Analysing the new site near Stonehenge

There have been two really interesting discoveries in the Stonehenge area over the past few weeks. One is at Marden, the great henge earthwork in the Vale of Pewsey near the source of the river Avon. I’ll write about that in my next post, but first I’d like to add a few words to the extensive media coverage of the geophysics survey that, according to National Geographic, revealed Stonehenge’s “long-lost twin” close by (otherwise identified as Stonehenge’s “sister” – the site is in front of the bank of trees in the photo above, on the horizon on the far right).

Sister or twin, that’s quite a claim, and it’s no surprise that it came hot on the back of a press release (from the communications department at the University of Birmingham), that was embargoed till June 22 when the story suddenly broke. Nowadays we hear about many new archaeological discoveries in this way: an organisation or an individual has something to promote (usually the organisation or the individual), and research is spun into something they think will appeal to the media. The media then repeat all the spin, and use the carefully vetted artificial quotes and the supplied illustrations, often (especially websites and bloggers) at several removes from the original release – confusing mildly-animated photocopying with journalism.

At its worst, it’s manipulative, lazy and patronising: but it’s better than the old system of archaeological reporting, which largely consisted of the excavator telling their small group of friends and noone else – and if asked, often flatly refusing to pass anything on to outsiders. Besides, there is no law against research, and the great thing about the press release is that it releases information. It’s not the system that’s at fault: it’s the peculiar reluctance of anyone to ask questions.

I was asked to comment on the Stonehenge geophys discovery for BBC News, which I did rapidly. I stand by what I first wrote about the find (“perhaps a henge, perhaps not, but an important discovery whose significance will be fully realised only with excavation”), but now I can dig a little deeper.

“A new henge discovered at Stonehenge”

That is how Birmingham University headed its press release, and you get the agenda from the opening two sentences:

“History is set to be rewritten after an archaeology team led by the University of Birmingham and the Ludwig Boltzmann Institute for Archaeological Prospection and Virtual Archaeology in Austria discovered a major ceremonial monument less than one kilometre away from the iconic Stonehenge. The incredible find has been hailed by Professor Vince Gaffney, from the University’s IBM Visual and Spatial Technology Centre, as one of the most significant yet for those researching the UK’s most important prehistoric structure.”

Great stuff, but what I’m interested in here is the archaeology. Is it a henge? We can’t possibly say. Is it so important? In itself, probably not, but for the issues it raises, I’d say it is. Gaffney goes on to say, “This finding… will completely change the way we think about the landscape around Stonehenge… [because formerly] people have tended to think that as Stonehenge reached its peak it was the paramount monument, existing in splendid isolation”. Engaging enthusiasm – but pure hyperbole.

Let’s start with the location. There’s a map on the web credited to Birmingham University apparently showing the site quite precisely:

I’ve superimposed on that the best barrow map in print (from the study by Desmond Bonney and Isobel Smith [1]). This shows the new site to be immediately south of barrow Amesbury G51, and north-west of barrow G50, both of which you can see in the air photo. (All the barrows – ancient burial mounds – around Stonehenge are numbered in a system created by Canon Goddard in 1914, grouping them by parish).

So the first point is that the site is in a large group of known barrows: we call it the Cursus group, after the long linear earthwork to the north, recently radiocarbon dated to 3630–3370BC [2]. This is around 1,500 years before people started to build the round barrows, so it’s possible the old earthwork attracted the barrows, or that the reasons for the siting of the Cursus’s western end and for the siting of the barrow group had something in common; though there are many other barrow groups apparently unrelated to the Cursus.

Henge or barrow?

The first working hypothesis, then, would be that what has been found is another barrow – perhaps not so attractive an idea to a press office, but clearly the most obvious one. So does it look like a barrow?

As a general rule, bronze age barrows are domed mounds built of turf, soil and rock, much of which came from a quarry ditch that encircled the mound. Dug into the centre below ground was a grave, and commonly people buried their dead in subsequent years in and around the barrow. In practice there is enormous variety in the ways these structures were put together, partly by design and partly because they were often altered over the years, and used for many generations and through changes in funerary ritual.

Around Stonehenge is one of the largest concentrations of barrows in north-west Europe. We know surprisingly little about them. Most excavation occurred in the early 19th century, when hundreds were dug in a search for graves and treasures, most famously by William Cunnington and Richard Colt Hoare. These men were pioneer archaeologists, and unsurprisingly we learn almost nothing about the barrows themselves from their records.

The other main period of excavation was in the 1950s and 60s. Archaeologists like Paul Ashbee, Patricia Christie, Ernest Greenfield, Nicholas Thomas and Faith and Lance Vatcher excavated dozens of barrows, often with immense skill. But there were two problems with this work. Resources were rarely committed to study and publication, so in many cases the results have only recently been published. Perhaps more significantly, most of these excavations were designed to “rescue” barrows that had all but been ploughed out. So we have a lot of graves, pits and ditches, but very little to tell us what the barrows looked like above ground.

This does not however, hinder our search for barrow analogues for the new site. The published geophysics plot seems to exhibit three significant features.

There is a ring of large dark blobs, probably deep pits; within this a ring of smaller and lighter circular blobs that look like postholes; and a darker area in the centre that may be the remnant of a mound, or old turf that had been protected by a mound since ploughed away.

All of that, taken with its scale, shouts “barrow” to me. We have a mound in the centre:

And an outer ring of posts that might also have been incorporated in the wider mound, surrounded by an irregular ditch that would have served as quarry for the mound:

If you know a little about these barrows, you might think that sounds like a rather odd one; and indeed, it would fall outside the several typologies that have been constructed. But that is a fault of the typologies, that use only a small part of the available data on barrows. There may not be an exact parallel for this as a barrow (and I’ll repeat the point that we know very little about the range of forms these barrows took), but we can find the components without looking too far.

Here, for example, are two barrows with rings of postholes.

Winterbourne Stoke G39 was excavated by the Vatchers in 1961 [3]. It was in a group just west of the Lesser Cursus (the “Silo group”), 1700m north-west of the new site. Faith thought the “stake circle” was contemporary with a central “mortuary house”; the cremated remains of five people were identified in the central grave pit, among them a neonate and an infant. The 57-post ring was 11.5m across, and the pits at least were within the mound which was enclosed by a ditch 30m across.

Snail Down site 15 (barrow Collingbourne Ducis G3) is more distant, on the north-east part of Salisbury Plain. This was dug by Charles Thomas in 1955 [4]. Like the previous barrow, little of the original mound survived. The postholes showed stains where the stumps had rotted underground, but the excavators again were unable to say whether the posts had risen up through the mound or been removed before its construction, though they did think it was likely to have been a complete ring and that missing holes had been eroded. Thirty six of the posts were on half a circle 7.4m across. Five skeletons were found buried under the mound, in pits marked I–V on the plan.

And the ditch?

The post ring at the new site has been compared by Birmingham archaeologists to the much larger site of Woodhenge, where there were six concentric oval rings of posts, some of them massive. The irregular ditch (composed of the large pits) is also claimed to be indicative of a henge; there would in that case have been a bank beyond the ditch, though there are no obvious signs of this in the geophysics plot.

While many barrows do have regular, circular ditches with no causeways, irregular ditches are not uncommon. As with the post ring, this in fact could easily be a feature of a barrow. A very similar arrangement, for example, is seen at Wilsford cum Lake G51, about 1700m to the south:

If you remove the “linear trench” (where a later bronze age land boundary cuts across the mound) and add a ring of posts, you have a pretty good idea of how the new site might appear if excavated. The barrow was dug by Ernest Greenfield in 1958; it had been almost destroyed by ploughing [5]. Interestingly, the “quarry pits” were dug in at least two different stages, and the history of burial and mound building was quite complex.

A similar ditch arrangement has been excavated at barrow Shrewton 24 [6], where again smaller pits were later expanded into larger ones (seven); there were two Beaker burials in the centre of the ditch ring, one above the other [6].

Also of interest is a curious site very close by, within Fargo Plantation (it lay between barrows 52 and 54 on the map above). Here a complex grave was excavated by JFS Stone in the 1930s. There was no apparent evidence for a mound, and the burials were enclosed by two arcs of ditch that have been called “hengiform” [1]. The gaps to south-west and north-east are analogous to the gaps in the new site:

Why it matters

So an explanation for the new site as an unusual, but not unbelievable barrow seems very plausible.

If, on the other hand, the site really is a henge, it would fit less easily into the prehistoric landscape as we know it. To date, the idea that timber henges are on a ridge beside the river Avon to the east, and stone henges out of sight and alone to the west, has been supported by recent excavation [7]. A timber henge on the new site would thus challenge that, and also the notion that the immediate Stonehenge landscape was dedicated to ancestors memorialised in stone [8, 9].

In that context, it would make more sense for the ring of pits to be stoneholes rather than postholes, a suggestion made by Josh Pollard, who notes that it is in that area that as-yet-unexplained fragments of bluestone (the material of the smaller megaliths at Stonehenge, imported from Wales) have been found on the field surfaces. Yet Gaffney (quoted on the same Nat Geo website) prefers the post idea.

I wouldn’t have written all this if I thought the new find was trivial. We are realising that survey and excavation within the world heritage site are very significantly improving our abilities to frame explanations for Stonehenge – why it was built and what it meant. This is one reason why the survey by Birmingham University and the Ludwig Boltzmann Institute for Archaeological Prospection and Virtual Archaeology – set to continue for three years – is so important.

But it also means that if you propose new ideas about sites in the area, you are challenging a great complex of information and thought. Despite claims to the contrary, if we are to know if the new site was a henge (3000–2000BC) or a barrow (2000–1200BC), it needs to be partially excavated. More generally, we desperately need a proper programme of research and excavation that addresses one of the great unsolved mysteries of the Stonehenge landscape – the barrows.


1 Royal Commission on Historical Monuments (England) 1979. Stonehenge & its Environs. Edinburgh University Press

2 Thomas, J, Marshall, P, Parker Pearson, M, Pollard, J, Richards, C, Tilley, C & Welham, K 2009. The date of the Greater Stonehenge Cursus. Antiquity 83, 40–53

3 Gingell, C, 1988. Twelve Wiltshire round barrows. Excavations 1959 and 1961 by F de M and HL Vatcher. Wiltshire Archaeological & Natural History Magazine 82, 19–76

4 Thomas, N 2005. Snail Down, Wiltshire: The Bronze Age Barrow Cemetery & Related Earthworks. Wiltshire Archaeological & Natural History Society

5 Smith, I F 1991. Round barrows Wilsford cum Lake G51–G54: excavations by Ernest Greenfield in 1958. Wiltshire Archaeological & Natural History Magazine 84, 11–39

6 Green, C & Rollo-Smith, S 1984. The excavation of eighteen round barrows near Shrewton, Wiltshire. Proceedings of the Prehistoric Society 50, 255–318

7 Parker Pearson, M, Pollard, J, Thomas, J & Welham, K 2009. Newhenge. British Archaeology 110, 14–21

8 Parker Pearson, M & Ramilisonina 1988. Stonehenge for the ancestors: the stones pass on the message. Antiquity 72, 308–26

9 Pitts, M 2000. Hengeworld. Century

8 thoughts on “Analysing the new site near Stonehenge

  1. I’m surprised more people haven’t compared this discovery with ‘Seahenge’ – it something of a closer fit. Gibson, in his ‘Stonehenge and Timber Circles’ actually dismisses Seahenge as an unusual palisaded round barrow.

  2. Dear Mike
    A really good post, but is there a certain amount of wishful thinking by both you and Josh Pollard in not wanting this to be a wooden henge?

    Ned Pegler

    1. Of course I can’t speak for Josh Pollard, but my argument is simply that on current evidence it seems more likely that the site is a barrow than a henge, especially a “woodhenge”, which is normally taken to mean something larger and more complex than a simple ring of posts. Look at Coneybury, shown by excavation in 1980 to be a henge with a post ring and (we now think) possible stone pits, c3000-2500BC. It’s 1300m south-east of Stonehenge and 40-45m across (the new geophys site is about 1km to the north-west and 25m across, and both are visible from Stonehenge, undermining silly claims that the new site is the first major ceremonial monument discovered close to Stonehenge in the last 50 years). I think Coneybury is better compared to the large “houses” that Julian Thomas excavated inside Durrington Walls than Woodhenge: these sites have a ring of relatively slender posts around the edge seeming to define a border, while at Woodhenge the posts are massive and cover the whole site. Limited excavation would tell us a lot more about the new site, and it might turn out to be something that denied a simple distinction between “henge” and “barrow”.

  3. It occurs to me that there could be a parallel in the methods of barrow construction to what has been seen in how big cairns were constructed. What was seen with the cairns was that the initial cist and surrounding was usually very well constructed, but once covered over the stones tendedto start off fairly small, and slowly increase in size until the last few were quite substantial.

    What was going on wasn’t a carefully-orchestrated build process by skilled artisans, but human nature pure and simple. Initially the cist gets built with everyone in the process watching; there’s a great imperative to make everything look as neat as possible. Then when the main construction starts, people want to be seen to be working, but also mostly don’t care to shift heavy, hard-to-lift stones when smaller, more easily moved ones will do just as well; they’re actively building the cairn but they ain’t putting themselves out too much either. Finally, when the cairn is nearing completion, the last few stones are big because that’s all that’s left lying about nearby.

    So, what if similar construction techniques were going on with building barrows (and with henges too)? With a barrow, you’re looking to dig a grave, interr the body, then build a mound over it. What you want it a mound, and you want it to stay visible, but you don’t much care about how it goes up as long as it goes up. So, the easiest way is going to be to cut turf from around the barrow for roofing it off when you’ve finished to stop it washing away, then pile up material from trenches around the barrowto put onto it.

    You’re not going to work to a set, fixed plan because you’re never going to be working from exactly the same composition of ground each time. Sometimes you might hit a nice layer of clay, so you just go downwards. Other times you hit bedrock, so the trench needs to be bigger, wider and instead of a ring of pits around the mound it becomes a complete ring-ditch. Remember that at the time barrows were constructed there weren’t any metal digging implements so getting through established turf was difficult, whereas digging down into clay or gravels was much easier when using a digging-stick.

    This is also why a lot of henges (including the one around Stonehenge) had ditches composed of a ring of pits; the ditch wasn’t the important bit and it was only there to provide material for the mound around the henge. That in turn was there for strictly practical reasons. The religion that produced henges also looks to have created stone circles, and the two classes of site both seem to have had something to do with rising and setting of sun and moon. Observing these is easy up on a hill; there’s no shrubbery around you to obscure things, but down on the wooded lowlands you need an artificial horizon, also known as a henge.

  4. Nice post, Mike — and good to see the wild excesses of the press release writers exposed for what they are. Why every new (or old) discovery from the Stonehenge environs needs to be hyped up to the eyeballs, goodness knows. And shame on the media, too, for lapping up assorted bits of nonsense without any application of critical faculties.

    Quote: “In that context, it would make more sense for the ring of pits to be stoneholes rather than postholes, a suggestion made by Josh Pollard, who notes that it is in that area that as-yet-unexplained fragments of bluestone (the material of the smaller megaliths at Stonehenge, imported from Wales) have been found on the field surfaces.” OK — all those fragments of bluestones — of many types — littering the surfaces can be referred to as “bluestones” — but do you have to repeat the old mantra that they were “imported from Wales” when we all now know that they have come from all over the place and that they are in all probability glacial erratics? Can you cite for me a single bit of evidence that the stones were IMPORTED FROM WALES rather than picked up and used more or less where they were found?

    Old mantras don’t get any truer simply by being repeated ad infinitum.

  5. I guess my research and understanding is your mantra! I can think of no way to show direct evidence either that stones were “picked up and used where they were found” or that that they were “imported from Wales”, and accusing me of parroting a factoid is just silly (if it were true, so by definition it would also be true of your position). But the hypothesis of human transport has grounds for support in widely agreed models of geology/geomorphology and archaeology/human societies. It’s always seemed to me that objections to the human transport theory, which come mostly not from archaeologists but from geologists and earth scientists, are driven by misunderstandings about the nature of societies and human culture, notwithstanding that the arguments are presented as geological: there is a scientistic unwillingness to recognise that people can do things that have no simply described practical purpose.

  6. I was simply seeking to suggest that you might have used the phrase “possibly imported from Wales” or some such thing. It’s just unscientific to write or speak with such utter conviction, when as we all know there is no firm evidence (as yet) to settle this argument either way. I would be interested to know which “widely agreed models of geology / geomorphology” support the human transport theory. And I’m not very convinced, either, when somebody says, in effect, that the people of Neolithic Britain probably did transport the stones, simply because people sometimes did things in the past that modern man cannot as yet find a rational explanation for. So that’s your research and your understanding?!! And if I don’t agree with you, then I am guilty of a misunderstanding? Hmmmm……..

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s