The Newall Boulder: the human transport hypothesis is as dodgy as ever

The bullet-shaped Newall Boulder now held in Salisbury Museum.  The faces have varying degrees of "freshness".  At the top of this photo we can see the damage done by geological sampling.

I should, of course, have been asked to review this paper prior to publication, but nowadays researchers effectively get to choose their own reviewers and to ban those who are deemed likely to be unfriendly........  Anyway, here is the peer review I would have written, if I had been asked.

Declaration of interest:  I have personally examined the "Newall Boulder" in Salisbury Museum, and have already expressed reservations about a previous report co-authored by Ixer, Bevins and Pearce (see note at the end of this review). 

Peer review

Richard Bevins, Rob Ixer, Nick Pearce, James Scourse, Tim Daw. 2023.
Lithological description and provenancing of a collection of bluestones from excavations at Stonehenge by William Hawley in 1924 with implications for the human versus ice transport debate of the monument's bluestone megaliths. Geoarchaeology 2023: 1-15
DOI:10.1002/gea.21971
https://onlinelibrary.wiley.com/share/author/YUUAUVRWBNTZTPSQVBGM?target=10.1002/gea.21971

This paper seeks to demonstrate that a small sub-angular boulder found in a Stonehenge excavation was knocked off the tip of a rhyolite monolith that no longer exists, having earlier been excavated from a known site in West Wales prior to human transport to Stonehenge for use in a bluestone setting.  The claim that natural processes were not at any stage involved in the entrainment, transport and emplacement of the boulder is an extraordinary one, which can only be supported by extraordinary and powerful evidence.  Has that irresistible evidence now been provided?  In the view of this reviewer, the answer is "No".

At the beginning of the paper the authors suggest that there are eleven known "bluestone lithologies" which must have been associated with specific numbered orthostats.  The labelling of these lithological groups is confusing, to put it mildly, and the authors fail to acknowledge that each group probably  involves stones from several different provenances.  Further, during the excavations at Stonehenge at least 46 "exotic" rock types have been turned up in excavations, many of them having no links at all with any of the known "bluestone orthostats".  To ignore them is to defy logic. Angular fragments, abraded pebbles and cobbles found in the sediments, which may be crucial for the interpretation of what happened at Stonehenge, are simply dismissed by the authors. This fact alone undermines many of the claims made here about the human transport of the bluestones.  If the bluestones were selected for their "special" qualities or indeed quarried from sacred sites, do the authors really believe that monoliths and smaller stones were extracted deliberately from 46 different places?

In their literature review relating the the glacial transport / human transport controversy, the authors should have made reference to the modern work relating to very extensive glaciation by the Irish Sea Glacier, and the modelling work suggesting that the glacial transport of the Stonehenge bluestones was "not impossible" (Hubbard et al, 2009).

In their examination of the 18 "Newall stones" from the Hawley excavations at Stonehenge, the authors make somewhat heavy weather of the suggestions that some, at least, might have come from North Wales.  My understanding is that the suggestions of North Wales / Lake District origins for the small boulder referred to as RSN18, by Harrison and other BGS geologists, were very tentative.  But they were all quite convinced that the boulder (referred to as a strongly welded acid vitric tuff or ignimbrite) was different, geologically, from any of the known rhyolites (at that time) from Mynydd Preseli.  I agree with that. The Newall Boulder rock is darker in colour, rougher and more "flinty" than the rhyolite of Rhosyfelin. 

The authors' investigations of the petrography of the thin sections from the Newall Boulder are interesting but inconclusive.  The authors claim that "petrographically the RS18 fragment matches rhyolitic tuff from Craig Rhos-y-felin", in turn claimed as the "dominant source of the Stonehenge rhyolitic debitage" and as the site of a Neolithic bluestone quarry.  This is not supported by the evidence presented.  The photomicrographs comparing the texture and lithology of RS18 and Craig Rhos-y-felin rhyolite show similarities, but not perfect matches, and there remains a strong possibility that RS18 has come from elsewhere on an extensive rhyolite outcrop, or even from another outcrop entirely unrelated to Craig Rhos-y-felin. Attempts by the authors to demonstrate that the foliations in the Newall boulder and on the rock face at Rhos-y-felin indicate that the one was derived from the other are unconvincing, since we are not told how many other outcrops of foliated rhyolite there may be in West Wales. The whitish weathering crust, also cited as an indicator of a Rhos-y-felin origin, is also unconvincing, since whitish weathering crusts are present on many other West Wales rhyolite outcrops.

The reported pXRF work on the Newall boulder, also reported in section 2 of the paper, is similarly inconclusive. The bivariate plots shown in Figure 6 were -- as is normal in papers of this type -- created specifically in order to demonstrate affinities,  but this does not mean that the demonstrated relationships are unique or significant.  The overlaps between the fields for the Newall samples and the Rhos-y-felin samples may be no more marked, for example, that the relationships between the Newall samples and the plots for Carn Alw,  Ty Canol Wood, or Maiden Castle, which the authors may have, but have chosen not to show us.

Slickenside features on one face of the boulder, which must coincide with a fault plane.  

On page 9 the authors accuse Kellaway of mistakenly interpreting slickenside lineations as glacial striations.  Kellaway was a good geologist, and I do not believe he could have made such a stupid mistake.  The striations which he interpreted as having a glacial origin are much more subtle and discontinuous, and can be seen only during a minute examination of the boulder surface. This matter is discussed again on page 10, where there is a serious misrepresentation of the characteristics of glacial striae.  Striae are NOT typically continuous over a large proportion of a facet surface.  The clast shown in Figure 9 may be an ideal text-book illustration, but the great majority of striated clasts which I have encountered during a lifetime of working with glacial sediments are not like this at all; many of them show just one or two surface scratches, which may or may not be sub-parallel or cross-cutting.  The Newall Boulder, which I have examined, has a few very subtle scratches that may be striae, and they are quite distinct from the slickenside features shown in Figure 10.

It is disingenuous of the authors to pretend that because slickenside features are present on the Newall Boulder and at Craig Rhos-y-felin, this demonstrates a source for the boulder.  Slickenside features including slickencrysts are common across West Wales, in all faulted lithologies and of all ages.

On page 12 the authors refer to the "consistency of lithologies" in the Newall bluestone assemblage as an indication of "human selection of the material" rather than "a random process of entrainment in glacial till".   The entrainment of debris in glacial deposits is not a random process.  The authors do not explain why or how human beings should have selected around 46 different lithologies (mostly from the west)  for incorporation into Stonehenge sediments.  Were they disinterested in lithologies from the north, east and south?   They claim that "several of the bluestone lithologies have been sourced to specific outcrops in North Pembrokeshire, namely Craig Rhos-y-felin, Carn Goedog, Cerrigmarchogion and Carn Ddu Fach."  That is a misleading statement; these sources have been suggested, but never adequately proved, and indeed in the papers referenced the authors themselves accept that there are NO definitive sources that are beyond dispute.  The fact that many of the bluestones come from a limited geographical area is NOT suggestive of the human selection of the bluestones; on the contrary, since so many different rock types are represented in the Stonehenge assemblage of bluestone clasts, the supposition must be that glacial transport was responsible for stone transport on a substantial scale.  It is vanishingly unlikely that small fragments and cobbles in the debitage (many of them quite unrelated to bluestone orthostats) were selected and then carried to Stonehenge by human beings.

The authors argue on p 11 that the "snub nose" shape of the Newall Boulder is typical of outcrops at Craig Rhos-y-felin.  I know the site, and I do not accept that. They also suggest that the boulder is probably the broken tip of a destroyed orthostat (maybe stone 32d) at Stonehenge.  On the contrary, the boulder's "bullet shape" is suggestive of glacial transport, with abraded edges and discernible facets.   They also suggest that the "fresh" surface of the boulder shows where it was broken from the complete orthostat.  However, the fresh surface of the boulder is of very limited extent, and does not coincide with a sizeable cross-cutting fracture scar.  Examination of the boulder shows that Kellaway was most likely correct in assuming that part of the exposure of dark blue "fresh rock" was the result of limited damage (probably man-made) inflicted on a boulder only slightly larger and heavier than the one we see today.  Also, it borders on the absurd to suggest that a flimsy and fractured rhyolite orthostat with dimensions no greater than 2m x 40 cm x 30 cm would have been quarried from Rhosyfelin, transported to Stonehenge, and incorporated into a bluestone setting without falling apart in the process.  Indeed, the unsuitability of Rhosyfelin rhyolite for use as standing stones has now been accepted by Bevins, Ixer and other authors who some years ago flagged up the site as "the Pompeii of Neolithic quarries."  The evidence against quarrying at this site was presented in detail in two peer-reviewed papers by John, Elis-Gruffydd and Downes which have been completely ignored by the authors of the present paper.

Finally, on the matter of glaciation and glacial erratic transport, the authors summarise the evidence relating to erratic trains and glacier extent in the Bristol Channel arena which has been presented by selected researchers.  They ignore the work of other researchers, and in doing so they fail to address the glaciological conundrum of how the ice of the Irish Sea Ice Stream can have reached the shelf edge in the Celtic Sea without also overwhelming the inner reaches of the Bristol Channel and the Somerset Lowlands.  They also fail to mention the modelling work which has shown that glacier extent at the time of the Greatest British Glaciation might well have involved an ice edge on or near Salisbury Plain.  They fail to mention the presence of very old glacial sediments in the Somerset Levels, near Bath and on the coasts of Devon and Cornwall, and the presence of high-level erratics on the coasts of the South-west of England which cannot be explained other than by an extensive glaciation. They claim (falsely) that there are no "bluestone erratics" in the vicinity of Stonehenge, while ignoring the fact that most of the Stonehenge bluestones are not beautiful pillars as portrayed in the textbooks but classic glacial erratics (boulders and slabs) with distinguishable facets, abraded edges and thick weathering crusts.  The evidence of Stonehenge glaciation is staring the authors in the face, without apparently being noticed.

The authors claim that Clark et al (2022) and Gibbard et al (2022) summarise the most recent evidence of Devensian and earlier ice limits in Britain, and that there have been no significant changes in proposed ice edge positions.  But those two authors would be the first to admit that there are large inconsistencies in parts of the evidence base, and some of the "accepted" ice limits in parts of the Bristol Channel /Celtic Sea arena do not actually make much sense.......... 

(https://brian-mountainman.blogspot.com/2022/09/pre-devensian-landforms-mis-6-late.html)

https://www.researchgate.net/publication/368426571_Was_there_a_Late_Devensian_ice-free_corridor_in_Pembrokeshire.      

It is clear that we are still some way from the telling of the full story.

An attempt by Gibbard and Clark to represent the ice edges for three glacial episodes.  In places these lines are matched by good "ground truthing", and in places they are highly speculative.  The Wolstonian ice edge shown for mid-Wales makes no sense at all....... 

However strong the evidence of glaciation and glacial transport may or may not be, it is far stronger than the evidence for human bluestone transport.  No evidence for human bluestone haulage from West Wales to Stonehenge has ever been found by the present authors or by anybody else, and the authors should have the good grace to admit this.  Indeed, there is no good evidence for substantial stone haulage with respect to any of the great megalithic monuments of the British Isles.

The contention that the Newall Boulder is not a glacially transported clast is unsupported by the evidence presented in this paper, as is the contention that it is a knock-off from a mysterious unknown Rhosyfelin rhyolite orthostat.  

Without considerable revision, this paper is not of sufficient quality for  publication in a serious scientific journal.

=============================

 

Declaration of interest: the following items are relevant:
https://brian-mountainman.blogspot.com/2022/09/the-newall-boulder-with-thee-samples.html
https://www.researchgate.net/publication/361939331_A_glacially-transported_clast_at_Stonehenge
https://the-past.com/feature/victorian-gifts-new-insights-into-the-stonehenge-bluestones/