Excavations on Wether Hill, Ingram, Northumberland, 1994–2015

By Peter Topping

The Northumberland Archaeological Group’s (NAG) Wether Hill project spanned the years 1994–2015 and was located on the eponymous hilltop overlooking the mouth of the Breamish Valley in the Northumberland Cheviots. The project had been inspired by the RCHME’s ‘Southeast Cheviots Project’ that had discovered and recorded extensive prehistoric and later landscapes.

The NAG project investigated several sites. Over the 11 seasons of excavation, NAG recorded evidence of residual Mesolithic activity (microliths), a burial cairn containing two Beakers in an oak coffin, which was superseded by a stone-built cist containing three Food Vessels, Iron Age cord rig cultivation and clearance cairns, a series of Middle/Late Iron Age timber-built palisaded enclosures, a cross-ridge dyke, which protected the southern approach to the Wether Hill fort, and sampled the multi-period bivallate hillfort.

The hillfort sequence on Wether Hill began with a succession of palisaded enclosures, which were later replaced by bivallate earth and stone defenses; both phases appear to have been associated with timber-built houses. Eventually the fort was abandoned, and three stone-built roundhouses were constructed in the fort. The 18 radiocarbon dates obtained from various contexts in the hillfort makes this site one of the better dated forts in the Borders.

The chronology of the Wether Hill fort spanned the Middle/Late Iron Age, which corresponded with dates from palisaded enclosures excavated elsewhere on the hilltop spur. Taken together, this evidence provides a snapshot of settlement hierarchies and agricultural practices during the later Iron Age in this part of the Northumberland Cheviots. The excavations also help contextualize some of the RCHME survey evidence, providing data to model chronology, potential prehistoric settlement density and land-use patterns at different time periods in the well-preserved archaeological landscapes of the Cheviots.

• Language: English
• Publisher: Oxbow Books
• Release date: Sep 30, 2023
• ISBN: 9781789259704

Peter Topping

Pete Topping was head of survey for English Heritage and is an expert in landscape interpretation. Following voluntary early retirement he returned to his main subject of research, undertaking a recently awarded PhD at Newcastle University on flint and stone extraction industries. He is on Oxbow's American Landscapes Editorial Board.

Book preview

1

Introduction

The hillfort on Wether Hill is one of the better preserved enclosures in the Northumberland Cheviots, exhibiting not only impressive bivallate defences and ephemeral traces of houses in its interior, but featuring a prominent cross-ridge dyke immediately to the south of the fort, and traces of field systems across the spur to the north. The detailed recording of this hillfort and its environs occurred during the late 1980s as part of the former Royal Commission on the Historical Monuments of England’s (RCHME) South East Cheviots Project (Topping 2008; Topping & Pearson 2008), which had mapped 66km² of these upland landscapes. The RCHME project area encompassed the Breamish Valley in the north to the Alnham Valley in the south, with an eastern boundary along the Cheviot escarpment, and the western defined by the high ground of Scaud Knowe and Hazeltonrigg Hill. The RCHME fieldwork produced significant results that demonstrated the survival of extensive prehistoric and later landscapes in the modern upland pastures. The project was able to map, characterise and identify locational preferences of all forms of settlement, the setting of field systems, the positions of burial monuments, and also determine the broad ebb and flow of colonisation and land-use over time based upon site morphology. Although the RCHME project was not the first to have discovered the survival of extensive archaeological landscapes in the Cheviots, previous work had largely been restricted to site specific surveys as a result of the limitations of contemporary survey equipment (e.g. Burgess 1970; 1980b; 1984; Jobey 1962; 1964; 1965; Topping 1981). Fortunately, the technology now available to the RCHME enabled a new and innovative approach that allowed the detailed, accurate recording of entire archaeological landscapes.

However, the site morphology approach available to the RCHME was a crude interpretive tool that only provided a generalised understanding of chronological developments and clearly needed to be refined with new and robust data, preferably from modern excavations. The scale of the problem was epitomised by the fact that when the RCHME embarked upon its fieldwork, only six radiocarbon dates existed from sites excavated within the RCHME project area – four from the later Bronze Age settlement at Standrop Rigg (Jobey 1983), one from the hillfort on Brough Law (Jobey 1971), and a final date from the palisaded settlement at Ingram Hill (Jobey 1971). Consequently, the RCHME team had to rely upon the field observations of relationships and associations between earthworks, combined with comparative site morphology, to attempt to differentiate broad phases of land use and establish a narrative of landscape change from the prehistoric period into the Modern Age.

A few years after the completion of the RCHME project, the Northumberland Archaeological Group (NAG) decided to try to address the fundamental issue of site and land-use chronology, and target excavations that would address such research questions. To this end, NAG designed and initiated the Wether Hill project in 1993, with fieldwork beginning in 1994. This project was intended to provide not only information about the nature of archaeological sites on the Wether Hill spur and their associated material culture, but importantly it was devised to produce a scientifically dated chronological framework for the hillfort and other sites on the spur. The overarching research aim of the NAG project was to provide modern, contextualised data that could be applied to the interpretation of the wider Cheviot archaeological landscapes. This could then be used to inform future conservation and research strategies.

The NAG project eventually became a partner in the Northumberland National Park Authority’s (NNPA) ‘Exploring our Hillfort Heritage’ initiative, alongside Durham University Archaeological Services (DUAS).

Site description

The hillfort on Wether Hill lies at 308m above Ordnance Datum (aOD) upon a locally prominent knoll overlooking the mouth of the Breamish Valley to the north-east (Figure 1.1). The spur is roughly aligned north-east to south-west, and, notably, higher ground rises beyond the spur to the south-west of the hillfort where, some 800m distant, the summit of Cochrane Pike rises to 335m aOD. The bulk of the Wether Hill spur shelves gently downwards in a north-easterly direction towards the floor of the Breamish Valley at Ingram village. Wether Hill forms part of a chain of hills comprising Cochrane Pike, Wether Hill, Ewe Hill, Brough Law, and a second Ewe Hill, which dominate the wide entrance to the valley. Overall, this chain of hills provided the setting for five hillforts and several contemporary settlements with field systems which are ranged along this eastern escarpment of the Cheviots.

The Wether Hill spur is physically separated from the rising ground of Cochrane Pike by a cross-ridge dyke, which lies some 75m to the south-west of the fort, and this acted as a boundary across a track that formed one of the major routes between the Alnham and Breamish valleys, and which passes just to the east of the hillfort through a gap in the dyke. The wide and gently sloping surface of the spur stretches north from the fort and retains evidence of episodes of prehistoric agricultural activity in the form of small clearance cairns and several truncated settings of cord rig ridges (narrow ridged cultivation generally 1m–1.4m wide; Topping 1989a) located around the southern and eastern parts of the spur. A stone-built round cairn lies on the eastern side of the spur along with what initially appeared to be a timber-built ring-groove house, which had been recorded on aerial photographs. A final fragmentary setting of cord rig cultivation lay on the north-western shoulder of the spur. As the spur gradually falls away on the east towards the Fawdon Dean Burn, the upper reaches of the valley sides are swathed in broad ridge-and-furrow of typical medieval to post-medieval date. In contrast, the western side of the Wether Hill spur overlooking Middledeanburn is uncultivated rough pasture.

The surviving earthworks of the Wether Hill hillfort comprise bivallate stone and earthen defences, which enclose a complex internal sequence of palisaded enclosures, timber-built ring-groove house foundations, and three stone-built roundhouses. Overall, the earthworks suggested a succession of broad constructional phases beginning with a summit cairn, followed by a sequence of palisaded enclosures and houses, which were then succeeded by the bivallate hillfort. The final phase was represented by a group of three stone-built roundhouses that partly overlay the abandoned hillfort defences in the south-east.

Previous archaeological research

Early antiquarian sources appear to largely ignore Wether Hill and its archaeological sites. For example, Captain Andrew Armstrong’s map of Northumberland (published 1769) provided schematic depictions of the local topography, but chose to only depict the most obvious archaeological sites such as Brough Law hillfort. Armstrong did, however, portray a bridleway/track that led south from Ingram village and ascended the spur, passing the hillfort on Wether Hill before descending to Prendwick and the Alnham Valley. During 1827–8 Christopher Greenwood (1831) mapped the county and some of its archaeological sites, particularly the ‘Roman or Pict’s Wall’, various ‘camps’, and the ‘Druids Temple’ at Threestoneburn. Sadly, Greenwood also largely ignored the archaeological remains in the Breamish Valley – even excluding previously recorded sites such as Brough Law hillfort. The only site shown in the vicinity of Wether Hill is a ‘camp’ located beside ‘Clench’ (Clinch), probably Castle Knowe. Also of interest is a cairn depicted on the Brough Law ridge on Ewe Hill to the east of the confluence of the River Breamish with the Chesters Burn. This lies in roughly the position of the now destroyed skyline cairn that had been recorded during the late 19th century, and was noteworthy for its massive size. It was subsequently dismantled to provide wall stone for the plantation enclosure situated adjacent to Middledeanburn promontory fort (Tate 1863).

The hillfort on Wether Hill was first surveyed by Henry MacLauchlan in June 1861, during a project sponsored by the fourth Duke of Northumberland (Figure 1.2). This project culminated in ‘Notes not included in the memoirs already published on Roman roads in Northumberland’, and was published and deposited in the Alnwick Castle Library during February 1867 (Charlton & Day 1984, 26). Alongside Wether Hill, MacLauchlan’s project had recorded 144 ‘native’ sites, of which 127 were surveyed to his usual high standards with the sites being depicted in their topographic settings. Unfortunately, only six of these plans appeared in print. Those recorded in the Breamish Valley catchment included:

Castle Hill (Castle Knowe), a multivallate hillfort near Fawdon (NU 0313 1468), surveyed in July 1861;

a ‘camp’ with the appearance of a fragmentary multivallate fort but now lost amongst the plantations lying between Snail Knowe and Brandon East Hill (c.NU 027 157), survey undated;

Great Ryle (Chubden), a multivallate hillfort lying to the north-east of the summit of Chubden (NU 0245 1354), survey undated;

Reaveley (Ewe Hill 2), a complex hillfort situated on the summit of the northern side of the Breamish Gorge, overlooking the valley (NU 0090 1669), surveyed during July 1861;

Figure 1.1: The location of Wether Hill in the Breamish Valley, Northumberland (illustration: T. Pearson).

Figure 1.2: The hillfort on Wether Hill looking west, with the D-shaped promontory fort of Middledeanburn in the middle distance (photo: R. Carlton).

Ewe Hill (Ewe Hill 1), also on the northern summit of the Breamish Gorge (NU 0040 1680), and located some 400m from the Reaveley hillfort (Ewe Hill 2), surveyed in July 1861;

Brough Law hillfort (NT 9984 1635), surveyed in June 1861;

Haystack Hill scooped settlement complex (NU 0055 1505), surveyed during July 1861;

Middle Dene (Middledeanburn) promontory fort, located overlooking the gorge of the eponymous burn (NU 0043 1464), surveyed in June 1861;

Chesters (Prendwick Chesters) hillfort overlooking the River Breamish (NT 9848 1488), surveyed during June 1861;

Hartside settlement complex located on the eastern summit of Hartside Hill (NT 9880 1580), surveyed in June 1861;

Meggrims (Meggrim’s Knowe) settlement complex overlooking the confluence of the Linhope Burn and the River Breamish (NT 9641 1590), surveyed during June 1861;

and Greve’s or Greaves-ash (Grieve’s Ash) hillfort, settlements and field system (NT 9652 38), surveyed in June 1861.

Of the Breamish Valley ‘hill sketches’, only Grieve’s Ash appeared in print.

MacLauchlan’s hill sketch of Wether Hill depicts the circular earthworks of the bivallate hillfort, complete with the north-eastern and north-western entrances, and two of the three stone-built roundhouses in the south-eastern part of the fort (Figure 1.3). The remains of the ring-groove houses in the fort’s interior are not shown, but this survey does significantly pre-date the first published recognition of such evidence. The existence of timber-built roundhouses was first noted during the Royal Commission on the Ancient and Historical Monuments of Scotland’s (RCAHMS) fieldwork in Roxburgh in the mid-20th century. This produced the first recognition of these ephemeral earthworks, summarised in various publications by the RCAHMS and their colleagues working in Roxburghshire and the Borders (e.g. Piggott 1948; 1949; Stevenson 1949; RCAHMS 1956). In addition, MacLauchlan’s fieldwork, undertaken during the summer months, would have been compromised by the growth of rough pasture grasses, which would have masked the slight traces of the foundation trenches of the palisaded enclosures and ring-groove houses. The small rectilinear lambing pen that now overlies the outer scarp of the outer rampart on the southern perimeter of the fort is not shown and must post-date the 1861 survey. To the south-west, MacLauchlan depicted a dashed line in the approximate location of the cross-ridge dyke, but it is unclear whether this is actually what he was surveying.

MacLauchlan originally named the site Corbie Cleugh after the small burn lying some 200m to the west of the fort, presumably as this was the nearest recorded place-name at the time of his survey. Wether Hill as a place-name appears to have developed sometime afterwards, and by 1935 it was used by Hope Dodds in Volume XIV of A History of Northumberland. Curiously, George Jobey (1965, 49) reverted to the MacLauchlan name of Corbie Cleugh in his 1965 survey of Northumberland hillforts, despite the fact that Wether Hill as a place-name was by then featured on the contemporary Ordnance Survey mapping. The name ‘Wether Hill’ continues a local tradition of sheep-related place-names and is joined by many ‘Ewe Hills’ in the area (a wether is a castrated lamb; cf. Wilson 2005, 196). MacLauchlan did eventually depict the hillfort on Wether Hill in the atlas volume that accompanied the publication of the eastern section of Watling Street from Hadrian’s Wall to Berwick (1864, Sheet III). Here he also illustrated the topography of the mouth of the Breamish Valley, which usefully provided the topographical setting for the proliferation of ‘camps’ and assorted archaeological sites at the entrance to the valley.

Figure 1.3: Wether Hill seen from the south-west, with the cross-ridge dyke in the foreground (photo: R. Carlton).

The Ordnance Survey (OS) systematically mapped the Cheviot Hills during the late 1860s as part of their fieldwork for the 1st Edition 6-inch map. Sheet XXX 10 (published 1866) portrayed Wether Hill as a bivallate enclosure with entrances in the east and north-west, but no other details. Although a prominent earthwork, the cross-ridge dyke was not shown by the OS.

In the early 1920s, Cecil Hedley (1924), having been inspired by the survey of early earthworks in Scotland by Christison (1898), collated a gazetteer of Northumberland sites based upon published sources found in county journals, the key antiquarian texts, and the 3rd Edition Ordnance Survey 1-inch maps. Hedley hoped that his list would form the foundations for future fieldwork in the county, but also sounded a note of caution:

It must be clearly appreciated that the records made by many of our early investigators are not accurate and are without details. In particular the deductions made from them are in most cases not to be relied upon in the light of more recent investigations. There is not necessarily a ‘camp’ where MacLauchlan says there is one. Many enclosures of earth and stone have been given a respectable antiquity to which they are clearly not entitled. In this connection imagination and romance must be strictly subordinate to accuracy. (Hedley 1924, 87)

Although Hedley’s approach to fieldwork and investigation is laudable, the fact that he claims MacLauchlan’s plans and identifications may be unreliable probably says more about his abilities in map reading and analysis than those of MacLauchlan’s fieldwork skills. Hedley (1924, 100) did list Wether Hill in his section covering Ingram Parish where it again appears as ‘Corbie Cleugh camp’, and his source was MacLauchlan’s (1867) Notes not included ... Clearly Hedley did not have an issue with the accuracy of MacLauchlan’s survey of Wether Hill, nor indeed the depiction shown by the OS.

A more systematic project designed to survey Northumberland’s archaeology and history began in 1935 as part of the County History programme, and was edited by Hope Dodds (1935). In this, Wether Hill was classified as one of the ‘Fortified Enclosures and Ancient Settlements’, and listed as a ‘Type B2: Forts on High Ground less dependent on Natural Slopes for Protection’, shown as site ‘45 Ingram – Wether Hill (Corbie Cleugh)’. The primary source referenced is MacLauchlan’s (1867) Notes not included …, but sadly no plan accompanied the listing of the hillfort in the County History (Hope Dodds 1935, 63).

In 1943 Hogg produced another review of ‘native settlements’ in Northumberland, drawing upon previous listings and excavations up to 1942 (Hogg 1943). This marked a shift towards more methodical approaches to produce classifications of the broad spectrum of prehistoric settlements in the county. Unusually, Hogg’s gazetteer of sites, which might have been expected to accompany the 1943 Antiquity paper, was published locally in the Proceedings of the Society of Antiquaries of Newcastle (PSAN) (Hogg 1947). In the first paper Hogg (1943, 138) classified hillforts into ‘contour and hill forts’, which he suggested were ‘not common in the area and the smaller sites may perhaps be more nearly related to the multiple ring forts’ (ibid.). He considered the largest hillfort in the county, ‘Yevering (sic) Bell’, the only site that might be considered a ‘town’ because of its large enclosed area and multiple roundhouses. The majority of hillforts were assigned to ‘ring forts and small enclosures’, with an internal area no more than 2 acres (0.8ha) such as the ‘multiple ring forts’ like Greaves (sic) Ash with its bivallate defences, but also including univallate enclosures such as that on the eastern summit of Hartside Hill. A group of ‘related sites’ also included the promontory forts at Middledeanburn with its D-shaped ground plan (Figure 1.4). Hogg also collated data for ‘rectilinear sites’, ‘villages’, ‘isolated huts’, ‘house types’, ‘associated large enclosures’, and burials, and produced a distribution map of site types across the county.

Hogg also produced a discussion of the putative cultural origins of certain site types, linking structural similarities between the small Northumbrian enclosed settlements, and those found in southern England of Bronze Age origin such as Plumpton Plain in Sussex. He also drew comparisons between the coarse pottery traditions in the two areas, and suggested that they may all share a common Late Bronze Age chronology. Hogg also observed that the rectilinear settlements of Northumberland, such as those found in Tynedale, ‘show a striking resemblance to the viereckschanzen … of the upper Rhine’, and proposed tentative links between the Rhaetian garrison stationed at Risingham Roman fort and their appearance in the Northumberland settlement record. The paper concluded with the observation that ‘in this area the culture of the Late Bronze Age may have survived up to Roman times and later. There seems no need to postulate any Iron Age invasion to account for the bulk of the pottery and other relics’ (Hogg 1943, 146).

In Hogg’s 1947 gazetteer in PSAN, he provided the data on which he had based his earlier paper in Antiquity (Hogg 1943). This gazetteer provided a revised classification scheme that he had developed using the Ordnance Survey’s (OS) 6-inch map depictions, which provided greater detail of the sites for a more informed analysis. In this, Wether Hill was again misnamed and listed along with an added alternative bracketed place-name as ‘Corbie Cleugh (East Cochrane Pike)’, despite the fact that the site name ‘Wether Hill’ had appeared in print since 1935 in the County History (Hope Dodds 1935). Hogg classified Wether Hill fort as a ‘Ring M’ type (‘Ring: Enclosures, more than ¼ acre [0.10ha] in internal area, enclosed by walls or banks roughly circular in plan’).

Figure 1.4: The D-shaped promontory fort of Middledeanburn seen from the south-east, with The Cheviot and Hedgehope Hill on the horizon. The Haystack Hill settlements lie centre right (photo: R. Carlton).

The OS Archaeology Division were once again active in the Breamish Valley area between 1956 and 1975 (cf. Bowden & Mackay 1999, 4, fig. 1), recording the archaeological sites that fitted within the tolerances of their mapping conventions, i.e. that banks or ditches survived to at least 30cm above or below the ground surface. These surveys were undertaken at 1:2500, which was the OS basic scale for mapping in this area, and were accompanied by brief monument descriptions that documented the principal dimensions of each site. This information was entered on record cards that ultimately formed a national archaeological record curated at the OS headquarters in Southampton. This baseline data was subsequently used to establish the various national monuments records in England, Wales, and Scotland, and many county-based Historic Environment Records.

During the late 1950s and 1960s, and overlapping with the OS Archaeology Division fieldwork, George Jobey (1965, 49) and his Adult Education students surveyed the hillfort on Wether Hill (calling the site ‘Corbie Cleugh’, following MacLauchlan). This survey was part of a project designed to produce baseline data of prehistoric settlements in Northumberland at a larger scale than previously attempted. Jobey’s surveys coincided with the recognition of the survival of timber-built structures, particularly palisaded enclosures and roundhouses as mentioned above. As a result, Jobey’s plan of Wether Hill depicted additional structural details for the first time, such as the internal surface quarrying for rampart material, and ten circular ring-groove