Heathery livrocky land: rethinking the stones of Neolithic Pembrokeshire

4 mins read
What is cultural, what is natural? On encountering the landscapes of upland Wales for the first time, early settlers could be forgiven for thinking that the hills were marked by the mounds and monuments of earlier inhabitants – perhaps a race of gods or giants. (Photo: Christopher Catling)
In a major new volume on the archaeology of Pembrokeshire, Tim Darvill argues that monument typologies do not help us understand how people viewed rocks and the landscape in the past. We need to think less like archaeologists and start asking questions about the meaning of stone and what these monuments might have signified to the people who built them. Chris Catling takes up the challenge.

Tim Darvill and Geoff Wainwright have been working on their SPACES project, mapping prehistoric monuments in the Strumble to Preseli landscapes of North Pembrokeshire, since 2001. Their recent joint paper in the newly published Pembrokeshire County History volume provides an up-to-date account of their latest thinking about such sites, covering a county that is now known, thanks to their work, to be well endowed with megalithic monuments, including dolmens and standing stones.

The Gors Fawr stone circle, shown here looking north-east over to Carn Menyn. (Photo: Tim Darvill)

While presenting their overview at the 2016 Pembrokeshire Archaeology Day, Tim Darvill broke off from his prepared presentation to issue a wake-up call to archaeologists wedded to taxonomy. ‘We have reached the limit of what we can learn by forcing these sites into monument typologies based on formal characteristics, and we must now try to make a leap of imagination and understand what these monuments might have meant to the people who built them’, he said, adding: ‘I come not to praise monument typologies but to bury them’, or words to that effect. This was archaeological heresy (even more shocking from the author of the Concise Oxford Dictionary of Archaeology, which is full of formal typological definitions and descriptions), but the audience at Pembrokeshire College in Haverfordwest loved it.

Tim went on argue that nobody in the past said: ‘OK, folks, let’s build a portal dolmen’. Monuments were not built to a strict template; there was no blueprint in the modern sense, nor any Platonic ideal to which everyone aspired. Instead, individual communities built individual monuments, similar but different; sometimes bigger and better than those made by others; always improvised according to local materials, resources, and circumstances. Builders not only reflected elements of their underlying beliefs but also, most likely, played out their desire to go one better, to show off. That being the case, we should celebrate diversity and difference, and recognise the human desire for one-upmanship, instead of classifying monuments according to their complexity and building a neat but misleading evolutionary sequence from the simplest to the most developed.

The types they are a-changin’

That is not to say that Tim entirely abandons traditional monument typologies in his contribution to Pembrokeshire County History, but he does argue for simpler classifications. For example, dolmens, he says, are simply raised stones, something that could be achieved in many different ways. In southwest Wales three main approaches were used, all of which share a common desire to lift a massive slab of rock out of the ground and suspend it in the air, in a manner once romantically described by Jacquetta Hawkes as ‘floating above the burial chamber’.

Pentre Ifan, Nevern, now stands proud and alone, but excavations have shown that it was a multi-phase monument with three major periods of activity. (Photo: Tim Darvill)

Not that all of these monuments were burial chambers, however: some are simply ‘propped rocks’, achieved by lifting one end of a large slab and supporting it using a second upright block of stone so that there is a gap between the stone and its bedrock matrix. In Pembrokeshire, the most impressive example is at Garn Wnda, near Llanwnda, where a massive slab of dolorite is supported on its downhill side by a pointed upright.

Propped rocks make a statement. Prehistoric people would recognise that this slab of rock had been positioned by human hands. Nobody could mistake them for a natural phenomenon – or could they? Everywhere you look in the upland landscapes of Wales, you will see rocks and mounds that make you ask ‘is it natural or is it the work of people?’. It is possible that monument-builders in the past were imitating what they saw all around them but with subtle differences. Perhaps they saw these pre-existing mounds and outcrops as the monuments of an earlier and now extinct race of giants, predecessors, or ancestors. Our categorical distinctions between ‘cultural’ and ‘natural’ may not have existed, or been more porous, back in the 4th and 3rd millennia BC.

This is especially true of the Preseli landscape, and of similar landscapes in the Cambrian mountains at the heart of Wales, or around the Harlech dome, where volcanic activity has created some very distinctive rocks and glacial erosion has left numerous features that we are now able to explain geologically, but that might have been seen in the past as evidence of supernatural forces: long ridges and terraces created by the deposit of morainic material; kettle holes; rock stacks and shattered rock outcrops; the formations known as roche moutonnée, pingos, and drumlins – all could be mistaken for the deliberate constructions of former inhabitants of the landscape. Glacial erratics can look like recumbent standing stones, and a roughly circular pattern of erratics at Llyn y Gorlan long ago entered Welsh folklore as a primitive Gorsedd circle, popularly believed to be the site of an early Eisteddfod back in the mists of time.

Coetan Arthur occupies an impressive setting and is visible from some distance away by sea and by land. (Photo: Tim Darvill)

A fascination with the meaning of these features may well have led to the ambition to imitate them, leading to a subtle interplay of natural and modified rocks. If so, this seems to have been an impulse widely shared, because propped rocks abound all around the Irish Sea basin and as far west as the Yorkshire Dales, not to mention on the near Continent. Many propped rocks occupy impressive settings and are visible from some distance away by sea and by land: Coetan Arthur, for example, is set on the cliffs above St David’s Head, while King’s Quoit, Manorbier, occupies a precarious perch at the end of a headland only 5m from the cliff edge. A propped rock on Skomer occupies a prominent skyline position at the north-eastern end of the island, moreover, while Carn Sian stands on the Preseli ridge. Only one of these – Garn Wnda – is associated with human remains. The early 19th-century antiquary who excavated the hollow beneath its slab found burnt bone and ‘very crude pottery’, but it is impossible to know now whether this was a primary burial associated with the construction of the monument, or a secondary interment dug by later visitors to the site.

This is an extract from a feature published in CA 324. Read on in the magazine. Click here to subscribe


  1. “For example, dolmens, he says, are simply raised stones”…LOL!! What we have here is academic frustration by the lack of understanding to this ancient civilisation. The archaeological norms tell us nothing, for they have failed to understand our ancestor’s motives and so modern usage of the words ‘ceremonial’ and ‘religious’ of meaningless content are now commonplace. To assist my fellow archaeologists, the reason that Dolmens are ‘non-standard’ is because they do not need to be as they are excarnation platforms and consequently, their design is immaterial.

  2. Carreg Samson is associated with St Samson of Dol rather than the Biblical Samson; Welsh folklore focuses on Welsh figures, such as St Gwyndaf of Llanwnda & Garn Wnda, and Arthur and his sons.
    To stress Welsh continuity, it is said that it took 2000 years of Welsh (language) culture to produce Waldo Williams, so it was a pity that Chris omitted to mention he wrote in Welsh! It was his own translation of Preseli that was quoted, with “livrocky” for the dialect “gelaets”.

Leave a Reply

Your email address will not be published.