Some of the ideas discussed in this blog are published in my new book called "The Stonehenge Bluestones" -- available by post and through good bookshops everywhere. Bad bookshops might not have it....
To order, click

Saturday, 30 November 2019

New paper on the Stonehenge sandstones: significant support for the glacial transport hypothesis

First of all, let's remind ourselves of the current glacial transport hypothesis:  The non-sarsen boulders, slabs, pillars, stumps, stones and "debitage" found in the Stonehenge area come from an assemblage of glacial erratics transported from West and South Wales towards Salisbury Plain by a powerful ice stream, and were later discovered and exploited by the builders of the stone monument.

Now, to the latest paper by Ixer, Bevins, Pirie, Turner and Power, and published in the Wiltshire Archaeological and Natural History Magazine with a 2020 date.    Here are the details:

‘No provenance is better than wrong provenance’: Milford Haven and the Stonehenge sandstones

by Rob A. Ixer, Richard E. Bevins, Duncan Pirrie, Peter Turner and Matthew Power.
Wiltshire Archaeological and Natural History Magazine 113, 20220, pp 1-15.


For over 70 years there has been confusion within the archaeological literature between the Stonehenge ‘Old Red Sandstone’ Altar Stone, the Stonehenge Ordovician-Silurian Lower Palaeozoic Sandstone debitage, and the previously postulated source rocks, the Old Red Sandstone (Devonian) Cosheston Group sandstones. However, petrographic data show that all three are very different lithologies with separate geographical origins. The Altar Stone is most likely to be from eastern Wales and the Lower Palaeozoic Sandstone from west or central Wales, north and east of the Mynydd Preseli; neither of these two Stonehenge-related sandstones is from Mill Bay, Milford Haven as has been suggested. The revised provenance determinations do not support the theory that the Stonehenge bluestones were shipped over sea from the Milford Haven area along the Bristol Channel.

This is a real curate's egg of a paper, containing some meticulous and interesting research, summarising a great deal of material already published, but demonstrating a singular lack of awareness of earth surface processes. It's also very convoluted,  and much more elaborate than it needed to be.  Like a number of other papers from Ixer, Bevins and others on the subject of the bluestones, I cannot imagine that there was even a cursory peer review process prior to publication.  If the Editor had asked me to look at this in its manuscript stage, I would willingly have given him quite a lot of help........ 

However, here it is in print, and in spite of its glaring defects as a scientific paper I have to say that I am rather delighted to see it, because it provides substantial evidence for the glacial transport of the Stonehenge bluestones.

Relevant sandstone outcrops investigated by the authors

In essence, the paper homes in on the Altar Stone (assumed to be Devonian) and the Lower Palaeozoic debitage in the Stonehenge landscape, and the question of whether any or all of the samples analysed actually came from the Cosheston Sandstones outcropping on the shore of Mill Bay. The research is framed as being necessary in order to decide whether the bluestones at Stonehenge were moved by sea or overland;  this is a completely ludicrous piece of research justification since all it does is to demonstrate the tunnel vision of the authors.  More of that anon.

In an extended analysis of the 3 sandstone groups (Altar Stone, debitage and Mill Bay) in the archaeological literature, the researchers focus on the famous thin section 277 (which may or may not have actually come from the big recumbent slab) and suggest that it is more likely to have come from the Senni Beds than from the Cosheston Beds -- although they are equivalent in the Devonian time sequence.  In their discussion of the "Lower Palaeozoic sandstone debitage" the authors point out that most of the samples obtained thus far do not come from within the original stone settings but from the Heelstone ditch, the Greater Cursus ditch and from Roman disturbed sites.  But there also appear to be four samples of the same rock type from genuine Stonehenge contexts.

The research core of the paper is a discussion of past work on sandstone samples from Mill Bay, and new analyses undertaken on fresh samples collected by me (duly acknowledged, with thanks!) and other fieldworkers.  The conclusion is that the mineral composition / petrography revealed in the samples is so different from that of the Stonehenge debitage fragments that Mill Bay can effectively be ruled out as a source area.  The authors say that Mill Bay sandstone fragments are "now considered not to be present within the Stonehenge landscape"  -- in my view they cannot say that, since thus far they only know about a small fraction of what lies beneath the ground surface.  However, it is sufficient -- and quite reliable -- to say that the known samples of Devonian sandstone at Stonehenge appear not to have come from Mill Bay or the Cosheston Beds, and that source areas within the Senni Beds - further to the east -- are much more likely.  That is an important advance.

re the 3 sandstone groups:  Each sandstone is distinctive in term of its minor accessory/heavy minerals (reflecting its source area), diagenetic history as determined by its authigenic cements and clay minerals, and its tectonic and metamorphic history as seen by its clay mineralogy and presence/absence of any tectonic fabric. (p 12)

With respect to the Senni Beds, the authors show a map of their distribution (reproduced above), and say this:

"........for much of this area the outcrop is thin, only widening towards eastern Wales and the Welsh Marches. The composition and relative abundances of the clay minerals in Devonian sandstones systematically varies from east to west Wales (reflecting a change in the metamorphic grade) (Hillier et al. 2006) and preliminary work on the Altar Stone clay mineralogy suggest that the sandstone may be from the east of Wales."

This is interesting, and it will be good to see where this leads -- but a word of caution.  In due course, geologists may well show that some of the Stonehenge fragments analysed may have come from the Senni Beds in eastern Wales.  But that does not mean that that is where the Altar Stone came from. We still do not have absolute certainty that any of the samples examined really did come from the Altar Stone (see also "The Stonehenge Bluestones", pp 172-176 for a more detailed discussion).

See also:
IXER, R. A., BEVINS, R. E., TURNER, P., POWER, M. and PIRRIE, D., 2019. Alternative Altar Stones? Carbonate-cemented micaceous sandstones from the Stonehenge Landscape. WANHM 112, 1–13.

Other posts:


In summary:

What does the paper do?  Well, it provides quite convincing evidence that the sandstone samples reputed to have come from the Altar Stone (stone 80) did not come from the Mill Bay area on the Daugleddau Estuary in mid-Pembrokeshire.

What does the paper NOT do?  Well, here goes:

1.  It does not prove that any of the analysed samples purported to have come from the Altar Stone did actually come from that source.  It follows that the Altar Stone might still be made of Cosheston Beds sandstone.

2.  It does not prove that the "sandstone stumps" numbered 40g and 42c are made of Devonian or Lower Palaeozoic sandstones.

3.  It does not prove the absence of Devonian sandstones from the Stonehenge environs, since the sampled fragments discussed in this and other papers have all come from the 50% or so of the land within the stone settings that happens to have been excavated, and from other sites in the wider landscape.

4.  It does not prove that any of the sandstone debitage has come from stumps 40g and 42c (just as the geologists have not proved that any of the foliated rhyolite debitage at Stonehenge has come from stumps 32d and 32e).  Some of the debitage might well have come from sandstone lumps that were too small to be used in the stone settings.


Entrainment and Transport

In their conclusions the authors wind things up by suggesting that all this somehow has a bearing on the arguments about whether the bluestones were "shipped out" from Milford Haven and along the Bristol Channel or by land along the "A40" route.  They say:  "The new studies instead strongly indicate a number of inland geographical origins for the bluestones; hence a land route is now firmly preferred over a sea route."

There is no mention whatsoever, even from this group of practising geologists, of the glacial transport theory -- even though people like them are supposed to know something about earth surface processes.  There are no citations of the work of Judd, Jehu, Geikie, Williams-Thorpe,  Elis-Gruffudd, Downes or myself, or anybody else who has proposed, with much supporting evidence, that glacier ice could have moved the bluestones.  Geoffrey Kellaway is mentioned, but only in the context of his rather weird idea of a Pliocene glaciation.  That is a strange citation -- but we can understand it as an attempt by the authors to suggest that the "glacial theorists" are fantasists who have lost contact with the real world.  But we are rather more grounded than they may think -- and it is completely indefensible for any serious authors dealing with bluestone provenance to pretend that their ideas are not disputed, and to refuse to consider or even cite a serious literature in the public domain which shows that there is abundant evidence that points to the glacial transport of the bluestones.

So the authors -- not for the first time -- are behaving, in this paper, in an academically reprehensible fashion, and I am amazed that the Editor of WANH magazine has allowed them to get away with it.  He is culpable too.

But hey -- life is too short to spend one's whole time being furious, and the silver lining in this case is that the article in question provides very substantial support for the glacial transport thesis. We know already that the bluestones at Stonehenge (of many different shapes and sizes) have come from around 30 different provenances, mostly in west Wales.  We also know that the "quarrying hypothesis", with respect to Rhosyfelin and Carn Goedog, does not withstand careful scrutiny.  The bluestones at Stonehenge (both orthostats and fragments) have come from a wide scatter of locations, and that in itself is an argument against quarrying for favoured rock types in special locations. Could there really have been 30 bluestone quarries?   Why quarry for a rubbish stone like foliated rhyolite down in a deep valley when better stones could have been picked up from the higher land surfaces round about?  Why quarry for spotted dolerite at the exposed and craggy tor of Carn Goedog, when there were boulders of all shapes and sizes dotted about all over the Preseli hill slopes?

Indeed, why would Neolithic tribesmen have bothered to quarry for monoliths at all, since there is no evidence of any preferential use of specific rock types (eg spotted dolerite, foliated rhyolite, Devonian sandstone, or Lower Palaeozoic sandstone) anywhere in the British Neolithic.  In Wales, cromlechs and standing stone settings were always made of whatever handy lumps of stone happened to be lying around in the vicinity.  It is entirely logical to assume that the same principle of monolith collection and use applied at Stonehenge.

Carreg SamsonSamson, Abercastle.  Just one principle of megalith construction:  use 
whatever stone you've got.  

As indicated in my book "The Stonehenge Bluestones",  entrainment in the compression zone of ice flowing up and over Preseli sits easily with glaciological theory.

Generalised erratic transport routes in South Wales, after many authors.  Note the crossing of arrows. Many erratics carried southwards from the Welsh ice cap were later incorporated into the east-flowing Irish Sea Glacier.

If -- as now appears likely -- the stones at Stonehenge included Lower Palaeozoic sandstones from somewhere in West Wales and Devonian sandstones from somewhere in South Wales, that again militates against targetted quarrying and supports the idea of ice entrainment and transport towards Stonehenge.  There is no problem at all with ice from the Welsh ice cap picking up large blocks of sandstone (or any other rock, for that matter) and transporting these blocks southwards prior to entrainment in the eastward-flowing Irish Sea Glacier at a later stage in a glaciation.  This was realized and commented on more than a century ago by the officers of the Geological Survey (including, ironically, HH Thomas).  

Finally, I am not in the least bothered about where the Altar Stone or the sandstone debitage at Stonehenge actually came from within the outcrops shown on the map of "possible source sandstones".  I am confident that ice did all of the heavy lifting and transportation, and that all our heroic ancestors needed to do, around 5,000 years ago,  was to collect them up from somewhere far to the east, with a view to turning them into an enigmatic monument.  As I have suggested many times, that monument was never finished because the builders ran out of steam and ran out of stones.


PS.  Thanks to Rob Ixer for correcting a slip of the pen.  Duly corrected.


a gee said...

The hunt for the sandstone quarry commences!!

tonyH said...

The later section of your Post, headed ENTRAINMENT & TRANSPORT, contains a paragraph starting

"But hey - life is too short.......

......The bluestones of Stonehenge (both orthostats and fragments) have come from a wide scatter of locations, and that in itself is an argument against quarrying for favoured rock types at special locations. Could there really have been THIRTY BLUESTONE QUARRIES [my capitals]."

As recently as November 28th, I said as much on a Comment to the 28th August 2019 Post "STONEHENGE: THE DAWSON NARRATIVE", where I point out the sheer variety of rock types, necessitating there to be many, many quarries, often in inaccessible locations for any hypothetical intrepid Neolithic rock collectors.

BRIAN JOHN said...

Quite so -- thanks tor making this point as well, Tony. The point is particularly important re Rhosyfelin, flagged up as the "Pompeii of Neolithic quarries" even though all we have at Stonehenge are assorted foliated rhyolite fragments (not necessarily from Rhosyfelin) and NO MONOLITHS AT ALL. As we have said many times before, certain archaeologists and geologists seem to inhabit a strange sort of fantasy land......

tonyH said...

The Honorary WANHM editor for 2017 and 2018 was Stuart Brookes, BA (Hons), PGDTiLL, MA, PhD, FSA. No idea if there was "even a cursory peer review process", as you are questioning in the Post.

tonyH said...

Dr Stuart Brookes has links with Birkbeck College and UCL Institute of Archaeology and has been involved with Stonehenge. Prehistoric Archaeology does not seem to be his prime subject.

BRIAN JOHN said...

Small journals tend to have minimal reviewing processes -- which is of course why this particular group of authors keep on publishing in WANHM rather than in proper geological journals where there will inevitably be much more critical peer review systems in place.....

tonyH said...

This WANHM article has now (May 2020) appeared in the published edition of the magazine [volume 113, 2020]. Stuart Brookes was again the Honorary Editor (see previous comments).

I noted, to my surprise, that a certain "John, Brian" appears in the volume's index [pages 7-8,10,14] in relation to the article.

As Brian knows, his earlier book "The Stonehenge Enigma", was reviewed in a previous volume of WANHM. We must make stringent efforts to obtain a balanced review for its 2018 updated successor, "The Stonehenge Bluestones".

BRIAN JOHN said...

That would be because in the article I'm thanked for collecting some sandstone samples from Mill Bay a few years ago. ....