Stonehenge a New Understanding (39 page)

Supervisor Chris Casswell stands with a scale behind tiny Stone 11 at Stonehenge. It is clearly too short to have supported a lintel and has not been dressed in a similar fashion to the other stones of the sarsen circle.

In strong contrast to the number of sarsen pieces found (6,500), only forty bluestone chippings were found by Colin in his five × five-meter trench. There were more bluestone chips in the avenue trench but only just over a hundred. Why are there so few fragments of bluestone outside the stone circle, when excavators digging
inside
Stonehenge have found many more bluestone chips than sarsen chips? It is most likely that the dressing of bluestones took place inside Stonehenge. That certainly makes sense if the bluestones were already there, in the Aubrey Holes, ready to be rearranged in the Q and R Holes. Not all the bluestones have been dressed: Of the forty-three survivors, only seventeen have been worked to create smooth surfaces.
¹⁶
Although moved around several times, presumably the bluestones have never left the premises since their arrival around 2950 BC.

The builders had to think very carefully about how to position the dressed sarsens. The lintels had to fit neatly on top of the uprights. The
difficulties of getting a perfect fit would have been compounded by the slight slope of the site. At Durrington Walls the ground was terraced flat before the construction of the Southern Circle, but at Stonehenge the architects chose to retain the natural slope even though they wanted the lintels on the stone uprights to be horizontal. To compensate for the slope, the uprights had to be shorter upslope to the southwest and taller toward the northeast.

The puzzle of how Neolithic people built such a complex structure has occupied many minds for centuries. Around 1640, Inigo Jones thought it was so precisely built that it must once have been a completely symmetrical monument with Classical proportions as set out by the Roman architect Vitruvius.
¹⁷
However, although Jones,’ stylized portrayal of Stonehenge’s ground plan is accurate to within 5 percent, his desire for it to be perfectly symmetrical led to his adding an extra, sixth trilithon that never actually existed. William Stukeley was the first to try to work out what unit of measurement had been used.
¹⁸
If it was the Roman foot of 0.96 feet (0.293 meters), then Stonehenge was Roman, as Jones had thought. Stukeley was convinced that the ancient Britons who had built it “knew nothing of Vitruvius” and, after taking 2,000 measurements, deduced that Stonehenge’s base unit of measurement was 20.8 inches (1.73 feet or 0.528 meters).

One of the reasons why William Flinders Petrie was determined to make an accurate plan of Stonehenge in 1872 was that he, too, hoped to work out the units of measurement used by its builders.
¹⁹
He found that the inner diameter of the sarsen circle amounted to 100 Roman feet (his calculation of the Roman foot was slightly longer than Stukeley’s at 0.973 feet, or 0.297 meters). He also deduced that the outer features—the ditch, bank, and Station Stones—were laid out using a completely different unit of 1.873 feet (0.571 meters).

In the 1960s, Alexander Thom stunned the world of archaeology by claiming that many megaliths, including Stonehenge, had been built by astronomer priests to measure astronomical events. He also claimed that they had all worked to a common unit of measurement, in use across the whole of Britain, to construct these monuments. This he called the Megalithic Yard of 2.722 feet (0.830 meters). In 1988, Thom and his son
published their finding that the average center-line diameter of Stonehenge’s sarsen circle is 37 Megalithic Yards.
²⁰
Although Thom’s work still has something of a following today among the wider public, archaeologists have never accepted wholeheartedly his concept of the Megalithic Yard (although it may be relevant for megalithic monuments in northern Scotland, where he first formulated the idea).

Most recently, archaeologist Tony Johnson proposed in 2008 that Stonehenge could have been planned without a unit of measurement, simply by laying out series of intersecting circles.
²¹
Using computer-aided design (CAD) methods, he showed how the Aubrey Holes could have been laid out in twelve moves. The positions of the seemingly unsymmetrical trilithons could have been established by marking the intersections of two sets of concentric circles, one centered between Stones 1 and 30 and the other between Stones 15 and 16. These are then combined with circles centered on Stones 1, 11, 20, and 30 of the sarsen circle to arrive at the positions of the trilithons. Johnson is able to show that the inside edges of the ten trilithon uprights (and the external edges of the great trilithon’s two uprights) can be plotted at chosen intersections of the six circles. Johnson calls this intersection of the two concentric circles a “diffraction grating.”

A few months later, John Hill, a research student at Liverpool University, demonstrated that Stonehenge could be laid out much more simply using lengths of rope, the sun’s shadow, and basic counting on fingers.
²²
To show how easy it could be, he enlisted the help of school-children from Northcote Primary School in Walton to lay out Stonehenge’s ground plan at the university sports ground.

I had always kept my distance from the complex issue of Stonehenge’s geometry. It was quite clear that there were all kinds of different ways of solving the same problem, and each originator was convinced that his solution was the correct one. How could anyone decide which was the actual solution used by Stonehenge’s designers and which were attractive possibilities but no more than that? For John Hill, the principle of Occam’s razor applied: They would have been most likely to choose the simplest solution. For Tony Johnson, a decider was the fact that his method could reveal that the great trilithon upright, restored under
Gowland’s direction in 1901, has been reerected in the wrong place, 60 centimeters from its original position (though this is exaggerated; for the midpoint of the stone’s face, the distance by which it is mis-set is actually about 30 centimeters).

Then my colleague Andrew Chamberlain took a look at the problem.
²³
As a specialist in human osteology and paleodemography, Andrew works with mathematical problems all the time. We had worked together some years before on Saros cycles (lunar-eclipse cycles) and their correlation with the timing of construction of Iron Age timber causeways; Andrew worked out how to compare dendrochronological dates for tree-felling with Saros cycles of full lunar eclipses. Here was a new challenge. Thom’s Megalithic Yard value of 37 for the sarsen circle’s diameter, for example, did not seem particularly convincing to Andrew. Had Thom come up with not one but a series of whole number measurements, with regular intervals in measurements—such as 40, 50, 60—for different parts of the Stonehenge plan, then the likelihood of the Megalithic Yard as a base unit would be more persuasive.

Instead of starting where everyone else had, with Stonehenge itself, Andrew looked at the ground plans of all the timber monuments at Durrington Walls, Woodhenge, and the other Wessex henges. These could provide an independent test of any concept of a Neolithic base unit of measurement at Stonehenge.

Since we knew that the wooden circles were laid out around the same time as Stonehenge, it should be straightforward to identify shared regularities in design. Andrew started with the Southern Circle, for which we now had a complete circular plan, at least for the larger posts. He employed one key principle from the beginning. When laying out the positions of planned features, people are likely to choose their centers and not their edges. To dig a ditch or raise a bank, it’s simpler to mark out the midline rather than the edges. The issue is even more important for uprights. A simple parallel is the erection of fenceposts in the garden—if you’re going to nail a trellis to the posts, you must try to get the distances right. You decide where you want the middle of one post to fall, and then measure off to where the middle of the next post will be; if you measure between the edges of the posts it will go wrong, and it’s no help at all to measure the distance between the edges of your postholes.

We know from excavations of postholes and stoneholes that it’s not always possible to control the erection of an upright so precisely that it fits exactly where it was intended to go. People didn’t measure to the edge of an imagined corner of a future post or stone that was going to be put up; they measured to roughly where the center of the upright would be and could thus mark the center of the hole to be dug for it.

Measuring the six concentric circles of posts of the Southern Circle’s second phase from the centers of the postholes, Andrew found that the diameters of the outer four circles can be matched with regular and incremental multiples of an ancient English unit of measurement called the “long foot.”

The distances came out as 70 to 90 to 110 to 120 “long feet.” The figures are not precise, with deviations varying from six centimeters to 50 centimeters, but Andrew thinks that this can be attributable either to imprecision in construction or, of course, to imprecision in archaeological recording during excavation. Nevertheless, he had identified a possible unit of measurement that could be tried out on other monuments.

The “long foot” (1.056 feet or 0.32187 meters) was one of the base units used in Medieval England. Andrew had been impressed by some of the research of John Michell and John Neal into ancient metrology, and wondered, as they did, whether some of our more recent units of measurement might have great antiquity.
²⁴
He was also aware from their work that ancient measurement systems often made use of multiple values of their base units.

The long foot was so called because it was longer than the short foot (0.96 feet, or 0.2926 meters)—and the two units were sometimes used together. For example, 11 short feet make 10 long feet. The ratio of 11:10 is useful when measuring out circles. A diameter of 7 short feet creates a circumference of 22 short feet or 20 long feet (22/7 is a close approximation of the value of π, or pi). In the old British system of “statute feet,” the lengths known as chains, furlongs, and miles are based on multiples of 11 statute feet, again reflecting this 11:10 ratio. Andrew and I are old enough to remember running the 110-yard hurdle race and using actual chains for surveying.

Turning to Stonehenge, Andrew examined the dimensions of its first phase. In addition to the Aubrey Holes, henge bank, and henge ditch,
he looked at the dimensions of a slight bank outside the ditch, called the counterscarp bank. This bank is earlier than the Stonehenge Avenue’s ditches, which cut through it, and probably belongs to the initial period of construction. The diameters of the four circular features are 270, 300, 330, and 360 long feet, with errors of between 12 centimeters and 45 centimeters. Thus, they could have been laid out with a rope running from a center point, marked with the radius lengths of 135, 150, 165, and 180 long feet.

The fact that these diameters are not only whole numbers but also built on regular intervals of 15 long feet is encouraging. Was 15 long feet a standard unit? Given that there are fifty-six Aubrey Holes, it is particularly intriguing that the circumference of the Aubrey Hole circle is 847.8 long feet, or 56.52 units when divided by 15. The centers of the Aubrey Holes lie on average just over 15 long feet apart.

Another intriguing measurement employed in Stonehenge’s first stage of construction is provided by the distance from the northernmost stonehole (Stonehole 97, found by Mike Pitts in 1980 close to the Heel Stone) to the center of Stonehenge. This distance measures 250 long feet and provides an alignment on the midsummer solstice sunrise from the center of the monument. Stonehole 97 might also have been used as a sighting for the rising moon’s northerly limit, aligned with Stoneholes B and C. Thus it may have combined both solar and lunar sightings in one.

The distance of 250 long feet was also employed in laying out the four Station Stones on northwest-southeast, northeast-southwest alignments. These stones also have astronomical associations: to the southeast toward the rising moon’s southerly limit (full in summer) and, to the northwest, toward the setting moon’s northerly limit (full in winter). Although they are essentially undated, the Station Stones were set within the circumference of the circle of Aubrey Holes and are probably later than them.

I asked Andrew to find out if the system of long feet works for the sarsen circle and trilithons, built five hundred years after Stonehenge’s first stage. The answer was no, not particularly well. The average diameter across the sarsen circle, midline to midline, is 95 long feet, making a radius of 47.5 long feet. We then realized that the circumference of the circle of sarsens is 300 long feet. With thirty uprights, the average spacing
between them is 10 long feet. Perhaps this distance of 10 long feet was more important than the radius, because the circle’s diameter was ultimately determined by the lengths of the lintels: If the lintels could have been longer, the circle could have been wider, but longer lintels would have been less stable because of their necessary curvature. It would have been most sensible for the builders to work out the optimum lengths of the lintels, decide where the uprights had to go and then determine the circumference and radius.

Perhaps the builders had aimed for a standard length of 10 long feet for each lintel. Looking more closely at the measurements for the uprights and surviving lintels, the distance between each pair of lintel mortises is seven long feet. Several of the lintels are four long feet by two long feet in width and height, the same proportions as the “two by four” that carpenters and joiners still buy from lumber yards.

Perceiving any regular arrangement of the stones inside the sarsen circle is more problematic. The various diameters of the bluestone circle, the sarsen trilithons and the bluestone oval don’t conform to any standard lengths in long feet except that the midline of the great trilithon is about 25 long feet from the center of the sarsen circle, and the length of its lintel is 15 long feet.