Harbottle Castle. Photo by Harry Rowland, 1971.

(NT 9322 0482) Harbottle Castle (Remains of). (1)

The castle started as a Motte and Bailey built by one of the Umfravilles in the 12th century. A stone castle was ordered by the King after 1157, but in 1173-4 the place was taken by the Scots (probably before the stone-work started). The stone castle was (probably) started soon after and comprised a shell-keep on the Motte and a castleyard occupying the western half of the bailey and defended by a curtain wall and tower, the eastern half of the bailey being cut off by a massive wall and used only as a barmkin. In 1318 the Castle was captured by Robert Bruce. It was ordered to be dismantled in 1319 but was restored in 1336. It was in ruins in 1351 but defensible again at the end of the century.
Much work on it was done between 1541 and 1551 when the western and northern sides of the shell-keep were rebuilt in the present form. Further repairs were made in 1563, and in 1585 the inner ward was 'reasonable' but the outer 'almost down'. There were no further repairs and the site was used as a quarry, suffering particularly when the new house 'Harbottle Castle' was built in the 17th century. There was possibly a pre-Conquest fortress on the site. (2)

The siting at one of the easy crossings of the Cheviots, was of critical importance for the control of the Middle Marches. (3)

OS 25 inch survey revised and annotated. Nothing seen to bear out the supposition that there was a 'pre-Conquest fortress'. (4)

OS 1:2500 survey revised. Building foundations on inner west side of curtain wall no longer visible. (5)

The stone foundations of a probable rectangular building, maximum measurements 22m by 6m, divided into 3 units are traceable inside the west curtain, and a small rectangular stone-founded enclosure measuring 7m by 5m is evident in the south east of the motte. Surveyed at 1:2500. (6)

Grade I Listed Building. Harbottle Castle remains. An important fortress with magnificent earthworks of the mound and bailey plan and also considerable structural remains. Plan and description in Northumberland County History XV. Scheduled as an Ancient Monument. (7)

Masonry remains are fragmentary, but the earthworks remain well preserved and their scale is still impressive. The major period of destruction and stone robbing seems to have been in the 17th and early 18th centuries. Nineteenth century illustrations show ruins used as rough pasture, much as they are now. Clear signs of ridge and furrow in the eastern half of the main enclosure but their date is uncertain. The destruction is currently accelerating due to a combination of natural erosion and land usage (cattle). Harbottle Castle ranks within the ten most important castles in Northumberland, and is a site of national importance. (8)

Detailed examination of the outer face of the curtain wall on the west side of the motte indicates the whole section is of one phase - the coursing is even and uninterrupted across the elevation. The lower section of the wall is probably broader for reasons of stability. At the east end of the wall are the remains of a parapet and the top of the wall walk, and the remains of the west edge of the splay, possibly for a door? Photographs in parish file A19. (9)

An area of collapse within the earthworks of the castle was visited in July 1991. This is located to the south-east of the motte and due to the presence of a sunken chamber of unknown purpose [dealt with more fully by HER 27432]. (10)

Harbottle Castle, Grade I. Ruined castle, started after 1157, reconstructed in C13 in stone. Extensively rebuilt in 1541. Medieval motte and bailey earthwork finest in Northumberland. Ruins include part of curtain wall, c.40yds long x 15ft high and remains of angle tower at north end. Extensive fragments of shell keep on motte. (11)

NT 9325 0481. Motte and bailey castle and shell keep castle at Harbottle. Scheduled RSM No 20959. The conical motte stands to a height of approximately 10m and measures 90m across at its base and 22m across its circular top. It is surrounded by a ditch which is on average 18m wide and has a maximum depth of 1.8m. The bailey lies to the N, E and W and is delineated by a massive rampart 2.4m wide standing 10m above the bottom of a ditch 12m wide. The bailey measures a maximum of 140m E-W by 100m N-S. Parts of the curtain wall stand to over 6m on the W. (12)

As part of a project of conservation works, in 1997 survey, excavation and consolidation work took place. A topographic survey of the motte and the north west half of the bailey was carried out by students of the Archaeology Department, University of Newcastle. A ground penetrating radar survey was conducted across the castle defences. Excavations on the motte were also carried out by Newcastle University students; this concentrated on two infilled embrasures and Tudor gun emplacements on the northern curtain wall of the motte. (13)

In 1998 excavation of the gatehouse situated between the inner and outer baileys of the castle, was carried out by Newcastle University.This had been trenched in 1935 by H L Honeyman and the objective was to re-excavate these trenches in order to assess the condition of surviving masonry. In June 1998 excavation began on the south side of the gate and remains of the outer gateway at the east end. (14)

Excavation continued in 1999, uncovering the middle gate and curtain wall between the inner and outer baileys, previously investigated by Honeyman. The gatehouse measured 10.3m by 9.35m with a gate passage 3.05m wide. The gatehouse is structurally later than the adjacent cross wall, with the drawbridge and postern a still later addition. Structures built after the gate was blocked, but before the abandonment of the castle, were found abutting the outer face of the gate and curtain. Distinctive mason's marks were seen on the external face of the gate tower and in the interior of the guard chamber.
Examination of the stonework by P Ryder identified moulded stones from Holystone Priory amongst loose stones from the castle.
The earliest phase of the castle is likely to have been of timber construction. The first stone castle is thought likely to date from before 1200, the date attributed to the construction of the cross wall and early middle gate. The gate was rebuilt on a grander scale in the 13th century. The extensive remodelling of the middle gate may be dated to the Anglo-Scottish wars when the castle was taken by Robert the Bruce in 1318 and demolished. It is possible this was the time when the outer bailey was abandoned and the middle gate was remodelled and became the main gate of the castle. The work at the drawbridge and barbican probably date to the end of the 14th century.
In 1999 the counterbalance drawbridge pit and the barbican at the middle gate were discovered; probably dating to the later 14th century. (15)

The ground penetrating radar survey in 1997 suggests that the outer castle ditch and the motte ditch were about 2m deeper than present day levels. The outer ditch shows signs of having been revetted. There were also signs of masonry in the outer bailey mound and inside the outer bailey. A possible stone-lined tunnel was noted within the outer bailey, but this was not confirmed. (16)

The scheduling was affirmed on 10 October 2001. The new description omits the phrase about the castle yard. (17)

A documentary study was commissioned by the National Park in 1998. The origins of the castle lie in the 12th century and the Umfraville liberty of Redesdale which was established in about 1120. The earliest reference to the castle records its capture by the Scots in 1174. Major construction occurred during Henry II's reign and may represent the initial construction of the castle, although the possibility that it replaced earlier timber defences cannot be excluded. The castle was rebuilt in the early 13th century and remains a baronial stronghold until the mid 15th century, when it briefly came into Crown control. It served as the seat of the Warden of the Middle March. It passed fully into Crown ownership in about 1546 and was repaired and converted to an artillery fortress. Lack of investment led to its decline and it finally abandoned around the end of the 16th century. (18)

The castle started as a motte and bailey castle and the earthworks, on which stone buildings were later constructed, are still extremely dramatic. On the motte, a shell keep was built in the 13th century, and below it a wall enclosed a large bailey and separated it from a still larger, unfortified bailey beyond. Extensive fragments of this shell keep and the bailey wall remain, including one dramatic stretch about 15ft high and about 40 yards long. The keep walls include one deeply-splayed 16th century gun loop. Apart from the interest of the ruins, the setting is superb, with the village nestling below and magnificent wilderness all around. (19)

Archaeological monitoring was carried out by Archaeological Services University of Durham throughout 1997 in conjunction with the consolidation of the ruined walls of the shell keep, the block of masonry tumbled from the keep, the cross wall and remaining enclosing wall of the bailey. This work augmented the previous recording of the walls carried out by Northern Archaeological Associates and identified some features not recorded earlier. These features included the remains of a staircase, depressions which may relate to temporary structures erected against the curtain wall, reused architectural fragments and a pair of metal hooks. A pair of gunloops have been conserved. One may indicate the presence of an arrowslit in this location as part of the original construction of the castle.
The walls on the western side of the castle appear to be contemporary as these are similar in construction and in choice of building materials. There is no visual evidence that this wall conceals or replaces an earlier wall. The walls of the western side of the stone keep also appear contemporary between themselves as they both contain reused architectural fragments and contain long rectangular yellow sandstone blocks which do not occur in any other wall. The composition of the mortar in these walls of the keep, as well as the tumbled block from the motte, is similar. The gunloops are likely 16th century when the site was undergoing repair by one of ther last garrisons. The cross wall may also be of this date as it is constructed from stones similar to Brinkburn Priory and perhaps referred to in an August 1537 document.
A series of reused architectural fragments and mason's marks were identified in this work. (20)

A geophysical survey of Harbottle Castle was undertaken as part of a Masters dissertation by Aidan Pratt. The magnetometry survey was carried out using a Bartington Grad601 Single Axis Magnetic Field Gradiometer System, with the resulting data plotted and interpreted using ArcMap 10.3.1. The survey covered a 1.2 hectare area including the inner and outer baileys of the castle and an area of flat ground to the east that has been suspected to be related to the development of the castle and its surrounding settlement. The trace plot of the data within the castle is difficult to read for a number of potential reasons detailed in the report. The survey identified various features, some of which were visible as surface features, including a well and the curtain wall of the inner bailey, ridge and furrow and a modern farm track within the outer bailey and the retaining stone wall for the eastern area. An extremely strong set of positive and negative anomalies, surrounded to the north by a rectilinear feature in the centre of the inner bailey may represent a large amount of ferrous material. The strong positive anomalies to the south-western corner of the inner bailey may represent a range of buildings. A sub-square feature at the southern extent of the outer bailey may also represent a building. The retaining wall in the eastern area was interpreted as a barmkin (a high wall around a yard or open space). Two strong negative anomalies on the northern limit of the barmkin could indicate an area of intense burning or a series of structures. Two rectilinear features in the south-eastern corner of the survey area were attributed to a structure recorded on the 1st Edition Ordnance Survey map of c.1866. (200