4 is in preparation and covers the west, mainly Gloucestershire, Oxfordshire and south Wales, and includes the largest mosaic found so far: the great Orpheus pavement at Woodchester.

The high quality book production has proved expensive, and the first half of the project depended on money donated by private individuals and organisations: we would like to take the opportunity to thank them all. We still need to raise money to complete the project. If you wish to make a donation, please send a cheque payable to ASPROM Corpus Fund to Dr DS Neal, Tylers, Little Brickhill, Milton Keynes MK17 9NP. The names of donors of £100 and 0 over are printed in the volumes.

Roman Mosaics of Britain, Volume 2: South-West Britain, by SR Cosh & DS Neal ISBN 0954791614/0954791612. Special price before June 30 £120 + £10 UK P&P; volume 1 (ISBN 0953784525) also at £120 + £10 if ordered at same time. Illuminata Publishers, The Annexe, The Red House, Old Valley Road, Barham Kent CT4 6QG or the Society of Antiquaries, Burlington House, Piccadilly, London W1J 0BE. See The Association for the Study and Preservation of Roman Mosaics,

CBA web:

British Archaeology

Jan/Feb 2005
Mar/Apr 2005
May/Jun 2005
Jul/Aug 2005
Sep/Oct 2005
Nov/Dec 2005
Jan/Feb 2006
Mar/Apr 2006
May/Jun 2006
Jul/Aug 2006
Sep/Oct 2006
Nov/Dec 2006
Jan/Feb 2007
Mar/Apr 2007
May/Jun 2007
Jul/Aug 2007
Sep/Oct 2007
Nov/Dec 2007
Jan/Feb 2008
Mar/Apr 2008
May/Jun 2008
Jul/Aug 2008
Sep/Oct 2008
Nov/Dec 2008
Jan/Feb 2009
Mar/Apr 2009
May/Jun 2009
Jul/Aug 2009
Sep/Oct 2009
Nov/Dec 2009
Jan/Feb 2010
Mar/Apr 2010
May/Jun 2010
Jul/Aug 2010
Sep/Oct 2010
Nov/Dec 2010
Jan/Feb 2011
Mar/Apr 2011
May/Jun 2011
Jul/Aug 2011
Sep/Oct 2011
Nov/Dec 2011
Jan/Feb 2012
Mar/Apr 2012

CBA Briefing

Courses & lectures
CBA Network
Grants & awards

CBA homepage