The Journal Of Antiquities

Ancient Sites In Great Britain & Southern Ireland


Ardoch Roman Fort(s), Near Braco, Perthshire, Scotland.

The defensive ramparts of Ardoch Roman fort(s), near Braco, Scotland. Photo courtesy of the late Patrick John Leonard.

NGR: NN 8389 0993. On the moorland a little to the north of Braco, Perth and Kinross, Scotland, are the very substantial earthworks of the Roman forts of Ardoch (ALAVNA), which are located beside the A822 Crieff road. The forts stand in the grounds of Ardoch House. These multiple Roman fort(s) here at Ardoch were all built on top of each other, and their defences extended and strengthened each time. The earthworks are in a well-preserved state, with the ramparts consisting of grassed-over banks, some up to 15 feet high, and ditches spanning a considerable area of ground in the usual sort of ground plan of a Roman fort – with rounded corners like a playing card. The whole site is roughly 300m x 217m (984 ft x 712 ft). However, the south and western sides of the site are now, sadly, almost devoid of any earthen banks and ditches, with trees taking their place. Further to the north and east of the site, there are the faint earthworks of another five or six temporary Roman camps, which together are even larger than the Ardoch fort, but they were obviously not as important. These six military sites and two signal towers were built after 80 AD, and continued to be garrisoned by thousands of Roman legionaries and auxiliaries into the late 2nd and early 3rd centuries AD, the soldiers probably coming from different parts of the Roman Empire. Ardoch Roman forts, temporary camps and signal towers stand in a north-easterly alignment and make up the Ardoch annexe at the southern section of The Gask Ridge. A Roman road runs in a north and then north-easterly direction to Perth. The Roman forts are in private ownership – part of the Ardoch House Estate.

Ardoch Roman Fort(s) near Braco, Scotland (as seen from above).

Peter Clayton, writing in 1976, tells us more about Ardoch Roman Fort and the surrounding area. He says that “Crossing over the Antonine Wall at Rough Castle and 18 miles directly to the north is the Roman fort of Ardoch (NN 839099), near to Braco. It is one of a series of forts thrown forward into hostile territory on a line going up into Angus. As such they must have resembled the French Foreign Legion Forts in Algeria in the 1930s. There is not a stone building or wall to be seen at Ardoch, but for all that it is one of the most spectacular Roman sites in Britain. The ramparts were built of turf on a stone base, like the Antonine Wall. They followed the usual playing card shape and had an area of 75 x 79 yards with the customary military buildings within. Outside the fort area on the north and east side are a series of five parallel ditches; presumably there was a similar system on the south and west sides, but here the evidence has been largely obliterated. A turf rampart further out encloses the whole defence system. The ground to the east and south of the fort was marshy, which meant that there was less chance of attack from those directions; consequently, the ditches on these two sides are closer together than those on the north side. The fort itself served as a permanent legionary base but beyond it, to the north, are the remains of five larger temporary camps of varying dates. Two of them – 13 acres and 29 acres respectively, the second one being an enlargement of the smaller – date from Agricola’s advance into Scotland. Even larger camps in the group can be ascribed to the campaigns of Septimius Severus in AD 208-9. All the camps overlie each other to a greater or lesser degree, and this does make their interpretation at ground level rather difficult.”  

Peter Clayton (1976) adds to the above, and says: “The Roman fort at Ardoch,  a forward base 18 miles – a day’s march – from the Antonine Wall.”

David E. Johnston (2002) says of the Ardoch Roman fort: “The impressive visible remains belong to the later second century − the Antonine successor to an Agricolan fort. The multiple earthworks are finest on the eastern side; at the north-east corner, the slight outermost bank may be medieval (like the remains of a chapel in the centre of the fort) and the two outermost ditches (with bank) are possibly Agricolan. The irregular spacing of the northern defences betrays a reduction of the fort from 3.6 to 2.3 hectares, both phases being Antonine; the three outermost ditches, with remains of corners at north-west and north-east, belong to the larger fort. Two later ditches cross the space between these and the main inner rampart, which originally had a stone facing. Internal buildings of masonry have been excavated but are not visible. The fort was built for a mixed garrison of auxiliaries and legionaries. 

To the north and north-east are very faint traces of a large annexe with a watchtower and no less than six overlapping marching camps – a complex covering the first to third centuries. The chronology is still disputed, and the traces are difficult to follow on the ground.” 

Ardoch/Braco Roman earthworks.

 

The AA (1963) tell us that: “To the east of the old bridge over the Kniak Water is the great Roman station of Ardoch, in the grounds of Ardoch House. The earthworks are of great strength, and are among the most striking of their kind in Great Britain. Coins excavated here date from the period of Hadrian. A little further to the north-west is another Roman Camp, known as the Great Camp, adjacent to which is another example, the two camps together being considered to have had space for nearly 40,000 troops at a time.”                                                                                                                                                                             

N.G.R. Coordinates for possible temporary Roman camps to the north and northeast of Ardoch:- A possible Roman camp at NN 83981021 and a possible military complex of Roman camps at NN 84101060. There is also a possible Roman fortlet or camp further to the northeast at NN 87071294, and several miles further to the northeast of Ardoch is the Parkneuk Roman signal station/watchtower NN 91671855, while a little further to the east, still, is the Ardunie Roman Signal station/watchtower at NN 946187.

Sources / References and Related Websites:-

Clayton, Peter, Archaeological Sites Of Britain, Weidenfeld And Nicolson, London, 1976.

Johnston, David E., Discovering Roman Britain, Shire Publications Ltd., Princess Risborough, Buckinghamshire, 2002.

The AA, The Illustrated Road Book of Scotland, AA Publishing, London, 1963.

https://en.wikipedia.org/wiki/Ardoch_Roman_Fort

https://www.undiscoveredscotland.co.uk/braco/ardochromanfort/index.html

https://www.roman-britain.co.uk/places/ardoch/

More info here: https://scarf.scot/regional/pkarf/perth-and-kinross-archaeological-research-framework-case-studies/the-gask-ridge-ardoch-fort/

https://portal.historicenvironment.scot/apex/f?p=1505:300:::::VIEWTYPE,VIEWREF:designation,SM1601

© Copyright Ray Spencer 2026.


2 Comments

Gilbeber Hill Earthwork, Near Greenberfield, Lancashire

Gilbeber Hill, near Greenberfield, viewed from the south.

Gilbeber Hill Earthwork near Greenberfield, Lancashire.

OS Grid Reference: SD 87627 48284. About halfway between Greenberfield and Bracewell, in Lancashire, is Gilbeber Hill (sometimes spelt as Gilleber), and an almost square-shaped earthwork consisting of a raised outer bank and, near the centre,   a small raised platform with a little stone. The rough, grassy earth-work has been considered to be a Roman camp, a Medieval enclosure, or perhaps a Civil War encampment from the 17th century! Bracewell was originally part of the West Riding of Yorkshire. The Roman road linking Ilkley in the east with Ribchester in the west runs close to Gilbeber Hill, the course of which is still to be found at Brogden Lane to the southwest. From the footpath (at the east side of Sewage works) on Greenberfield Lane: head across the field to a second wooden gate, then continue on this footpath to the footbridge over Stocks Beck. Gilbeber Hill is the low, grassy hill on your left; the footpath runs around the base of the hill towards a third gate and then onto Gisburn Road. A similar earthwork is to be found on Hawber Hill, ½ a mile to the northwest.

Gilbeber Hill Earthwork near Greenberfield. The Western side.

Gilbeber Hill Earthwork. The small raised feature at the SW side.

The rectangular-shaped earthwork atop Gilbeber Hill (drumlin) is a raised, grassy area with a visible bank around its edges, which measures between 1-2 feet in height, and there are slight traces of an outer ditch. Nearer the centre at the W side is a small, raised square feature (platform) with a little stone standing upon it, but whether this ‘white stone’ was part of the earthwork structure is not known; it may have come from somewhere close by. At the E side is what might be an entrance? The earthwork is roughly 24m x 26m and 632-756 square yards with a parameter length of 101m covering 0.156 acres of the flat hilltop site. But was this a temporary camp from the late Roman period, or perhaps a Romano-British farmstead (similar to Bomber Camp near Coal Pit Lane)? or was it a medieval settlement/enclosure? or could it have been a Civil War encampment from the 1640s? – this we don’t know with any certainty. There don’t appear, however, to have been any Civil War camps in this part of west Craven, and so that theory must be dismissed.

John & Phillip Dixon (1990) say of the earthworks: “Near the settlement of Stock are two hill-top earthworks; one is sited above Stock upon Hawber Hill, the other is sited upon Gilbeber Hill, roughly 30 yards square and 27 yards square respectively. The Gilbeber earthwork displays signs of an outer ditch, and…….commands a view of the surrounding area that is bisected by the Roman road. There are over a dozen such hill enclosure sites in the western central Pennine area and none, other than the Bomber Camp/Primrose Hill sites, have been subject to archaeological study. At this stage it may be reasonable to view the Hawber and Gilbeber site as having their possible origins in the late Roman period. And, in the case of Hawber, may be related to some of the earthwork features around Stock Green that we consider may be a Romano-British settlement similar to that of Bomber Camp.”

John Dixon goes on to add: “Whitaker in his “History of Craven”, states that “tradition holds that they (the Hawber and Gilbeber earthworks) were constructed by the Royalist forces of Prince Rupert during their march through Craven in 1644. They consist of small square encampments and are strengthened by long rectilinear fosses, which descend along the slope of the hills on each side to the plane beneath.”

The PastScape / Historic England Monument No. is: 45403.

Sources / References and related websites:-

Dixon, John & Phillip, Journeys Through Brigantia Volume One: Walks in Craven Airedale and Wharfedale, Aussteiger Publications, Barnoldswick, 1990.

https://www.pastscape.org.uk/hob.aspx?hob_id=45403

https://www.pastscape.org.uk/hob.aspx?a=0&hob_id=45398

© Ray Spencer, The Journal of Antiquities, 2019

 


1 Comment

Cobers Laithe Earthwork, Swinden, Near Nappa, North Yorkshire

Cobers Laithe Earthwork, near Nappa (looking north-west).

Cobers Laithe Earthwork, south-side (looking north-west).

Cobers Laithe Earthwork, near Nappa (looking north-east).

Cobers Laithe Earthwork, south-side (looking north-east).

    OS grid reference: SD 8669 5340. Sadly next-to-nothing is known about the oval-shaped earth-work or enclosure near to Cobers Laithe farm on  Swinden Moor, which is also known as Swinden Earthwork and Swinden Roman Camp, just 1 mile east of Nappa, North Yorkshire. Was it a Roman camp, as a few have suggested, or was it perhaps a Brigantian settlement – at the time of the Roman invasion? Or was it a more typical Iron Age settlement or enclosure? We don’t know with any certainty. And also it is a bit of an odd sort of earthwork-enclosure as it is intersected through the middle by a stream, and there are a numble of circular pits (bell pits), especially at the N side. The earthwork lies on private land. It is located in a field close to Mill Lane at Swinden – between Bank Newton and Nappa, just 380m to the west of Swinden Moor Head farm. On the earthwork side of the lane Ash Tree Farm is about 530m up the fields to the east.

Cobers Laithe Earthwork, south-side (looking north-west).

Cobers Laithe Earthwork, south-side (looking north-west).

Cobers Laithe Earthwork, south-side (looking east)

Cobers Laithe Earthwork, south-side (looking east)

    It is quite a large earthwork measuring roughly 107m x 87m and it used to have a smaller inner earthwork with a circular bank but this seems to have disappeared altogether, maybe due to farming and the stream. At the time of my visit the outer bank and ditch of the earthwork were deep in grass and reeds in several places, but despite that they are quite pronounced at the SE and SW sides. And at the N, NW and NE sides the bank and its associated ditch are still quite well-defined, and there is a possible entrance at the NW. But as to whether they were ramparts designed for warfare or security, we don’t know, but in my opinion I would think they were non-defensive.

Cobers Laithe Earthwork (the enclosure south-side). (the inner enclosure).

(The enclosure at the  southern-side).

Cobers Laithe Earthwork (grassy bank and ditch)

Cobers Laithe Earthwork (grassy bank and ditch)

    The oval-shaped layout of the earthwork or camp does ‘not’ look particularly Roman to me. We know the Romans always built square-shaped fortifications. Although it could have been a ‘temporary’ Romano-British camp. However, there does not appear to have been much, if any, Roman activity in the area; and the theory that there was a temporary Roman fort or camp at Long Preston, in the field near St Mary’s church, seems to have all but died a death. In all probability the earthwork here at Swinden was a Brigantian camp, settlement, or enclosure. But another distinct possibility being that this was a more typical Romano-British farmstead.

Cobers Laithe Earthwork (southern-side grassy earthworks.

Cobers Laithe Earthwork (southern-side grassy earthworks.

    The relatively flat part of the earthwork, nearest the lane and fence, is on the south-side of the little stream while the northern and, by far the largest part of the earthwork, lies just beyond the stream and up the slight slope of the field beyond. But unfortunately the stream in between the two sections has made for some very muddy and boggy conditions, and so it is not easy to reach that ‘northern’ section unless you have wellies! Its quite obvious that the steam is a ‘more recent’ feature, having gouged out the channel through the centre of the ancient earthwork. To put it another way: the stream was not here when the earthwork was constructed. It was formed from a spring further up the field over hundreds of years, but certainly ‘not’ thousands of years.

    The northern part of the earthwork is pock-marked by holes or depressions (bell pits) in the ground, which are probably the result of quarrying for coal a few hundred years or so back. This poor or cheap coal substitute being used by local farms. Some of the holes have now become ponds.

    The Pastscape website has the site of an alleged Roman camp (monument no. 45530) about a ¼ of a mile to the north-west at Swinden (OS grid ref: SD 8617 5440), though this is perhaps an error? See the website link below.

Sources and related websites:-

Click on this Geograph link for a good photo of the enclosure on Swinden Moor:-     http://www.geograph.org.uk/photo/2837370

http://www.pastscape.org.uk/hob.aspx?hob_id=45530

http://www.pastscape.org.uk/hob.aspx?hob_id=45517

http://www.megalithic.co.uk/article.php?sid=28080

https://en.wikipedia.org/wiki/Swinden

http://www.gisburn.org.uk/nappa/

http://www.longpreston.info/history/history.html

http://mapio.net/o/1975548/

                                                        © Ray Spencer, The Journal Of Antiquities.


Werneth Low and Hangingbank, Greater Manchester

NGR: SJ 9709 9315. Werneth Low is a hill 279 metres high located just west of Uplands farm near Hyde in Tameside, Greater Manchester. It was the site of an Iron-Age settlement or enclosure on the south-west facing side of the hill, dating from the 1st century BC. A flint knife and a stone mace were excavated here. The Brigantes tribe who held the territory just before the Roman army arrived would have come here to the hill to celebrate both the winter and summer soltices. And what a wonderful sight that must have been!

Werneth Low near Hyde, Greater Manchester, on ...

Werneth Low near Hyde, Greater Manchester (Photo credit: Wikipedia)

NGR: SJ 9647 9352. Some 2 miles to the north-east of Werneth Low is another ancient settlement or farmstead called ‘Hangingbank’ just south of the A660. It is a low hill covering 3 acres that was inhabited from the middle to late Bronze-Age through to the Iron-Age. A number of Bronze-Age artefacts were found at the site. The earthworks at Hangingbank consist of a double-ditched enclosure with numerous crop-marks indicating where ancient field boundaries were. There may have been a Roman camp here in the 1st-2nd century AD as a shard of pottery was dug up from one of the ditches and a post hole was excavated. The course of the Roman road from Melandra Castle to Astbury supposedly crossed the site. Today a modern-day war memorial stands in the centre of the earthworks of the prehistoric settlement.

Copyright © Ray Spencer, The Journal of Antiquities, 2012 (updated 2025).