SHETLAND: S6. Papil, North Yell · Introduction At the north-east tip of the large island of Yell is the once-important site of Papil, witnessed by the many Papil-topographical names around. Close to the head of Papil Bay, on the south side of the large headland of Papil Ness and just to the east of the southern end of the Loch of Papil, is the settlement of Papil (NB, Shetland, Book 19, 51). Papil Ness marks the entry to Bluemull Sound, which divides Yell from Unst and is noted on several of the earlier maps (Blaeu 1654; Preston 1781; Depot Generale de la Marine 1803b; Arrowsmith 1807; Thomson 1827; Blachford 1846), while the settlement of Papil and Papil Bay are first named on Thomas’s map, dated 1838. The first documentary reference to Papil in Yell is in a famous series of depositions concerning the inheritance of Gudrun Sigurdsdotter in Faeroe, Norway and Shetland, a transcript of which was drawn up in 1407 (B.and S, 1999, nos.10-13). One of these concerned the control of the Faeroese property by another woman, Ragnhild Havardsdottir, and it had been written at Papil on 9 November, 1405 (ibid.no.12). Presumably one of the actors in this legal tangle lived at Papil, indication enough that it was a place of status. A century later, in 1506, xxii marks of land in ‘Pappale’ were left to Thorvald of Brough and his wife and heirs in the remarkable will of Sir David Sinclair, bastard son of the last earl of Orkney (Crawford, 1978,). This holding was probably one of the many purchases of land which we know Sir David made, one purchase of land in Unst being conveyed at Brough (B.and S.no.33). This is some good documentary evidence that Papil in Yell was a place of some value in the medieval period.48 On the OS First edition map (1878), Papil is depicted as two roofed buildings, an enclosure and a well, lying just to the north-east of another (unnamed) group of buildings within a rectilinear enclosure. Papil is described as a house in the Name Book (Shetland, Book 19, 51); one of the buildings at Papil is shown as unroofed on the Second edition map (1900), but both clusters are still in occupation today. The unnamed cluster, which is situated immediately behind the beach, is annotated as ‘Site of chapel and burial ground’ on both of the earlier editions of the OS map (see below). A number of small farms lie to the west of the Loch of Papil, but, to the east, the headland itself - though crossed by field walls - is depicted as rough pasture. The standing stone, which once lay to the north west of the chapel, is described by Irvine as lying on common land known as the Garths of Papal (1887, 216-7; RCAHMS 1946, no. 1749; NMRS no. HP50SW5, HP 5378 0425). This stone had, in one corner, the initials ‘H.I. 1674’ and ‘M.W. 1683’, alongside ‘an old merchant’s mark or something of that description’, while there is a tradition it marked the burial place of a queen who had come in a ship to Papal (Irvine 1887, 216-7). On the modern OS map, Papil Bay is shown as a largely sandy bay extending south as far as the Ness of Cullivoe.
The OS Name Book records that traditionally a chapel of considerable dimensions stood at Papil: it was at the south end of the Loch of Papil. Although no trace of the building remained at the time of the surveyors’ visit they did observe that great quantities of human remains had been found from time to time on the site (NB, Shetland, Book 19, 51; Brady 2000a and b, Site 7; these bones are suggested in the NMRS to have come from the kitchen midden described below). Irvine of Midbrake, in the early nineteenth century, noted that the Kirk at Papil was ‘held very holy’, while stone coffins, containing skulls and other bones, had been discovered from 1820 onwards, at ‘the lowest depth of the soil’ (Irvine nd, quoted in Brady 2000a, 36). Candles were apparently burnt in the church on Candlemas night (ibid.). J.T. Irvine, later in the nineteenth century gave the dedication as St. Ninian (ibid.,11). The RCAHMS visited the site in 1931 and described the midden material next to the chapel site as representing domestic refuse rather than erosion of the burial ground and noted the foundations of a possible building, ‘too much covered with vegetation and blown sand to admit its character being determined’ (RCAHMS,1946,no.1732). In the twentieth century, this chapel site at Papil has been described in more detail. Macdonald and Laing recorded a north-south oriented building in the corner of a triangular enclosure which is the traditional site of the chapel (1969, 127, no. 1). The footings were set on a slight natural ridge and completely turf covered, making it difficult to distinguish natural from artificial (ibid.). The building measured 28ft N-S by 17ft E-W, over walls c.1ft high, although the east wall was no more than footings. The enclosure walls were turf-covered ridges 1ft to 18ins high, with much stone visible on the east side (ibid.). The building occupied the bulk of the enclosure, the long east side of which abutted the beach. The midden, comprising limpet shells and animal bones, along with traces of crude walling, was visible in the adjacent cliff edge at a lower level, indicative perhaps of the fact that it represents quite early occupation (information from OS surveyor, 1969). The midden formed a mound and following considerable erosion after a storm, ‘pre-broch’ and ‘broch’ period levels were exposed (Beveridge 1973, 50). These layers contained abundant shellfish and faunal remains, rims and body sherds of ‘broch period’ pottery, worked bone and quartz, including a parrallelopiped bone dice and a perforated ox bone scapula shovel (ibid.). Further stone walls have been noted in the vicinity of the supposed chapel (information contained in SMR, PRN 2090 where the mound is not believed to be structural in origin). This large mound is still subject to the activities of burrowing mammals and is eroding on the seaward side (Brady 2000b, 10); rough pottery, bone, shell and burnt stone from the midden levels were noted by the HS monument warden in 1995 (information contained in SMR). The rapid coastal erosion in the area noted by the warden had resulted in finds of human skulls on the beach, which had been allowed to wash away by the owner. There is nothing to firmly identify the building footings as a chapel and, given their stratigraphic position on top of the mound the most recent survey by University of Glasgow suggests that they are of post-medieval origin (Brady 2000b, 9-10). Although in the vicinity of the chapel there appears to have been little change since the visit by Macdonald and Laing in the late 1960’s, the landholder reported to Brady that 1-2m of the beach was being eroded each year along the bay (ibid., 9). He believed the foundations of the chapel were located some distance out into the bay and were only visible at low water (ibid., 10), although it is also noted in Brady’s report that he was opposed to any archaeological work taking place on the site. There is clearly an important multi-period site here, and the Glasgow survey makes recommendations concerning future investigations. Further south along Papil Bay and just north of St. Olaf’s at Cullivoe (see below) is Kellister or Killister; in local tradition there was also a chapel here, thought to be associated with St. Ninian’s at Papil (NMRS no. HP50SW12, HP 5421 0308; Brady 2000a and b, Site 6). The name itself has nothing to do with the Gaelic element -cill (= church)(Stewart, 1987, 230). Only clearance cairns, one containing worked stones, a stone scatter and a longhouse-like building are noted in the SMR at this location, while there appears to be no current local knowledge for a chapel site in the vicinity (Brady 2000a, 10-11; 2000b, 9). The old church of North Yell at Breckon, two miles north-west of Papil - dedicated to St. Olaf - is the potential successor of an earlier chapel at Papil (Muir 1885, 139-40; Dryden 1896, 151-7; RCAHMS 1946, no. 1732; Cant 1976, 15). Muir describes it as lying in the burying ground of Toft (1885, 72; Jakobsen 1936, 109 for general meaning of Toft). There is both an unroofed building and an extensive area of prehistoric settlement, along with the site of a broch, cairns, and a cist, extending from Toft north westwards onto the Ness of Houlland. The church, lying on the west shore of the Kirk Loch, was commonly known as the ‘Kirk o’ Ness’, perhaps because of its proximity to the Ness of Houlland. The church is described by Scott as being located near to the Voe of Papil (FES, Vol. VII, 304), although it would seem to be located too far to the north for this to be a accurate description. The form and fabric of the church indicate a medieval date. It is built of random rubble and consists of a nave and chancel, separated by a (now fallen) Romanesque arch (RCAHMS 1946, no. 1712; Brady 2000a and b, Site 8; a detailed description of the building is given by Dryden 1896, 151-7, while both he and Muir 1885, 138-40 have drawings of the church, the former showing the chancel arch still extant). The church has a number of square headed windows and two recesses in the thickness of the wall close to the east end, one in the north wall and one in the south. According to local tradition, the building remained thatched until the beginning of the nineteenth century (Brady 2000a, 13), but was roofless by the end of that century, although the bellcote on the west gable still survived (Tudor 1883, 551; Dryden 1896, 151). The church lies in the centre of a rectangular enclosure, which had been recently enclosed by a ‘neat iron railing’ in Tudor’s time (1883, 551), but which is now surrounded by a stone wall (Brady 2000b, 10); underneath this are the faint traces of an earlier wall (ibid.). The burial ground remains in use, although sand continues to encroach upon it (ibid.; Brady 2000a, 14). Alongside a well-carved armorial slab, the graveyard contains over fifty small upright stones, some of which may be in their original locations, although others have undoubtedly been re-used (Brady 2000b, 10). A single rude cross-shaped headstone remained to be noted by Irvine in the nineteenth century (Irvine, quoted in Brady 2000a, 34). Irvine records that the church was supposed to have been built in the twelfth or thirteenth centuries, adding;
Irvine also provides an account of the collapse of the chancel arch in January 1863 (ibid.). The successor of Kirk o’Ness in 1750, after it had become swamped by sand, was St. Olaf’s at Cullivoe, which itself went out of use in 1825-6 and is now a ruin (NMRS no. HP50SW13, HP 543 027; Brady 2000a and b, Site 5). Although Muir implies that North Yell had no church until a mission chapel was constructed at Cullivoe after the parish’s unification with Fetlar later in the nineteenth century (1885, 72), according to the OSA, a church - along with one on Fetlar - was newly built in the 1790’s (Gordon 1793, 287); it seems to have been rebuilt in 1832 (Watson 1841, 32). The tradition of a chapel at Brekon or Daall, immediately to the east of Kirk o’ Ness, provides another candidate for the predecessor of St. Olaf’s (Irvine, quoted in Brady 2000a, 36; Brady 2000a and b, Site 9; NMRS no. HP50SW2, HP 528 049). Although no longer locatable, the chapel is described by Irvine as standing ‘at the western end of the sand in Daall prior to the erection of St. Olaf’s church at Ness in the fourteenth century’ (Brady 2000a, 15). The recorded NGR for this site indicates that it lies at the back of the sand dunes in an area of ungrazed meadow, in the vicinity of a deserted fishing settlement, itself now buried under the sand (ibid.; Brady 2000b, 11). Structures, plus artefacts of all periods, in this area indicate its continued importance for settlement (e.g. NMRS no. HP50NW1, HP 528 054). A list of the divisions of the run-rig lands, compiled in 1772, suggests that part of the arable lands, by then part of the farms of Brekkon and Huland, may once have pertained to the church (Jakobsen 1936, 194-5; Houll is marked on the modern OS map at HU 5385 0455). In a commentary on this list, Irvine states that ‘de Kilnategs’ formerly belonged to St. Olaf’s church at Ness. This name appeared to Irvine to be a compound of ON teigr, a measured strip of land, and either the ON kíll, a narrow inlet, or the Gaelic cill-, church (Irvine 1894, quoted in Jakobsen 1936, 194-5). In addition, part of the same lands is called ‘de Papelsflot’ or ‘Papels flute’, where flotr means a strip of arable or grassland (ibid.). Jakobsen also derives ‘Turmanlands’, another part of the field, from the OI termon, originally a fenced piece of land belonging to the church (this is unlikely-ed.)
The island of Yell was divided into three parishes, which Cant suggests may have formed one priest’s district (Cant 1976, 15). Gudbrandur Magnusson, parish priest (soknar prestr) of Yell is mentioned in 1405 (ibid., 15, n.64), while John Chalmer was vicar in 1477 and John Fallowsdale in 1542 (ibid., n.65 & 72). Half of the corn teinds of North Yell, also known as the Glupe, belonged to the bishop of Orkney, with the residual teinds being reserved to the vicar (Goudie 1910, 306; Cowan 1967, 212). At the Reformation, Fetlar was also joined with the three parishes of Yell, remaining so for some length of time after that (Sibbald 1711, 30, 69; FES, Vol. VII, 295). In Pitcairne’s day there were three churches on Yell (1605-1617, 157), which Sibbald a hundred years later described as ‘3 Churches for Sermons, each distant eight miles’ (1711, 30, 69). In addition, there were ‘about twentie Chappells’ (ibid.), of which two or three were still in use at the beginning of the nineteenth century (Hibbert-Ware 1822, quoted in Muir 1885, 136). Sibbald states that, only a short time prior to when he wrote and presumably following the disjunction of Fetlar, two ministers had been appointed to the charge of Yell (1711, 30). In North Yell, in addition to the churches described above, there were pre-Reformation chapels at West-a-firth, Gloup, near Tofts, Kellister, St. John’s at Gutcher and St. John’s at Kirkabister (FES, Vol. VII, 304; RCAHMS 1946, no. 1735; Cant 1976, 15, n.12, 50); at Gloup, Cant says there were chapels at Kirks and Graven (identifiable with Scott’s West-a-firth), while there was also a chapel on Linga in Bluemull Sound (Cant 1976, 15, 50; see also RCAHMS 1946, nos. 1730 and 1733; Macdonald and Laing 1969, 131, no. 10). A stack site has been identified at Aastack, off the west coast of North Yell (Lamb 1976, 146-7; Brady 2000a and b, Site 13; NMRS no. HP40SE2). Mid Yell, also known as Reafirth, was dedicated to St. John and appears to have remained united with South Yell or Hamnavoe, dedicated to St. Magnus, even after North Yell and Fetlar were disjoined (FES, Vol. VII, 301, 305).
No early Christian carved stones are connected with the site of Papil, although one was recorded as having been found at South Garth, 4 km to the south (Fisher, 2002, 55). However Ian Fisher brings to our attention the remarkable number of early grave-stones carved with simple crosses which have survived in grave-yards on both Yell and Unst (ibid.). Their cultural affinities seem to be with Scandinavia in the early centuries after the conversion of the Vikings rather than from the pre-Viking period.
This very important church site would appear to have been both a secular and eccclesiastical centre of importance. The documentary evidence points in that direction, although the archaeological evidence needs a great deal more focus, and investigation, before any certainty about the age of the structures can be ascertained. The survival of folk traditions is also significant. This location has been a centre of population since prehistoric times and we can assume from the name itself, despite the lack of sculptural remains, that it was associated with an early Christian community. The present Land Capability value of 52 might be mis-leading as to the arable potential of the land in times past.
· Soil Survey of Scotland Land Capability Class 52
OS maps:
|
|||||||||||||||||||||||||||||||||||
)
