First Look at Clarke et al. (2026) – From Highlands to Henge

A refined provenancing study with a new statistical “best match”, but it still feels like identifying one possible needle in a very large and varied haystack.

Clarke, A. J. I., Veness, R. L. J., Kirkland, C. L., Clark, C. D., Gandy, N., Emery, A. et al. (2026) From Highlands to Henge: Refining the Provenance and Transport Pathways of Stonehenge's Altar Stone. Journal of Quaternary Science, 1–8. https://doi.org/10.1002/jqs.70080

The paper uses Kolmogorov–Smirnov tests and MDS plots on the limited published detrital zircon datasets and finds the strongest match at Sarclet in Caithness (p = 0.96), with other northern mainland sites (Braemore, Kirtomy, Portskerra) also statistically compatible. More southerly Orcadian Basin and Grampian outlier samples fare worse. This is a genuine step forward from the 2024 basin-level result, yet the authors themselves stress that the Orcadian Basin is enormous (up to 10,000 km²) and lithologically heterogeneous, with only sparse published zircon coverage. So while Caithness now looks the most likely area among the data we have, it remains one plausible candidate among many unsampled possibilities.

On transport, the paper dismisses any direct Late Devensian glacial delivery to Salisbury Plain. Using time-integrated ice-flow modelling (Veness et al. 2025 method on BRITICE-CHRONO geometries), the favoured Caithness sources mostly disperse north and east; only specific sensitivity scenarios (or slightly more southerly seeds) allow material to reach as far as Dogger Bank.

The authors float an intriguing hybrid idea: glacial transport could have delivered the stone part-way to Dogger Bank (~400 km from Stonehenge), after which humans might have collected it and moved it the remaining distance, perhaps while the area was still accessible or as post-glacial sea-level rise began to inundate Doggerland. They note Dogger Bank was exposed dry land with moraines until roughly 8–7 ka BP.

However, several problems stand out even on a first reading:

• No supporting evidence on the ground (or seabed): No other Altar Stone-like erratics or matching detrital signatures have been reported from Dogger Bank or the surrounding North Sea floor in the right context. The companion Clarke & Kirkland (2026) fingerprinting study found no glacial detrital zircon–apatite signal from northeast Scotland in Salisbury Plain sediments at all.
• Timing difficulties: Any Dogger Bank staging post would require the stone to be removed and transported onward (or cached) across a gap of several millennia before erection at Stonehenge. The paper acknowledges this creates a complicated multi-stage history rather than a neat solution.
• Modelling limitations: Ice-flow reconstructions are models, not direct observations. They have known issues reconciling with the actual geological record of erratic distributions in places, and the present simulation has relatively coarse spatial (2.5 km) and temporal (1,000-year) resolution. The authors are open about the high sensitivity to exact starting position near the former ice divide and that only the Devensian can be modelled usefully here.

Conclusion Even in its most favourable reading, this paper does not rescue a primarily glacial explanation. It narrows the likely source area statistically while showing that direct glacial transport to Wessex is not supported, and that even a partial glacial assist to Dogger Bank still leaves substantial anthropogenic transport necessary — plus extra chronological and logistical complications. The absence of corroborating erratics or detrital evidence further weakens the hybrid staging-post scenario.

On balance, the work inadvertently strengthens the case that Neolithic people organised and executed the long-distance movement of this 6-tonne block, whether from Caithness directly or via some intermediate location. It is a careful, data-driven paper, but it does not move the glacial needle very far. More targeted zircon sampling from the Caithness area and higher-resolution modelling (or new seabed data) would be needed before any Dogger Bank idea could be taken seriously.