Home » Celtic
Category Archives: Celtic
Bury Hill’s earthworks are quite prominent on the summit of a low rise at the confluence of the rivers Anna and Anton, near Andover. An Early Iron Age hillfort of 10 ha (24 acres) occupies the site but was superseded by a much stronger circular fort of half the size. The earlier enclosure, tested by geophysical survey and excavation, had no evidence of occupation, but it was a different picture in the later camp, which was densely packed with features.
What was striking about Barry Cunliffe’s work here in 1990 was the remarkable haul of horse harness and chariot fittings. The excavation produced six copper alloy terret rings, three strap unions, two small strap rings, a linchpin and an antler side piece capped with a copper alloy disc, as well as numerous iron horse bits and nave rings. Actual horse remains comprised 48% of the animal bones found, about ten times higher than at nearby Iron Age sites, suggesting highly specialised activity at Bury Hill.
Calculations show an average height of eleven hands for a horse population aged between one and fourteen years. With less than 5% juvenile mortality it is likely that the animals were not kept for breeding and they may have been semi-feral, similar to New Forest ponies today. With peak mortality at age six or seven years it is also unlikely that they were exploited for meat. The probability is that the herd was carefully managed, with the emphasis on horses for riding and chariot use.
It seems clear that the community viewed the chariot as a symbol of prestige. It is likely that the vehicles were manufactured at the site and the necessary specialised metalwork made there by craftsmen working for the local elite. This was a group potentially more warrior-dominated than in previous Iron Age generations and ties in with Caesar’s descriptions from 54 BC, when the Late Iron Age forces ranged against him included up to 4,000 chariots!
The Wessex Hillforts Survey has identified an oval enclosure of approximately 1.6ha (4 acres) just outside the later camp at Bury Hill, with its largest entrance facing towards the fort. Is this where the elite warriors – the royal house guard or comitates – strutted their stuff, leaving the artisans to their carpentry workshops and smithies inside the defences?
The allegiance of this Late Iron Age ‘arms factory’ is not clear. The local Atrebates tribe were pro-Roman, but suffered incursions from the Trinovantes and Catuvellauni. Whoever it was they followed, there’s no doubt that the later 1st century BC and the early 1st century AD were turbulent times, if the importance given to the production of weapons of war is anything to go by.
Royal Blood Postscript: In later times Polydore Vergil (16th century) and Sir Richard Colt Hoare imagined Edmund Ironside and the Danish King Canute occupying Bury Hill and Balksbury respectively, as they slogged it out for possession of England in the early 11th century, but there is no hard evidence for this episode from either site.
Cunliffe, B & Poole, C. 2000. The Danebury Environs Programme: The Prehistory of a Wessex Landscape Vol 2, part 2. Bury Hill, Upper Clatford, Hants, 1990. English Heritage and Oxford University Committee for Archaeology Monograph no. 49.
Series by: Dave Allen, Sarah Gould, Jane King, Lesley Johnson, Peter Stone
In the autumn of 1994 a metal detectorist found a Late Iron Age decorated bronze mirror, the first from Hampshire, at Latchmere Green, near Silchester. It was associated with the cremation burial of a woman and a child, and tells of a tradition of high status metalwork ‘reflecting the British nobility’s contacts with, and travels in, Italy in the late 1st century BC and early 1st century AD’.
The mirror was unearthed just within Bramley parish, at the southern edge of a known Roman site close to the junction of the Roman roads from Silchester to Winchester and Chichester. This settlement had been previously surveyed (Corney, 1984) and had yielded pottery finds of the late 1st to early 2nd century, through to the 4th century AD.
The mirror itself was badly corroded, with the handle and plate separated, and was in generally poor condition. Assuming circularity, dimensions were estimated as 170 mm diameter, giving a 227 sq cm surface area; it was 1.1 mm thick. The overall length (mirror and handle) was estimated to be about 263 mm.
Basket engraving on the reverse side of the mirror plate was found to be in the form of a whirligig or triskele, with the lower arms extended at right angles to the axis of the plate in pelta (or shield-like) loop patterns, giving the impression of a pair of eyes.
This elaborate style of decoration, known as ‘irregular oblong block’, is not unlike that of the 1904 ‘Colchester mirror’ and other similar finds and is thought to date from the late 1st century BC to the early 1st century AD, while the plate area links the mirror with the south-eastern group* of mirrors, centred on East Anglia and the London Basin. Up to 60 mirrors of the period are now known, and many of these finds have been made over the past three decades.
Samples of metal taken from the handle and plate of the Latchmere Green mirror consisted wholly of ~88% copper and ~12% tin with a trace of phosphorus, which showed that the mirror was not of Roman manufacture. Roman mirrors always contain more than a trace of lead and a lower proportion of copper with tin.
Following the initial discovery, a small controlled excavation (5 x 5 m) unearthed a late Iron Age pedestal jar, lying on its side in a shallow pit. The jar contained a quantity of cremated bone and the evidence suggested that the mirror had been placed as a ‘lid’ closing the jar. Also present were fragments of iron pin and other pieces associated with brooches. These latter were found to be comparable with finds at Silchester and Thetford and again can be dated to the very late 1st century BC and the early 1st century AD. The wheel-thrown jar, although in a very fragmented condition, is of a type identifiable with the pre-conquest period.
The cremation was unusual, in that the small fragments proved to be very probably of a female, aged 30 or more, with a child who was not newborn or an infant. That the adult was most likely to be female is evidenced by the fact that such mirrors have never been unambiguously associated with Iron Age male burials. The small fragments of cremated animal bone present were found to be of pig – again an occurrence consistent with animal bone finds in other Iron Age burials.
In summary, the evidence points to a late pre-conquest or early post-conquest date for this Late Iron Age high-status burial, possibly of a mother and child (but see below). As to the triskele design with loops it can only be said that this is of unknown origins but no object associated with it has been datable to earlier than 1st century BC. The ‘masterly’ and ‘mature’ embellishments, as they have been described, would appear to be unique to Britain.
*As the Latchmere Green discovery is an outlier to the south eastern group, it was included in the Dating Celtic Art programme of radiocarbon dating (Garrow et al, 2009). The dates realised (360 – 50 and 360 -110 cal BC) seem too early (2nd century BC) for the associated brooches and pottery vessel, and there is the intriguing possibility that one of the individuals (the one dated) had been cremated some decades before the double burial was actually made.
M Corney (1984) A Field Survey of the Extra-Mural Region of Silchester, in M Fulford, Silchester Defences, 1974-80.
M Fulford & J Creighton (1998) A Late Iron Age Mirror Burial from Latchmere Green, near Silchester, Proc Prehist Soc, Vol 64, pp331-342.
D Garrow et al (2009) Dating Celtic Art: a Major Radiocarbon Dating Programme of Iron Age and Early Roman Metalwork in Britain, Arch J, Vol 166, pp 79-123.
Archive A1994.26, held by Hampshire Cultural Trust
The Latchmere Green mirror is currently on display at the Museum of the Iron Age, Andover.
Series by; Dave Allen, Sarah Gould, Lesley Johnson, Jane King, Peter Stone.
When the much-needed Andover bypass swept around the southern reaches of the town in the late 1960s it sliced through the southwest corner of the 18 ha (45 acre) plateau enclosure of Balksbury. From that moment on, the earthwork’s days were numbered. Housing developments eventually arrived and now carpet nearly all of the interior, but at least the disappearance of the archaeology was well-documented, and six episodes of excavation are on record.
In 1939 the southern side of the defences was examined by Jacquetta Hawkes, while husband Christopher was busy excavating at nearby Bury Hill. Twenty years later the northeast corner of the enclosure was looked at during house-building. More substantial opportunities came in 1967 with the arrival of the bypass and again in 1973, when the same excavator, Geoff Wainwright, undertook one of his celebrated ‘big digs’, uncovering 10 ha of the interior. The Central Excavation Unit investigated a further 2 ha in 1981. Finally, between December 1995 and April 1997, another 5.5 ha of the interior was examined, additional work took place outside the enclosure, and a section of bank and ditch was looked at in detail.
Viewed as a whole, the site produced evidence from Neolithic to Late-Roman date, although the earliest activity is represented only by stray finds, including a Beaker burial of an adolescent female.
Around the 8th-9th centuries BC (Late Bronze Age) large-scale clearance of woodland preceded the construction of the enclosure. A single entrance was found to the southeast as well as three phases of bank and ditch, some involving timber strengthening. If there was any occupation at this period, it apparently took the form of a few four and five-post structures, scattered around the periphery. In the Early Iron Age the evidence is more substantial, with three round houses and 27 storage pits identified. Pits also dominated in the Middle Iron Age, with 90 attributable to this phase, but there were no recognisable structures.
In the Late Iron Age and Early Roman periods a number of pits and gullies were dug but the type of activity they represent is elusive. It does, however, appear to merge into the later Roman occupation which involved a substantial building with ovens (and painted wall plaster), a corn-drier, small enclosures, pits and burials. Finds included both local and imported pottery, worked bone and stone, coins and jewellery, including 11 brooches, four bracelets, five rings and a buckle plate. Part of a scabbard for a La Tene 1 type dagger, a rare find from a settlement site, is of Middle Iron Age date.
In the mid-1990s, the final phase of excavation concentrated on the nature and economy of the enclosure, rather than its wider landscape setting. The enclosure, at 18 ha, was too big to be defended effectively and its description as a ‘hillfort’ is wrong. It probably marked out an important place in the landscape, a focal point for communal activities, which may have included feasting, exchange of goods, and the means to build social relationships. Later activity took place in the centre of the site and the impression is of a succession of small farming settlements taking advantage of an existing, but redundant, enclosure.
A1978.12 Archive held by the Hampshire Cultural Trust.
Balksbury Camp, Andover, Excavations 1973 & 1981, Wainwright & Davies (1995), English Heritage Archaeological Report 4.
Excavations at Balksbury Camp, Andover 1995-97, Ellis & Rawlings (2001), Hants Studies Vol 56.
Series by Dave Allen, Sarah Gould, Lesley Johnson, Jane King, Peter Stone
Today, the area around Sandford Springs, near Kingsclere, is occupied by a golf course and attendant hotel. In prehistoric and Roman times it seems likely that the springs were the focus for religious and ritual activity, probably involving a temple or shrine.
When the A339 Kingsclere bypass was constructed, in the early 1980s, a number of Bronze Age artefacts were recovered, including a tanged chisel and socketed axe, presumably from a dispersed hoard. Somewhat previously, the enlarging of the eponymous pond had thrown up a large number of Iron Age and Roman coins and other offerings, including a fine, enamelled, military style Roman belt buckle.
When work began on landscaping the golf course in the mid 1980s, archaeological monitoring took place, although features were few, being limited to a shallow gully and patches of burning. The practice of the then landowner, to always have his metal detector at the ready, resulted in a spectacular find, however, when he discovered the ‘Kingsclere crock’.
As one of the grading machines was smoothing the contours, a large flint nodule detached itself from the soil and rolled obligingly down to his feet. A pass over it with the detector resulted in a loud signal and removal of a clay ‘plug’ revealed seven Iron Age gold staters, neatly filling a natural cylindrical hole. The coins were subject to a Coroner’s Inquest (the flint wasn’t – although it would be today) and declared to be Treasure Trove. Hampshire County Council was happily in a position to acquire them, and the find is on display at the Willis Museum, Basingstoke.
The ‘triple tailed horse’ issue dates to around 50 BC and the coins were probably struck by the Atrebates at Calleva (Silchester). The design, a stylised horse straddling a wheel, is derived from Macedonian gold coinage which showed Philip II (d. 336 BC) father of Alexander the Great, driving a two horse chariot in which he was victorious in the Olympic Games.
The discovery brings to mind a find made some 23 km to the west at Chute, Wiltshire, in 1927. A 13 year old boy, Victor Smith (living at The Forge!) was acting as a beater for a Chute shoot, when he threw one large flint against another and a shower of gold spilled out. After one or two return visits he had collected 65 coins and these were declared Treasure Trove. In 1986 an enterprising detectorist searched the same site and found a further 55 coins. These could not all have fitted inside the flint (which is in Devizes Museum) but were struck with the same range of dies (stamps) and must have been part of the original deposit.
There have been other discoveries of flint ‘crocks’, from Rochester in Kent, for example, and more recently (2004) Henley in Oxfordshire, where 32 coins very similar to the Kingsclere examples, came to light.
Series by Dave Allen, Sarah Gould, Lesley Johnson, Jane King, Peter Stone, with help from Stacie Elliot.
Many carved stone heads have been found in Britain, mostly in the north and west of the country. A great number of them originate from the religious practices of the native (Iron Age) people living in Britain both before and during the Roman occupation of the first four centuries AD. They are essentially cult objects. Hampshire has little natural stone (other than flint and chalk) but a fine example of a head comes from the parish of Boldre in the New Forest. It is made of Bembridge limestone, which is found nearby on the Isle of Wight and also on the Isle of Purbeck, Dorset.
The head was discovered in a disused gravel pit, then being used as a rubbish dump, by a boy who was staying in a cottage in Portmore, Boldre, and he took it back to the village. In the 1960s it came to the attention of Sidney Jackson, an archaeologist from Yorkshire, who was keen on cataloguing as many stone heads as he could. At that time, it was resting on a makeshift base beside the gate of a cottage. Permission was granted for the stone to be taken to Southampton for examination and recording and, following a number of adventures including a long spell up north, it is now in the care of the Hampshire Cultural Trust.
The head is of a ram-horned man. The eyebrows extend across the head in an unbroken line – a typically Celtic feature. The eyes are bulbous and the moustaches sweep upwards. These features give the face a ferocious appearance. Unfortunately, the lower part of the face is damaged, so we cannot be sure of the full expression of the mouth. Ram’s horns are a symbol of virility and strength and are associated with the cult of a horned god, known throughout the Celtic world. This suggests a warlike, even if pastoral, attitude for the people who made the object. The human head was thought to have evil-averting powers (apotropaic), and the fierce expression would have added to its protective potency.
The closest parallel to the Boldre head is one found in County Armagh, Northern Ireland, possibly connected with a nearby pagan sanctuary. Boldre is close to the presumed boundary between two Iron Age tribes – the Atrebates (to the east) and Durotriges (to the west); and the head may have played a role in defining the limits of their territories or been associated with a native shrine or temple.
Pagan Celtic Britain, Anne Ross (1967)
Celtic and other stone heads, Sidney Jackson (1973)
Proc Hants Field Club 26, 57-60 Anne Ross (1969) A Romano-British Cult Object from Boldre.
Series by Dave Allen, Sarah Gould, Lesley Johnson, Jane King, with help from Stacie Elliot.