Dowsing for Roman Roads in Wales

Ian Pegler

"Seventy percent of all archaeology is done in the library" - Indiana JonesTM

Introduction

Map dowsing has been used successfully in the past to find archaeological relics and sites[1], but in the past I've never been able to get it to work to my own satisfaction. Part of the problem is always the difficulty of verification - I am not a qualified archaeologist and I know very little about excavation techniques. It occurred to me that finding a Roman road by map dowsing might be something that I could easily verify by library research (or so I thought!), so I decided to give it a go. I also decided that I would visit any site located by my map-dowsing to see if there was anything worthy of note and do as much follow-up work as I possibly could. This has proven to be a very fruitful exercise of value in its own right.

In previous articles I have demonstrated how a bit of digging around in a library can be of some help in verifying results from dowsing[2]. It can sometimes be more practical especially when physical digging is not an option. It is very often the case that you will not find definitive answers and this can be frustrating of course, however some aspects of historical and archaeological work are, by nature, speculative. Very often this speculative work can become verified as the passage of time unravels fresh evidence.

Where to start?

Tal-y-bont is a large village in Mid Wales, seven miles to the north of Aberystwyth[3]. A recently published book written by the villagers contains a map which shows the area peppered with old mine-shafts. Some of these shafts date back to Roman times and some as far back as the Bronze age. During Elizabethan times there was an influx of miners from Germany and later on from Cornwall into the Tal-y-bont area. The mines continued to be exploited until the 19th century, when the industry fizzled out[4]. Knowing that there was some Roman activity in that region in ancient times, I decided to map dowse the area.

Map dowsing

I've been reading Dowsing One Man's Way[5] recently so I decided to try out one of J. Scott-Elliot's methods. His "detailed area search" involves keeping the pendulum steady and moving one's hand back and forth over the area to be searched in a "ploughed field" pattern. The idea is that the pendulum will seem to pull towards the object of search when the pendulum is near or over it. I had distinctly mixed results and eventually decided that "one man's way" was not the right way for me. I reverted to the "divide and conquer" method (appropriately enough for the ancient Romans!), looking at each of the squares on the map in turn. The pendulum gave positive responses for two squares containing a minor road to the south of the village, and I was later able to confirm by research that this is generally accepted as having been a Roman road[6]. For once I had got it right, or at least so it seemed.

I then decided to change my frame of reference slightly and look for undiscovered Roman roads, i.e. those roads which have not been rediscovered in modern times. The pendulum indicated that there were such roads to be found somewhere on the map, so I examined each square in turn, and subdivided those squares where I got a positive response, to narrow my search. Eventually I came to focus on two small squares which seemed to indicate a short, straight length of footpath.

Myth-busting time

Everyone "knows" that the ancient Romans always built neat, metalled (i.e. surfaced) roads which always ran in dead straight lines, with ditches either side and a raised centre. Unfortunately this isn't quite true and in a mountainous, rugged land like Wales this would not always have been an option. Neither is it true that the roads were all built in the same way. On boggy terrain different techniques would have been employed such as laying down straight beams of wood across the track below the foundation[7].

The purpose of the road would also have been an important factor. A main road for transporting troops over long distances would have been built with far more care and attention than a minor track for accessing a mine-shaft. However, from a purely archaeological view-point, even a minor track might be of some interest if it could be shown to be ancient, and might even open up the possibility of more finds.

Ground-work

The centre of Tal-y-bont has a quaint old-world charm with its village green overlooked by two large inns - the Black Lion and the White Lion - which stand side by side. At one time a building belonging to the Temperance movement stood incongruously between the two. Eventually this small building was squeezed out of business, rather like a puny wrestling-referee that dared to get between Big Daddy and Giant Haystacks![8]

A minor road from the village green takes you towards Allt-y-Crib woods that lie to the west. This is now Forestry Commission land, evidenced by the regimented planting of deciduous trees on one side of the track and conifers on the other. There they stand - in neat ranks and files like Roman soldiers poised for battle.

There was some walking to be done before I reached the site. All the while I dowsed with my nylon Y-rod but got no reaction. I noticed a considerable slate vein by the side of the road and pondered whether it might have been quarried in ancient times. What does your pendulum tell you?



Photo 1 - exposed slate vein


Eventually the single-lane track tails off into a footpath, and a right-fork takes you along parallel with a stream called Nant-y-Cwm. The path leads through a gate into some older woods until you finally emerge near a derelict cottage. An ancient holly tree stands nearby. Behind the cottage in the neighbouring field is a mysterious mound. The path leads across the stream and up a steep incline into open fields and by this time I was losing hope of finding anything. I crossed the stream and finally my Y-rod started to get lively. I reached the top of the brow and suddenly the footpath broadened out into an eleven-foot wide, straight track lined with deciduous trees on both sides.



Photo 2 - the trackway


What was the purpose of this short stretch of road? Was it built in Roman times or more recently? It could merely have served as a horse-and-cart track to access the cottage at the bottom, but there was no sign of an out-house, neither did the track lead into an area of the field where there might have been one. If my track was Roman it could have been used for accessing mines or quarries in what is now the wooded area behind the cottage. The track stretched as far as a gate, but it became an impassable quagmire, even with stout walking boots, so I could not progress any further.

Testing the track

I used my home-spun geodetic rod to test for the presence of "track-lines" and there were none. I then decided to test the road using my long-cord pendulum[9].



Photo 3 - testing the track with a long-cord pendulum.


For those readers who may not be familiar, long-cord pendulum dowsing differs in its modus operandi from ordinary pendulum dowsing. Instead of asking yes/no questions, the cord - which must be at least forty inches long - is wound around a spindle or windlass. The dowser lowers the pendulum over the object in question by unwinding the windlass slowly with thumb and forefinger. At some point the pendulum will start to rotate. After a certain number of rotations, the pendulum will switch to a back-and-forth swing. The length of the cord and the number of rotations taken together give a "rate" which is characteristic of the object in question.

I take rates as being personal, and the road gave a rate that was the same as my rate for quartz. I did in fact find a massive quartz lodestone in the wall later on. There seemed to be a lot of quartz around. Using the pendulum in the more conventional manner I dowsed the age of the track to be about 1600 years old.

Initial research findings

I have spent some time researching this project. This has taken the form of:
  1. Looking at books and surveys to see if the track is already recognised as Roman.
  2. Looking at old O.S. maps, tithe and sale-maps to see if the track was marked on the map and to see if it had a name.
  3. Examining aerial photographs.
So far I have discovered that:
  1. The track dates back to at least 1837 [10].
  2. There was no out-house near the bottom of the field marked on even the largest scale maps[11].
  3. There is an old quarry directly in line with the road, behind the mound mentioned earlier, in the neighbouring field[12].
  4. The area was thought to be mineral rich, especially in relation to lead[13].
I contacted the Royal Commission of Ancient and Historical Monuments in Wales and made an appointment to visit the library, based in Aberystwyth. I was able to view aerial photographs of the field dating back to the 1940s and 50s. One of these photos - taken on May the 2nd, 1946 - seems to show a footpath connecting the trackway to the quarry area behind the cottage, but it is very difficult to make out. I am now even more certain that there was never an out-house towards the bottom of the field as the photos show a short line of trees which would have made it tricky for a horse-and-trap to manoeuvre around.

In the National Library of Wales I found a Roman-road survey[14] which covered the whole area, produced by John Rigg who worked for the Ordinance Survey for seventeen years as an archaeological field investigator. This survey consists of a large number of black-and-white maps covering North and Mid Wales, a few accompanying photos and written information. The survey utilises colours to highlight the roads using the following key:



"RED : Actual Roman remains; surveyable course or strong air photo evidence.

VIOLET : Certain course which has no reasonable alternative.

YELLOW : Preferred course where options exist.

GREEN : Physical barriers. 'No go' areas."[15]



The main road (A487) to the north of Tal-y-bont is coloured in violet. The minor road to the south, which Davies and Kirby[16] say is a Roman road, is coloured in green. The main road (A487) to the south of Tal-y-bont is coloured in yellow. There is not a scrap of red on the map. In fact, there is very little red in the whole of the survey. One wonders how John Rigg drew so many conclusions from so little physical evidence?! Would this be admissible in a court of law? I wonder why scientists can get away with this kind of thing and dowsers can't?

By favouring the main road over the minor road as being the true Roman route, Rigg is flying in the face of local knowledge and tradition. As I've mentioned elsewhere[17] this is something that the archaeologist (or dowser) does at his peril, but on the other hand he might be right to have concerns about the traditional attribution in this case.

Dating and purpose

At a meeting of the West Wales Dowsers I showed photographs of the track to one of our most experienced members, Rob Hemming. Rob dowsed the age of the track to be about 1800 years old - a similar result to my own. He suggested that the track was not Roman built, although his dating puts the road in the Roman era. He also said that the road was used as a drover's road but that this was not its original purpose, which he felt was to connect places. Terry Wilson also did not think it was a Roman road.

Contacting Archaeologists

As well as contacting the RCAHMW (as mentioned above) I have also e-mailed Cambria Archaeology (the Dyfed Archaeological Trust) but so far they have not replied.

I have successfully made contact with the Aberystwyth Archaeology Society and have attended one of their meetings. The historian David Palmer gave a talk on the Romans in Wales. I showed him the photos of the track. Unlike Rob and Terry he was far from dismissive. It was pointed out that the raised ridges/walls running along side might be later additions and that I need to survey a cross-section of the area to see if the central portion of the track is raised in relation to the surrounding fields. Even if this turns out not to be the case, it doesn't mean that the track isn't Roman. If it is raised then it becomes a much more likely candidate for a Roman road. Note however that not all raised roads are Roman either[18].

Surveying method

This is a method of surveying tracks to discover if they have the right profile which would suggest (but not prove) Roman origin. It's simple to do, the equipment required is fairly inexpensive and two people could do it in about an hour. Since the method is non-invasive, dowsers or amateur archaeologists could do this themselves, with permission from the land-owner or farmer.

You will need: Since you will be using a laser-level out of doors it is best to do this when the weather is dry but not too sunny.



Figure 1 - Surveying using a laser level


The laser does not have to be absolutely level because all the measurements will be relative to each other, however the tape measure needs to lie as flat as possible. The laser level should not be moved once it is in position. The idea is to examine a number of positions across the track (noting these horizontal positions from the tape measure) and to read off the points at which the beam hits the metre rule, which must be held vertically so that the bottom just touches the surface of the track. This makes it possible to work out the relative height at that position. These measurements can then be plotted on a graph which will illustrate the cross section of the road. Use the X axis for the horizontal positions, and the Y axis for the vertical readings. The Y-axis can be exaggerated if needs be without invalidating the results.

The metre rule should be oriented so that zero is at the top. That way the more raised positions will give you larger numbers and your cross-section will not be upside down!

Note that the ditches on either side of a Roman road can become filled up with soil and stones over time. Also the raised central section of the road, known as the agger, can become flattened through the action of walkers, farm animals, ploughing and other machinery. Hopefully though, you might end up with a graph that looks something like this:



Figure 2 - levelled cross section


If your graph does look like this you may be in luck. Note however, that some later roads emulated the style of the earlier Roman ones, so having the characteristic cross section with two ditches and a raised centre is no guarantee of a Roman road.

If you have some photos and a levelled cross-section like the one above then this might help you in persuading your local archaeological society to take an interest.

Surveying the track will be my next step, which will depend on whether or not I can get the requisite permission of course. I will need to subtract the walls on either side from the "equation", in order to get a good reading for the road.

But is it Roman?

That's the big question and I shan't know the answer to that one for some time. Assuming for a moment that my map-dowsing is correct, then what function might this road have served? One possibility is that it literally served no purpose at all. It was common practice for Roman troops to build small practice roads to keep them busy. However, my best guess is that it was to access mine-shafts and quarries in the area. Most of the mine-shafts listed in a database held at RCAHMW seem to be labelled as "Post-Mediaeval", but in truth it is very difficult to date mines, especially if they become re-used in later eras.

The trackway is on a slope, which means that the surface of the road would probably have been something like gravel. The reason for this is that brakes on wagons were poor or non-existent, so the gravel surface would have been employed to slow the cart down. Such a surface would have been destroyed over time.

There was a Roman fort at Erglodd, which is only 2 miles NNE of my track. It could be the case that any raw materials mined/quarried from the area would be taken up north. This being the case there might be more tracks waiting to be discovered to the north of the one I dowsed. If the road was associated with Erglodd, then that makes Rob Hemming's dating more accurate than my own.

Summary

Discovering Roman roads might be a very worthwhile avenue (pun!) for those dowsers wishing to get into archaeology. We've looked at a variety of dowsing techniques that might be employed for doing this type of work and even seen how to indulge in a little amateur archaeology without damaging a site. However, as far as my own investigations are concerned the work is just beginning. I may well fail to prove my road to be Roman but I feel I have already learned a lot from doing this project.

Acknowledgements

My thanks go to David Palmer and Peter Bewers of the Aberystwyth Archaeological Society for their input and support.

Postscript

The day after I submitted this article to the BSD office the BBC reported that archaeologists had discovered evidence of lead-smelting activity underneath a mediaeval track-way near Llancynfelin. The BBC described it as a "Roman Industrial Estate"[19] although dating on the ash had not been completed at the time. The discovery was made within about 200 yards of the Roman Fort at Erglodd. It's a funny old world.



[1] See for example http://www.kingcharles-wrex.co.uk/ for the search for the Blessing of Burntisland, a ship-wreck that was located by dowser Jim Longton.
[2] See my articles on Dowsing Ysbyty Cynfyn stone circle, published in the journals of the British Society of Dowsers (December 2002 and June 2003) and the West Wales Dowsers Society (June and August 2003).
[3] See http://www.tal-y-bont.org/ for more information.
[4] Ein Canrif / Our Century, various contributors, limited edition printed by Y Lolfa press.
[5] Dowsing One Man's Way, J. Scott Elliot, reprinted by the British Society of Dowsers.
[6] Cardiganshire County History, vol 1 P309. J. L. Davies and D. P. Kirby.
[7] A good book to read is Roads in Roman Britain, Hugh Davies, Tempus 2002.
[8] Two famous British heavyweight pro-wrestlers. Good book: The Wrestling, Simon Garfield. Faber & Faber 1996.
[9] Long-cord pendulum dowsing was pioneered by the late T. C. Lethbridge. c.f. The Power of the Pendulum, Routledge and Kegan Paul, first published (posthumously) in 1976.
[10] Sale-map entitled A plan of Penrhyn Gerwyn, Allt-goch and Fron Goch farms... dated 1837. [11] For example Tithe schedule Llanfihangel Genau'r Glyn, dated 1847.
[12] O.S. map, 6" to 1 mile, dated 1886.
[13] See A plan of Penrhyn Gerwyn, Allt-goch and Fron Goch farms... op. cit.
[14] Roman Roads in North Wales, Surveys and Photographs, by John Rigg.
[15] ibid.
[16] Cardiganshire County History, op cit.
[17] Dowsing Ysbyty Cynfyn stone circle, Part 2, op. cit.
[18] Roads in Roman Britain, op cit.
[19] See http://news.bbc.co.uk/1/hi/wales/mid/3786093.stm



© Copyright Ian Pegler 2004

Contact details:
22 Erw Goch
Waun Fawr
Aberystwyth
Ceredigion
SY23 3AZ
U.K.