Introduction
A strand of research, looking at buildings associated with the routes taken by the army of James IV on its way to and from the Battle of Flodden, came to focus largely on ecclesiastical buildings, largely because it was found that the majority of genuinely medieval buildings in the region are churches and monasteries, and it is inconceivable that such establishments, in particular monastic houses, would not have played a significant role at the time of the battle, and especially in its aftermath, when many wounded were trying to make their way home.
In this regard there is a tradition that the nuns of Coldstream priory carried wounded Scottish soldiers back to Coldstream for treatment and it is also of note, in this context, that King James’ cousin, John Stewart Duke of Albany, who became governor of Scotland after Flodden on behalf of the young James V, ‘marched to Eccles a monastery about six miles distant’ (Buchanan, History 2: 287-8) following the failure of his campaign late in 1523 to take the newly fortified stronghold of Wark Castle.
Following the initial visits a number of ecclesiastical sites were chosen for more detailed investigation. In May 2016 a search was made for the site of St Ethelreda’s Chapel, on the ancient Staw Road in the Halterburn valley, near Yetholm, based on earlier research by Flodden Project volunteer, Tom Broad. Exploratory excavations on a raised mound in the narrow flood plain uncovered structural remains and medieval pottery, suggesting it as a possible site of the chapel, or perhaps an associated residence. 
Longformacus grave slabs with cross of pain motif similar to those at Coldingham
Coldingham grave slabs - with cross of pain motifs - of possible 16th century origin
At Abbey St Bathans, site of a former late medieval Cistercian nunnery, test pitting revealed settlement remains immediately east of the current church and churchyard, while the site of the purportedly earlier chapel of St Bathan to the south-east, was excavated in order to reveal and record structural remains last seen, and perfunctorily recorded, in the 1860s. In addition, two days in May 2016 were spent recording the complex structural remains of the multi-phase medieval church of Preston. Finally, a visit to a private burial ground in the grounds of Wedderburn Castle revealed the head and base of a medieval cross reputedly linked to members of the Hume family killed at Flodden and in a preceding campaign of 1497, but subsequent excavations, led by John Hume-Robertson, failed to find any evidence of burials on the site.
In addition to surviving medieval and early postmedieval church remains, a possible tangible link to the battle was found in a series of grave slabs, several at Coldingham Priory and single examples at Bassendean, Edrom, Greenlaw and Longformacus, which are probably of early 16th century date and have a distinctive cross motif, identified in a 1911 Berwickshire Naturalists Club article as a heraldic ‘cross of pain’. In addition, all have marginal panels as if for an inscription, although there is no sign of any ever being carved.
Crosshall Cross near EcclesA link to Flodden cannot be ruled out since, stylistically these are slabs of a type which would normally cover high-status burials and could well be of the early 16th century, but some look to have been produced in a hurry and never finished, tenuously suggesting a possible link with the aftermath of battle. Some features of the incised crosses on the Crosshall Cross also appear later medieval (15th or early 16th century) and may be secondary, with triangular terminals similar to some at Coldingham.
The medieval Church and church architecture in Scotland
The tides of history have erased the medieval ecclesiastical landscape of Scotland far more thoroughly than that of England, where a parish church dating back the better part of a thousand years is still present in most villages. In the whole of the country of Scotland less medieval churches - around 50 in all - remain in use than in any one of the English border counties. Virtually all English churches remained in use after the Reformation, albeit with a reordering of their internal fittings and furnishings; Victorian restoration - which often entailed a return to an idealised medieval arrangement - has been the main threat to their fabrics. Even in industrialised areas such as West Yorkshire only a handful of churches do not retain something of their medieval predecessors.
In Scotland, where the Reformation took a far fiercer hold, it is a very different story. The ripples of John Knox’s famous sermon in Stirling, in which he encouraged mobs to wreck and loot churches, spread across the country. Churches were rebuilt or remodelled so as to be more effective houses for the reception of the Word rather than the Sacrament, the pulpit being moved so as to be the central focus rather than the altar. Often it was set in the centre of the nave, with an extension - here usually termed an ‘aisle’ although in English terminology more like a transept - opposite it giving the main body of the church a characteristic T-plan. This extension might often accommodate a laird’s pew or loft, or simply additional congregation members more readily in earshot of the preacher. Chancels, once the preserve of the priest, were sometimes abandoned or converted into a burial vault. Such vaults, or burial aisles, can be seen as succeeding chantry chapels in medieval churches, now becoming simply a private place of sepulchre, not a chapel where masses could be offered.
Preston - Phased plan of Preston medieval Church
Despite these conversions and rebuildings, a great many old churches were abandoned and replaced by new buildings on new sites, often closer to the main centres of population. Church buildings were abandoned far more readily than burial grounds, which would remain in use for many generations after the old church was demolished or left to crumble. Many abandoned churches were simply converted into burial aisles or enclosures by local families; Preston near Duns is a good example. With the rise of ‘resurrectionism’ in the late 18th and early 19th centuries, with the recently-deceased being illegally exhumed to cater for the demand of new medical schools, mort houses (where a corpse could be kept under lock and key until no longer desirable) and watch houses (to house guards) were often built into ruins.
Where medieval fabric does survive in a church that is still in use, it can be difficult to identify as all original architectural features may have been lost. An unusually elongate shape may indicate Pre-Reformation origins, as will a nave-and-chancel plan. 19th and early 20th century restorations sometimes saw a church returned to its medieval form, and might well include the reclamation of a chancel adapted or rebuilt as a burial vault. It is possible that medieval fabric survives unrecognised in quite a number of churches, especially those in which wall surfaces are obscured by render and plaster.
A far greater number of medieval churches are now simply sites, marked by grass-grown foundations - which might represent either the church or perhaps a later burial aisle - and perhaps a few decaying gravestones. These are generally of post-medieval date, but the odd medieval stone sometimes survives, often in poor condition and rarely properly recorded. Coped-and-tegulated slabs (often termed ‘hogbacks’) are a case in point. One at Ancrum is now disintegrating and urgently in need of conservation; another at Nisbet was broken up and buried on the grounds that it got in the way of the motor-mower. It was disinterred and reassembled by a local historical society and given to Hawick Museum, who then lost any account of its provenance, and indeed were considering it a Roman artefact... A similar stone at Edrom has been lost at some time in the 20th century, as have other medieval slabs at Hume and elsewhere. An important collection of slabs at Coldingham, set up against the wall of the ruined south transept in the early 20th century, have suffered considerably from weathering, despite conservation measures.
Map showing Ecclesiastical sites investigated as part of the Flodden Routeways project
The Churches and Sites Investigated
Descriptions of the majority of ecclesiastical sites visited as part of survey work associated with the Roads to Flodden project can be found on the Flodden 1513 website. These did not include all of the medieval ecclesiastical sites in the Borders region but focused on those considered most likely to have a connection with the events preceding and following the Flodden campaign. In addition to the monastic sites of Abbey St Bathans, Coldingham, Dryburgh, Eccles, Jedburgh, Kelso and Melrose, the sites of two memorial crosses, at Crosshall and Wedderburn, and the following church sites were visited and investigated: Ancrum, Ayton, Bassendean, Benrig, Bunkle, Chirnside, Cockburnspath, Cranshaws, Crosshall Cross, Eckford, Edrom, Ellem, Fogo, Greenlaw, Hume, Kirk Yetholm, Ladykirk, Legerwood, Lempitlaw, Lennel, Linton, Longformacus, Maxto, Minto, Morebattle, Mow (or Molle), Nenthorn, Nisbet, Old Cambus (or Aldcambus), Oxnam, Polwarth, Preston, Halterburn (site of St Ethelreda’s Chapel), Simprim, Smailholm and Wedderburn (Cross site). Following are descriptions of sites where more detailed recording and excavation work took place.
Abbey St Bathans Kirk
Abbey St Bathans - excavations south of the kirkNow privately owned and no longer a place of worship – leaving Coldingham as the only former priory church in the Borders to remain in use for worship - this was originally the church of a small Cistercian nunnery, in plan a simple rectangle remodelled in the 18th century (when its western third was removed) and again in the mid-19th when the small north-east tower and south-west porch were added. The thick east wall is medieval, with a two-light window with mid-wall tracery, and inside a 19th century recess containing a fine effigy said to be a prioress, of 15th century date, which had been found re-used in another wall. The east gable has an interesting shaped footstone at its south end, and the jambs of blocked doors are visible below 19th century windows near the east end of the south wall and in the centre of the north wall. Outside the church the lower courses of the north wall of the demolished western part survive for around 5 m, visible at the foot of the external face of the churchyard wall. There is also toothing for a continuing metre-thick wall at the north end of the east wall. The cloister is thought to have been on the north, where there is an artificial raised platform between the church and the Whiteadder Water.
Abbey St Bathans - domestic remains exposed south of the kirkExcavations in May 2016 focussed initially on the pasture field on the east side of the churchyard wall, since this was the only piece of open ground adjacent to the kirk where permission was granted for excavation. Several trenches here produced little of interest, although metal-detecting in this area produced numerous finds, including two spectacle buckles in cast copper alloy dating to c.1500 -1650, a lead disc with reverse impression of a Mary half-crown and four Charles I copper ha’pennies of mid-17th century date. Excavation of a small trench immediately adjacent to the field boundary on a slight ridge which continued westwards into the cemetery, running south of the line of the kirk, was more productive, however, revealing part of a cobbled surface, a large post-hole and tumbled building remains amongst which were large fragments of late medieval pottery, including cooking jars. This suggests that there was a significant focus of domestic activity in a building, or buildings south-east of the kirk. Whether these were directly linked to the Cistercian priory or part of an adjacent village is open to discussion and should be the subject of future investigation, as should the presumed cloister area on the north side.
Abbey St Bathans, Chapel of St Bathan
Abbey St Bathans chapel - shaped stone probably from a window in the south wallThe remains of a chapel, optimistically associated with the 6th century St Bathan were discovered in the 1860s some 300 metres from the medieval monastic church, and preserved inside a fenced enclosure planted with rhododendrons. The exposed stonework had become buried again, but the site was located and partially cleared in May 2016. Rubble footings define a rectangular building with a shallow recess in the internal face of the east wall and a possible splayed window near the west end of the south wall. Further excavations in October 2016 failed to find any traces of surviving floor surfaces below the lowest courses of walling and showed that a possible doorway in the west end is more likely to be a gap in the wall created by the drainage works 
Abbey St Bathans chapel - view looking east towards remains of a late medieval stone coffin and grave slab
which originally located the chapel, as evidenced by a surviving infilled gulley running eastwest through the opening. Documentary sources suggest that a plain tapered grave slab and the fragmentary remains of a stone coffin, now inside the chapel, had been moved there from adjacent positions by the Victorian excavators.
Abbey St Bathans Chapel Plan
Halter Burn - St Ethelreda’s chapel fieldwork location map
St Ethelreda’s Chapel
The location of a chapel dedicated to the 7th century St Ethelreda, who spent a year as a novice at Coldingham Priory between 671-672 AD, has long been debated and several putative sites suggested, most recently in a survey of sources carried out with the benefit of detailed knowledge of the local landscape by Yetholm resident and Flodden volunteer, Tom Broad. Tom Broad suggests, in his exhaustive examination of the sources (A Halter Burn Chapel: the search for St Ethelreda’s, Berwick upon Tweed: BAS) that the chapel may have been built as a place where the faithful could rest and pray on their way to Lindisfarne, with its origins in the shelters made by people receiving Cuthbert’s ministrations in remote and mountainous areas. The sparse references to the chapel, however, are all in later medieval and post-medieval documents, the first known being in a land grant of c. 1230 described in the Kelso Cartular. 
Halter Burn - Linear earthwork and mound on the flood plainThe other, later references do not refer to St Ethelreda’s chapel but to its location: thus, a Helter Chapel is included in Bowes survey of 1583 and Johnson and Goodwin’s survey of the Border Line in 1604 and a Shotten Chapel (apparently then in England, but perhaps the same site) is referred to in communications between Lord Dacre and the Earl of Surrey in 1523. Four sites were examined as part of the current survey, including the site at the foot of Shotton Hill, on the English side of the Border, currently recognised by the Ordnance Survey and another, at the confluence of Humbleton Sike with the Halter Burn which had been recognised until the 1930s. Principal focus was upon the latter which, on the basis of topography and, in particular, the 1604 survey, seemed most plausible.
Halter Burn - Building remains on the earthen mouldA total of nine test pits were excavated over a three day period in May 2016, of which seven were close to the Humbleton Sike site with two others at another of the four sites suggested by Tom Broad, ‘a ruinous barn known as Halfway House, formerly an inn, some 800 metres up the Halter Burn.’ The latter produced pottery, roof tile and clay pipe fragments of relatively modern origin, certainly no earlier than 18th century, but the restricted area of excavation means that the possibility of earlier remains on the site can not be dismissed. The investigations at Humbleton Sike focussed on test pitting the site of a suspected earthwork mound and related linear bank within the flood plain, with three other pits positioned to examine topographically-promising sites in the vicinity. 
Halter Burn - Stone wall footings below linear earthwork feature
While the latter produced no finds of significance, the four pits excavated on the visible earthworks all produced finds of great interest. The linear earthwork, examined in two places, proved to be the remains of a drystone wall, lost to a palaeo-channel at its south end but surviving well at foundation level to the north where it was some 0.8 m wide and associated with sparse medieval pottery. The earthwork mound on the west side of the linear feature, closer to the Halter Burn, was investigated by means of two pits, both extended to 2 m in length. Made deposits were found to survive in each pit to over a metre in depth while abundant rubble suggested the remains of a building, confirmed in the most northerly test-pit where the remains of a wall were uncovered, along with 20 sherds of largely unabraded medieval pottery. No other finds of note were made, but these are sufficient to suggest the domestic occupation of a medieval building on this site. Given the position given for St Ethelreda’s chapel in the 1604 survey it is reasonable to suggest that the remains discovered are those of the chapel or, perhaps more likely given the amount of pottery recovered, those of an ancillary building, perhaps a rectory or rest house. In view of these findings it is concluded that this site certainly merits more investigation by excavation.
Wedderburn Cemetery and Cross
Wedderburn Castle lies 2 km south-east of Duns, and in a field 450 m east of the castle is a small walled enclosure c 15 m square traditionally said to have been the burial place of several members of the Hume family; several old accounts described a free-standing cross. Clearance revealed the head of a late medieval cross, of trefoil form, with an incised foliate cross, differing slightly in design, upon each face. Excavation showed evidence of quite recent disturbance; several boulders and a roughly-shaped cross base were unearthed, although the shaft of the cross remains missing. No evidence of any burials was found; if indeed internments were made here, it is thought that they might have been relocated to the family burial vault close to Duns parish church at a relatively early date.
Wedderburn - socket stone for memorial cross found revealed by excavation
Wedderburn Cross head
