Figure 1
Distribution map of clay dabbins on the Solway Plain (after Jennings, forthcoming).
Clay houses have been built on the Solway Plain since medieval times. In this paper the materials and construction of a typical medieval clay dabbin are described, and an attempt made to quantify the materials and man-hours required for its construction. This is compared with that needed for a particular type of Stone Age house..
Now that the wisdom of squandering the earth's resources is at last being questioned, it may be useful to examine some of the methods used for house-building in less prodigal times. On the Solway Plain clay houses, as durable as any mud houses in the country, survive from the Middle Ages and later. Known locally as clay dabbins or daubins, the first extensive survey of them was made in 1962 (Footnote 1). Others have followed, but they are still little known outside the immediate area (Footnote 2). The Plain stretches from the edge of the northern fells to the Solway Firth and for some miles east and north of Carlisle, and also along the Scottish border of the Firth. The map (Fig. 1) shows the distribution of dabbins known to the writer, while Ray Harrison has published a more extensive map showing the distribution of surviving mud buildings throughout the UK and Eire (Footnote 3).
Figure 1
Distribution map of clay dabbins on the
Solway Plain (after Jennings, forthcoming).
The earliest dabbins may have originated as single-storey longhouses in the fifteenth century, and both they and later examples have been adapted to serve the changing needs of their occupants and are still in use as comfortable working farmhouses. Unlike dwellings built today they were designed and built by the people who were intending to live and work in them, so that they are fit for their purpose.
The aim of this paper is to estimate the cost of building the clay dabbins. However, even during the relatively short period of history in this country when written accounts survive, and taking into account the effect of inflation, currency changes and changes in lifestyle make direct comparison in monetary terms impossible. As we have seen in our own time, a single cost-of-living index soon becomes outdated as consumption patterns change; candles no longer figure significantly in most household budgets but electricity bills do. To compare the cost of building a clay dabbin with the cost of a modern, cavity-walled, brick, three-bedroom semi, we must calculate the amount of work and materials which have gone into the building of the dabbins, since this is the only valid measure of the real, that is social, cost of building. From the Egyptian pyramids to the Mesopotamian ziggurats, from the pyramids of pre-Columbian Mexico to a Mongolian yurt or an Innuit ice-house, the cost must be estimated from the number of man hours required, not from the money paid out. Another valid comparison between the cost of medieval and modern buildings is the amount of maintenance required; the former were low-cost, high-maintenance buidings, whereas the latter are high-cost, low-maintenance. This is due to the change from labour intensive to capital intensive working. We could also consider the total cost, both erection and maintenance, over the occupiers' lifetimes as a proportion of their lifetime income (not necessarily monetary). But neither of these points will be pursued here.
So far as the writer is aware, a calculation of the man-hours required to build a vernacular house has not been published for other parts of the country. However, it has been undertaken for a particular type of historic building. W. Startin has attempted to reconstruct the houses of the Stone Age Linear Pottery Culture, to estimate the amount of materials used in their construction, and to calculate the man-hours required to build them (Footnote 4). These people are thought to have lived from the middle of the fifth millennium BC, originating to the north of the Hungarian Plain and eventually extending as far as the Netherlands. Startin based his reconstruction on the archaeologists' discovery of postholes and a small amount of wall planking and daub, together with a knowledge of the materials available to the builders in that area at that time, and the rather scanty information gleaned from surviving pottery models of houses of a slightly later period. The range of tools available to Stone Age builders is known; they had stone versions of most of the medieval carpenter's toolkit, including axes, adzes, augers and chisels, but no saws. Startin also used ethnographical data such as that from New Guinea, which studied the rate of working of native woodmen using the stone tools with which they were familiar (Footnote 5). Inevitably, as with reconstructions of much later Viking buildings (Footnote 6), there is a large measure of error, particularly when trying to deduce the roof structure. However, it is true to say that human beings the world over, faced with similar problems and with similar resources tend to find similar solutions; the writer has seen timber-framed houses in China not wildly dissimilar to those in Surrey.
Startin arrived at the figure of 2200 man-hours for his model, i.e., about one man-year, and we shall here attempt a similar calculation for the medieval Cumbrian clay dabbins.
We are on much firmer ground than Startin, as the model proposed is taken from standing clay and cruck-framed buildings which may go back to the fifteenth century. Our standard farmhouse stands on a cobble plinth 0.4m (1ft 4in) high, and its outside dimensions are 15m x 6m (49ft x 20ft). Its walls and ridge are 2.3m (7ft 6in) and 5.3m (17ft 5in) high respectively. The walls are 0.6m (2ft) thick, constructed of pebbles and sand in a clay binder. Unlike some houses of mass clay walling elsewhere in this country, the straw fibre necessary to distribute cracking as the clay dries out is added every 0.05m-0.08m (2-3in), as well as to the mix. The building is of five bays and has four oak cruck pairs, two side purlins and a ridge beam. The rafters are of cleft oak, about four to a metre, with no laths, and fixed with oak pegs. The covering consists of stapple thatch, the local style, about 0.4m (1ft 4in) thick. There is no loft, and the floor is of beaten earth. There is no chimney or firehood, but instead an open hearth in the middle of the floor. There are opposed entrances at the ends of a cross passage, and possibly a couple of small windows. Window openings would have been shuttered and unglazed. The materials in this house were nothing but oak, mud, cobbles, wheat straw and turf.
A number of clay-walled, cross passage houses survive on the Plain, and they fall into two groups. In the majority the firewall is a bay length from the nearest crucks, as one would expect; life was not so easy that our forebears would have wasted their energies putting up unnecessary supports for their roofs. Brunskill refers to this type as longhouse derivatives and they probably date from the late sixteenth century onwards (Fig. 2) (Footnote 7). However, eight houses have a firewall inserted into an existing bay (Fig. 3).
Figure 2
A longhouse derivative.
Figure 3
A true longhouse with inserted firehood, based on Orchard Cottage, Thursby.
As J. T. Smith has pointed out, this must mean that these were originally true, not derived, longhouses (Footnote 8), built in the style of the Middle Ages and housing family and stock in a single, undivided building open to the roof, with an open fire in the floor of the domestic end and no chimney. Tree-ring dating has given felling dates of 1489, 1491 and 1615 for the crucks in three of these houses. Another has reused crucks felled in 1382, -10, + 16 years. It is the usual practice to repair and patch the clay walls as necessary, and this can be observed during major repairs when the plaster has been removed, in barns with unrendered internal surfaces, and in derelict buildings. As Wrathmell remarked, it is the timbers, not the walls, which are the permanent part of the structure (Footnote 9). This is obvious in the case of framed houses, where not infrequently the original wattle-and-daub has been replaced with bricks.
Needless to say, not all the standing buildings conform precisely to the dimensions and features chosen, but they represent the majority. Cumbrian houses tend to be much more alike, both in plan form and dimensions, than those in many parts of the country. Most of the standing clay buildings of the Plain, including barns and cottages, have bays of remarkably similar length and width, although not all were five bays long. Recent dendrochronology has shown that medieval crucks are also found in some two- and three-bay farmhouses with no cross passages or adjoining byres, although these were added later. Stapple thatch was not universal; turf roofs have been found at Meadowbank, West Curthwaite, and at Hitchens Onset, Scaleby.
Apart from a few special cases such as Abbeytown, where hewn stone was readily available from derelict ecclesiastical or military buildings after the Middle Ages, clay dabbins are the only form of vernacular building to survive on the Solway Plain before the eighteenth century, and cottages and outhouses continued to be built this way until less than a hundred years ago (Footnote 10). The reason mud was used for the walls rather than timber or stone was of course the shortage of these materials; most of the Plain has been overlain by a thick layer of boulder clay since the last Ice Age (Footnote 11).
As was usual in the Middle Ages, all the materials for these vernacular buildings came from nearby. Until stone door and window dressings came into use, probably in the late seventeenth or early eighteenth centuries, all the materials and labour for the houses was local. Apart from the felling and shaping of roof timbers, and the collection of materials, all work was done on site. Just as today most fell farmers can build a drystone wall, so the skills needed to build mud walls on a cobble plinth would have been common knowledge. Work was done by neighbours and friends and no outside professionals were required, except possibly for making the crucks. This neighbourly tradition continues to this day with the communal running of the saltmarsh pastures.
In the Middle Ages villeins had certain rights to materials for the construction of their houses, hedges and ploughs, known as houseboot, hedgeboot and ploughboot respectively, as well as the right of turbary, the taking of peat for fuel. From the roaring inflation of the sixteenth century onwards, lords attempted by various means to increment their incomes and restrict the rights of tenants, but the latter continued to enjoy these rights until the abolition of copyhold in 1924. Thus from the Middle Ages onwards, the tenants had a right to free materials for house building, underlining the point that monetary value is not particularly relevant when calculating the cost of erecting a new house.
A cruck consists of two blades, usually curved, which carry the load of the roof. The earliest buildings were earthfast, but nearly all surviving examples rest on a stone plinth or padstone. The walls need not be load-bearing unless there is an upper floor. The blades are joined in various ways at the top where they carry the ridge beam. Figure 4 shows Alcock's type C, which is one of the most common apex arrangements (Footnote 12). At a lower level they are joined by a tie which may also support purlins.
Figure 4
Exploded view of a cruck pair
with Alcock’s apex type C (after Jennings, forthcoming).
Archaeologists have shown from the evidence of pollen analysis that timber has been in short supply on the Solway Plain since at least 800AD (Footnote 13). This scarcity continued into post-medieval times, as shown by the seventeenth-century records of Holm Cultram parish (Footnote 14). As a result, no more wood was used than was absolutely necessary. Except for obviously later modifications, no wall plates or sill plates are found. Few thatched roofs remain, but from the evidence of those stripped in the last few years it appears likely that medieval roofs did not have horizontal laths; with closely spaced oak rafters they were unnecessary. There are no gable crucks, and in fact the shrinkage of the clay on drying would make it difficult, if not impossible, to have full crucks actually in a clay gable where the quick-build method was used. If a building became derelict or outlived its usefulness, the crucks or purlins were often recycled in the replacement house or barn, as is indicated by the empty slots for halvings or mortices of the many reused roof timbers in standing buildings.
Damien Goodburn, of the Museum of London, has found that an axe and boring tool are sufficient for the construction of crucks (Footnote 15). A medieval farmer would have possessed these, but it is not known whether some or all of the crucks were in fact made by professional carpenters. Wills surviving from 1661-1750 include thirty-five carpenters from the Plain, as well as seven farmers who left carpentry tools, three of whom were yeomen (Footnote 16).
The plinths usually consisted of stones cleared from the fields or the shore. In non-coastal villages, such as Moorhouse, tradition has it that cobbles for plinths were the result of clearing the fields for ploughing, as was also the case with the drystone walls on the fells after the Enclosure Acts. Not all the stones are cobbles, as glacial erratics can also be found in boulder clay. In some places dressed stone was available from derelict churches or the Roman Wall, and in Thursby near Wigton, red freestone may have been used from a nearby sandstone quarry (Footnote 17).
Where early plinths have undergone recent repair, as at The Cottage, Newton Arlosh, it was found that lime mortar had not been used; instead the stones were bedded in clay.
Materials for the walls (mud, pebbles, straw or occasionally peat or reeds) came from nearby. The only tools used would have been farming ones such as spades. Farmers at Baldwinholme and Hayrigg Hall have remarked that there used to be a pond in every field, and that it was from these pits that the clay had come. In this connection it is interesting to note that the grandmother of the now retired blacksmith of Moorhouse used to relate that the clay for the village houses there was brought from Bank House two miles away, as the local soil was too sandy. Two miles with a horse and cart seems a long way to bring all the materials for a mud village, but perhaps only enough was brought to provide a suitable mix.
The method of building the walls was described in the late eighteenth century:
In the first place, they dig out the foundations of the house, and lay a row or two of stones, then they procure, from a pit contiguous, as much clay or brick-earth as is sufficient to form the walls and having a good quantity of straw, or other litter to mix with the clay, upon a day appointed, the whole neighbourhood, male and female, to the number of 20 or 30, assemble, each with a dung fork, a spade, or some such instrument. Some fall to working the clay or mud by mixing it with straw, others carry the materials; and 4 or 6 of the most experienced hands build and take care of the walls. In this manner, the walls of the house are finished in a few hours; after which they retire to a good dinner and plenty to drink which is provided for them, where they have music and dance, with which, and other marks of festivity, they conclude the evening. This is called a 'daubing' and in this manner they make frolic of what would otherwise be a very dirty and disagreeable job (Footnote 18).
Or again:
These houses are generally made up in a day or two; for, when a person wants a house, barn, etc. built, he acquaints his neighbours, who all appear at the time appointed; some lay on clay, some tread it, while others are preparing straw to mix with it. By this means, building comes low and expeditious: - and indeed it must be owned that they have brought the art of clay building to some perfection (Footnote 19).
The Reverend George Williamson mentions the practice in his diary of 1743 (Footnote 20), while Pennant describes a slightly different method used in Canonbie, Dumfriesshire (Footnote 21). The neighbourly custom of communal building survived in Brittany and other parts of Europe until recently (Footnote 22), and in country districts of Mexico to this day. In Cumbria it lasted into the twentieth century, probably as long as the practice of clay building itself (Footnote 23).
However, this 'quick-build' method was not the only one current, and some buildings were certainly constructed in several 'lifts', with time allowed for drying and shrinkage between successive lifts. Harrison noted that when the walls of Lamonby Farm barn collapsed they fell in half metre blocks, although they contained the usual thin bands of clay and straw, and he saw similar 'lift lines' in the now-demolished, clay-built, Green Lanes barn at Powick, Dalston (Footnote 24). At Midtown Cottage, Burgh-by-Sands, the stone back and front walls are bonded to the clay gables so the work must obviously have been allowed to dry between lifts. In a barn near Abbeytown the top of the gable is clearly of a different build from its supporting wall. Thus it would appear that two traditions of clay building survived side by side. The reason for this is not known; possibly the 'quick-build' method required more skill and experience, possibly the other way was more suitable if helpful neighbours were not available.
James Jackson was the bailiff of Holm Cultram estate. In 1662, when he built his 'new barn', twenty-eight friends 'made..[his].. new barn walls' and brought 'a great sile from Souterfield', i.e. a pair of crucks. But in 1667, at the building of his stone house, he gives long lists of gifts from neighbours in lieu of 'boon labour' or personal service, and his diary gives an account of payments to the stonemasons (Footnote 25). It would appear that the neighbours were willing to build in clay but not in stone.
Wheat stapple thatch is the local style, and is quite different from southern thatch. Grass or heather divots (similar to the lawn turf supplied by garden centres today) were laid in overlapping courses like slates, directly onto the rafters without laths. The bundles of straw about 1m (3ft) long were bent double (Fig. 5), and often tied at the bent end, as seen at Monkhill Cottage, Burgh-by-Sands. These were thrust, bent end first, between the divots (Footnote 26). Rope fastenings were not used. Whether there was a standard ridge finish is not known, as this may have been damaged when the protective sheeting on all the remaining examples was installed. Such evidence as remains suggests a capping of mud.
Figure 5.
Stapple thatch from Monkhill
Cottage, Burgh-by-Sands.
We have evidence for stapple thatching because a few houses have, or had until recently, their original thatch. Usually this is in poor condition, but protected by iron sheeting. Most of the turf divots on standing buildings are of grass, but at Howard's Cottage, Warwick Bridge, and probably at Baldwinholme Farm, heather divots were found. This must be a matter of availability since both houses are near sources of heather. Howard's Cottage is about four miles from Bampton Moor, and Baldwinholme is not far from the northern fells. Despite this, some haulage costs would have been incurred in both cases. The thatch itself was usually wheat straw, a by-product, although for those houses near moorland heather might be used and the method was somewhat different. The roofs have a much lower pitch than southern thatch in order to avoid the underthatch slipping off, and this has the advantage that less straw is need for the same thickness (Footnote 27). The method lent itself to patching as required.
Another variation was the use of turf as the main covering. At Hitchens Onset, Scaleby, and Meadowbank, West Curthwaite, the roofs have recently been completely stripped and replaced, and at both houses it was found that over the firehouse, the oldest part, the covering consisted of neatly laid layers of turf, grass side downwards, with only a skimming of straw on top.
No evidence of original rendering has been observed. In Devon, some of the older houses have a coat of mud plaster under later limewash (Footnote 28). In the Middle East today, for example in the Yemen, mud is the only rendering on adobe (mud-brick) houses. It seems reasonable to think that this was probably also the case on the Plain, which has no limestone for limewash.
The walls would not have been plastered until they had had time to dry out, a matter of weeks or months depending on conditions. The process would have needed repeating from time to time, so it comes under the heading of maintenance, and is therefore not considered here.
It is probably safe to assume that the medieval clay dabbins all had mud floors. Old Sandstead, in Burgh-by-Sands parish, was a sizeable forty-acre farm in the nineteenth century, and the marsh book (in the possession of the reeve) shows that in 1813 the sum of £15 was allowed by the marsh committee, in order to board the parlour floor and flag the kitchen; presumably the floors had previously been of mud. In order to reduce the dust the mud for the floors was bound with fresh bullocks' blood (Footnote 29).
No Viking houses or clay dabbins have been excavated on the Solway Plain, so we must rely on information from archaeological digs in other areas for information regarding the hearth (Footnote 30). Stone ones have been found in Viking houses, and during recent repairs the owners of High Hill Cottage, Scaleby, found what may have been a stone hearth about 2m (6ft 6in) from the present one. Another possibility is that, as in the Scottish 'black houses' which were open to the roof, the hearths were of clay, renewed every three years or so. This information was vouchsafed by the custodian of one, now a museum, at Arnol, on Lewis; she lived in this house as a child.
We are now in a position to estimate the man-hours and quantity of materials required to build our 'model' longhouse. We shall assume the working day was ten hours, and, since most of the material lay to hand, will not take account of haulage.
Cruck buildings have recently been constructed by Damien Goodburn and Ross Noble, and others have been re-erected at Ryedale Folk Museum and St Fagan's Museum, using traditional methods. There are also accounts given by local people with long memories, as well as a few contemporary descriptions. These will be used as the basis for calculations.
Ross Noble has described the experimental building of a cruck, turf-walled house at the Highland Folk Museum, Kingussie (Footnote 31). Trees with suitable natural curves were selected, and out of seven trees, four cruck pairs were made, including the eight collars and some of the purlins. The project was funded by the Manpower Services Commission in order to provide work for the long-term unemployed, so the labour was either semi-skilled or unskilled. Instead of machinery, adzes, hand axes, two-handed cross-cut saws, and augers were used. As the work progressed and experience was gained, the speed of working increased, and although the first four pairs of crucks and the purlins took nearly a month to make, he estimated that a fifth pair of crucks would probably not have needed more than fifteen hours or two days. Unfortunately, in the present context, they used pine not oak, as pine is much easier to work; so these times must be scaled up for Cumberland.
Another account from Scotland was written in 1885 by Robert Dinnie, a working mason, who describes building three or four-bay thatched farmhouses in 1765:
'The couples [crucks] were sometimes of whole trees, squared a little with the adze or axe, sometimes with trees cleft down the middle called half tree...',
both of which forms have been observed on the Solway Plain. He goes on to say:
'The doors were seldom above five feet six inches in height, and the windows about two or two feet six inches in width, which was considered a fair size, but many were smaller…six or seven days of a carpenter were sufficient to do all the woodwork of an ordinary farmhouse….The cost of a house of this description at the present time [1885] would not exceed £20, and about 100 years ago, £5, while several farmers' houses in this locality have recently cost £300 sterling (Footnote 32).
The owner of one clay dabbin, who has done extensive repairs to his house, but does not wish to be quoted on the subject of time taken, since this depends heavily on the quality of the timber, believes that finding suitable trees would have taken longer than felling and constructing the building. However, this choice would probably have been made a few years earlier. On the other hand, Damien Goodburn thinks that two experienced men could make a simple oak cruck pair in a week, including selecting the tree, felling it, and cutting up the waste for sale, and that two experienced men could make all the timbers for a two- bay house in three-four weeks (Footnote 33). However, this is for a fully-framed Welsh Border house, with end crucks, wall plates and sill plates. Given that clay dabbins are simpler and have fewer timbers, Goodburn's upper figure of two men for four weeks may be a reasonable approximation for constructing the timbers for a five-bay dabbin.
Some of the local crucks are obviously made from scraps, but for better quality work a timber might be cleaved in two, along its length. Four, possibly five, such trees would suffice to make the crucks for a five-bay house, together with the ridge beam, side purlins, window and door frames, shutters (glass would not have been available), and doors. We may reasonably assume that the trees would already have been allocated by the lord's bailiff, so that two skilled men should have been able to do all the cutting and fashioning of the timbers for a five-bay house in four weeks or 480 man hours.
The rearing of the cruck frames of fully-framed houses in Worcestershire has been discussed by F. W. B. Charles (Footnote 34), and his reasoning can be extrapolated for our clay dabbins. We may assume that for a standard five-bay house the cruck pairs, the ridge beam and the purlins would have been assembled elsewhere, disassembled, and brought to the site where the cruck frames would have been reassembled. That is to say, each of the four pairs would already have had its blades pegged together and to the tiebeam. The first pair would be laid on the ground and then its feet would be levered onto the plinth or padstones. The frame would then be raised to an upright position by means of shearlegs and a pulley. During the raising the pair/frame would be stabilised by men with ropes on each side. It would have been possible to use poles instead of ropes for stabilisation, but evidence for this in the form of notches in the blades has only been found in three examples on the Plain.
The ropes would then be tied to posts fixed in the ground, and the adjoining cruck pair treated in the same way. Using long forked branches the ridge beam would be lifted onto the top joint between the blades, and the purlins onto the tiebeams. The same sequence would then be followed for the next two cruck pairs. The purlins reach from only one bay to the next, and each succeeding one is laid on top of the previous one. As the ends are shaped to form very blunt wedges they may perhaps qualify as scarf joints. Usually, but not invariably, they are pegged to each other or to the tiebeam. If the ridge beam was not long enough to reach the full length of the house it was scarfed above one of the cruck pairs. The weight of the beams, together with the ropes, would give enough stability to enable men on ladders to do the pegging.
Two open-air museums have recent experience of raising crucks. At Ryedale Folk Museum machinery was not used and the task required well over thirty people (Footnote 35). At St Fagan's in Cardiff, twenty men were needed unless shearlegs were used; this reduced the number to five. Using shearlegs, five men re-erected the timbers of Hendre Wen barn: 'in one day the carpenters, probably with the assistance of anyone who was close at hand or could be roped in to help, would raise one pair of crucks…they would also, on the second and subsequent days, fix the purlins' (Footnote 36). This suggests that five men, probably working an eight-hour day, could have raised the timbers of a five-bay clay dabbin in five days or 200 man-hours, which would have been twenty man-days at ten hours a day.
In these northern roofs, the feet of the rafters rest on the outside of the top of the walls, aiding stability. Some rafters are full length, reaching from ridge to eaves, while others are half rafters, perhaps inserted later. The remaining evidence is inconclusive, but in most cases the rafters appear to have been pegged to the ridge and to the purlins. Four rafters are required per metre, and 120 full length rafters would therefore be needed for the two roof slopes. Cleaving, drilling of peg holes, and pegging into position, could probably be done at the rate of two rafters per hour, so this would require another 60 man-hours.
The construction of the cobble plinth may be estimated from the building of drystone walls, which are very similar and are still being maintained and rebuilt today. Building with rounded cobbles is not easy - the owner of Fairfield House, Moorhouse, likening it to doing a 3-D jigsaw puzzle. Modern figures for the length a single waller can build in a day vary between 4.9m-5.5m (5yd 1ft-6yd) (Footnote 37), and 5.5m-6.4m (6-7 yd)(Footnote 38) per day. A house plinth is lower but thicker than a drystone wall, and assuming that these factors cancel out, a house plinth of 39.6m (130ft), would take 6-8 man-days to build.
In the days of railway construction it was estimated that a navvy, using a pick, shovel and wheelbarrow, could dig fifteen cubic metres (19.6cu yd) a day. Since the walls of our model dabbin total 54 cubic metres (70.3yd3), one man digging 15 cubic metres a day would take 32/3 days (and leave a useful pond for watering stock), while using Jackson's account of building his new barn in 1662, and assuming that the quick-build method was used, we arrive at the figure of 28 man-days to put up the clay walls.
In addition to the clay, we may estimate that 635kg (1400lb) of straw would have been needed for the walls. Details relating to obtaining the straw are discussed below.
In Scotland, a good man could cut up to 200 divots a day (Footnote 39). It is a long and tedious procedure, and Ross Noble estimated that 1200 man hours are needed in modern times to cut about 226sq m (270sq yd) (Footnote 40). Bruce Walker points out that Ross Noble's men were using modern steel tools to cut turf from good farmland, so this figure must be scaled up (Footnote 41). Estimating 33 per cent longer, this gives 1600 man hours to cut 226sq m (270sq yd) in the Middle Ages. The area of the roof is 15 x 4.25 x 2 = 1275.5 sq m. The divots are laid in overlapping tiles, i.e. double thickness overall. Therefore 255 sq m (2775sq ft) are needed, requiring 1805 man- hours.
Until the early nineteenth century wheat was usually cut with a sickle, not a scythe. This took two or three times as long, so that a man could reap only about a quarter of an acre a day. However, the method was less likely to damage the straw and render it unfit for thatching (Footnote 42). Thatching by the long straw method requires 28kg of straw per square metre (75.8lb/sq yd) to give a thickness of 0.4m (15.8in) (Footnote 43). At a technical day school run by the Society for the Protection of Ancient Buildings (SPAB) and the Council for Small Industries in Rural Areas (COSIRA), it was stated that two acres of straw are need to thatch a cottage - probably a two-bay, hipped-roof, house.
COSIRA's figures are for a thicker thatch than that found on the Solway Plain. Reducing the thickness to take account of the underthatch, 2677.5 kg (5903 lb) would be required for 127.5 sq m of roof, plus 635 kg of straw for the layers in the clay walls, making a total of 3312.5 kg. Joe Holliday of Hall Farm, Moorhouse, reckons today to get a ton to a ton-and-a-quarter of barley (1016-1270 kg) per acre. Assuming that it is the yield of grain, not straw, which has risen so spectacularly since the Middle Ages, and that barley and wheat straw require the same acreage, this would mean cutting 25/8-31/4 acres of wheat for the thatch and walls, taking 105-130 man hours. The average is 117.5.
A standard work on estimating for building contractors states that one thatcher and one labourer can cover 9.3sq m (100sq ft) with 0.305m (12in) thickness of plain wheat straw thatching in ten hours (Footnote 44). This figure is for southern thatching, and includes the time taken to fix battens, but nothing for the underthatch. If one can assume that these times would cancel each other out, then 127.5 sq m of 0.4m (15.8in) thickness of stapple thatch, would take 365.5 man-hours.
The making of mud floors in nineteenth-century Wales has been described by Eurwyn William. After it had been rammed down, the maker would walk over it in a special pair of clogs. It seems likely, however, that the evening party which followed the communal raising of the walls of a clay dabbin would have served the same purpose. Clogs were standard wear on the Plain until after the Second World War.
The construction of a stone or clay hearth on the floor would have taken a couple of hours work at the most.
The figures presented above suggest that in order to build a five-bay medieval longhouse on the Solway Plain, about 3400 man-hours would be needed. Some of this took place over a period of time, some during the actual construction process. This is about 1.6 x Startin's figure for a prehistoric house which was about 1.7 times as large in plan area and built 6000 years earlier with stone tools. It can only be regarded as an approximation, since the figures are imprecise and a number of assumptions have had to made. However, the method provides a means of comparing house-building costs, and to some extent standards of living at different places and times. It would be interesting to compare this result with estimates for medieval house-building in other parts of the country.
This paper is based on a dissertation for the Postgraduate Diploma in Local History at the University of Lancaster. For helpful suggestions and constructive criticism, thanks are due to Dr Angus Winchester during preparation of the dissertation, and to Ray Harrison for his comments on an earlier draft of this paper. Thanks are also due to Vernon Barrow, Roy Brigden, Damien Goodburn, Joe Holliday, B. C. Jones, John Letts, Ross Noble, John Thorp, and Bruce Walker, as well as to Geralt Nash of St Fagan's, and the Ryedale Folk Museum for their help, and to Bob Bartle, Peter Kidson and Edward Peters for checking the figures, but any remaining errors are the author's.