Archaeology and the invented narrative

From the ever-reliable and objective rag called the Daily Mail comes a story that should at least make us stop and think for a few moments -- a tale of an archaeologist (Sir Mortimer Wheeler, no less) inventing a spectacular story on the basis of very little evidence which is now shown to have been complete nonsense.

https://www.dailymail.co.uk/news/article-7826865/Archaeologist-accused-inventing-AD43-battle-digging-skeletons-ancient-tribe.html?fbclid=IwAR2vE7kYTIBq_z0O8eCuJn8DAyxyuxjk-1gC6ohymrOBjO5xFeNH90sERA0

The story of a vicious assault (by Vespasian and a Roman legion) on Maiden Castle in Dorset in the year AD43 followed an excavation by Sir Mortimer and his wife in 1936-7.  It was based on the discovery of multiple "British warriors" with terrible injuries in hurriedly dug graves  -- and the "leap of faith" by Sir Mortimer in concluding that he had discovered a "war cemetery."  He also concluded that the slaughter of the local warriors by the Romans led to the abandonment of the site.

This exciting and appealing tale was picked up and repeated in many other papers and texts, and remains to this day a part of the EH narrative for this site and for the events of the Roman Conquest.

This is an interesting extract from Wikipedia:

After ending his work at Verulamium, Wheeler turned his attention to the late Iron Age hill-fort of Maiden Castle near to Dorchester, Dorset, where he excavated for four seasons from 1934 to 1937. Co-directed by Wheeler, Tessa, and the Curator of Dorset County Museum, Charles Drew, the project was carried out under the joint auspices of the Society of Antiquaries and the Dorset Field Club.  With around 100 assistants each season, the dig constituted the largest excavation that had been conducted in Britain up to that point, with Wheeler organising weekly meetings with the press to inform them about any discoveries. He was keen to emphasise that his workforce consisted of many young people as well as both men and women, thus presenting the image of archaeology as a modern and advanced discipline. According to the later historian Adam Stout, the Maiden Castle excavation was "one of the most famous British archaeological investigations of the twentieth century. It was the classic 'Wheeler dig', both in terms of scale of operations and the publicity which it generated."

Wheeler's excavation report was published in 1943 as Maiden Castle, Dorset. The report's publication allowed further criticism to be voiced of Wheeler's approach and interpretations; in his review of the book, the archaeologist W. F. Grimes criticised the highly selective nature of the excavation, noting that Wheeler had not asked questions regarding the socio-economic issues of the community at Maiden Castle, aspects of past societies that had come to be of increasing interest to British archaeology.  Over coming decades, as further excavations were carried out at the site and archaeologists developed a greater knowledge of Iron Age Britain, much of Wheeler's interpretation of the site and its development was shown to be wrong, in particular by the work of the archaeologist Niall Sharples.

Back to the latest news story.  To quote from the Mail:

......... Dr Miles Russell, Professor of Archaeology at Bournemouth University believes the battle tale is ‘misleading’.   He said: ‘Most archaeologists know there is absolutely no evidence for such a “great battle” at Maiden Castle, a site which in any case had been largely abandoned a century before the arrival of Rome… another case, I guess, of not wanting an epic myth to be slain by ugly facts.’ Sir Mortimer wrote the first report on Maiden Castle in 1943. Dr Russell, in the latest volume of the Oxford Journal of Archaeology, said that since then ‘countless books, papers and television documentaries have treated a speculative Roman assault upon the hillfort as definitive fact.  ‘The account of a furious but futile defence of property, family and land by the local tribe of the Durotriges, leading eventually to their slaughter or enslavement, is undeniably powerful and remains one of the more potent stories relating to the demise of British prehistory.’

We now know that there were only 52 bodies in the graves, and fewer than 40 of those had signs of traumatic or severe injuries which might have been sustained in conflict. The graves were in fact quite carefully prepared, and radiocarbon dating now shows that the corpses held within them were placed in position over a 50-year period, ‘suggesting the population had lived through multiple periods of stress, competition and conflict’.

Since Sir Mortimer did not have access to sophisticated dating techniques in 1943, some forgiveness is in order, but there is nonetheless a feeling that he was one of the "glamour boys" of the archaeological establishment at the time, and that notoriety and press coverage were, in his mind, perhaps more important than fastidious research and cautious interpretations of digging results.

There are close parallels with the behaviour (and the reputation) of Prof Mike Parker Pearson today -- a man with a high media profile, intent on popularising and promoting his discipline at every opportunity.  For Sir Mortimer, read MPP:  "..........He was keen to emphasise that his workforce consisted of many young people as well as both men and women, thus presenting the image of archaeology as a modern and advanced discipline."  As we all know, MPP has an unfortunate tendency to develop narratives that are wildly out of kilter with the evidence unearthed during his digs.  (Like Sir Mortimer, he is based at the London University Institute of Archaeology.)  Like Sir Mortimer, he displays a tendency to ignore inconvenient evidence and scientific opinion.  His theories about bluestone quarrying in North Pembrokeshire, the human haulage of monoliths to Stonehenge, and the development and dismantling of "proto-Stonehenge" on the flanks of Mynydd Preseli have received wide media coverage and very little critical scrutiny from other archaeologists -- but they are hugely unreliable, and are unsupported by any solid evidence.  

Over-enthusiastic storytellers whose archaeological digs are conducted in an unprofessional manner may enjoy short-lived notoriety, but they are always -- eventually -- exposed either as charlatans or as incompetent field workers by younger and more careful professionals, and by the use of new research techniques. 

Watch this space......

Nadolig Llawen!

For Christmas this year, one of my paintings.  A happy Christmas and jolly New Year to everybody, and thanks for all your stimulating -- and sometimes irritating! -- contributions.

2019 has not been one of the best years in living memory, with much nastiness and divisiveness in evidence across the nation -- let's hope and pray for a more peaceful and productive 2020!

All the best

Brian

The Mynachlogddu bluestone quarrying experiment

Back in prehistory -- in the spring of 1986, to be precise -- I was paid a few quid by a Japanese TV company to conduct a scientific experiment into a quarrying technique that might have been employed at Carn Meini or somewhere else during the quest for bluestone monoliths.  This was the fire and water method  -- heating the rock with fire and suddenly cooling it with cold water.  (This all presumes that people wanted to QUARRY monoliths from the living rock rather then picking them up from the ground surface, where they lie in profusion -- but no matter.  Science is science, and a few quid always come in handy.....)

We brought in a huge pile of mature dried logs, built the funeral pyre, and placed the stone (which was as big as we could manage without a JCB) in the middle of it, and lit the fire.  Up it went in flames, and there were several loud cracks from the midst of the inferno.  It was several hours before the fire had died down into a pile of glowing embers and before we could approach close enough to throw bucketfuls of cold water onto the stone.  That was rather dangerous, and we were not sure whether splintered fragments of spotted dolerite might shoot off in all directions. In the event, nobody was injured, and when the scene of the experiment had cooled off enough, we recovered the stone and found that it had indeed been fractured in various (unpredicted) places by the heat of the fire -- and maybe by the water treatment as well.  Anyway, the Japanese TV crew went off reasonably happy -- the bonfire was very spectacular, anyway.......

Carn Meini (Carn Menyn), thought by Thomas and Atkinson to have been a major source for the spotted dolerites at Stonehenge.  It was suggested many times that fires against the rockface would have been set in order to "extract" suitable bluestone pillars........

Lessons?

1.  If this sort of method had been used in a "bluestone quarry" to break off monoliths from the rock face, the amount of timber required would have been phenomenal, and huge amounts of labour would have been needed as well.

2.  The fracturing of the stone was anything but predictable, and I doubt that the process would have been controlled enough to have extracted "marked pillars" with selected dimensions.

3.  Throwing cold water in large quantities onto the heated rock would also have been very difficult, given the residual heat of the fire and the dangers in getting too close too early. First, the water had to be carried in containers up to this difficult location.  A bucketful of water at a time is all that one could reasonably manage -- and throwing water into the glowing embers from above (from the clifftop) or from below would have involved great hazards. 

On balance, the logistics do not really make any sense -- especially since there are pillars and slabs of dolerite lying all over the place across the upland landscape.

That was one of my contributions to experimental archaeology...... actually it was quite good fun.......

A lesson from prehistory

Well well -- now here is a funny thing.  I have taken a bit of stick from assorted Stonehenge "experts" for daring to float the notion that Stonehenge was never finished, and that the builders simply ran out of stones.  I was looking for some photos on my file (nothing to do with Stonehenge) when I happened upon this extract.  No idea whose words these are, or what the publication was, but it all sounds perfectly sensible to me........

The Stonehenge bluestones -- pontification is not a substitute for evidence

There has been quite a discussion about bluestones and glacial transport on on Austin Kinsley’s Facebook page, starting on 1st Dec 2019, with many comments from the great and the good. I won't use any names here, because that would make the discussion rather personal, but below we itemise and discuss the seven points initially posted by somebody who presumes to know quite a lot about Stonehenge and about the supposed human transport of the bluestones.

As we know from our ongoing discussions over the years, on and off this blog, there are still people out there who are prepared to take all this human transport mythology seriously. The trouble is that many of them don't know the territory and tend to read things rather selectively........ as Tony has pointed out in the discussions on Auston's page. And let's remind ourselves -- there is not a shred of hard evidence in support of the idea of long-distance human transport of bluestones from Wales to Stonehenge, either by land or sea.

So here we go:

1. Where's all the other bluestones on the Salisbury Plain? Surely a glacier wouldn't have brought only the precise number of them required at Stonehenge -- numbers that are significant to the structure. The various other examples of full-sized stones in the area have clearly been robbed from the site over the centuries.

Response: this is an absurd point, and I cannot for the life of me understand why people are still making it, after all the advances of recent years. Forget the "immaculate Stonehenge" with 82 bluestones and 82 sarsens. There is no evidence that Stonehenge was ever completed.  I'm not alone in saying this.  As Cleal and others have pointed out over and again, we have no idea how many bluestones or sarsens were originally present at the site. As far as the bluestones are concerned, we know there are 43 -- represented as boulders, pillars, slabs, and stumps. There may be a few more stumps, not yet discovered.   We know that they were moved about and repositioned many times.  It is far more logical to suggest that there never were enough bluestones to finish "the Stonehenge project", and that after playing around with assorted stone settings over many centuries, the builders (maybe the descendants of the originators) just gave up and walked away......   This argues for the use of glacially transported erratics which were systematically collected up and used -- until there were none left in the Stonehenge landscape.

2. While sampling the various types of stone isn't allowed at the site, the volume of chips found in the wide area would easily correspond with the presumed number of present and missing stones. There's lots of visual evidence that points to stone-bashing throughout the life of the monument. See also the 'Stonehenge Layer'.

Response:  The suggestion that the "volume of chips" somehow corresponds with the presumed number of missing stones does not survive a moment's scrutiny. Some mathematics please!  What is the volume of chips that we are supposed to accept?  Remember -- we have no idea how many "missing or destroyed" monoliths there may be. What would be the total mass or volume of those hypothetical missing stones? Figures given by Ixer and Bevins for "debitage volumes" are seriously (and deliberately?) misleading, since they are based on a limited number of excavations and sampling exercises.  About 50% of the surface area within the stone settings has never been excavated, and until that area has been investigated speculation about "debitage volumes" is meaningless.  In any case, we have no idea how much of the debitage may have come from smaller stones -- mauls, hammer stones, packing stones and scattered small erratics -- rather than from destroyed monoliths.

3. The vacant stone-pipe at Rhosefellin more than suggests it was removed by humans -- no glacier would select a single example from that face and leave the others intact. Despite what Dr Johns says, the site was a well-used quarry from as far back as the Mesolithic. The petro-chemistry of that pipe matches Bluestone-44, that stone having been sampled before the present rules applied.

Response:  Ah -- Rhosyfelin!  A lovely spot.  Right on my doorstep.   Now we are seriously into the realms of fantasy.  "The vacant stone pipe" or "monolith extraction point" (as MPP likes to call it) does not exist.  It is a MPP invention, conjured up out of thin air by the good professor when he was told that some of the debitage at Stonehenge was made of a foliated rhyolite which came from within a few square metres of a geological sampling point at the tip of the Rhosyfelin rocky spur.  That piece of "spot provenancing" is hotly disputed, as described in my book -- and I do not know of a single independent geologist or geomorphologist who accepts it.  And of all the senior academics who have visited the site with me, not one of them accepts that the configuration of the rock face at the "extraction point" gives any indication that a bluestone monolith was taken away from here.  On the contrary, detailed surface research shows that there have been several phases of rock breakage off the face, controlled by intersecting fractures, over many millennia.  Some of the fracture edges are fresh, and others are very abraded and weathered.  See this post:
https://brian-mountainman.blogspot.com/2018/04/that-famous-monolith-extraction-point.html

The so-called "extraction point" at Rhosyfelin.  It has had a complex history, with small chunks and slabs of rock breaking off on many different occasions.  In this respect this site is no different from any other part of the rock face. Click the image to enlarge.

On my blog, if you enter "monolith extraction point" into the search box, you will find other entries as well.

The assertion that this was a well-used monolith quarry back into the Mesolithic is, as Mr X knows perfectly well, hotly disputed in my book and in two peer-reviewed scientific papers which he and all of the MPP team know all about, but have studiously ignored.  This is not just bad manners -- it is bad science too, on the part of a group of academics who now have so much invested in their spectacular quarrying hypothesis that they cannot abide the thought of anybody questioning -- let alone disproving -- their "evidence".

The bit about bluestone 44 is an invention, as others have pointed out.  Very careless.  The writer should have checked his facts more carefully.

Brian John, Dyfed Elis-Gruffydd and John Downes (2015a). "Quaternary Events at Craig Rhosyfelin, Pembrokeshire." Quaternary Newsletter, October 2015 (No 137), pp 16-32.
https://www.researchgate.net/publication/283643851_QUATERNARY_EVENTS_AT_CRAIG_RHOSYFELIN_PEMBROKESHIRE
https://www.academia.edu/19788792/Quaternary_Events_at_Craig_Rhosyfelin_Pembrokeshire

Brian John, Dyfed Elis-Gruffydd and John Downes. 2015. OBSERVATIONS ON THE SUPPOSED “NEOLITHIC BLUESTONE QUARRY” AT CRAIG RHOSYFELIN, PEMBROKESHIRE". Archaeology in Wales 54, pp 139-148. (Publication 14th December 2015)
https://www.academia.edu/19788912/Observations_on_the_supposed_Neolithic_Bluestone_Quarry_at_Craig_Rhosyfelin_Pembrokeshire 

https://www.youtube.com/watch?time_continue=9&v=qBTEcByYUME&feature=emb_logo

4. In the 1920s HH Thomas was wrong about a possible source being Carn Meini -- they are now known to include Carn Goedog and probably Bedd Arthur. There is a lot of archaeological evidence surrounding these sources.

Response: The geological evidence does indeed suggest that Carn Meini (Carn Menyn) was probably not a source of Stonehenge bluestones -- although Profs Darvill and Wainwright (the latter sadly no longer with us) refused to accept the geological evidence.  The dolerites (spotted and unspotted) are now thought to have come from other outcrops including the Carn Goedog sill -- but most definitely not Bedd Arthur.  The latter is not a rock outcrop but a stone setting including small locally-derived monoliths; nobody has ever claimed that it was a source for Stonehenge monoliths.   More care, please.

5. There is a marked difference in size and shape between the outer bluestone ring and the inner horseshoe. This strongly suggests they arrived at different times -- the outer ring almost certainly near-original, with the taller versions being installed after the Trilithons went up, much much later. How likely is it these were collected from deposits of the near-environs in such precise order?

Response:  I would agree that the bluestones in the outer ring and the inner horseshoe look different.  In the former, there are pillars and slabs, some of them shaped quite carefully.  In the latter the stones are mostly rough, weathered and abraded boulders which look and feel like an assortment of glacial erratics.  Most of them would not look out of place near the snout of a modern Arctic glacier.  I strongly disagree that the stones arrived at different times, as a result of two distinct stone-collecting expeditions.  There is no evidence to support that contention.  There was no "precise order."   I agree that the stones have been rearranged many times, and my reading of the evidence is that in the last re-setting the "best" of the bluestone assemblage (including the tallest and most elegant pillars) were selected for the horseshoe, and some of them were carefully worked and embellished.  I have no problem with these selected stones being more "special" or more "revered" than the others.

6. I'm pretty familiar with the work of Rob & Richard -- Rob being a personal friend, as you know. Neither has published anything that suggests agreement with the glacial transport theory, while both privately scoff the idea. See also Rob's recent work on the Altar Stone.

Response:  I admire the faith that Mr X has in Ixer and Bevins.  Neither of them has published anything in support of the glacial transport hypothesis because they are both embedded in the MPP team that has set its face steadfastly against the use of any geomorphological evidence.  Sadly, it appears that neither of them has any understanding of glacial processes, or of the literature relating to the glaciation of the western parts of Britain.  In all of their articles they have not once cited the 2015 articles written by me, Dyfed Elis-Gruffydd and John Downes.  Quite frankly, that is reprehensible.  If they had bothered to read the abundant research articles which I have been citing on my blog for the last ten years or more, they might have learned something.  I'm sad that they scoff in private; I would prefer a reasoned academic debate.  On the matter of the Altar Stone, much has been written by Ixer and his colleagues, but the origin of the stone has still not been resolved.

https://brian-mountainman.blogspot.com/2019/12/stonehenge-more-on-devonian-sandstones.html

7. Show me evidence of a glacially entrained Welsh Bluestone south of Bristol.
The idea of glacial transport has been thoroughly examined and found to be implausible. It's not the conspiracy of prevailing thought -- it's very well established.

Response:  If you don't mind me saying so, that is an arrogant and dismissive statement put out by somebody who does not know the literature.  Can Mr X please tell us which experts have found the glacial transport idea to be implausible?  He can take it from me that among the geomorphology / glacuiology experts qualified to consider the matter, there are far more who consider glacial transport to have been perfectly feasible, as against two (Scourse and Green) who take the "implausible" view.  Mr X needs to do some enlightening Christmas reading.  On the matter of glacially entrained bluestones south of Bristol, he just needs to buy a copy of my book -- or if he wants to read something more "authoritative" he could always go to the official Geological Conservation Review volume for SW England, written by a multitude of expert geomorphologists, who lay out all the evidence for Quaternary glaciation and erratic transport in the Somerset area.
Campbell, S., Scourse, J.D., Hunt, C.O., Keen, D.H. and Stephens, N. (1998) Quaternary of South-West England, Geological Conservation Review Series, No. 14, Kluwer, London, 439 pp.

It appears to me that Mr X, like most of those whose advice he takes, pretends that there is no dispute in progress, and refuses to engage in constructive debate.  Instead, he simply insists on repeating ad infinitum a parallel and very tired narrative that happens to be subscribed to by just a small group of academics who have very powerful vested interests and who seem to be trapped in some fantastical Neolithic bluestone quarry..........

If academics and other Stonehenge enthusiasts are not prepared to have their ideas scrutinized by others, they should be doing something else with their time -- and stop wasting ours.

The Stonehenge sandstone bluestones

To summarise the contents of earlier posts, I am now convinced that in the Stonehenge debitage (as explored in excavations across about 50% of the ground area within the stone settings) there are traces of at least four different sandstones.  Although Ixer and his colleagues are understandably reluctant to admit this, the evidence presented in their "sandstone" papers does NOT suggest one Devonian sandstone source and one Lower Palaeozoic source for bluestone material at Stonehenge.    We still do not know where the Altar Stone came from, or stump 40g, or stump 42c.    Mill Bay and the Cosheston Sandstone appear now to have been eliminated from the frame, and that represents good progress!  Although the papers contain accurate descriptions of a number of sandstone samples from assorted collections, we are short of tabulated information and short of graphic representations which might help the verify the suggestions of the geologists involved in these studies.

So there are at least two Devonian sandstones and at least two from the Lower Palaeozoic (Upper Ordovician or Lower Silurian).  It is quite possible that none of the rocks examined has come from Pembrokeshire -- more research is needed to work out where they have come from.  Possibilities include the Ordovician outcrops of the Tywi Valley, and the Senni Beds on the fringes of the South Wales Coalfield.

As I have said before, the more detailed the research on the geology of the bluestones becomes, the greater the number of convincing provenances are identified or suggested.  At the same time the idea of glacial entrainment and transport becomes more and more convincing, as the idea of Neolithic bluestone quarrying in West Wales appears more and more absurd.

Welsh Ordovician (?) sandstones at Stonehenge

This is the biggest chunk of Lower Palaeozoic sandstone yet found at Stonehenge.  Photo courtesy Rob Ixer.

In a previous post I looked at the evidence for the provenances of the "Stonehenge sandstone bluestones" (including the Altar Stone) which appear to be of Devonian age.  Let's now take a look at the evidence for the provenancing of other sandstone fragments that appear to be much older -- from the Lower Palaeozoic.  See this too:

https://brian-mountainman.blogspot.com/search?q=The+petrography,+geological+age+and+distribution+of+the+Lower+Palaeozoic+Sandstone+debitage+from+the+Stonehenge+Landscape+

See also the following paper:  ‘No provenance is better than wrong provenance’
https://www.academia.edu/41105834/Mill_Bay_Milford_Haven_and_Stonehenge
For reasons that are obscure, this paper has two titles, one formal and the other informal.  (It's on Academia, a web site that seem to enjoy spreading confusion......)

Back to 2017. Re this paper:

"The petrography, geological age and distribution of the Lower Palaeozoic Sandstone debitage from the Stonehenge Landscape."
by Rob Ixer, Peter Turner, Stewart Molyneux, and Richard Bevins.
Wiltshire Archaeological & Natural History Magazine, vol. 110 (2017), pp. 1–16

https://www.academia.edu/32048879/LOWER_PALAEOZOIC_SANDSTONE_DEBITAGE_FROM_THE_STONEHENGE_LANDSCAPE_The_petrography_geological_age_and_distribution_of_the_Lower_Palaeozoic_Sandstone_debitage_from_the_Stonehenge_Landscape

Rob Ixer and his colleagues demonstrate that nearly all of the "sandstone debitage" (from excavated contexts)  in the Stonehenge area is of Lower Palaeozoic age, not Devonian. The use of palynology as well as detailed petrography (mostly thin section work) showing a "slight metamorphic fabric" is fascinating and quite convincing. This points to an Ordovician age.  The inclusion of microfossils and other older debris pushes the age of the sandstones (note the plural) towards the Upper Ordovician.

Quote: The Darvill and Wainwright May 2008 excavation within the Stonehenge circle found a total of 74 samples of Lower Palaeozoic Sandstone with a total weight of 10440.9g, representing 2.00% of the bluestone debitage by number but 24.55% by weight. Three of the heaviest 20 bluestones from all contexts are Lower Palaeozoic Sandstone.

Using palynology on some samples, the authors suggested that the biostratigraphical ranges and preservation of acritarchs suggest that they were derived ultimately from the Upper Ordovician of the Welsh Basin (Caradoc to Ashgill), or, possibly the Silurian if all the acritarchs are reworked.  This age was also suggested by the inclusion of other debris derived from lower in the Ordovician sequence.

The authors argue that the samples examined from the Stonehenge collections have come from the northern or north-eastern corner of Pembrokeshire, to the north of the Preseli Hills. The samples are not all identical, and seem to show at least two sandstone types. Quote: Although this might suggest two separate sandstone sources, the petrography of the two lithic samples (and indeed all of the other debitage samples) suggests that they are part of a single sourced lithology and the apparent age discrepancy is a sampling issue. It is hoped that further sampling of the Lower Palaeozoic sandstone will help to determine this.

I'm not sure what is meant here -- if the samples have come from "a single sourced lithology" they could of course have come from widely separated localities or provenances wherever that lithology outcrops -- and how can the "apparent age discrepancy" be a sampling issue? If there is an age difference between one sample and another, does that not mean that one sample is older than the other, given that there are always some statistical / confidence issues? Whatever the truth of the matter, it is clear that either the Devonian or the Lower Palaeozoic debitage needs to be matched with sandstone stumps 40g and 42c -- if those two stones really have provided some of the debitage fragments, more progress will have been made. And if they don't match, things get even more interesting………

From the abstract: "The lithology is believed to be from an unrecognised Ordovician (or less likely Silurian) source to the north or northeast of the Preseli Hills.”

For obvious reasons, the authors are keen to demonstrate that the fragments have all come from one source -- they are after all quarrying enthusiasts, and the more sources there are, the less likely it is that quarrying was involved. They do not make their case at all convincingly, and from my reading of the evidence, it looks as if there are at least two sources for Lower Palaeozoic fragments. Neither am I entirely convinced that the source area is to the north of the Preseli uplands. That assumption is based on the idea that since the igneous / volcanic bluestones at Stonehenge seem to have come from Mynydd Preseli, the sandstones must also have come from somewhere nearby. But if the Altar Stone and the other Devonian sandstone fragments have indeed come from the Senni Beds, maybe more than 100 km from Preseli, the same disconnect may well apply to the older sandstones as well. Ordovician sandstones, shales and mudstones outcrop across a very wide area, and there are many outcrops on the flanks of the Tywi Valley, which was the route followed by one of the largest outlet glaciers from the Welsh Ice Cap. It is perfectly possible that erratics of Upper Ordovician or Lower Silurian sandstone have been carried by Tywi Glacier ice down into Carmarthen Bay, where they were later picked up by the Irish Sea Glacier and transported eastwards towards Somerset and Wiltshire. This possibility has not been considered by Ixer and his colleagues.  Another possibility is that the sandstone erratics have come from North Wales, and that they were carried westwards into Cardigan Bay by outlet glaciers from the Welsh ice cap before later incorporation into the southward-flowing Irish Sea Glacier.

Generalised geological map of Wales. The Ordovician outcrops are shown by the light mauve colour -- and by "Ord" on the map. Most of the rocks (sedimentary and igneous) are of Ashgill age, but there are many other outcrops of Caradoc, Llanvirn and Arenig age, particularly in north-central Pembrokeshire to the north of the Trefgarn Gorge. The igneous rocks of the Fishguard Volcanic Group are associated with Upper Llanvirn sediments, including the rocks of the Aber Mawr Formation, around 460 million years old.

My conclusion from looking at the three most relevant papers on the Lower Palaeozoic sandstones is that there are two -- and possibly more -- provenances involved, and that no evidence has been presented that would tie any of the provenances to north Pembrokeshire.

PS.  One last point. Could any of the sandstone fragments discussed in these papers have come from the sedimentary rocks in the vicinity of Carn Goedog, Carn Meini, Rhosyfelin or Waun Mawn?  It's extremely unlikely. The sediments into which the Fishguard Volcanics are intruded and within which they are interbedded are from the Aber Mawr formation of Middle Ordovician age, attributed to the Llanvirn -- around 464 million years old.  If the sampled fragments have been correctly attributed to the Upper Ordovician, and are of Ashgill or Caradoc age, they must be around 450 million years old, or younger than that.  That points not to the uplands of Mynydd Preseli, but the the lower land along the North Pembrokeshire coastal strip.

The age of the Fishguard Volcanics in the Preseli area has been confirmed by Bevins et al (2016) as Darriwilian (c 460 - 465 million years) by the U-Pb zircon dating of five rhyolite samples.  See the report here:

https://brian-mountainman.blogspot.com/2016/10/new-dates-confirm-mid-ordovician-age.html

One of the rhyolite samples, thought to have come from Rhosyfelin, was shown not to have come from there, but from some other related foliated rhyolite outcrop as yet unidentified.  This reinforces the point which I have made frequently -- namely that both the igneous / volcanic monoliths and the sedimentary monoliths at Stonehenge, and the debitage fragments, have not come from carefully selected quarries but from multiple sources across quite a wide area.

https://www.academia.edu/34770710/U_Pb_zircon_age_constraints_for_the_Ordovician_Fishguard_Volcanic_Group_and_further_evidence_for_the_provenance_of_the_Stonehenge_bluestones