ARCHAEOLOGICAL RESEARCH GROUP
GADARG
|
GLOUCESTER AND DISTRICT ARCHAEOLOGICAL RESEARCH GROUP |
GADARG |
||||
|
by Nigel Spry |
||
|
The recent publication of Tony Wilmott's interesting and authoritative book The Roman Amphitheatre in Britain1 has reawakened an interest in the question: where was Gloucester's amphitheatre? He noted that Gloucester did not have one. However, by reporting that at York the likely location of its lost amphitheatre is suggested by the medieval names of two streets and by their forms, he provides further impetus for identification work at Gloucester. In the case of York, a 1993 paper by Patrick Ottaway2 proposed that the street names Hungate and Aldwark alluded to amphitheatre structures. These streets curve in a way that suggests they may have originally bounded the large characteristic oval shape of a Roman amphitheatre - the current Hungate regeneration excavations at York may yet reveal physical evidence. In a similar way, the site of the northern half of the legionary masonry amphitheatre at Chester is respected by Pepper Street and Little St.John Street.3
In both places, not surprisingly, the locations are outside the Roman military defences but, particularly in the case of the potential site at York, are within the area of the medieval city. At Gloucester the subjective evidence for an amphitheatre, just outside the urban area, is provided by field boundaries recorded on a map of 1780 accurately surveyed by Hall and Pinnell4 (fig 1). This present note, intentionally, does not consider any alternative locations.
 Fig. 1: Rigney Stile Grounds on the Hall and Pinnell map of 1780. (Part only)
 Fig. 2: The Hall and Pinnell map overlaid onto the current Ordnance Survey map (scale 100m). In Gaudy Green 225m south of the south-east corner of the Roman defended area (at SO 834 181) a suggestive nearly enclosing curvilinear boundary was shown. If this irregular enclosure did derive from the earthworks of an amphitheatre, then its length of 108m and approximate width of 60m is consistent with the overall dimensions of the ones at Caerleon (80m x 66m)5'6'7, Carmarthan (91m x67m)8.9, Chester (c.100m x c.85m)10,11,12, Cirencester (c.70m x 58m)13'14, Dorchester (c.100m x c.100m)15,16,17, London (c.100m x c.85m)18, Silchester (c.78m x c.78m)19,20,21 and the poorly understood one at Chichester (c.70m x c.58m).22'23 In the latter case the dimensions are from crest to crest of the slight earthwork. The Gloucester boundaries of 1780 could also have run along the crest of banks rather than their degraded outer limits.
 Fig. 3: Legionary & Urban Civil Amphitheatres in Roman Britain, with alignments. Llanthony Priory's Terrier of Gloucester, when read in conjunction with the Gloucester Rental of 1455,24 makes it clear that a way called Rykenel Style was on the line of the present Brunswick Road and that the gate, which appears on the Hall and Pinnell map in front of the later site of Bastion House, was its termination.25 The destination of this route running south from the south-east corner of the city defences may have been the feature under discussion. As shown in fig. 2, the continuation of Brunswick Road, laid out linking Gloucester's spa development of the early 1820s with the city centre, runs directly over the site of the Hall and Pinnell enclosure. According to Smith's Place Names of Gloucestershire26 the name Rykenel is probably a transference from Icknield or Rykenild Street, the Roman road in south-east England. But why is uncertain. Stile is an Old English word with the same meaning then as now. The area-name Gaudy Green is itself intriguing because of the fair/festival connotation of the word Gaudy.
 Fig. 4: The Walburton Survey of 1731. The postulated Gloucester amphitheatre would have had its major axis alignment, between entrances, about NE to SW. Most British urban civil amphitheatres have their alignments between NNE to SSW and ENE to WSW. The exceptions are Chichester NNW to SSE and London E to W. The known legionary examples, both with stone built outer walls, those of the 20th legion at Chester and the 2nd legion at Caerleon, have NNW to SSE alignments (fig. 3). There is no consistency about the location and distances of the British amphitheatres from their town's defended areas. In the case of urban civil ones they are usually positioned on the east of the town between 90m and some 500m away. At Chester the legionary amphitheatre is on the south-east at a distance of 50m and at Caerleon it is south-west at a distance of only 18m from the wall. The position of the Hall and Pinnell enclosure does not invalidate the amphitheatre suggestion, but no archaeological evidence has been found to support it. Unfortunately nothing was observed during construction of the Ecclesiastical Insurance Office (fig. 2), a building that would be over the arena and eastern bank of any amphitheatre on the site. A resistivity survey of the southern corner of the Brunswick Square garden might be worthwhile; in 1823 Christ Church was built on a rise opposite here and so a similar survey of its front burial area, to the west, could also be considered (fig. 2). Documentary evidence for the area is not helpful in resolving problems of interpretation. John Rhodes has edited the Llanthony Priory Terrier of 144327 and determined the history of Gaudy Green and the adjoining area from the 13th century.28 The name Rykelde Style occurs in the time of Edward I (1272-1307), it was Rykenel Stile in 1443, Rigners Style in 1629 and Rignor Stile in 1648.29 In 1780 it was shown as Regnum Stile.30 Gaudy Green (as Jordan's Croft), a four-acre former Llanthony Priory property was purchased by Thomas Bell in 1539. He settled it upon the borough in 1542. The area was used as a battery by Royalist troops during the Civil War siege in 1643. In 1648 it was let by the borough as a seven-acre pasture. Sold by the borough in 1800, it was developed as Brunswick Square from 1822.31 John considers it significant that the area increased from four to seven acres after the borough took it over. He suggests that the additional three acres represent the feature that may have been an ancient property of the borough, probably claimed as wasteland; the shape of the property boundary antedates the siege 32 The incomplete circuit shown on the Hall and Pinnell map is likely to have resulted from the annexation. Warburton survey of 1731 also shows these city properties 33 (fig. 4). Having considered location, two further linked questions arise: what type was it and what might have been the circumstances of construction? From the 20th legion's transfer to the Gloucester `City Centre Fortress' site in the late AD 60s an amphitheatre would have been needed for training and military display. In the initial phase of the amphitheatre at Chester, it was provided with a timber outer wall, so it is unlikely that at Gloucester, before moving into a new base at Chester, the legion would have had a masonry outer walled version. (The advantage of a rear wall is that, for the same width of bank, the wall more than doubles the seating capacity of the structure compared to simple earthwork ones where seating is confined to the inner face of the bank). As befitted its status, the primary phase of Roman London's amphitheatre was built with an outer wall, of timber but later remodelled in stone. All other known urban civil amphitheatres, for civic and public entertainment in civitas centres, were of the more simple earthwork design, although nevertheless their entranceways, arena walls and internal chambers were solidly built. If, surprisingly, Gloucester did not have an amphitheatre in the military period, one certainly would have been required to satisfy municipal pride when the Colonia was founded towards the end of the 1st century. Local urban rival, the Dobunnic capital at Cirencester, had its earthwork amphitheatre from about AD 11034; it would have been unacceptably humiliating for Gloucester not to have had one by at least this date. Acknowledgements I am grateful to Philip Greatorex of the Gloucester Historic Environment Team for scanning the Hall and Pinnell map and for overlaying images, and to Les Comtesse for processing them. Comments by John Rhodes on the historical evidence have been particularly appreciated, as has his editorial assistance. The County Archivist's permission to reproduce part of Thomas Warburton's survey is acknowledged. I am also indebted to Phillip Moss for his imaginative illustration of the Gloucester amphitheatre Fig.5.
 Fig.5: Conjectual reconstruction of the amphitheatre at Gloucester by Philip Moss.
References
Published in Glevensis, Journal 41, 2008. | ||