British Archaeology banner

Cover of British Archaeology Issue 61

Issue 61

October 2001



Ice Age hunting camp found in Hampshire

Medieval hospital found under modern hospital in Stoke-on-Trent

Early leadmill uncovered in Sheffield

Mystery of ‘old soldier’s homestead’ in East Yorkshire

Lakenheath cemetery focused on Bronze Age mound

Iceman’s murder, Iron Age boats, Roman mosaic


Years of pestilence
Tom Beaumont James on archaeology of the Black Death

The medieval fleet
Gustav Milne on the medieval shipbuilding industry

Great sites
Stephen Aldhouse-Green on Paviland Cave in South Wales


On Bronze Age necklaces, roofing materials and quarries


George Lambrick on proposals to alter the planning system

Peter Ellis

Regular column


Copperopolis by Stephen Hughes

Debating the Archaeological Heritage by Robin Skeates

Safe Sanctuaries by Christopher J Brooke

Blood Red Roses edited by Veronica Fiorato, Anthea Boylston & Christopher Knüssel

Seahenge by Francis Pryor

What the Romans did for Us by Philip Wilkinson

CBA update

favourite finds

Tom Williamson’s was a prehistoric landscape in Norfolk


ISSN 1357-4442

Editor Simon Denison


Great Sites: Paviland Cave

Stephen Aldhouse-Green looks at new research on the site of Britain’s best-known Palaeolithic burial, a young man who died some 26,000 years ago

Goat's Hole cave, Paviland, on the Gower Peninsula in South Wales, has long been renowned as the site of one of the best-known prehistoric burials in Britain - the notoriously misnamed 'Red Lady of Paviland' which was discovered in 1823. Yet it has taken archaeologists nearly two centuries to unravel the mysteries of this remarkable site, with the definitive report published only last year.

The story begins not in 1823 but during the previous year, when Daniel Davies and the Rev John Davies, respectively surgeon and curate at Port Eynon on the south coast of Gower, explored the cave and found animal bones, including the tusk of a mammoth. The Talbot family of Penrice Castle was informed and Miss Mary Theresa Talbot, then the oldest unmarried daughter, joined an expedition to the site and found 'bones of elephants' on 27 December 1822.

William Buckland, Professor of Geology at Oxford University and a correspondent of that well-connected family, was contacted. He arrived on 18 January 1823 and spent a week at Goat's Hole - a week in which his famous discovery took place. He later wrote:

[I found the skeleton] enveloped by a coating of a kind of ruddle . . . which stained the earth, and in some parts extended itself to the distance of about half an inch [12mm] around the surface of the bones . . . Close to that part of the thigh bone where the pocket is usually worn surrounded also by ruddle [were] about two handfuls of the Nerita littoralis [periwinkle shells]. At another part of the skeleton, viz in contact with the ribs [were] forty or fifty fragments of ivory rods . . . [also] . . . some small fragments of rings made of the same ivory and found with the rods . . . Both rods and rings, as well as the Nerite shells, were stained superficially with red, and lay in the same red substance that enveloped the bones. (Buckland, Reliquiae Diluvianae, 1823)

In the field, Buckland had identified the skeleton as male, suggesting that the bones were those of a Customs Officer murdered by smugglers. By the time of publication later that year, however, the gender had changed with a new and better story.

The ochre-stained skeleton had become a 'painted lady' who serviced the needs of the Roman soldiers garrisoned in the camp on the hill above the cave. It was a good story. But by the early years of the 20th century, it could be seen not to add up: the burial was male, the mammoth ivories were Palaeolithic and not modern products made from fossil ivory as Buckland had claimed, and the camp was an Iron Age promontory fort.

Buckland had found a burial, representing a single event in the human history of the site, but found no more than a few flints, even though the site has since yielded thousands. Indeed, in subsequent excavations the site was to produce diagnostic or dated material - flint, ivory and bone artefacts, and the burial itself - spanning more than half a dozen Palaeolithic events over at least the period 35,000-11,000 BP (before present). Buckland did not know this. What he did know, as Dean of Westminster and Curate at Christchurch - or strictly speaking what he believed - was that the bones of such animals as mammoth and woolly rhino found in the cave could not be contemporary with the burial since such species, he thought, had not made it onto the Ark and so had been drowned in Noah's, or an earlier, flood. His belief in such a deluge is shown by the title of his book, Reliquiae Diluvianae ('Evidence of the Flood').

He therefore regarded the 'Red Lady' as intrusive, a reasonable inference given the intellectual context of his day and the undeveloped nature of archaeological excavation as a technique. But did he really miss the flints because of his mindset? The answer may be yes, but Buckland's elevation drawing of the site suggests that the burial may have lain at the lowest level then exposed. The many thousands of lithics, now interpreted as earlier than the burial, were not found until 85 years later through the excavations of William Sollas, also holder of Oxford's Chair of Geology.


The 'Red Lady' would now be interpreted as a ceremonial interment of the Gravettian period of the Palaeolithic (c 28-21,000 BP), such as are now known across Europe from Paviland to Moscow and south to Portugal. But at the time when the 'Red Lady' was unearthed she - or rather he - was not only the first such burial to be found but also the first human fossil ever to have been recovered anywhere in the world.

The radiocarbon dating technique was not invented until the late 1940s, so neither Buckland nor Sollas could have known the true age of the interment. Sollas did, however, work on the assumption that the burial was Palaeolithic. He confirmed the burial site by finding a spread of ochre associated with ivory rods parallel to the cave wall, and added to our understanding of how the body - which was incomplete at the time of discovery probably because of marine erosion - had been interred.

The bones were deeply stained with red ochre, and the grave goods - ivory rod and bracelet fragments, and perforated periwinkle shells - were all similarly stained. In addition, small limestone blocks may have been placed at the head and feet. Perhaps, too, the skull of a mammoth found nearby may have been part of the grave furniture - this was the interpretation of the Abbé Breuil who had joined the Sollas expedition in the role of lithics analyst. Sollas correctly identified the body as that of a man.

Sollas's fieldwork at Paviland in 1912 had been prompted by a visit to Oxford of the French scholar Emile Cartailhac, then preparing his magnum opus on the caves of Grimaldi in Liguria. Cartailhac dated these burials as Upper Palaeolithic and suggested that the mode of burial at Paviland was comparable. It was not until the 1960s, however, that an attempt was made to date the burial scientifically, when Kenneth Oakley published a radiocarbon determination made on the actual bones of the 'Red Lady'.

The result of 18,460 ± 340 BP coincided with the peak of the last Ice Age when the edge of the ice lay only an hour's walk north of Goat's Hole. The date conjured up a picture of great charm and power, with a later suggestion that the body could have been transported from somewhere further south to a distant, venerated site at the edge of the ice for a summer burial when temperatures rose.

Things were beginning to hot up, however, for John Campbell's 1977 study of the Goat's Hole lithic assemblage showed convincingly that it belonged to the later part of the Aurignacian period of the Palaeolithic (c 40-28,000 BP) and was perhaps 30,000 years old. He suggested, however, that the burial might be younger, specifically Gravettian, on the basis of the dating of comparable European ivory bracelets.

At much the same time, hitherto little known material from Belgium was being studied by Marcel Otte and became widely known through his publication of a synthesis on the earlier Belgian Upper Palaeolithic. This was complemented, in 1980, by Roger Jacobi's ground-breaking study of the British Upper Palaeolithic. Jacobi undertook a rigorous analysis of both the 'Red Lady' burial and human presence at Goat's Hole and concluded that parallels could be found within the Belgian Aurignacian for the ivory artefacts associated with the interment and that the 'Red Lady' was therefore likely to be of that age.

He dismissed the 18,460 bp date on the grounds that human presence was simply lacking from north-western Europe at this time. Jacobi pointed also to continental parallels to artefacts termed 'leaf points' that should be contemporary with or, even, predate the Aurignacian and reaffirmed the presence there of a flint 'Font Robert' Gravettian spearpoint and of Late Upper Palaeolithic artefacts that related to the recolonisation of the British peninsula after the peak of the last Ice Age sometime after 13,000 bp.

In other words, it was quite clear by 1980 that Goat's Hole had been the scene of a number of apparently discrete phases of human presence spread over 20,000 or more years of the Palaeolithic. But all this knowledge depended largely upon typological parallels and on one very suspect radiocarbon date.


The stage was clearly set for a scientific re-evaluation of Paviland. In 1989, a new and far more plausible result of 26,350 ± 550 BP was produced by the Oxford Radiocarbon Accelerator on bone powder residual from the original Oakley sample. Because contamination of ancient samples normally results in ages that are too young, it was reasonable to assume that the 'Red Lady' died around 26,000 years ago in radiocarbon years (calendar years are possibly several thousand years older than radiocarbon years at this date). Even so, this dating was at least a couple of millennia too young for the Aurignacian.

Could he be yet older? A visit that I made myself to the site in 1989, following a massive storm that had exposed apparently in situ deposits, convinced me that further work would be useful. There the matter rested, however, until 1995 when I received an unexpected letter from Erik Trinkaus revealing that he had made a comprehensive study of the 'Red Lady' skeleton some years previously. Since then, Erik had been sidetracked into Neanderthal studies to the great benefit of palaeoanthropology. But now he was once more involved with the study of anatomically modern humans and the 'Red Lady' paper was just awaiting a publication outlet. This was the final spur to action.

The new study began with a radiocarbon dating programme and resulted in the dating of some 40 radiocarbon samples of fauna, artefacts and the bones of the 'Red Lady' himself. The skeleton was re-dated to 25,840 ± 280 BP and an age of the order of 26,000 years confirmed. None of the ivory or shells associated with the 'Red Lady' was dated because of problems of potential contamination by preservatives, but charred bone dates are earlier and centre on 28,750, and so are plausibly Aurignacian.

Of the ivory pieces, 75 per cent are ornaments, virtually all associated with the burial of the 'Red Lady', although the well known perforated ivory pendant made from a growth in a mammoth's tusk is later at 24,000 BP. Bone artefacts include three bone spatulae dated to 23,000 BP. The latest phase of human presence with a firm radiocarbon date is represented by ivory-working of 21,000 BP.

At the time when the young man was ritually interred, there is no substantive evidence in this remote part of Europe for a human presence that was other than episodic. Indeed, faunal compositions and densities probably oscillated over time and space. Human presence in the British early Upper Palaeolithic may plausibly be linked to a 'biomass expansion', an overall increase in the availability of animals and other forms of food, centred on the 29th millennium. The coincidence of the dating of burnt bones to this period, combined with the presence of burnt Aurignacian artefacts, supports this as the most likely time for Aurignacian presence at Paviland. Radiocarbon dating of an Aurignacian bone spearpoint to around 28,000 bp at nearby Uphill lends additional weight to this interpretation.

Gravettian visitation is attested by a scatter of large tanged points occurring across southern Britain, including Paviland. Such points are generally dated to 28-27,000 BP, although their use may possibly extend down to the time of the 'Red Lady' burial.


As part of the radiocarbon dating process, the 'stable isotope values' of carbon and nitrogen within the bone sample were measured. These values provide important information about ancient diets and show that the 'Red Lady' had a penchant for seafood - either collected when living on the coast then 100 kilometres distant, or in the form of salmon fished out of the palaeo-Severn, which bears from Paviland are also known (from stable isotope values) to have eaten.

The 'Red Lady', when alive, was a healthy young adult male - aged 25-30, about 5' 8" (1.74 metres) in height, and possibly weighing about 11 stone (73 kg) - but less robust than might be expected for this period. Whilst the earliest anatomically modern humans in Europe were characterised by tropically-adapted body proportions, arising from their African ancestry, this is not reflected in the skeleton of the 'Red Lady', probably because the Paviland individual was a product of perhaps 10,000 years of evolution of modern humans within Europe.

Molecular biologist Bryan Sykes has shown that the 'Red Lady' skeleton has a DNA sequence corresponding to the commonest extant lineage in Europe. As such, the Paviland evidence lends support to the argument that the roots of modern Europeans lie not with Neolithic farmers but with the ingress into Europe of human populations who were to replace the Neanderthals.

Restudy of the Goat's Hole lithic collections has confirmed material ranging from about 40,000 BP to about 13,000 BP (including Mousterian, leaf point, late Aurignacian, early Gravettian, Creswellian, and Final Upper Palaeolithic phases), although the earliest and latest phases are not dated by radiocarbon. Aurignacian finds form the dominant element. These artefacts were made from a range of imported and local raw materials. It is interesting that analysis of the ochres is consistent with a local origin, probably within Gower.

Clearly, the people responsible for the interment possessed considerable local as well as more far-flung knowledge. Preferential use of imported flint for 'busqué burins', a specialist kind of engraving tool, and blade blanks, suggests the import to the site of curated items. A type of Aurignacian inversely retouched scraper is special to the site and may reflect the long term isolation of a social group.

The ceremonial burial of the 'Red Lady' involved the interplay of art and consciousness which combine in an act that is simultaneously creative and symbolic. The rite possesses features replicated, in regionally changing modes, across Europe in other ceremonial Gravettian burials. These include an extended burial position, positioning of the corpse along the cave wall, the presence by the grave of large herbivore remains, the placing of stone slabs at head and feet, the use of ochre, the deposition of personal ornaments, and the possibility that the body may have been headless when interred. No head was found at Paviland, and other headless Gravettian burials are known.

In chronological terms, Paviland is early in the European series of burials and is actually the earliest with a firm radiocarbon date measured directly on human bone. These burials are resonant of a complex early European society in which status may have been inherited rather than acquired by merit - as evidenced by several very rich child burials elsewhere.

In the ancient world, the sacred and profane were inextricably intertwined. Paviland cave was occupied by the hunters of the Gravettian mammoth steppe as a functional shelter; but there may also have been an aura of sanctity attached to the place, explaining the burial here of the 'Red Lady'. We may wonder whether one reason for visits to Paviland, as the climatic downturn accelerated and the British peninsula was increasingly abandoned, may have lain in its status as a special place.

Stephen Aldhouse-Green is Professor of Human Origins at the University of Wales, Newport. His report on the cave, 'Paviland Cave and the Red Lady', was published by Western Academic & Specialist Press last year at £40.00.

CBA web:

British Archaeology

Jan/Feb 2005
Mar/Apr 2005
May/Jun 2005
Jul/Aug 2005
Sep/Oct 2005
Nov/Dec 2005
Jan/Feb 2006
Mar/Apr 2006
May/Jun 2006
Jul/Aug 2006
Sep/Oct 2006
Nov/Dec 2006
Jan/Feb 2007
Mar/Apr 2007
May/Jun 2007
Jul/Aug 2007
Sep/Oct 2007
Nov/Dec 2007
Jan/Feb 2008
Mar/Apr 2008
May/Jun 2008
Jul/Aug 2008
Sep/Oct 2008
Nov/Dec 2008
Jan/Feb 2009
Mar/Apr 2009
May/Jun 2009
Jul/Aug 2009
Sep/Oct 2009
Nov/Dec 2009
Jan/Feb 2010
Mar/Apr 2010
May/Jun 2010
Jul/Aug 2010
Sep/Oct 2010
Nov/Dec 2010
Jan/Feb 2011
Mar/Apr 2011
May/Jun 2011
Jul/Aug 2011
Sep/Oct 2011
Nov/Dec 2011
Jan/Feb 2012
Mar/Apr 2012

CBA Briefing

Courses & lectures
CBA Network
Grants & awards

CBA homepage