Fforest yr Esgob

an archaeological survey of Deserted Rural Settlements.


Carmarthen Bar 

and its shipwrecks


Carmarthenshire Place-name Survey



The study of early and historic upland settlement in Wales

Deserted Rural Settlements have recently been the subject of detailed study by numerous archaeologists in Wales. In the last decade or so it has become apparent that the Welsh uplands contain a wealth of evidence in the form of simple habitations, many of which survive only as low stony earthworks. The siting is often in narrow valleys close to streams. Their simple form has lead some archaeologists to suggest that they are medieval in date - perhaps originating during the period of massive population increase in the 13th century. However it is equally certain that these uplands were again heavily settled during the late 18th and first half of the 19th century, as a result of pressures to grow more crops during the Napoleonic Wars and then as a result of the Enclosure Movement. In the mid 18 hundreds these uplands were almost exclusively used for keeping sheep . 

The Four Forests :: Fforest yr Escob alias ’Rescob Forest

Fforest yr Esgob lies in the corner of Llanddewi Brefi parish in SE Ceredigion right on the borders of Carmarthenshire and Breconshire. From at least the high Middle Ages it was part of the estates of the Bishop of  wpe1.gif (70441 bytes)St. David's. Today this beautiful and remote area is only partly covered by modern forest, and has large areas of moorland worked by three sheep farms. This was once a much more densely settled landscape of upland farmsteads, cottages and small lead workings. Some of the mountain sites were probably once transhumant habitations.  In the 14th century there were 15 timber buildings in the forest, and it is assumed that most of these were habitations used by upland farmers tending cattle. The forest, as well as producing honey, had a mill, a lead mine and could support about 240 cattle. The date-range of these habitations cannot as yet established, because simple earthworks or ruined stone structures, such as the one shown below, were still in use in the 1900s.

It must be emphasized that simplicity of form, apparent ‘antiquity’ or remoteness cannot be taken as evidence for great age - or of any age. However we have to assume that there is a high probability that the farms that survive, or were in use into the 20th century, may be on medieval sites, because these are the best from the point of view of aspect, shelter and closeness to good drinking water. One such is the site at Dinas. This appears to have a medieval hall at its core. The site was abandoned by 1891.

     By comparing evidence for ownership and tenure shown in maps of 1812, 1817 and 1838 with information contained in the census returns of 1841, 1851 and later, it is possible to build up a picture of a proportion of the 88 or so habitation sites in the townships of Doethïe-Camddwr and Doethïe-Pysgotwr townships that were occupied in the 19th century. Many sites, like that on Banc Hendre’r Dail, are not recorded in any of the documents consulted so far.

Evidence shows that  large families were occupying the tiniest of cottages: it is now beyond doubt that even the simplest of earthwork sites can be late. Through a process of elimination it may be possible to suggest which deserted sites may be medieval. One of the four holdings in the ‘Four Forests’ -- Cnwch Eithinog -- has extensive evidence for cultivation at an altitude of 350-400m (1150-1300ft), which have been roughly mapped from oblique aerial photography. One of the associated habitation sites was occupied in the 19th century and its land used as a ‘sheepwalk’. We have no historical evidence for this cultivation - presumable for oats - it could be medieval or Napoleonic in date.

  Cnwch Eithinog also appears to have the largest number of habitation sites, a great proportion of which have not been identified in 19th century sources. This inability to connect monuments with historic sources does not mean that all these sites must be early; some are very likely 18th century. Throughout the study area a number of examples of what we might loosely call hendre/hafod relationships have been noted, where valley floor or major settlements in naturally protected locations are linked to tracks leading to simpler sites on the mountain. This is certainly the case with the four ‘caputs’ of the four forests and at least one other site in the Henfaes holding (arrows point from ‘hendre’ to ‘hafod)’.

Over 40  habitations ranging from simple single-celled rectangular earthworks to developed farmsteads with medieval halls at their core have observed in the two townships. Slight but convincing evidence for wickerwork chimneys in the pine end walls of standing stone ruins, which would have looked like the one depicted below, has been noted. By studying the plans of these, and the rich documentary and place-name evidence, it is hoped to explain the development of land-use in the forest. The richness of the sources, both archaeological and historical, may provide a sufficiently sound sample for drawing conclusions about typology and chronology. Howver it must be remembered that since wpe11.gif (61071 bytes)the function of many habitations did not materially change over many centuries it seems that the morphology of farms, houses and cottages had little need to change. The poverty of upland settlers probably increased in the 18th and 19th centuries; the raw materials, certainly wood, would have been far more scarce than in the Middle Ages, so that later cottages may very well have been cruder.

To find out more about Welsh archaeology including work on the Uplands, see  Cambria Archaeology and the Royal Commission's web site:  http://www.rcahmw.org.uk 

Home Page