Printed from the APC web site: navigation and non-essential images removed.
Please view on-line for full content (URL at end of document).

Excavation: Period 2 Roman Field Systems & Cemetery

Cecily Spall and Nicky Toop

The Roman Road

There are two principal suggestions for the route southward from the legionary fortress from either the Porta Praetoria or the Porta Principalis Sinistra. The first is that proposed by the Royal Commission (1962, 1-2), who suggested a main road exiting at the Porta Praetoria, the main gate, turning southwards at a crossroads within the colonia, and becoming Road 2 (1962, Fig 3) (see figure below). Just beyond the site of Clifford's Tower the postulated route turned eastwards towards the line of Walmgate ultimately towards Brough, and then spurred southwards again, cutting across the area of Paragon Street, Kent Street and through the site of York cemetery (RCHM 1962 fold out map). This model was rejected by Richard Kemp who reviewed the local topographic evidence and suggested that a more likely route, avoiding low marshy ground, may have lead from the Porta Principalis Sinistra, the southeast gate (1996, 94). Kemp's postulated route continued the line of the Via Principalis along Colliergate, crossing the Foss at Foss Bridge, whereafter the road forked, leading east along Walmgate and also south, reflected in the line of George Street and Fawcett Street. Excavation at Blue Bridge Lane appears to support the latter view, manifest in the form of a continuous and straight gully following the modern route of Fishergate and dated by pottery to Period 2. Although no direct contact with a road surface was clearly made, the feature represents strong circumstantial evidence that the southwards route was actually aligned more north-south, and is probably reflected in the modern A19 to Fulford village, which exploits the high ridge atop the eastern riverbank. In addition, further contact with the Fishergate Roman cemetery suggests that it was laid out in relation to the road as early as the late 1st century, and the nearest burials are located within c.10m of its postulated route.

road reconstruction

Postulated road network leading east from the fortress

The roadside gully was encountered in three separate sections (F389B/F43B/F283F), situated c.10m away from the current line of Fishergate to its immediate west, and their projected alignments appear to belong to a continuous feature at least 60m in length, with a width of between 1.07m to 1.29m. The features are considered to represent the western road flanking ditch (see Figure & Plates below).

F389B
F43B
F283F
F389B post-excavation F43B post-excavation F283F post-excavation

Location of Period 2 non-burial features (right click to zoom)

As well as the roadside ditch, a cobbled surface was located within an evaluation trench undertaken by YAT (Trench C). When positioned in relation to ditch F283F, the surface lies immediately inside the line of the ditch and it appears that it may represent the original road metalling, which seems possible given that the archaeology in this area of the site was far better preserved than anywhere else. When encountered at Blue Bridge Lane, the ditch was more truncated and no evidence for the road surface was present. The area immediately to the east of the flanking ditch, although it had suffered truncation, did not contain any features, and this may also suggest the area belonged to the route of a thoroughfare, rather than consisting of open ground for other activities.

Roman Field System & River Access

A number of associated boundary ditches were also encountered throughout both sites, largely in a much more truncated state. F203B/F128B/F144B/F233B appeared to be sections of the same NNE/SSW narrow gully visible for c.28m within the main excavation area at Blue Bridge Lane. Parallel and aligned to this gully was a second gully located c.4.0m to the south (F110B/F519B). These features appear to define a narrow corridor of land inside which no activity of Roman date was identified; this may indicate the presence of a trackway heading southwest towards the Ouse. The site is known to be a good beaching point, and these ditches suggest exploitation of the eastern riverbank during the Roman period.

In the vicinity of the road flanking ditch and possible trackway, several lengths of linear features were encountered. Projections of these linear features indicate further subdivision of the land into relatively small rectilinear enclosures.

Within the main excavation area at Blue Bride Lane such features were restricted to an undated short truncated length of shallow gully F412B. Elsewhere, four further gullies were identified in Interventions 19, 22 and 25 at Blue Bridge Lane, dated by pottery to the Period 2 and would appear to represent subdivision of the land adjacent to the road (F438B, F474B, F529B and F537B). F438B appeared as a truncated north-south orientated gully and the post-excavation hachure plan of the feature shows two sub-rectangular depressions in the ditch-base and suggests the ditch may have held a series of posts within it along its length. To its southwest, within Intervention 19, another truncated gully of possible Roman date was identified, F474B. To its north in Intervention 25, a length of narrow east-west gully F537B was defined and excavated as well as a length of north-south gully, F529B.

At Fishergate House, evidence for the subdivision of land was also encountered in the form of narrow linear features within Intervention 4. F297F, an east-west ditch, can be projected to join flanking ditch F283F. In addition, a small length of north-south gully encountered in Intervention 2, F199F, can be projected to form a possibly continuous boundary with F412B to the north and F297F to the south.

Although these features were not obviously part of the same system, due to the interruptions between trenches and their truncated state, the pottery recovered from their backfills suggests they were contemporary and their projected alignments form a possible field system.

Evidence for land cultivation comes from the small rectilinear field enclosures to either side of the trackway, but was also present in the form of remnant ploughsoil allocated C2040B, C2211B, C2133B, C1341F and C1501F, all of which contained Roman pottery (some exclusively so) in a condition consistent with having arrived at the site through manuring and having been abraded by plough action and exposure. Apart from the intact soils at Fishergate House (C1501F), the buried soil layers were encountered downslope from Fishergate, predominantly in the western area of Intervention 15 and also within Interventions 9 to 11, 16 and 24.

The small lengths of boundary features belonging to Period 2 have allowed a surprisingly high level of landscape reconstruction, given the truncated state of most features. The sections of ditch that did survive betrayed a typical rectilinear field system to the east of a major road, but one which also incorporated access to the Ouse. The riverward trackway suggests that the eastern bank of the Ouse may have been busier in the Roman period than previously anticipated. The field system is likely to have been much more complex and extensive, and probably extended to the western side of the road. Excavations to the north of Blue Bridge Lane encountered 'sinuous ditches thought to relate to drainage', alongside ploughsoils, and a disturbed subsoil or tilth thought to relate to cultivation (Kemp 1996, 68). While no actual ploughmarks have been recorded anywhere at Fishergate House or Blue Bridge Lane, the condition of most Period 2 pottery is consistent with that of plough abrasion. The land regime has thus been determined, although the small pocket of very well-preserved Period 2 archaeology within the grounds of Fishergate House, and the position of the cremations in relation to boundaries, suggest that at least some of the rectilinear enclosures within the system were set aside for burial and possibly pasturage.

The Roman Cemetery

The earliest archaeological features encountered at the site were five cremation burials, four of which have been dated by the pottery urns and accessory vessels, including glass unguentaria, to the late 1st to early 2nd century (see The Roman Pottery and The Glass), and the fifth to between the early 2nd and early 3rd century. The early burials were encountered towards the southern boundary of Fishergate House in a small cluster (F267F, F319F to F321F), during a watching brief on the excavation of new service trenches allocated Intervention 4. The fifth burial was encountered in Intervention 22 at Blue Bridge Lane, close to the Fishergate frontage (F394B). The grave had been disturbed by an Anglian cess pit (F427B) to its north, but survived as a partially intact smashed Grey Ware jar containing a matrix of silty-clay and calcined human bone.

Location of Roman cremations and possible Roman inhumation (right click to zoom)

If projections of the boundary features are correct, the cremation burials would appear to have been placed within small rectilinear enclosures. Indeed, if the burial plot to the south of the F297F was never cultivated, this would explain the good levels of preservation for the Roman horizon in this area of the site.

A single inhumation was thought to belong to this period, since it was cut by a Period 5 pit containing an assemblage of Gritty Ware pottery and the Roman cemetery is known to have included inhumation burial. The inhumation was submitted for radiocarbon dating and was returned unexpectedly as 920AD +/-35 years (see Dating). While no intact Roman inhumations were encountered, disarticulated unburnt human remains were recovered from Period 2 features demonstrating inhumation burial in the area. Disarticulated human remains from Period 3 features both at Blue Bridge Lane and at 46-54 Fishergate (O'Connor 1991, 259) are also likely to derive from the disturbance of Roman inhumations.

Cremation catalogue

Fishergate House cremation group (right click to zoom)

Cremation 1

Cremation 1This burial was first identified in the machine-excavated southwest section of the NE-SW service trench of Intervention 4 and consisted of an urned cremation burial in an unusually large Grey Ware jar placed in a sub-circular pit. Two glass unguent bottles, one fused and burnt, were found close by, in the overlying cemetery soil, and may derive from a graveside ceremony associated with burial.

Osteological analysis of the cremated remains found them to belong to a juvenile individual of approximately 3 to 5 years of age (see The Human Bone). The calcined remains of pig bones, which had been included with the body on the pyre, were identified with the primary cremation deposit (see The Animal Bone).

Cremation 2

Cremation 2First identified during hand-cleaning of the southern extension of Intervention 4, c.3.5m to the south of Cremation 1. The grave consisted of a primary urned burial in a fragmentary Grey Rustic Ware jar; an accessory Ebor 1 flagon had been placed to the west of the urn. Both vessels sat in a shallow scoop cut into the cemetery soil, and which bottomed onto the surface of the natural subsoil; the urn displayed evidence of ancient disturbance and damage.

Osteological analysis of the cremated bone identified the remains of two individuals, a juvenile individual aged between 5 and 7 years and a young middle adult. Zooarchaeological analysis also identified burnt pig bones among the calcined remains.

Cremation 3

Cremation 3This burial was identified during hand-cleaning of the southern extension of Intervention 4, c.0.50m to the south of Cremation 2. The grave was well-defined and consisted of a shallow scoop containing a primary urned burial in a Grey Rustic Ware jar and an accompanying Ebor 1 flagon placed to its southeast. To the south of, and between the two ceramic vessels, was a complete glass unguent bottle.

Osteological analysis of the cremated remains found them to belong to a juvenile aged between 5 and 12 years and a young middle adult. Zooarchaeological analysis identified cremated bones from a medium-sized mammal.

Cremation 4

Cremation 4This burial was identified during hand-cleaning of the southern trench of Intervention 4, and was the southernmost burial to be encountered. It consisted of a shallow sub-oval pit containing an urned burial in an Ebor 1 jar, which had been placed towards the eastern side of the grave in a primary deposit of a dark brown sandy-silt, C1608F. During excavation of C1608F, several iron nails were recovered alongside multiple conjoining sherds from several vessels, including a local Grey Ware jar, a White Ware vessel and an unusual Grey Ware jar, all which may have been associated with a grave-side ceremony. Around the neck of the urn, several lengths of charred timber had been deliberately placed, forming a square 'collar' of charcoal. The timber remains and iron nails are likely to derive from a coffin or bier, burnt along with the body on the pyre.

Osteological analysis of the remains contained within the urn determined that they derived from a juvenile aged between 4 and 6 years. Zooarchaeological analysis noted a quantity of calcined pig and medium mammal bones, as well as the calcined remains of bird bones.

Cremation 5

Cremation 5This burial was located within Intervention 22 and located close to the Fishergate frontage. It had been badly damaged by the cutting of a large Anglian cess pit immediately to its north, but the grave and urn were both partially preserved, albeit in a disturbed state. The grave consisted of a shallow scoop into which a single Grey Ware urn with acute burnished lattice decoration had been placed; the urn is dateable to between c.AD120 and c.AD225. In addition, sherds of a white-slipped Ebor 1 flagon and a local rough cast beaker were recovered from inside the urn, although these are considered to represent residual material from the backfill. Iron nails and charcoal were recovered, suggesting a coffin or bier had been placed on the pyre.

Osteological analysis found the cremated remains to belong to an adolescent aged between 16 and 20 years.

The five Roman cremations form a small but significant group. Not only were the cremations remarkably well-preserved in an intact Roman burial soil, but the similarities between the Fishergate House group and those encountered in the 19th century on the western side of the road are noteworthy. There are a range of factors which appear to indicate a similar burial rite and therefore some sort of meaningful grouping. The choice of urn may well have been dictated largely by the domestic pottery available, but the presence of two rusticated Grey Ware jars, accompanied by Ebor Ware flagons in cremations a few metres apart, may be significant. Admittedly, the similarities between the rites of Cremation 2 and 3 may seem significant within what is in fact a small group. However, the cremations recorded from the eastern side of Fishergate can be added to the assemblage, and include a remarkably similar range of vessels in the burials (RCHM 1962, 68-9). As well as the ceramic vessels, the presence of unguent bottles, not only within Cremation 3, but contained within the burial soil close to Cremation 1, as well as within the Antiquarian collection and residually in many features at Blue Bridge Lane, suggest that during burial and in graveside ceremonies, the anointing or libation that the small phials represent was also commonplace.

The most striking and objective indication of uniformity of burial rite comes from the osteological analysis. The examination of the burials at Fishergate represents the first scientific contact with the cemetery, and the profile of the individuals present is the most powerful indicator of homogeneity in the group. The four Fishergate House cremations contained the remains of juveniles (aged between three to twelve years), and two also contained the remains of Young Middle Adults; the Blue Bridge Lane burial contained the remains of an adolescent. In addition, three of the four Fishergate House cremations contained calcined pig bones in the primary cremation deposit, while the fourth contained Medium Mammal bones (which does not preclude pig). Whether these shared characteristics, both in the choice of accompanying objects and in the selected calcined remains, reflect a family plot in which the association of relatives within burials was important, a burial ground for children who are often conspicuously absent from cemetery groups, or the common habits of a burial club, must remain mere speculation, but the similarities are striking.

The land enclosures were almost certainly contemporary with the use of the site for burial, since the burials appear to respect areas and boundaries. Material recovered from all of these ditches dates from the late 1st to 3rd century and may include vessels used in the cemetery, either as urns, or in grave-side ceremonies. Fragments of unburnt human bone were recovered from the excavated section of F43B, suggesting that the ditch was open during the use of the site for inhumation burial, and a local Grey Ware lid from ditch F297B at Fishergate House is thought to derive from a cremation burial rather than manuring. Several sherds of a head pot were recovered from ditch F438B, and sherds from smashed pots, including a burnished Grey Ware bowl and a red-slipped Ebor Ware jug, were recovered from ditch F389B. These vessels are likely to have been in use in the cemetery and redeposited in the ditches when backfilled some time in the mid-3rd century. Overall, apart from scant residual later 3rd and 4th century ceramic, the latest pottery dates to the mid-3rd century and it appears that the use of land declined at this time.

The truncation of Period 2 features resulted in a large body of residual Roman material in later features. Despite being redeposited, the information gained has made a contribution to the landscape model. The fused and vitrified unguentaria, fragment of mosaic glass bowl, the sherds of head vase and fragment of glass bangle all probably derive from a funerary context, and have been considered part of the cemetery collection. The disarticulated unburnt human bone in Period 2 and 3 features confirms the suggestion that the cemetery also included inhumation burial. The presence of seventy-seven fragments of disarticulated human bone from Period 3 features at 46-54 Fishergate (O'Connor 1993, 255) also demonstrates the presence of inhumation burial that far to the north. In addition, the nature of residual Roman pottery recovered there is reminiscent of that used for cremation burial at Fishergate House and it would seem that the mixed cemetery extended to 46-54 Fishergate, but had been almost completely obliterated by the intensity of later activity.

Roman Occupation

Alongside the road, cemetery and field system, some evidence for contemporary occupation has been recovered. During the evaluation programme, contexts in Intervention 9 (C1114B), Intervention 10 (C1090B) and 11 (C1099B) at Blue Bridge Lane were seen to represent a cobbled surface, which was dated by pottery to the 2nd century. Widespread areas of the site retained small spreads of what were presumed to be the original Roman ground surface, including C1773B, C2040B (Intervention 15), C2139B and C2133B (Intervention 16), C1306B (Intervention 21), C2184B, C2205B and C2211B (Intervention 24), C1136B (Intervention 12), C1341F (Intervention 2) and C1501F (Intervention 4). As well as the surviving surface and buried soils, the range of recovered ceramic building material of Roman date suggests Romanised building at or near the site (see The Building Materials). The range of forms include roof tile (tegulae and imbrices) and fragments of chimney pot, box flue tile, hypocaust tile and opus signinum,which indicate a building of some sophistication. The Roman brick recovered included a legionary stamp and tally marks, both of which have been associated with military production.

Cobbled surface C1090B
Cobbled surface C1099B
Cobbled surface C1090B Cobbled surface C1099B

The Wider Landscape

The presence of residual Roman ceramic within the current area of investigation, and its presence in excavations at 46-54 Fishergate, allows the areas used for burial and the wider landscape to be mapped (see figure below). The finds spots of residual pottery thought to have been used in the cemetery, intact Roman burials and antiquarian contact with the cemetery suggest, unsurprisingly, that the areas favoured for burial were those along the summit of the ridge, flanking the route of the road, and on the plateau above the slope down to the banks of the Ouse. The cemetery once stretched along both sides of the road, but was subject to erosion by Roman ploughing and by increasingly intensive medieval and later development. The significance of this distribution lies not only in the understanding of Roman land use, but moreover, suggests hot spots of later activity, at least some of which can be surmised to be Anglian in date. In addition, areas such as the enclosure at Fishergate House, where the Roman cemetery horizon was intact, demonstrate areas into which Anglian activity did not extend with any intensity.

extent of known and postulated roman cemeteries

The material considered to have been imported in manure is typical of the domestic rubbish that was exported to extra-mural locations all around the city. Most significantly, in the present context, this material provides an indication that land use had declined significantly by the mid-3rd century, when the boundary features were levelled, since only a handful of later 3rd and 4th century pottery was recovered. The two late Roman coins found within later features at Blue Bridge Lane may be better explained as material intended for recycling during Period 3 than as coins lost casually or imported in manure during Period 2.

Also removed from its original context is the body of Roman building material, but which nevertheless denotes Romanised building nearby. This assertion is in some way supported by the intact cobbled surfaces overlain by occupation rubbish encountered towards the western side of the area of investigation. Unfortunately, few of the CBM forms were indicative of date; nonetheless, the range of forms suggests building of some sophistication. Standard tegulated roofing systems, alongside evidence for heating systems in the form of chimney pot, hypocaust and box flue tile and the fragment of opus signinumsuggestive of high-quality wall or floor finishes are represented in the assemblage.

The site of investigation represents an unusually large area over which to examine a Roman suburban hinterland. The range of land use identified - some occupation, a road, and an extra mural cemetery surrounded by fields - is hardly surprising. Still, the opportunity to examine them all at once on the outskirts of a major Roman city is rarely encountered.

apc > monographs > blue bridge lane & fishergate house > excavation: roman field systems & cemetery