About sixty people turned up in the field outside Stonehenge, wearing what I think was religious costume—a variety of robes and antlers—banging drums and shouting, to the alarm of both the diggers and the visiting public. Two small girls burst into tears and had to be calmed by their parents. Tolerance was apparently a one-sided affair: Julian Richards was publicly condemned with a Druidical curse. Finally, the proceedings calmed down and a Pagan blessing was provided next to the parking lot by some of the more moderate celebrants.
11
THE AUBREY HOLES
__________
By the early 1930s, mass tourism had reached Stonehenge. There was a pressing need for visitors to have somewhere to park their cars. In January 1935, in consequence, Robert Newall (Lieutenant Colonel Hawley’s assistant) and William Young (who had worked at Woodhenge for archaeologist Maud Cunnington in the 1920s) were digging in advance of the construction of this first parking lot just to the northwest of Stonehenge. They took the opportunity to re-open Aubrey Hole 7 and bury in it all the cremated human bones that had been found by Hawley at Stonehenge. They kept a written record of their work but didn’t record why they chose Aubrey Hole 7—though its large size and proximity to the road might have been the reason. Young recorded in his diary that they dumped four sandbags of bone in the re-opened hole; so that future archaeologists would know what these bones were—and in case Young’s diary did not survive—they also left an inscribed lead plaque in the pit on top of them.
Working out what the Aubrey Holes were originally used for is tricky. Before working at Stonehenge, Atkinson had dug a series of pit circles on the river terrace gravels at Dorchester-on-Thames in Oxfordshire, so he had some experience of this type of Neolithic feature.1 After digging two Aubrey Holes in 1950, he decided that the Aubrey Holes “were never intended to hold any kind of upright, either the bluestones . . . or wooden posts.”2 They were just a circle of pits. Later archaeologists tended not to agree with him. By the time our project started, the usual interpretation given in guidebooks and elsewhere was that the Aubrey Holes had once held wooden posts. Josh Pollard, for example, could easily interpret the old excavation records and the beautiful sections drawn for Atkinson by Stuart Piggott to see that the compacted chalk rubble found in all the holes was best explained as packing around a central upright.
A section of Aubrey Hole 32, drawn by Stuart Piggott, showing the filled-in void where a bluestone once stood (5) and the chalk packing material for the stone (4, 6, 7) from which bones of a cremation burial were recovered (4). To the left (at 3), the side of the pit has been crushed and the packing layer displaced where the stone was removed.
Going back through Hawley’s notes and reports of his excavations in the 1920s, I discovered that he had made careful observations about how the Aubrey Holes had been used. He noted that the chalk bottoms of at least two of them had been compacted and crushed. Many more, Hawley wrote, had their edges “shorn away, or crushed down, on the side toward the standing stones of Stonehenge, this being apparently due to the insertion or withdrawal of a stone, probably the latter.” In 1921, he concluded that “there can be little doubt that they [the Aubrey Holes] once held small upright stones.”3
For some reason, Hawley didn’t have the courage of his convictions and later changed his mind.4 In his diary he wrote that Maud Cunnington and her husband had convinced him that the Aubrey Holes were postholes. Their excavation at Woodhenge between 1926 and 1928 had uncovered evidence that part of this Neolithic monument consisted of massive wooden posts, not just standing stones. These Woodhenge postholes were about the same diameter as the Aubrey Holes so superficially looked very similar.
In 2007, I began to wonder if Hawley had been right the first time. In all the discussion and mulling over what the Aubrey Holes could have once held—timber, stone, or nothing at all—no one had thought to make a very careful comparison of the dimensions of these intriguing pits with those of known postholes or stoneholes. There are plenty of excavated examples to make such a comparison possible—from Stonehenge itself one can examine the records of excavated sarsen holes, the bluestone Q and R Holes, and other bluestone-holding holes. From our own experience at Woodhenge and Durrington Walls, and its avenue, we knew that stoneholes are much shallower than postholes of equivalent diameter. The stoneholes we had dug also had a thin layer of crushed chalk at their bases, which is something not found in even the largest of the postholes.
The results of a metrical comparison between Neolithic postholes and Neolithic stoneholes are very clear.5 The Aubrey Holes are too narrow to be pits, and too shallow to be postholes. They are, on average, half a meter shallower than postholes of equivalent diameter. They are also narrower than either sarsen holes or, for example, the Neolithic pits found at Dorchester-on-Thames by Atkinson. The Aubrey Holes are, in fact, identical in width, depth, and shape to the bluestoneholes located elsewhere in Stonehenge.
One might wonder whether the Aubrey Holes held small, short posts in shallow sockets, thereby contrasting with the normally deep postholes at other Neolithic sites, such as Woodhenge and the Southern Circle. But that cannot be the case because such short posts would not have been heavy enough to cause the crushing and compaction that Hawley noted within the Aubrey Holes.
Returning to Piggott’s section drawings of the Aubrey Holes, I could see how the crushed chalk rubble that fills the holes had been deposited and then displaced. It seems that the chalk fill had initially been packed against the base of a stone in each hole and was then displaced on one side when the stone was pulled out. Looking at the drawings, it is evident that in Aubrey Hole 32, for example, not just the rubble but also the side of the pit itself was crushed by withdrawing a stone, just as Hawley had observed on other examples.6 From the records and drawings alone, a very strong case can be made for the Aubrey Holes having held small upright stones—presumably the bluestones—and for these to have formed a stone circle right at the beginning of Stonehenge’s sequence.
Removing a bluestone from an Aubrey Hole, showing how the shape of a pit is altered when a stone is removed.
I am puzzled as to why Atkinson said that the Aubrey Holes had held no uprights of either wood or stone. If a pit is dug and then filled in, an archaeologist can see the clear edges where the pit was cut into the soil. If a timber post is left to rot in situ, you see a parallel-sided “pipe” within the pit, where the timber has decayed. If an upright is pulled out of a pit, however, it changes the shape of the pit’s sides, because of the movement of soil caused by levering it out. These differences can be hard to spot for someone with little experience of digging postholes and stoneholes.
To work out what happened at Stonehenge, we needed to know more about the Aubrey Holes, to find out both what they were for and when they were dug. Everyone agrees that the construction sequence at Stonehenge begins with the ditch and bank. These were constructed at some point within 3000–2920 BC, on the basis of Bayesian-modeled radiocarbon dates. Some archaeologists reckoned that the Aubrey Holes belonged to this earliest phase, because they are set in a ring just inside the circles of the bank and ditch; others thought they must be part of a later phase.
To date the Aubrey Holes, we turned to Salisbury and South Wiltshire Museum again, which has responsibility for many of the prehistoric finds from Wessex. Given the amount of archaeological investigation that has been carried out at Stonehenge, and the huge amount of effort that has gone into unraveling the construction sequence, it is truly surprising that there is material still waiting to be dated. The curators in Salisbury take care of three cremation burials excavated by Atkinson and gave us permission to take samples for radiocarbon dating.
One of these cremation burials was found in Aubrey Hole 32, excavated during Atkinson’s very first Stonehenge season in 1950. Piggott’s impeccably drawn section shows that the bone fragments were found scattered within the layer of compacted chalk rubble packed into the bottom sides of the hole; these bones were therefore deposited i
n the initial fill of the hole, and not inserted later.
We sent a small sample of cremated bone off to the lab. Material for radiocarbon dating (C14 dating) has to go to one of a handful of specialist labs with accelerator mass spectrometers. The two labs that we use are in Glasgow and Oxford. Radiocarbon dating is a slow process that takes many months. On completion, the results are sent to the excavator by letter—until it arrives you don’t know the date of your sample. When our result finally came back, we learned that this Aubrey Hole cremation burial dates to 3030–2880 cal BC, the same period as the antler picks found at the bottom of the Stonehenge ditch, so we were confident that the Aubrey Holes did belong to Stonehenge’s initial phase of construction.7
We also sent for radiocarbon dating two other cremations found by Atkinson, not from Aubrey Holes but from different layers within the ditch. These dated to around 2900 BC and around 2500–2300 BC. As well as carefully buried cremated bones, loose human bones were also being scattered at Stonehenge: Two fragments of unburned human skulls from the ditch date to 2800–2600 BC.
We now had some new dates to add to the existing chronology for Stonehenge’s construction and use. The 1995 book Stonehenge in Its Landscape by Ros Cleal and her team had finally provided the full and definitive account that Atkinson, Piggott, and Stone had envisaged, though only Piggott lived long enough to see it. Cleal’s team had examined all the existing radiocarbon dates, working out whether they were from secure contexts or not, and English Heritage had paid for a new suite of dates from antler picks and animal bones. Here was a solid framework for Stonehenge’s chronology, but there were still some problems to be ironed out. How did our new dates for the Aubrey Hole and for the other human bones fit into the chronology?
The new dates affected the chronology in two ways. Firstly, the orthodox view, that Stonehenge was used as a cemetery for just a short period of time around 2600 BC, was wrong. The new radiocarbon dates showed that Stonehenge had started as a place of burial, since the Aubrey Hole cremation dates to the moment of Stonehenge’s construction and initial use. Secondly, the new date showed that the Aubrey Holes were definitely some of the first constructions at Stonehenge. If we were right that they once held bluestones, this had significant implications for the sequence of construction and the dates of the different stages of the monument. If Stonehenge had actually started as a bluestone circle shortly after 3000 BC, these stones must have arrived more than five hundred years earlier than anyone had previously reckoned.
12
DIGGING AT STONEHENGE
__________
The dig to retrieve the 1935 deposit of cremated bones from Aubrey Hole 7 began, and all except a small handful of Druids went away. Now we could start trying to find out who those people buried at Stonehenge were. Hawley had stripped the entire area around Aubrey Hole 7 and its neighbors, and Newall and Young had dug it out for a second time, so we didn’t expect to find very much in the hole other than the reburied bones.
This Aubrey Hole is one of the largest and it was one of the most prolific in finds. Hawley recovered fifty-five sarsen hammerstones from this one hole, together with more than eighty bluestone and sarsen chips, and an ax-shaped bluestone. The bones of a disturbed cremation (of what he thought was a young adult) were scattered from top to bottom through the layers filling the southeast side of the pit. On the bottom of the pit, among chalk rubble, he found a small deposit of wood ash. Apart from three shards of Roman and Bronze Age pottery, and a single worked flint, Hawley did not bother to collect the smaller finds. In 1935 Newall and Young found an unfinished oblique arrowhead in his backfill. This was just the tip of the iceberg in terms of what he either missed or deliberately left behind: We found more than thirty worked flints, more chippings of sarsen and bluestone, and more shards of Roman and prehistoric pottery.
In the thirty-two Aubrey Holes that he excavated, Hawley found thirty cremations, more than four hundred hammerstones, and more than a thousand chips of bluestone and sarsen. Just like in the Q and R Holes, there was not a single antler pick. He noticed that most of the cremation burials had been disturbed by the removal of what he reckoned were standing stones, but he also recorded that one or two cremations had been inserted after stone removal, because they remained intact and had not been crushed or otherwise disturbed. Cremation burials had also been placed around the edges of some of the holes.
Why all the hammerstones, stone chippings, and cremations? Atkinson believed the Aubrey Holes were neither structural nor sepulchral in their primary purpose. He thought that the cremations were introduced to the holes during their refilling, in the same way that ritual libations were made in pits as entrances to the underworld in Classical Greece. Yet Hawley was very clear that most of the burials had been included in the initial filling of the pits, and then disturbed later.
Hammerstones are fist-sized cobbles that have been used for pounding the surfaces of the sarsens. They are often used subsequently as packing for standing stones—Gowland found twenty-two of them slipped into the south side of the great trilithon upright to help fix it in place. Perhaps they were used for a similar purpose in the Aubrey Holes, or perhaps they were introduced after the stones were pulled out, to fill in the consequent depressions.
At the bottom of the loose soil tipped back in by Newall and Young in January 1935, we found the lead plaque that they had left there for us, the archaeologists of the future. It reads:
MOST OF THESE BONES WERE DUG UP IN THE YEARS 1921 1922 1923 FROM THOSE HOLES JUST INSIDE THE BANK OF THIS MONUMENT AND CALLED AUBREY HOLES BY THE SOCIETY OF ANTIQUARIES OF LONDON IN CONNECTION WITH HIS MAJESTYS OFFICE OF WORKS SOME BONES WERE FOUND IN THE DITCH THE HOLES WERE CALLED AFTER AUBREY BECAUSE HE SUGGESTED THEIR EXISTANCE IN THE YEAR 1666 REBURIED 1935
We were all a bit disappointed with this truly prosaic inscription—we’d secretly wished for something more archaeologically informative, or even more poetic. And it has a spelling mistake.
Once we’d carefully lifted the plaque, we could see a mass of cremated bones beneath. Scraping away the last of the loose soil on top of them, it was obvious that they had been dumped in a single heap. Young had written in his diary of their having been laid in four separate sandbags but there was no sign of any such division. It looked more likely that the bones had been poured into the hole, perhaps simply to keep the sandbags for re-use.
All of us had hoped that these archaeologists of the 1920s and 1930s had understood the value of context—and might therefore have appreciated the need to keep the bones from each burial separate. In the months before starting work, we’d been hoping that the bones from each burial had been kept separate in tins, paper bags, cloth wrappings—anything that could allow us to distinguish between the separate burials. We’d even fantasized that Newall and Young might have written indelible labels detailing from which context—or at least which Aubrey Hole—each burial had originally come.
Jacqui McKinley (top) and Julian Richards (right) excavating the undifferentiated mass of prehistoric cremated bones deposited in Aubrey Hole 7 in 1935.
The lead plaque left on top of the cremated bones in Aubrey Hole 7 by Robert Newall and William Young.
Instead, everything was mixed together. It was possible that the whole lot had not been utterly shaken up before being put in the ground, and that the bones from one burial might be packed next to the bones from another. If we took them out of the ground using a very precise three-dimensional grid of five centimeter blocks, it was just possible that specialist laboratory analysis could determine which bones belonged to which burials. The gloom induced by this archaeological carelessness was lifted by the optimism of Jacqui McKinley. She is the osteoarchaeologist (human bone specialist) for Wessex Archaeology, and has probably analyzed more cremation burials than anyone else in the world.1 Not only were there far more bones than we had anticipated, but she was very pleased to see that they were generally large pieces in good condition.
When a body is cremated on a pyre
, the fat and flesh burn away over a period of four to eight hours (depending on fuel and the tending of the fire) so that only the bones remain. These become calcined, turning blue and white, shrinking, warping, cracking, and splintering. Once the fire cools, they can be gathered off the pyre surface, which has become a mass of charcoal, ash, and calcined bones. Colin Richards has seen the whole process on an open-air pyre at a cremation ceremony on the Indonesian island of Bali. Many years ago, when I was doing some research on modern British funerary practices, I spent some time with the staff at a British crematorium, peering through the furnace spy-hole to watch the flaming, bubbling fat and flesh burning off to leave a glowing red skeleton. The crematorium operator then takes a long iron tool and pokes the bones so that they fall in chunks down a chute, to be collected in a metal container. After they’ve cooled, the burned bone fragments are put into a “cremulator” (a machine that looks rather like a clothes-drier) to be reduced to tiny dust-sized particles.
Of course, the cremulator is a twentieth-century invention and prehistoric cremated bones never received this kind of final treatment. That said, archaeologists sometimes find that the bones from a cremation have broken into very small pieces and have been heavily weathered and eroded. Jacqui has also found that the average archaeological cremation weighs less than two pounds, whereas the weight of an adult’s burned bones should be around five pounds. Most prehistoric cremation burials have lost some of the bones along the way—partly, she thinks, because retrieval from the pyre would not have been particularly efficient, and partly because bones could well have been divided up so that not all were buried. For example, the three cremation burials excavated by Atkinson from Stonehenge vary in weight from 77 grams to 150 grams to 1,546 grams, with only the last one likely to comprise most of the recoverable bone.
Stonehenge—A New Understanding: Solving the Mysteries of the Greatest Stone Age Monument Page 19
