Stonehenge a New Understanding (47 page)

Richard Atkinson was also well aware of the power of earthworms to move small objects, not only from reading Darwin and others on the subject, but also from his own experiences of digging at Stonehenge. In a famous paper on “Worms and weathering,” published in 1957, he declared that “the excavator who ignores the capacity of worms to displace small objects downward, either through ignorance or wilfully, does so at his peril.”
⁵

It was clear that the criteria that Mike Allen had set for establishing reliable dates from the small pieces of charcoal in Geoff and Tim’s trench could not be met (see Chapter 11 if you want to remind yourself of the detailed rules). Unfortunately, explaining radiocarbon dating is difficult, and the BBC’s reporters just hadn’t understood what was (and what was not) a secure and reliable date. Certainly each piece of charcoal had a date—we knew how old every piece was—but this didn’t tell us anything about the layers of soil in which they’d ended up. Without knowing about the dates of the soil layers, we can’t say anything about the construction sequence at Stonehenge. At the very best, these dates were simply evidence that charcoal was dropped on the ground at that time, or formed in a small fire. Nobody really cared that someone had burned a bit of wood at Stonehenge in 2300 BC—we were all trying to find out when they built the place! So the BBC was wrong and Stonehenge didn’t date to 2300 BC after all.

The one really interesting date from Tim and Geoff’s samples is a piece of charcoal from 7330–7070 BC, the Mesolithic period. This coincides nicely with the date of the later of the two pine postholes in the Stonehenge parking lot. It may mean that somewhere very near Stonehenge—even closer to it than the Mesolithic campsite we have identified in the field to the west—there lie further as-yet-undiscovered traces of Early Mesolithic activity. It’s certainly a possibility but caution is needed, because single small pieces of charcoal can be moved about on people’s shoes or in clods of transported earth, or even by being blown from place to place.

A year later, after all the Stonehenge hype had died down, Geoff, Tim and I together worked out a new chronology that accounted for all the Stonehenge radiocarbon dates ever obtained. The previous scheme had ignored one of the dates because it didn’t seem to fit, but now we had a chronology that worked.

The first stage of Stonehenge (in the period 3000–2920 BC) was undoubtedly the construction of its ditch and bank, and the digging of the Aubrey Holes, which I believe held a circle of bluestones. Then, apart from various cremation burials, there was no major building work until around 2500 BC. At this point, Geoff and Tim had a radical proposal. A tremendous result of their excavation in the interior of the stone circle is that they’re now certain that Atkinson was wrong about the Q and R Holes: He had said they were earlier than the sarsen trilithons and circle; everybody believed this interpretation and it was all a mistake.

This seemed, at first, to be impossible—there is even a photograph taken by Atkinson that we all thought showed the hole for Stone 3 of the sarsen circle definitely cutting Q Hole 4.
⁶
Tim and Geoff explained that the most important new information to come out of their excavation is that many of these stoneholes had been cut into, dug out and refilled again millennia after they had been originally dug by the Neolithic builders, with this later interference taking place mostly in the Roman, Medieval, and early modern periods. As a result, none of the stratigraphic observations from Atkinson’s excavations—in which he said which stonehole cut which other stonehole—could be trusted.

I cast my mind back to the contradictory section that Tim had showed me when I visited his team in their first week of digging at Stonehenge.
Their trench had uncovered a Q Hole that actually cut the hole for a fallen stone of the sarsen circle, the complete opposite of what Atkinson thought he saw. It’s clear from Atkinson’s photograph that Sarsen Hole 3 cut Q Hole 4—but what no one had asked themselves was whether this cutting of one hole by another could be dated. It was just assumed to be prehistoric, but we actually don’t know when it happened. That particular sarsen’s hole is uncharacteristically large, so it’s likely that this enlargement was made much later: It’s most likely not a Neolithic intersection between the two holes at all. Furthermore, Tim and Geoff’s findings show that the Q Holes have been placed at the wrong point in the construction sequence.

Geoff explained that one of the problems with studying Stonehenge is that it can be so difficult to put aside our taken-for-granted assumptions. We cling to what we think are certainties and it can be difficult to recognize when a mistake has been made earlier, back down the line, because it has taken on the status of an incontrovertible fact. Atkinson’s stratigraphic observation of the relationship between bluestone arc and sarsen circle was precisely one of those apparently unchallengeable facts, until new information revealed that we could no longer assume that this stratigraphy had been formed in prehistory. Tim and Geoff found out that the interior of Stonehenge had been hacked about very badly before modern times.

Geoff and Tim suggested that the Q and R Holes belong in the main phase of construction, the same phase as the sarsen trilithons and circle. Both spatially and architecturally, they fit within this stage dating to 2620–2480 BC. It then makes sense that the bluestones in the Q and R Holes were later rearranged into the bluestone circle of Stage 4.

19

The new work at Stonehenge—in terms of both excavations and the examination of old records—has produced a sequence of construction that takes into account, for the first time ever, all the spatial relationships between the different stoneholes, and all the radiocarbon dates that have been acquired over the years. We don’t all agree on every detail, but these are the main stages that we can identify.

The first features to be constructed were the
outer bank, ditch, inner bank, and Aubrey Holes with bluestones
in them.
Cremation burials
accompanied these early features. Arrangements of
posts
and perhaps three or more
standing stones
were set up within the central area of the circular enclosure; some of these post settings were aligned roughly on the southern major moonrise to the southeast. To the south there was an entrance through the banks and across the ditch. A post-lined passageway led from this south entrance to the center. There was a wider entrance to the northeast. In this northeast entrance and just beyond there were two settings of posts aligned approximately on the northern major moonrise. On the same axis, beyond the entrance and running southwest-northeast, was a line of standing stones (surviving as Stoneholes B, C, and 97). The most distant of these (97) provided a sightline on the midsummer solstice sunrise from the center of the circle. Together with stones B and C, it provided a second approximate alignment toward the northern major moonrise.

It seems most likely that the five
trilithons
were the next element to be erected. For engineering reasons, the trilithons must have gone up before the sarsen circle around them. The building of the trilithons was followed by a rearrangement of the bluestones into the double arc of
Q and R Holes
and the construction of the
sarsen circle
.

Beyond the northeast entrance, the erection of the
Heel Stone
may belong to this stage (moved from its former position in Stonehole 97), as may the erection of the
Slaughter Stone and its two partners
within that entrance.
Two D-shaped buildings
were constructed (beneath the North and South Barrows) against the inner bank, and two of the four
Station Stones
were erected within them. English Heritage’s survey team has suggested that the northern D-shaped structure was built before Stage 1 because the bank and ditch appear to kink around it, but the Aubrey Hole ring shows no such deviation. Only future excavation will reveal for certain which construction stage this D-shaped building belongs to.

While the axis of the trilithons was northeast-southwest, on midsummer sunrise/midwinter sunset, the rectangular plan of the Station Stones provided approximate alignments both on this same axis and on another, on its southeast-northwest axis, to the southern major moonrise as well as the northern major moonset.

The
avenue
, running from the northeast entrance along the natural, solstice-aligned periglacial ridges, was built by digging out two parallel ditches with inner and outer banks, in 2500–2270 BC. Stonehenge’s circular
ditch was cleaned out
in 2560–2140 BC but
cremation deposits
continued to be buried in the ditch even after it subsequently filled up. In the center of Stonehenge, there may have been a circular arrangement of
bluestones
inside the trilithons, indicated by an arc of four stoneholes on the west side of the interior.

Then a
large pit
was dug into the north side of the great trilithon, cutting this bluestone arc as well as the outer circle of Q and R Holes. This is the pit that Atkinson mistakenly identified as a ramp, and it can be dated to the period 2470–2210 BC. The Slaughter Stone’s two companions (
Stoneholes D and E
) were probably taken down to widen the
northeast entrance. The
avenue ditch filled up and was cleaned out
at some time in 2290–2120 BC.
Mounds
were raised to form the North and South Barrows, on top of the D-shaped buildings.

The bluestones in the Q and R Holes were rearranged into an outer
bluestone circle
(2270–2020 BC) and the central arc/circle of bluestones was remodeled into a
bluestone oval
(2210–1930 BC). Atkinson thought that the bluestone oval was later modified by the removal of four of its northeastern monoliths, leaving a horseshoe-shaped arrangement of stones. Geoff and Tim now know that Atkinson was unaware of the full extent of Roman and later destruction and removal wreaked on the monument. He made a fair assumption that the removal of four bluestones from the oval to create a horseshoe was an event planned and carried out in prehistory but, in light of the 2008 excavation, it’s more likely that this happened millennia later. Out goes the bluestone horseshoe—there is no archaeological justification for regarding it as a separate prehistoric entity: It is more probably simply the degraded remains of the bluestone oval.

Two rings of rectangular holes
were dug outside the sarsen circle. These pits are known as the Y and Z Holes, and they appear to be arranged in a double circle. As well as antler picks, some of these holes contained other antlers that were antiques—perhaps centuries old—when they were put in the holes. Atkinson suggested that the Y and Z Holes were dug to hold some of the bluestones but were never used, and no one has ever come up with a better idea. The holes were left open and filled in gradually with fine silt blown in from surrounding fields. Two circular, low ridges outside each ring have been tentatively identified as being the remains of prehistoric hedges—“Stonehedge”—but they may just be the spread remnants of the heaps of chalk produced when the pits were dug, and then just left lying outside the holes.
¹

This period of 1680–1520 BC, in the middle of the Bronze Age, was a turning point in prehistory. No more major monuments were constructed in Britain, and it seems that collective labor switched from
tomb-building to the laying-out of field boundaries all over the country. Laboring for the ancestors gave way to laboring for the living. The abandonment of the Y and Z Holes and their filling-up with soil blown in from newly opened, nearby fields encapsulates this transition very neatly. This was the last time that any attempt was made to modify Stonehenge’s form other than by robbing its stones. Thereafter, it was a relic of a bygone age, surrounded by a landscape of fields laid out around the cemeteries of Early Bronze Age round barrows.

Pottery shards found in excavations at Stonehenge span the period from the Middle Bronze Age to the modern era. There is pottery of the Middle and Late Bronze Age, Iron Age, Roman, Saxon, Medieval and early modern periods from various parts of the site. Certain moments of activity stand out. The many shards of pottery and occasional coin from the third to fourth centuries AD indicate a marked degree of interest in the site during the Late Roman period, shortly before the departure of the Romans in AD 410. After that date, there is little evidence for any centralized authority in Wessex during the Dark Ages until the rise of the West Saxon kingdom in the early seventh century.
²
A piece of human skull found at Stonehenge dates to AD 340–510, on the cusp of the Dark Ages.

Geoffrey of Monmouth’s twelfth-century “history” of Britain says that the British king Aurelius Ambrosius (elsewhere referred to as Ambrosius Aurelianus) and his brother Uther Pendragon (the father of King Arthur) were buried at Stonehenge, at a date that can be estimated around AD 500. Ambrosius Aurelianus is named as a Romano-British war leader of the fifth century by the Dark Age historian Gildas.
³
Gildas was a cleric writing in the early sixth century whose
De Excidio et Conquestu Britanniae
(
On the Ruin and Conquest of Britain
) is the most reliable of our few written sources from the Dark Ages; despite its shortcomings—it was written as a sermon and not a history—it is certainly more reliable than what was written by Geoffrey of Monmouth, who seems to have fabricated other aspects of Ambrosius’s life, including quite probably his place of burial. The imaginative reader could fantasize that this fragment of skull is all that is left of Ambrosius Aurelianus but this is very unlikely.