Provenance of the Stonehenge Altar Stone: a puzzling development

 Curiously, two weeks after my previous post about Stonehenge, a wider geochemical study of the Devonian sandstones and a number of Neolithic megaliths in Orkney seems to have ruled out the Stonehenge Altar Stone having been transported from there (Bevins, R.E. et al. 2024. Was the Stonehenge Altar Stone from Orkney? Investigating the mineralogy and geochemistry of Orcadian Old Red sandstones and Neolithic circle monumentsJournal of Archaeological Science: Reports, v. 58, article 104738;   DOI: 10.1016/j.jasrep.2024.104738). Since two of the authors of Clarke et al. (2024) were involved in the newly published study, it is puzzling at first sight why no mention was made in that paper of the newer results. The fact that the topic is, arguably, the most famous prehistoric site in the world may have generated a visceral need for getting an academic scoop, only for it to be dampened a fortnight later. In other words, was there too much of a rush?

The manuscript for Clarke et al. (2024) was received by Nature in December 2023 and accepted for publication on 3 June 2024; a six-month turnaround and plenty of time for peer review. On the other hand, Bevins et al. (2024) was received by the Journal of Archaeological Science on 23 July 2024, accepted a month later and then hit the website a week after that: near light speed in academic publishing. And it does not refer to the earlier paper at all, despite two of its authors’ having contributed to it. Clarke et al. (2024) was ‘in press’ before Bevins et al. (2024) had even hit the editor’s desk. The work that culminated in both papers was done in the UK, Australia, Canada and Sweden, with some potential for poor communication within the two teams. Whatever, the first paper dangled the carrot that Orkney might have been the Altar Stone’s source, on the basis of geochemical evidence that the grains that make up the sandstone could not have been derived from Wales but were from the crystalline basement of NE Scotland. The second shows that this ‘most popular’ Scottish source may be ruled out. To Orcadians and the archaeologists who worked there, long in the shade of vast outpourings from Salisbury Plain, this might come as a great disappointment.

Cyclical sediments of the Devonian Stromness Flagstones. (Credit Mike Norton, Wikimedia)

The latest paper examines 13 samples from 8 outcrops of the Middle Devonian Stromness Flagstones strata in the south of the main island of Orkney close to the Ring of Brodgar and the Stones of Stenness, and the individual monoliths in each. On the main island, however, there is a 500 m sequence of Stromness Flagstones in which can be seen 50 cycles of sedimentation. Each cycle contains sandstone beds of various thicknesses and textures. They are fluviatile, lacustrine or aeolian in origin. So the Neolithic builders of Orkney had a wide choice, depending on where they erected monumental structures. Almost certainly they chose monolithic stones where they were most easy to find: close to the coast where exposure can be 100 %. The Ring of Brodgar and the Stones of Stenness are not on the coast, so the enormous stones would have to be dragged there. There is an ancient pile of stones (Vestra Fiold) about 20 km to the NW where some of the mmegaliths may have been extracted, but ancient Orcadians would have been spoilt for choice if they had their hearts set on erecting monoliths!

In a nutshell, the geological case made by Bevins et al. (2024) for rejecting Orkney as the source for the Stonehenge Altar Stone (AS) is as follows: 1. Grains of the mineral baryte (BaSO) present in the AS are only found in two of the Orkney rock samples. 2. All the Orcadian sandstone samples contain lots of grains of K-feldspar (KAlSi3O8) – common in the basement rocks of northern Scotland – but the AS contains very little. 3. A particular clay mineral (tosudite) is plentiful in the AS, but was not detected in the rock samples from Orkney. Does that rule out a source in Orkney altogether? Well, no: only the outcrops and megalith samples involved in the study are rejected.

To definitely negate an Orcadian source would require a monumental geochemical and mineralogical study across Orkney; covering every sedimentary cycle. Searching the rest of the Old Red Sandstone elsewhere in NE Scotland – and there is a lot of it – would be even more likely to be fruitless. Tracking down the source for the basaltic bluestones at Stonehenge was easy by comparison, because they crystallised from a particular magma over a narrow time span and underwent a specific degree of later metamorphism. They were easily matched visually and under the microscope with outcrops in West Wales in the 1920s and later by geochemical features common to both.

But all that does not detract from the greater importance of the earlier paper (Clarke et al., 2024), which enhanced the idea of Neolithic cultural coherence and cooperation across the whole of Britain. The building of Stonehenge drew people from the far north of Scotland together with those of what are now Wales and England. Since then it hasn’t always been such an amicable relationship …

See also:  Addley, E. 2024. Stonehenge tale gets ‘weirder’ as Orkney is ruled out as altar stone origin. The Guardian 5 September 2024.

Geology cracks Stonehenge mysteries

High resolution vertical aerial photograph of Stonehenge. (Credit: Gavin Hellier/robertharding/Getty)

During the later parts of the Neolithic the archipelago now known as the British Isles and Ireland was a landscape on which large stone buildings with ritual and astronomical uses were richly scattered. The early British agricultural societies also built innumerable monuments beneath which people of the time were buried, presumably so that they remained in popular memory as revered ancestors. Best known among these constructions is the circular Stonehenge complex of dressed megaliths set in the riot of earlier, contemporary and later human-crafted features of the Chalk downs known as Salisbury Plain. Stonehenge itself is now known to have been first constructed some five thousand years ago (~3000 BCE) as an enclosure surrounded by a circular ditch and bank, together with what seems to have been a circular wooden palisade. This was repeatedly modified during the following two millennia. Around 2600 BCE the wooden circle was replaced by one of stone pillars, each weighing about 2 t. These ‘bluestones’ are of mainly basaltic igneous origin unknown in the Stonehenge area itself. The iconic circle of huge, 4 m monoliths linked by 3 m lintel stones that enclose five even larger trilithons arranged in a horseshoe dates to the following two-centuries to 2400 BCE coinciding with the Early Bronze Age when newcomers from mainland Europe – perhaps as far away as the steppe of western Russia – began to replace or assimilate the local farming communities. This phase included several major modifications of the earlier bluestones.

It might seem that the penchant for circular monuments began with the Neolithic people of Salisbury Plain, and then spread far and wide across the archipelago in a variety of sizes. However, it seems that building of sophisticated monuments, including stone circles, began some two centuries earlier than in southern England in the Orkney Islands 750 km further north and, even more remote, in the Outer Hebrides of Scotland. A variety of archaeological and geochemical evidence, such as the isotopic composition of the bones of livestock brought to the vicinity of Stonehenge during its period of development and use, strongly suggests that people from far afield participated. Remarkably, a macehead made of gneiss from the Outer Hebrides turned up in an early Stonehenge cremation burial. Ideas can only have spread during the Neolithic through the spoken word. As it happens, the very stones themselves came from far afield. The earliest set into the circular structure, the much tinkered-with bluestones, were recognised to be exotic over a century ago. They match late Precambrian dolerites exposed in western Wales, first confirmed in the 1980s through detailed geochemical analyses by the late Richard Thorpe and his wife Olwen Williams-Thorpe of the Open University. Some suggested that they had been glacially transported to Salisbury Plain, despite complete lack of any geological evidence. Subsequently their exact source in the Preseli Hills was found, including a breakage in the quarry that exactly matched the base of one of the Stonehenge bluestones. They had been transported 230 km to the east by Neolithic people, using perhaps several means of transport. The gigantic monoliths, made of ‘sarsen’ – a form of silica-cemented sandy soil or silcrete – were sourced from some 25 km away where Salisbury Plain is still liberally scattered with them. Until recently, that seemed to be that as regards provenance, apart from a flat, 5 x 1 m slab of sandstone weighing about 6 t that two fallen trilithon pillars had partly hidden. At the very centre of the complex, this had been dubbed the ‘Altar Stone’, originally supposed to have been brought with the bluestones from west Wales.

The stones of Stonehenge colour-coded by lithology. The sandstone ‘Altar Stone’ lies beneath fallen blocks of a trilithon at the centre of the circle. (Credit: Clarke et al. 2024, Fig 1a)

A group of geologists from Australia and the UK, some of whom have long been engaged with Stonehenge, recently decided to apply sophisticated geochemistry at two fragments broken from the Altar Stone, presumably when the trilithons fell on it (Clarke, A. J. I. et al.2024.  A Scottish provenance for the Altar Stone of Stonehenge. Nature v.632, p. 570–575; DOI: 10.1038/s41586-024-07652-1). In particular they examined various isotopes and trace-elements in sedimentary grains of zircon, apatite and rutile that weathering of igneous rocks had contributed to the sandstone, along with quartz, feldspar, micas and clay minerals. It turned out that the zircon grains had been derived from Mesoproterozoic and Archaean sources beneath the depositional site of the sediment (the basement). The apatite and rutile grains show clear signs of derivation from 460 Ma old (mid-Ordovician) granites. The basement beneath west Wales is by no stretch of the imagination a repository of any such geology. That of northern Scotland certainly does have such components, and it also has sedimentary rocks derived from such sources: the Devonian of Orkney and mainland Scotland surrounding the Moray Firth. Unlike the lithologically unique bluestones, the sandstone is from a thick and widespread sequence of terrestrial sediments colloquially known as the ‘Old Red Sandstone’. The ORS of NE Scotland was deposited mainly during the Devonian Period (419 to 369 Ma) as a cyclical sequence in a vast, intermontane lake basin. Much the same kinds of rock occur throughout the sequence, so it is unlikely that the actual site where the ‘Alter Stone’ was selected will ever be known.

To get the ‘Alter Stone’ (if indeed that is what it once was) to Stonehenge demanded transport from its source over a far more rugged route, three times longer than the journey that brought the bluestones from west Wales: at least 750 km. It would probably have been dragged overland. Many Neolithic experts believe that transport of such a large block by boat is highly unlikely; it could easily have been lost at sea and, perhaps more important, few would have seen it. An overland route, however arduous, would have drawn the attention of everyone en route, some of whom might have been given the honour of helping drag such a burden for part of the way. The procession would certainly have aroused great interest across the full extent of Britain. Its organisers must have known its destination and what it signified, and the task would have demanded fervent commitment. In many respects it would have been a project that deeply unified most of the population. That could explain why people from near and far visited the Stonehenge site, herding livestock for communal feasting on arrival. Evidence is now pointing to the construction and use of the ritual landscape of Salisbury Plain as an all-encompassing joint venture of most of Neolithic Britain’s population. It would come as no surprise if objects whose provenance is even further afield come to light. It remained in use and was repeatedly modified during the succeeding Bronze Age up to 1600 BCE. By that time, the genetic group whose idea it was had been assimilated, so that only traces of its DNA remain in modern British people. This seems to have resulted from waves of immigrants from Central Europe, the Yamnaya, who brought new technology and the use of metals and horses.

See also: Gaind, N. & Smith, R. 2024. Stonehenge’s enigmatic centre stone was hauled 800 kilometres from Scotland. Nature, v. 632, p. 484-485; DOI: 10.1038/d41586-024-02584-2; Addley, E. 2024. Stonehenge megalith came from Scotland, not Wales, ‘jaw-dropping’ study finds. The Guardian, 14 August 2024.