Excavations on the Site of the Baths Basilica at Wroxeter 1966-1974: An Interim Report Author(s): Philip Barker Source: Britannia, Vol. 6 (1975), pp. 106-117 Published by: Society for the Promotion of Roman Studies Stable URL: http://www.jstor.org/stable/525992 . Accessed: 24/06/2014 21:22 Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp . JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected]. . Society for the Promotion of Roman Studies is collaborating with JSTOR to digitize, preserve and extend access to Britannia. http://www.jstor.org This content downloaded from 185.44.77.82 on Tue, 24 Jun 2014 21:22:02 PM All use subject to JSTOR Terms and Conditions
Transcript
Excavations on the Site of the Baths Basilica at Wroxeter 1966-1974: An Interim ReportAuthor(s): Philip BarkerSource: Britannia, Vol. 6 (1975), pp. 106-117Published by: Society for the Promotion of Roman StudiesStable URL: http://www.jstor.org/stable/525992 .
Accessed: 24/06/2014 21:22
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .http://www.jstor.org/page/info/about/policies/terms.jsp
.JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range ofcontent in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new formsof scholarship. For more information about JSTOR, please contact [email protected].
.
Society for the Promotion of Roman Studies is collaborating with JSTOR to digitize, preserve and extendaccess to Britannia.
http://www.jstor.org
This content downloaded from 185.44.77.82 on Tue, 24 Jun 2014 21:22:02 PMAll use subject to JSTOR Terms and Conditions
Excavations on the site of the Baths Basilica at Wroxeter
1966-1974: An Intrim Report
By PHILIP BARKER
HE complex of the public baths at Wroxeter (that part of the site at present open to the public) was entered from a great basilican hall through doors in the Old Work, the standing north wall of the frigi-
darium of the baths (A on FIG. I). Earlier excavations, beginning with Thomas Wright's in 1859 and extending at intervals up to the present day, have shown that the basilica was an aisled hall (B) 73 m (240 ft.) long and 20 m (65 ft.) broad with a portico (C) extending the whole length of its northern side. At its east end was an annexe with a small additional room in one corner. At the west end (D) Thomas Wright found evidence of an entrance 'with a considerable amount of architectural ornamentation'. The south aisle and the nave were paved with tiles arranged in a herring-bone pattern, but the north aisle was divided into compartments with mosaic floors, some of which were removed in the nine- teenth century. North of the portico ran a street which bounded the baths insula.
The present excavations began in 1966 at the far eastern end of the field in the small area (Site 68, FIG. I) between the modern fence, which lies along the eastern edge of the north-south street (F), and the baths precinct-wall (E).
OUTSIDE THE PRECINCT
Immediately beneath the ploughsoil were the remains of a bow-sided wattle- and-daub building (I, FIG. 2) some I2m (40 ft.) long, divided into two or perhaps three sections, one being a clay floor with a hearth, separated by a cross-passage from a hard-standing of broken tiles set on edge (FIG. 3). This building, which lay parallel to the street and the precinct wall, had been preceded by two others, each with associated hearths, the evidence for all three being contained within layers only a few cm thick (FIG. 4). None of these buildings could be closely dated. There was no obvious occupation-debris, though the clay floor of Build- ing I contained a sherd of fourth-century pottery, showing that the building must be fourth century or later, though how much later is quite uncertain.
Io6
This content downloaded from 185.44.77.82 on Tue, 24 Jun 2014 21:22:02 PMAll use subject to JSTOR Terms and Conditions
Between these three late buildings and the baths precinct-wall there had been a succession of flimsy outbuildings and back-yard sheds, (d on FIG. 2). How- ever, these buildings were not the latest structures on this part of the site, as the southern end of Building I had been destroyed by a pit (PI FIG. 3), since identi- fied as a clay-and-timber-lined cistern, presumably for the storage of water. This cistern itself was not closely datable.
WROXETER o Baths Basilica MUSEUM
Site of Excavations 1966-73
000-ite,68 lo-J
F:
II
L metres 0 ~ 40
FIG. I
THE PRECINCT AREA
Lying between the eastern precinct-wall and the eastern end of the basilica were two structures (XXIV and VI on FIG. 2). Building VI was a large rectangu- lar building with sandstone footings, covered probably with a wooden floor ex- cept within the entrance, where the sandstone was worn smooth. There was evidence also of a wide porch and an entrance at the eastern end; the building seems likely to have been a barn, probably built as a lean-to against the still- standing precinct-wall. A coin of c. 367, found sealed in the rubble make-up of this building, gives the latest terminus post quem yet discovered on the site.
This content downloaded from 185.44.77.82 on Tue, 24 Jun 2014 21:22:02 PMAll use subject to JSTOR Terms and Conditions
Under the building a further rectangular building of post-hole construction was discovered in 1974.
Building XXIV, which was probably earlier than Building VI, was a large lean-to structure against the eastern precinct-wall and with a porch marked by post-holes and with a row of posts supporting the roof ridge. There were no traces of a hearth or other occupation-debris, and this building, like VI, seems to have been a barn.
Between the line of postholes of Building XXIV and the precinct-wall a series of features (h) lay on a slight platform or terrace. They were parallel to the wall and consisted of a shallow gulley with possible post-holes flanking the remains of a reused timber beam. This had disappeared completely except for three lengths of lead flashing along its sides, and an iron clamp at each end. It was disconcerting to realize that it was most improbable that we should have detected this beam lying within the gulley without these metal remains, since no soil colour-changes were discernible in spite of the most careful observation under all variations of soil humidity. This series of features may be connected with a small lean-to building of different date from XXIV, or they may be part of its internal arrangements.
These two buildings were preceded by a number of rectangular pebbled sur- faces which lay at 45 degrees to the precinct-walls. Work is still proceeding on these surfaces, whose purpose is unknown. Their totally different alignment may reflect the fact that a branch of the aqueduct bringing water to the baths enters the precinct about here. Two breaks in the foundations of the precinct-wall had worn rubble outside them (a,a on FIG. 2) though elsewhere outside the wall the rubble was incoherent and unworn. This suggests that access to the basilican area was through these gaps in the wall.
THE SITE OF THE BASILICA
The whole of the rest of the site has now been stripped of its covering of top- soil (Layer I) and the underlying rubble meticulously cleaned, drawn and photo- graphed stereoscopically. It is daunting to realise that this rubble spread, cover- ing almost an acre, is quite meaningless in areas of less than about 2,000 square metres. Over 1,5oo vertical photographs of the site have been taken. Not one of them looks more enlightening than a spread of rubbish. Put together one can see the phantoms of buildings, some of them vast, emerging from apparent chaos.
It became apparent early in the excavation that the rubble areas were not uniform nor, except in some cases, were they random, but that most of them had been deliberately laid. They were roughly rectangular in shape and changed in their composition from area to area. This could hardly be accidental over such a large site. In addition, the east-west street bounding the insula on the north, covered with incoherent rubble at its eastern end, had been completely
This content downloaded from 185.44.77.82 on Tue, 24 Jun 2014 21:22:02 PMAll use subject to JSTOR Terms and Conditions
removed from the eastern end of the basilica westwards for a distance of about 65 m (200 ft.). This must have involved the digging out of hundreds of tons of consolidated pebbles, the street being c. 7 m (c. 22 ft.) wide and at least I m (c. 3 ft.) thick. The trench thus dug was filled with what appeared at first sight to be very fine gravel, but which on examination proved, surprisingly, to be sifted rubble of the ruins of the basilica, being composed of tiny fragments of tile, mortar, sandstone and pebbles. This must have been a cul-de-sac entered from the Watling Street to the west, since no vehicle could have got on to the rubble overlying the old pebble street surviving at the eastern end, rubble which is clearly undisturbed either by vehicles or foot traffic except in two places where it is worn. Elsewhere it must quickly have been protected by a covering of earth and plants or grass.
The principal clues to the function of the rubble-laid areas on each side of this 'sifted rubble street' came from the north-eastern corner of the site (Build- ings VII and VIII). The evidence for Building VII (FIG. 5) consisted of a sym- metrical arrangement of two sandstone post-pads, four large shallow post- sockets, a clay-lined timber slot and a short length of clay-packed wattle walling. There can be little doubt that this represents the foundations of a fagade built of wood but following classical models, rather like the timber houses of colonial America. The post-sockets, though very large, are extremely shallow and must imply a framed building capable of standing without its uprights being sunk into the ground. This view is reinforced by the fact that the builders have gone to the trouble of cutting the post-holes deeper on the slight uphill slope, so that the building was level. There is no trace of a floor in this building or in any of those to the west of it, nor is the rubble which flanks the street worn. The in- ference must be that we are dealing with wooden-floored buildings, perhaps of considerable pretensions-the size of the timbers of Building VII could easily support a two-storied structure. Building VIII has a similarly symmetrical fagade, and another is implied by the two post-pads of Building XXII.
Close examination of the rubble bounding the northern side of the street showed that its character changed every few metres as though each rectangular area was laid at different intervals of time, though the unity of the whole lay-out may mean that the intervals were short, perhaps only a matter of days. We seem to have here building-frontages, some with alleys between them, bounding the street on the north.
South of the street it became apparent that the rubble was laid symmetrically about an axis through a 'portico' formed of two large column-fragments (b,b on FIG. 2) with rectangular 'wings' (c,c, FIG. 2) of heavier sandstone rubble with much mortar lying each side of the 'portico', and with two yellow standstone mortar areas lying behind and a little to one side of each 'column base' (b,b). On the assumption that the two rectangular'wings' marked the limits of the southern fagade of the building two perpendiculars were taken across the area to the robber-trench of the drain (D on FIG. 2), and it was then seen that the raft of small, selected rubble noticed in I971 ended at these two points (e,e on FIG. 2).
This content downloaded from 185.44.77.82 on Tue, 24 Jun 2014 21:22:02 PMAll use subject to JSTOR Terms and Conditions
What, quite astonishingly, appeared to be emerging was the plan of a very large rectangular building (X), with a southern fagade which had a central portico and wings which lay symmetrically each side of this portico. It is not clear whether or not the wings were thrust forward, since earlier archaeological excavations had destroyed most of the evidence, but there is a hint at one point (f on FIG. 2), where the destruction has not been total, that they may have projected a metre or more in front of the fagade.
The dissection of one of these 'wings' showed conclusively that rubble from parts of the demolished basilica had been carefully laid to form a raft on which the principal symmetrically-planned timber-framed building of the last occupa- tion had been constructed. The rubble was in places laid in pitched rows and packed with painted wall-plaster. Since the paintings seem all to be of one decor- ative style, of large circles containing stylized flowers with green lanceolate leaves surrounding the circles, all above a red dado, it is probable that the wall from which the rubble was derived was standing, still carrying its painted plaster, until just before the last-period buildings were begun. It is also possible that this building had a northern facade opening on to the street, but there is no evidence for this. However, robber-trench D may have cut into the edge of the rubble foundation here, destroying what evidence there was.
Since there were no post-holes in the whole area, the building must have been timber-framed, lying on its prepared platform of selected rubble. Because of the rubble hardcore and the highly permeable subsoil below, this platform would have been very well drained and therefore almost as suitable as the more usual dwarf walls for the foundations of a timber-framed building lying on sill-beams. The rubble platform was notably horizontal across its width but sloped down gradually from east to west.
As the excavation proceeded westwards it appeared that there was a long narrow extension (XI) to this major building. The foundations of this extension were of different construction, consisting of a platform of heavy rubble with regularly-spaced post-pads made of flat stones or very large tiles. Building X measures 38-5 m (125 ft.) long and 16-o m (52 ft.) wide, and its extension (XI) measures 25-0 m (8o ft.) by 5-0 m (16 ft.).
The 'sifted rubble street' came to an abrupt end west of Building X. The evi- dence is very complex, but the vertically-cut original pebble street was so per- fectly preserved that it must have been protected from weathering and wear at all times, and the differential scattering of small pebbles from its surface on to the 'sifted rubble street' suggest strongly a narrow passage past a timber Build- ing XXVI erected on the surface of the pebble street when its full width was no longer required. The evidence and its possible interpretations will be fully des- cribed and discussed in the forthcoming definitive report.
The area between Building X and the baths basilica party-wall (of which the Old Work forms part) is strewn with random rubble up to and a little beyond the robber-trench of the south colonnade of the basilica. Along the front of the wall, which still stands a few courses high, lie sub-rectangular areas of rubble
This content downloaded from 185.44.77.82 on Tue, 24 Jun 2014 21:22:02 PMAll use subject to JSTOR Terms and Conditions
and at least two apparent floors. The whole complex strongly suggests a series of narrow buildings (XVI-XXI, XXVII-XXIX, FIG. 2) built as pentices against the baths basilica wall. This theory is reinforced by the fact that though the underlying herring-bone tiled floor of the south aisle of the basilica slopes down from east to west, the rubble has been built up horizontally along its whole length (to within I cm over one stretch of 30 m). This in turn suggests that the roofs of the buildings were aligned with one of the tiled bonding-courses of the baths basilica wall (which, unlike the floor, are nearly horizontal), and it is possible that this wall was lowered to the second bonding course, which is 2-50 m (8 ft.) above the platform levels. This would give a series of buildings of shed-like proportions along the wall.
When the surface of the 'sifted rubble street' was examined closely it was seen that there was a series of slightly raised sub-rectangular platforms lying along its southern edge (1-6, FIG. 2). There were also a highly detailed bow-sided structure, XII, and a small structure, 7, lying in the centre of the 'street' at its western end. These structures, whose purpose is unknown but which were per- haps facades or booths or sheds, imply a second phase of building here, since it is unlikely that the 'sifted rubble street' would have been laid initially for these somewhat minor and irregular projections.
The picture which emerges is of a very large timber-framed building with a porticoed facade and with wings which perhaps projected forward linked to the Watling Street by a narrow western extension, the whole forming the central structure of a complex of elaborate buildings which may have incorporated some of the bath-buildings reused.' Within the eastern precinct area were two large barn-like buildings, VI and XXIV while beyond, outside the precinct-wall, lay the latest of a series of bow-sided 'peasant houses', I.
A great deal of the site has been destroyed by previous excavations, some of them disastrously extensive. Even so, enough has been left to show that there were many periods of occupation of the basilica or its site after it had gone out of use as one of the city's major buildings. Fragments of the west end of the basilica were revealed in these earlier trenches, though there was little of the 'architectural ornamentation' found by Wright. That end of the site is exten- sively disturbed, but enough remains to show that buildings, including what looks like a small guard room (XXVI, FIG. 2), overlay the street close to its junction with the Watling Street and that in the penultimate phase a furnace or oven of the type found by Atkinson in the Forum2 had been built between the basilica and the Watling Street. Emptying of some of the earlier trenches has shown that there are in places as many as seven pebble floors overlying the herring-bone floor of the basilica proper. Whether these were refloorings of the building while it was still standing and roofed, or whether they are the floors of structures built
1In particular, the frigidarium, in which grain was found by Thomas Wright in the nineteenth century.
2 D. Atkinson, Report on excavations at Wroxeter (the Roman City of Viroconium) .
1923-1927 (Oxford, 1942), io8-I3.
This content downloaded from 185.44.77.82 on Tue, 24 Jun 2014 21:22:02 PMAll use subject to JSTOR Terms and Conditions
within the ruins remains to be seen; but an area at the south-western end of the basilica, opened in 1967, showed that a very large rectangular timber building (XXX, not illustrated) had been built into the bedding mortar of the basilica floor after the floor had been removed, presumably for its tiles.
A most interesting situation seems to be emerging at the eastern end of the basilica (at g, FIG. 2). There is evidence from an earlier excavation, re-examined, that for nine periods, during and beyond its life, the extreme east end of the nave was an area of special significance, either being floored differently from the rest of the nave, or having a raised or (at a very late stage), being enclosed by a small building. This special significance seems to have persisted even after this late building was demolished. One can only hazard a guess at the reason for this: perhaps the most likely explanation is the presence here of altars or cult figures.
A crucial point in the history of Roman towns seems to be that beyond which mortar is no longer used. From then on all building is in timber or dry stone or clay etc. This thought is prompted by the discovery that one of the latest floor levels (below XVI) close to the Old Work, which appears at first sight to be a mortar or plaster floor, is in fact merely crushed wall-plaster, made per- haps into a slurry and spread to form a rather soft white surface. By the time this floor was laid the ability or the wish to make proper mortar seems to have gone. Whether this is the reason for, or the consequence of, building in wood is a de- batable matter.
Outside the precinct wall, on the site of the earlier portico which ran between the wall and the street, lie a series of pebble floors, surrounded by stake-holes, and crude hearths of many periods. Comparatively little pottery came from the areas around these late hearths; there was very much more bone than pottery, and it is probable therefore that there was a long aceramic period at Wroxeter before the end, or that supplies of pottery fell off drastically, leaving only a few types available. The very great quantities of residual pottery make these types difficult to isolate but the picture will become clearer as lower, sealed, levels are reached. The external surface of the precinct-wall still carried small areas of
plaster painted with broad red stripes. This painted plaster was probably shel- tered originally by a colonnade.
These glimpses of the development of the site of the basilica after its aban- donment as a great public building point to a decline in the life of the city, with meat bones strewn among the rubble, and fires lit on the pebble floors. At first sight this seems to support the generally accepted view of the end of many Roman towns, with a steady decline into squalor leading to eventual abandon- ment. However the evidence, outlined above, from the very latest period of occupation of the basilican area shows that it was completely redeveloped with a planned complex of timber-framed buildings, some of them very large and looking back to classical models. They must, in fact, be among the last classi- cally inspired buildings constructed in Britain before the seventeenth-century revival.
It is almost certain that the 'sifted rubble street' was a pedestrian precinct,
This content downloaded from 185.44.77.82 on Tue, 24 Jun 2014 21:22:02 PMAll use subject to JSTOR Terms and Conditions
inaccessible to wheeled traffic, and covered over at roof-height; presumably the roof joined the large Building X to the facades north of the street.
This drastic reorganisation of the city-centre needed wealth, strong motiva- tion, and a high degree of organisation. This is not the work of a peasant village, nor can it be attributed to incoming Irish or Saxon invaders. It has all the hall- marks of Roman public works, translated into timber. We must surely be look- ing at a complex of religious or public buildings, or the private demesne of a great man.
There is a complete lack of burning in the last period, nor are there any signs of hurried abandonment in the shape of indestructible iron-work such as hinges, locks or other fragments. Wroxeter, or at least this part of it, did not die a linger- ing, poverty-stricken death nor was it overwhelmed by a horde of pillaging barbarians. The implications of the evidence are inescapable. Wroxeter ends not with a whimper but with a bang.
The final phase of all seems to be a deliberate dismantling of these major buildings. Why? Wroxeter is very large-it must have covered some 2oo acres before the Severn eroded its western defences. Its perimeter, some two miles long, would be extremely difficult to defend without a very large fighting force. The discovery just outside the city ramparts of the tombstone of an Irish king commemorating Cunorix, son of the son of the Holly, and dated by Professor Kenneth Jackson to the second half of the fifth century3 implies the presence of an Irish mercenary force in the city at that time; but at some point, presumably after this, Wroxeter was abandoned for a smaller, more easily defensible site. Perhaps this was the Berth at Baschurch, connected by an old Welsh poem with the sixth- to seventh-century prince of Powys, Cynddylan, or perhaps the site of Shrewsbury itself.
THE DATING OF THE LAST PERIOD
The date of this last major phase is still very uncertain. There is nothing but Roman material, and the symmetrical plans of Buildings VII, VIII, and X strongly suggest classical prototypes. The basilica seems to have gone out of use and been demolished by 350.4 At the eastern end where an earlier archaeological trench was emptied, there were fifteen major layers between the floor of the basilica and the latest building-level. However, at other points further west there appear to be far fewer intermediate layers. The finds of latest date include six plumbatae (lead-weighted javelin-heads),5 fragments of a decorated 'Romano- Saxon' pot, some late Roman metal and quantities of shell-gritted pottery of a type which seems to be late in the pottery sequence on the site. The horizontal distribution of this pottery is interesting. There is little of it in the baths precinct area, but its incidence increases very markedly between Building X and the baths
3 JRS Iviii (1968), 206-7; Antiq. Journ. xlviii (1968), 296-300. 4 This is the date suggested by Dame Kathleen Kenyon in 'Excavations at Viroconium 1936-
37', Archaeologia lxxxviii (1940), 175-227. ' Britannia ii (1971), 261, fig. 8.
This content downloaded from 185.44.77.82 on Tue, 24 Jun 2014 21:22:02 PMAll use subject to JSTOR Terms and Conditions
themselves. A transect across the site shows that its incidence falls off very con- siderably within Building X. This suggests that it was one of the principal pottery-types in use during the life of this building, whose wooden floors would presumably have been swept clean. In addition, the 'Romano-Saxon' pot also came from the area between Building X and the baths.
A date range between c. 350 and c. 500 must almost certainly cover the whole sequence from the destruction of the basilica to the abandonment of the area, and it therefore seems likely that the major rebuilding took place some time after 400 but not necessarily long after this. The final abandonment of the city centre seems to have been orderly, with unhurried dismantling of the build- ings after the removal of their furiture and contents. We shall probably never know who was responsible for this massive injection of energy, money and man- power into what seems to have been a declining town, but the most likely candi- date is one of the minor rulers or tyranni who struggled for control of the king- doms of sub-Roman Britain, though Vortigern himself has been suggested.
THE SEQUENCE OF ROBBING
It was clear early in the excavation that the great basilica had been syste- matically demolished before the buildings of the last period were laid out. This is proved conclusively by the fact that the rubble foundation of Building X lies undisturbed over the line of the destroyed north wall of the basilica (C (produced westwards) on FIG. 2). However, there are three large and continuous robber- trenches running the whole length of the basilica and cutting through the last period foundations (D, E, F, FIG. 2). This stone-robbing must therefore have taken place after the abandonment of the site, i.e., some time after c. 500. It is significant that the stone-robbers left the Old Work standing and did not dig out the north or south walls of the basilica, but went for the central colonnade foun- dations, or stylobates, and the portico/drain foundations. They were clearly looking for the very large stones of which these were built rather than the small blocks which make up the still visible walls.
When did this robbing take place, and why? There is little, if any, trace of reused Roman stone in the four major Shropshire abbeys or the medieval churches, and only a comparatively small amount is used in local farm walls and buildings. The major reuse of large Roman blocks is in the Saxon parts of the churches of Wroxeter and Atcham, and it is therefore very tempting to assign this robbing of the basilica foundations to the masons who built these churches, probably in the seventh or eighth centuries. This poses a most inter- esting question. How did the stone-robbers know that these large stones were there?-only the walls made of small stones could have been visible. Can we envisage Saxons trial-trenching across the site, looking for the masonry they wanted? It seems the only answer. Whether it will be possible to isolate Saxon trial trenches from the welter of other trenches of all kinds which criss-cross the site remains to be seen, but it is an engaging possibility.
This content downloaded from 185.44.77.82 on Tue, 24 Jun 2014 21:22:02 PMAll use subject to JSTOR Terms and Conditions
A full list of thanks to all whose help was indispensable would be longer than this short report. However, special thanks must go to the Inspectorate of Ancient Monuments for their understanding of the nature of the problems posed by the site and their provisions of finance and services to help solve them; to the site super- visors, technicians, domestic staff, volunteers and prison labourers, and especially to my assistant director, Mrs. Kate Pretty; to my wife for organising the volunteer force, and to my colleagues both on and off the excavation whose criticisms and suggestions have formed the basis for much of the interpretation offered here.
4 St. George's Square, Worcester
This paper is published with the aid of a grant from the Department of the Environment.
This content downloaded from 185.44.77.82 on Tue, 24 Jun 2014 21:22:02 PMAll use subject to JSTOR Terms and Conditions
