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: The Cemetery

Cecily Spall and Nicky Toop

The Cemetery & its Identification with St Helen's, Fishergate

The cemetery on Fishergate was initially contacted during evaluation work within the grounds of Fishergate House in 1994 (YAT 1995), and at Marlborough Grove in 2000 (FAS 2000). On the site of Fishergate House, a single inhumation burial was identified cutting subsoil, orientated east-west. At Marlborough Grove, a 2m x 2m trench (Trench A) encountered the remains of three inhumation burials.

Burials were therefore anticipated within the grounds of Fishergate House; although, the density of burial was unexpected. During excavation within an area of c.400m 2 on the site, a total of 250 inhumation burials were defined, of which 244 were recorded and excavated; six were not threatened by development and were left in situ.

Clearly, the burials were not scattered or isolated, but represented a medieval cemetery of some longevity. Nor can the remains be attributed to the neighbouring Priory of St Andrews; the burials lay outside the known extent of the priory precinct, and osteological assessment demonstrated that the cemetery population was typical of a lay community, comprising men, women and children (see The Human Bone). This population would be typical of a parish church; however, no such foundation had previously been identified with this site, and so identification of the cemetery and its associated institution required further investigation.

The Cemetery: Problems of Interpretation

In attempting to ascertain the identity of the Fishergate cemetery using documentary sources, a number of candidates present themselves; although none of the documents associate the site with a particular institution, and all have been located elsewhere by antiquaries, St Helen's, Fishergate, emerges as the strongest contender.

A number of religious institutions are known to have been situated on Fishergate, including the churches of St Stephen, St Andrew, All Saints, and St Helen, and chapels dedicated to St Catherine and St George. Of these examples, St Stephen's, St Andrew's and St George's are all securely located elsewhere, and as such can be discounted from the outset. Of the remaining candidates, St Catherine's is poorly documented, mentioned only indirectly in a deed describing land 'near to the chapel of St Catherine's' (Ousebridge Master's Rolls; Raine 1955, 301-2). Raine mentions that the chapel occurs in wills dating to 1445, 1468 and 1472.

All Saints' Church is likely to have been a pre-Conquest foundation, having been granted to Whitby Abbey in the 11th century (Rollason 1998, 207). The church is documented throughout the 14th and 15th century (an anchoress dwelt in the churchyard in 1428 and 1444), and is thought to have been evacuated by Whitby during the Reformation, in 1536. The site of the church is known to have existed on Fishergate, and has been located approximately 140m from Fishergate Bar, where, in 1724, 'a large building resembling a church' was discovered (Raine 1955, 300). Many burials and some stone coffins were also identified. In 1826, beneath the nearby site of the Cattle Market, more human bone was found; further discoveries of skeletal material were made in 1855 and again in 1946. Since the Second World War, inhumations have been encountered on four separate occupations in the vicinity of these finds (YAJ 1978, 16; Medieval Archaeology 1988, 291; 1992, 274; YAT Gaz 1991.15).

Together, these finds strongly suggest the presence of a church and associated cemetery in the area of what is now the Barbican Centre. The identification as All Saints' is accepted widely, although some reservations have been expressed. Sarah Rees Jones (1987 I, 57) has drawn attention to a document dated to between 1265 and 1278, recording an agreement made by the community of All Saints' over the diversion of water in their 'area' in Fishergate. Rees Jones suggests that this may reflect a location closer the river, on the western side of Fishergate. However, such a suggestion would leave the physical remains of the cemetery and possible associated church building near the Barbican requiring an identity.

This leaves the final possibility of the Church of St Helen, which was located somewhere on Fishergate and is believed to have been one of the earliest foundations in the area. Although traditionally considered to have been situated to the east of Fishergate, there is very little evidence to support such a premise; other widely accepted church locations have previously been proved wrong.

Documentary history of St Helen's Fishergate

From an early period, the Church of St Helen appears to have been associated with Holy Trinity Priory, Micklegate (Torre 1690, 133), to which it was granted by Ralph Paynell in 1086 (Wilson and Mee 1998, 83). Holy Trinity Priory, and the Church of St Helen and 'toft in front', were granted to the monks of St Martin Marmoutier c.1090-1100 (Benson 1911, 77; Raine 1955, 302; Wilson and Mee 1998, 48). The church would therefore have been founded some time prior to the late 11th century; Tweddle, Moulden and Logan go so far as to suggest that St Helen's may have been an Anglian foundation (Tweddle et al 1999, 179).

Few documentary sources survive relating to this church, and so little can be reconstructed of its history. Drake refers to testamentary burials and a list of rectors compiled by 'Mr Torre', but does not elucidate this information further (Drake 1736, 250). Although apparently not available to Raine (1955, 301), these documents, collected by James Torre in 1690, are extant (Torre 1690). Torre records a list of eleven rectors, between 1275 and 1475, most of which appear to have been patronised by Holy Trinity. A single testamentary burial is also recorded, of one John Walwyn; the testament was proved on 12th July 1442, and stated that 'his body to be buried in this church of St Elene in Fishergate, Ebor' (Torre 1690, 133). The church is considered to have been 'a small unimportant church', which may explain the general deficit of recorded grants. As one of the poorest foundations in York, St Helen's was taxed at the lowest rate (Raine 1955, 301): in 1397 with a group of seven churches is was deemed to be worth £3; in 1423 it was worth 20s; by 1492 it was assessed at only 2s 6d (Wilson and Mee 1998, 83). Raine (1955, 301) records the presence of a female recluse dwelling in the churchyard in 1435 (potentially confused with the anchoress of All Saints'). The churchyard fell out of use in the mid-16th century; in 1549 the churchyard was rented out (Wilson and Mee 1998, 83). In 1585 or 1586, the parish was united with St Lawrence, the church was demolished, and the land subsequently sold (Tillott 1961, 366; Palliser 1974, 97).

The Church of St Helen has been associated with a hospital, also situated on Fishergate, and referred to as a 'spittle house in Fishergate, beside St Helen's' (Widdrington 1663). This house is considered to have been in existence in 1399, and is documented into the 15th century, being demolished finally in 1622; at this time, 'timber and stuffe' from St Helen's hospital was transported to St Anthony's for their use (Raine 1955, 301).

Location of St Helen's

Following its demolition, the precise location of the Church of St Helen appears to have been lost. Drake (1736, 250) states that 'near the further windmill, where some coffins have been lately dug, stood once the parish church of St Elene or St Helen'. Drake does not directly link the coffins to the medieval church, although it seems likely that they prompted his mention of the church; moreover, it seems probable that the coffins belonged to the Roman cemetery. Although Drake depicts a windmill on the western side of Fishergate (after Benedict Horseley 1694), he places St Helen's Church to the east of the road. Skaife, possibly following Drake, situates the church about half way down Winterscale Street (Raine 1955, 301), although the reason for this is unclear.

There are good reasons for questioning the accuracy of Drake's positioning of St Helen's, and the namesake church of St Helen-on-the-Walls provides an example; in many ways, the life of this church reflects what little is known of St Helen's, Fishergate. St Helen-on-the-Walls was a small, poor, but active medieval church, which was unified with another parish church in the 16th century, quickly demolished and all but lost from memory, being misplaced later by several antiquarian scholars led by Drake in 1736 (Palliser 1980, 20). The church was closed following union with St Cuthbert's, and demolition appears to have occurred soon afterwards. The church remains were apparently not visible to Speed in 1610, and owing only to folklore that St Helens was where Constantine's tomb was located (St Helen was canonised as the mother of Constantine the Great), Drake positioned the church, knowing only that its stood in the parish of St Helen's Aldwark. Subsequently, Drake's assumption had been reiterated by later scholars, until excavation in advance of development on the site of the Ebor Brewery encountered a church and cemetery, and a strong case was built by the excavators for the remains being that of the church and cemetery of St Helen-on-the-Walls (Magilton 1980).

Although inaccurately located by antiquaries, the references to St Helen's, Fishergate demonstrate a consensus that the church of St Helen, Fishergate was situated in the general vicinity of Fishergate House, albeit on the opposite site of the road. In the light of definite evidence to suggest otherwise, the cemetery on the site of Fishergate House can reasonably be attributed to this foundation.

The Date of the Cemetery

Dating evidence for the cemetery site was not abundant, but through detailed consideration of the stratigraphy, ceramic material, rare grave goods and a targeted sequence of radiocarbon dates, the date parameters of the cemetery have been broadly established (see The Stratigraphic Diagram and Dating). Initially, it was assumed that the cemetery was of later medieval date; ceramic recovered from graves dated from the 10th to the 17th century, but suggested a floruit from the late 14th into the 15th century.

An unexpected result was produced by one of the stratigraphically earliest burials (Inhumation 131), which returned a radiocarbon date of AD 730±40 (SUERC-5359: GU-12719); previously, it had been assumed that all burials were medieval, with the exception of one possible Roman inhumation. In the field, Inhumation 131 was thought to have cut a Period 4 pit, F125F. F125F was dated by a reliable assemblage of Torksey Ware to the late 10th century, prompting reassessment of photographic records, which indicated that the relationship was not well-defined on site, and that the true northward extent of F125F was not clear. It appears more likely, therefore, that Inhumation 131 was cut by F125F; remarkably the skeleton was not disturbed.

The presence of Anglian inhumations at the site is perhaps not surprising, given the remains of the adjacent settlement identified at Blue Bridge Lane and 46-54 Fishergate (see Period 3). The settlement has been dated by coins and pottery to between the late 7th and mid-9th century; the burial would fit neatly within this date range. Although comparable cemeteries have been excavated in the vicinity of Anglian settlements at Ipswich, London and Southampton (Scull 2001), little or no evidence survived for burial grounds that may have served the settlement at Fishergate. The relationship between Fishergate and Anglian burials situated over 800m away at Belle Vue House and Lamel Hill, has generally been considered tenuous; Christopher Scull warned against assuming that these cemeteries would have served the riverside settlement (Scull 2001, 72-3). The identification of a contemporary inhumation burial in the immediate vicinity of the settlement would suggest that the population at Fishergate may have been served by a much closer cemetery. At Southampton, it has been noted that 'the sequence appears to show the abandonment of the earliest cemeteries in favour of churchyard burial within the settlement at some time in the eighth century' (Scull 2001, 73). Although an Anglian church cannot be surmised from a single inhumation of 8th century date, this remains an attractive possibility, particularly given the documented early date of St Helen's.

Inhumation 250, also one of the earliest burials, was considered during earlier phases of post-excavation work to have been Roman in date (and allocated Roman Inhumation 1). However, a radiocarbon date indicates that this burial, and three further examples, can be securely placed within the Anglo-Scandinavian period, producing dates of AD920 ±35 (Inhumation 250; SUERC-5365: GU-12722), AD1010 ±40 (Inhumation 208; SUERC-5367: GU-12724), AD1025 ±40 (Inhumation 189; SUERC-5366: GU-12723) and AD1035 ±40 (Inhumation 156; SUERC-5360: GU-12720). These results provide evidence for a more substantial cemetery at the site rather than dispersed burial, and would fit neatly with the documented pre-1100 date of St Helen's. However, when non-burial features are considered, the sequence becomes more complicated, and more difficult to explain. The 8th century burial was at least overlain by the late 10th century pit F125F, which appears to belong to a period of occupation represented by an SFB (Structure 1) and associated hearths, postholes and rubbish pits (see Period 4). In addition, Period 3 cesspit F64F, situated a few metres away from Period 3 Inhumation 131, was in use during the mid-9th century. This sequence of burial and intermittent occupation seems too complicated, and further dates are required to clarify the sequence.

The latest dates for the site were provided by radiocarbon-dates from two of the stratigraphically latest burials. These burials, Inhumations 64 and 14, produced dates of AD 1420 ± 40 and AD 1545 ± 40 respectively, and confirm the impression of a long-lived cemetery given by the deeply stratified and intercutting burials on the site. A 16th century date for the end of the cemetery would, again, fit with documentary evidence for St Helens, and for many smaller churches in York, which declined following their union with neighbouring parishes.

The Extent & Development of the Cemetery

During excavation at Fishergate House, significant evidence was revealed for the demarcation of cemetery boundaries and for the changing extent of the cemetery over time. A number of linear features were defined and excavated which appear to delimit burial at the site.

Early Cemetery Boundary

A ditch encountered in two sections (F263F and F282F) was allocated to Period 7 on the basis of the ceramic contained within its backfill. F263F consisted of an east-west orientated ditch, identified within Intervention 4, and measured 2.00m wide and 1.20m deep. The composition of the lowest backfill (C1514F) suggested slow accumulation in water, indicating that the ditch had remained open for some time, before more rapid backfilling represented by later fills. Ceramic material within the backfill included large quantities of plain and peg-tile, with sherds of Humber Ware and Brandsby-type Ware, suggesting that the feature was backfilled soon after the mid-14th century.

Fishergate House, early phase cemetery plan (Interactive SVG image)

In the eastern extension of Intervention 4, ditch F282F was identified and excavated. F282F was orientated east-west, and upon excavation proved to be 0.60m deep; again, the composition of the lower fill appears to indicate gradual accumulation of material, before the ditch was backfilled and levelled. The latter fills contained York Glazed Ware, York Gritty Ware, York Splashed Glazed Ware and Brandsby-type Ware, as well as plain roof tile and no brick. When projected, F263F and F282F would appear to represent two lengths of the same boundary ditch.

Significantly, no inhumation burials were encountered to the south of ditch F263F/F282F. On the northern side of this feature, the slightest suggestion of an accompanying bank might be interpreted from the offset of the closest burial to the north (F275F); the distance of 1.20m is equivalent to the width of the ditch. The final backfill of F263F consists of a mixture of soil and subsoil, what can be interpreted as its original upcast and may represent bank make-up. Together, therefore, these features would appear to represent the demarcation of the southern boundary of the cemetery in the form of a bank and ditch, at some time prior to the 14th century.

Cemetery contraction

Excavated evidence suggests that the cemetery had contracted by Period 8. A second ditch (F288F) was identified to the north of the earlier boundary. Unlike F263F/F282F, this ditch cut into, and removed, parts of preceding burials; to the north of this apparent boundary, burials were found to be much denser. Together, this evidence strongly suggests that the cemetery contracted, and that F288F represents the demarcation of the more restricted cemetery area.

Fishergate House late phase cemetery (Interactive SVG image)

Consequently, burials encountered to the south of ditch F288F, but to the north of F263F/F282F, have been assigned to the earliest phase of burial in the cemetery; burials cut by the digging of F288F have also been included. In Interventions 1 and 2, where this type of sequential relationship was lacking, only the stratigraphically earliest graves have been included in the early phase cemetery plan, albeit somewhat arbitrarily. As a result, the apparent density of burial between F263F/F282F and F288F, when compared to Intervention 1 and 2, is artificial, resulting from the enhanced stratigraphic and spatial information available for the area surrounding the shifting boundary ditches.

The likely area of the cemetery can be reconstructed using the available evidence for burials at Fishergate House and Marlborough Grove, the projections of boundary ditches, and the frontage represented by Roman flanking ditches and possible medieval surfaces identified by YAT. This area covers approximately 1400m 2 ; occupying a thin strip of land running perpendicular to Fishergate, westwards to the river. Initially, this might appear to represent an unusual form; if the urban landscape of Fishergate is considered, however, the site can be placed within a more comprehensive context.

Fishergate cemetery and postulated position of St Helen's Church (Interactive SVG image)

In c.1100, the grant of St Helen's to Marmoutier included both the church and the 'toft in front' (Raine 1955, 302), suggesting that the church was established within a pattern of medieval landholding, in which the land would have been divided into tofts. Rees Jones notes that by the 13th century, much of Fishergate was partly developed with housing (Rees Jones 1987 I, 68); some plots of land are specifically referred to as being in the vicinity of St Helens. The husgable roll of c.1284 suggests that Fishergate contained a minimum of 75 tofts, owing 7s 4d to the crown. The evidence indicates that from an early period Fishergate would have been divided into tofts, which presumably occupied linear plots fronting onto the street. The plot occupied by St Helens would have fitted into such a plan; the cemetery may have contracted as land was sold off to raise revenue for what is known to have been an impoverished church (Dr Lawrence Butler pers. comm.).

The contraction of the cemetery was certainly followed by sequences of pit-digging and postholes, indicative of domestic activity and attributable broadly to Period 5 to 8. Such features either interrupted graves directly or redeposited already disarticulated remains suggesting some depth of sequence in the small window of activity encountered. The earliest features to interrupt burials were of possible Period 5 date; two Period 5 pits, F214F and F258F, were stratigraphically later than burials radiocarbon dated to the early 10th and early 11th century. This might suggest contraction of the cemetery during the 11th century, although the early boundary was certainly not disused until the 14th century, and this seems the most likely date. In any case, the evidence suggests that the pressure on urban space was felt this far out of the city from as early as the 11th century, and the toft granted to the monks of Marmoutier probably reflects Anglo-Scandinavian land holding patterns; certainly by the 13th century the toft system was firmly established.

Orientation, Burial Rite & Layout of the Cemetery

The early phase plan of the cemetery demonstrates some degree of order in the arrangement of burials. Some of the burials appear regimented, with the western end of burials respecting the eastern end of the next 'row'. Although no evidence for grave markers was recovered, this would indicate that the burials were marked or visible in some way. Notably, some of the earliest burials within the cemetery are complete, although this can potentially be attributed to protection from the accumulating soils from burials.

No change in burial orientation has been identified during the later phase of burial; most burials conform to a west-east alignment, deviating by only a few degrees and probably influenced by the form of the toft. Those that do differ markedly do not appear to demonstrate any particular significance in terms of time and space, nor do they appear to have been caused by local features such as paths or boundaries. The shifting cemetery boundaries, which occupy slightly different alignments, do not appear to have affected burial orientation in any way. Later burials do, however, exhibit more evidence for intercutting, which is likely to be owed to increasing pressure for space.

Neither the distribution of burials by age, nor by sex (based on the distinction of adult/subadult) demonstrated any specific segregation during the early phase of burial. The burials ascribed to the early phase comprised twenty males, fifteen males, twenty-eighth sub-adults and ten undetermined; this reflects the overall demography identified for the cemetery as a whole. In some cases, juxtaposition of burials, or a particular sequence, can indicate some defined groups or associations, tentatively assigned to familial associations. Two baby skeletons, Inhumations 162 and 163, were identified side-by-side in Intervention 2, and osteological analysis found that they belonged to foetuses both aged 34 weeks in utero. Although no grave cut was defined, osteological analysis suggested that the burials represent premature twins, likely to have been interred together. A double burial, possibly of a mother and child, was also identified. Inhumation 40, a mature adult female, was found to have been buried with the remains of a child aged 1-2 years (Inhumation 41) laid across her chest. Other familial groupings may have been present, but the dense, intercutting burials mean that any combination of age and sex can potentially be identified, and without more conclusive associations, such conjecture has been avoided.

Inhumations 162 and 163, twin foetuses grave plan (Interactive SVG image)

Generally, the burials within the cemetery exhibited a remarkable uniformity, being in an extended supine position with extended arms placed to either side, crossed over the abdomen or placed on the pelvis. The presence of shrouds was often implied by burials, which appeared particularly confined or 'tidy' with the knees and feet together, or hands placed over the pelvis. Only one possible copper-alloy shroud pin was recovered from Inhumation 21; other examples, therefore, would seem to have been sewn, tied or otherwise fastened with organic material. Of 250 burials, ninety-nine appeared to have been buried with a shroud, and showed no clear evidence for specific distribution.

Within the cemetery, a number of 'special burials' were identified, which provided evidence for higher investment in burial within certain areas of the graveyard. Six burials have been ascribed to this group, including those inhumations which appear to have been coffined, and those buried with associated grave goods.

Coffined Burials

Two coffined burials were identified during excavation, one of which appears to have been interred within a wooden coffin (Inhumation 88), and another which was buried within a rectilinear arrangement of upright stone blocks forming a composite stone coffin (Inhumation 244).

Inhumation 88 was found to contain the remains of an old- middle adult burial, aged between 35 and 45 years old. Forty-six iron coffin nails were located; the nails were predominately found to have been located on the southern side of the grave. The distribution of the nails is suggestive of a planked lid, held with nailed wooden batons, which appears to have collapsed southwards during decomposition.

Inhumation 81, coffin burial (Interactive SVG image)

Inhumation 244 was not fully excavated; only the eastern end was identified within the area of investigation, which negated the need for mitigation excavation. The presence of an inhumation was confirmed, but further excavation was not carried out. The upright, squared blocks that formed the composite stone coffin were found to represent reused building material, suggested in particular by the presence of mortar adhering to the stone.

Inhumation 244, composite stone coffin burial (Interactive SVG image)

Grave Goods

Four graves were found to contain grave goods, including a copper alloy ring (Inhumation 139), a copper-alloy cross mount (Inhumation 123), an iron key (Inhumation 214), and a pierced scallop shell (Inhumation 108). A second key was recovered from the backfill of another grave (Inhumation 143), but cannot be securely associated with the skeletal remains, and as such must remain only a possible example.

The copper alloy ring was found on the left middle finger of a female aged between 35 and 45 years (Inhumation 139), identified within a well defined grave-cut. The ring, of a type known as a merchant ring, comprised a welded band of copper alloy with a soldered oval bezel on which the initial 'I' was cast with primitive sprigs of foliage incised to either side. Although not precisely dated, the ring is believed to date to the 15th century; as Inhumation 139 represents one of the latest burials stratigraphically, this would seem to represent a reliable date.

Inhumation 139, female burial with ring (Interactive SVG image)

Inhumation 139, ring detail, and ring

Inhumation 139; Inhumation 139 ring detail; Copper alloy ring from Inhumation 139

Inhumation 123, that of a male skeleton, was found to have been badly truncated and disturbed by three successive burials. During disturbance, the sacrum of the skeleton had been flipped over; once lifted, this was found to have covered a copper-alloy mount, which would originally have been located, face-up, over the pelvis. The object comprised an ornate, cast copper-alloy mount, diamond in shape, and adorned with cast decoration in the form of an expanded-arm cross with straps and ropework around its edges. At two opposing corners, an arrangement of three clenched rivets is preserved, and mineralised wood was found to adhere to the back of the object. The mount had clearly been mounted onto a wooden object, which may have been a mortuary cross. Such objects are not uncommon in Christian burials, although they are frequently made of lead. A comparable example to the Fishergate mount was recovered at Coppergate identified as a lead matrix (Mainman and Rogers 2000, 2476, fig 1998). The Ringerike decoration on the object has been used to ascribe a date in the early 11th century; if comparable, this would place Inhumation 123 in the earliest phases of burial at Fishergate. The grave cut only one other burial, so this would seem to be a reasonable assumption.

Inhumation 123, male burial with possible mortuary cross (Interactive SVG image)

Find no 371, front
Find no 371, back
Cross mount from Inhumation 123, front & back

Inhumation 214 was well-defined during excavation, but had suffered some horizontal truncation from overlying Inhumation 226 and had been cut away to the west by a modern drain. The remains of a mature female (45+ years) were identified, and the position suggested a shrouded burial. In addition, an iron slide key was found with the skeleton, close to the waist on the right side. This form of key dates to the late 11th to early 13th century and the burial also lay within the early cemetery. Again, Inhumation 214 was found to cut only one other burial and allocation to the early phase using the date of the key would seem to be reasonable.

Inhumation 214, female burial with iron slide key (Interactive SVG image)

Inhumation 214

Inhumation 214

Of the grave goods, perhaps the most informative example was a pierced scallop shell, identified close to the waist of a mature female inhumation (Inhumation 108). The shell may originally have been suspended on a belt from her waist or sewn onto a bag or similar.

Inhumation 108, female Santiago pilgrim burial (Interactive SVG image)

Inhumation 108

Inhumation 108; Close-up of pilgrim scallop shell; Pilgrim scallop shell from Inhumation 108

The find of a pierced scallop shell (Pecten Jacobaeus) in the grave of adult female Inhumation 108 provides a link between the cemetery in Fishergate and the phenomenon of medieval pilgrimage. The scallop shell is widely recognised as the badge of pilgrims who had travelled to the shrine of Santiago de Compostela in northwest Spain, and would have been instantly recognisable as such, allowing the wearers to gain alms, hospitality, exemption from tolls and general protection (Dr Peter Yeoman pers. comm.). The scallop is known to have been symbolic of pilgrimage to the shrine of St James from at least the 12th century, although the origins of this association are unclear (Hohler 1957). These shells are known to have been sold in booths around the courtyard of Santiago de Compostela; the Archbishops had 'received papal approval of their monopoly on souvenir production, and the sale of such badges elsewhere was forbidden on pain of excommunication (Yeoman 1999, 116).

Although pilgrim badges could be obtained elsewhere (albeit illegally), this particular shell has been identified to species (Pecten jacobaeus), which is distributed around the Mediterranean sea, off the Canary and Cape Verde Islands and therefore may indeed originate from the northern coast of Spain (Dr Nicky Milner, pers. comm.). Pilgrims are often treated differently in burial grounds and evidence for interment fully clothed is often identified, a special dispensation otherwise reserved for the clergy. The presence of the scallop shell allows identification of a pilgrim and this may also indicate she was buried clothed; the position of the shell within the grave suggests it may have been fixed to a pilgrim's leather bag or scrip (Dr Peter Yeoman, pers. comm.).

Accounts of pilgrimage are common during the medieval period, and Santiago is one of the main European destinations, considered to have been second only to Rome, until political tensions of the late 14th century made such journeys more problematic. The majority of recorded pilgrims are male; females, like clergy, were not required to account for such activities, and are therefore frequently absent from documentary records. Female pilgrims were not, however, uncommon; in the early 15th century, Margery Kempe of Lynn is known to have travelled to Rome, Jerusalem and Compostela (Webb 2000, xiii, 209), and the fact that Chaucer's Wife of Bath had been to Bologne, Compostela and Cologne suggests that pilgrimage would not have been an unfamiliar undertaking for women.

Evidence for the Church

The visibility of the church building, by comparison with the cemetery, has been negligible. As a small low tax church St Helen's Fishergate is likely to have been built as a small one- or two-celled building with little by way of expansion. Again, reference to St Helen-on-the-Walls might suggest a building no larger than 5.8m x 4.4m internally. In addition, the physical remains of the church at Aldwark that survived demolition were no more than cobble- and rubble-filled foundation trenches. If all that remained of St Helen's Fishergate were features of similar composition, the remains would be almost invisible, particularly to the builders of Fishergate House. Considering Fishergate House was built in 1837 and no mention of a cemetery occurred during construction, the church remains themselves are unlikely to have been mentioned if they were visible in any way.

It seems probable that the assemblage of building material recovered from the cemetery ditches, and occasionally from graves, belonged to the church building (see The Building Materials). The contrast between the ceramic profiles of Fishergate House and Blue Bridge Lane was noted for several periods, indicating the likely independent sources of material at both sites; the building material need not, therefore, be considered as having derived from the Priory. Indeed, the only length of lead came from the entire site investigation was recovered from the backfill of a Fishergate grave. In addition, three sherds of window glass, one painted, and all dateable to the 12th to 15th century, as well as structural ironwork are likely to have derived from the building itself. The large quantity of medieval plain roof tile and fragment of architectural stone recovered from just a small length of the levelling of ditch F288F, probably in the 16th century, are exciting indicators of the character of the elusive church building. Just where the church or chapel would have lain in relation to the cemetery cannot be postulated easily, although the area used for burial without evidence for structures can be roughly surmised, which provides a guide to where the building certainly was not.

The only possible evidence for structural activity on the Fishergate House site was represented by a robber trench or cobble-filled foundation trench (F237F), identified in the north-facing section of Intervention 2. The feature appeared as a sub-rectangular profile filled with mixed voided rubble, comprising large rounded cobbles and limestone fragments set in a loose matrix of brown clayey silt. The feature had been cut (but not disused) by an encroaching inhumation burial, and has therefore been allocated to Period 6 (since burial appears to be densest during Periods 7 and 8), but no dateable material was recovered during excavation.

St Helen's-on-the Walls, Aldwark, is the best comparison from York and in its early phases measured a mere c.7m x 6m externally, expanded to c.13m x 6m with walls c.0.70m wide (Magilton 1980, Fig. 5 and 6); both are dimensions which could easily have been lost beneath the footprint of the early 19th century great house.

The dogma and superstition which prevailed in the choice of burial spot, provides some clues as to a possible site for the small building. The east end of the church both internally and externally was considered to be the high end, and preference for burials close to this area would have been the privilege of the wealthier members of the congregation. The 'special' burials, most notably the only stone coffin, were encountered towards Fishergate, pushing the probable location of the east end of the church, to the west of these burials. In addition, a pan-European superstition resulted in the burial of the unbaptised and suicides on the north side of the church (Magilton 1980, 10). Intervention 2 at the northern extreme of Fishergate House was the location for a small tightly buried 'nest' of five foetuses and four neonates who were, by virtue of their age, most probably stillborn or very young at death, and therefore the least likely candidates to have received the sacrament. These glimpses of order, in the otherwise apparently random burials of the cemetery, if guided by the prescriptions of medieval burial geography, suggest the location of the church beneath the stable block footprint of Fishergate House within the projected line of the burial enclosure.

Disuse of the Cemetery

The combined evidence of ceramic deposition, radiocarbon dating and historical references suggest that burial did not continue much into the 16th century; although the lack of activity on the site (restricted to the small assemblage of pottery and a single trade token), suggests that the sanctity of the cemetery was respected for some time following its disuse. Documentary references mentioned by Palliser suggest the cemetery land of St Helen's was rented out in 1573, but then reclaimed by Queen Elizabeth who gave the land to William Wentworth (Palliser 1974, 97). Given the lack of ceramic deposition the cemetery is likely to have been given over to open land, possibly pasture. It is unlikely that local knowledge of the site would have disappeared instantly; the accumulation of soils across the site, and eventual turfing over of existing traces of burial, would have resulted in the gradual loss of the cemetery from communal memory. Like the Priory, the church fabric is likely to have been robbed comprehensively and apparently no significant traces of the church were visible to Speed in 1610.

Discussion

By drawing together the available strands of evidence for burial and building activity on the site, it has been possible to chart the development and contraction of a cemetery on the site of Fishergate House, potentially from the 8th century onwards. Conclusions drawn from a single, Anglian burial must be necessarily tentative, but neither can the potential significance of this discovery be overlooked. The proximity of the burial to the settlement at Blue Bridge Lane provides an obvious context for its existence, and potentially for the establishment of an early church, which is already known to have existed from at least 1086. The link with Holy Trinity provides an even more tantalising link to the somewhat fragmentary history of Anglian York. Holy Trinity has been linked with the church and associated monastery of Alma Sophia, described by Alcuin in 780 (Wilson and Mee 1998, 38). The link documented during the refounding of Holy Trinity in 1090 may have been a perpetuation of an existing relationship between an Anglian minster, and an outlying foundation that served the riverside community.

More substantial evidence survives for a later pre-Conquest foundation, represented by at least four burials dated to between AD920 and AD1035. This provides further support for identification with the church of St Helen, known to have been situated on Fishergate prior to AD 1086. Using the distribution of burials, in particular those of special graves, neonates and foetuses, it has been possible to propose a hypothetical location for an associated church; this would appear to be supported to a limited degree by the location of a possible foundation trench.

The sequence is, however, complicated by the intermittent non-burial features that are interspersed chronologically with burials during the 8th to 11th century. Further dating would be required to clarify this situation. Based on information currently available, a broad sequence indicates burial associated with the 8th century wic; once the wic declined, the cemetery was used for refuse pits, before being used again for burial in the early 10th century. Notably, the early 10th century burial is thought to have been that of a near-decapitated individual; this appears to represent an aberrant burial, and would not necessarily have been buried in consecrated ground. This was succeeded by a period of occupation in the late 10th century, before the foundation or refoundation of a church on the site in the 11th century, which continued to the 16th century.

By the 13th century, it would seem that the landscape surrounding the cemetery had been divided into individual tofts, which left the cemetery occupying a linear plot fronting onto Fishergate. The pressure on space within this developing urban area appears to have impacted what is known to have been an impoverished church, and the cemetery contracted; the external area was then given over to domestic activity. It is possible that this was undertaken at the decree of Holy Trinity, who are known to have provided the church with rectors from at least 1275.

Finally, the church fell into disuse following the union of its parish with St Lawrence. The land was leased and eventually sold, and appears then to have become open land, possibly used for pasture or arable purposes.

apc > monographs > blue bridge lane & fishergate house > excavation: the cemetery