Brian and John Howells. National Library of Wales journal, Vol XXI/4, Winter 1980.

Extracted onto the pages of GENUKI with the kind permission of the National Library of Wales

This is a complete extract of this article [Gareth Hicks August 2003].

KNOWLEDGE of the domestic architecture of early modern Wales is at present based firmly upon the work of archaeologists and students of folk-life, and in particular upon a number of important studies made between the appearance of Iorwerth Peate's pioneering work The Welsh house: a study of folk culture in 1940, and the publication of Peter Smith's monumental Houses of the Welsh Countryside in 1975. 1 Now there is a lively awareness of the importance of an aspect of the Welsh past which was once greatly neglected, and in most counties steady progress is being made by small groups of people dedicated to the task of identifying old houses and recording their characteristics. In studying vernacular architecture there is no substitute for the field study of individual buildings, for the painstaking measurement, photographing, drawing and close analysis of plans and constructional detail and, following on all this, careful consideration of the distribution and possible dissemination of house types and specific building features. Yet field-work necessarily has its limitations. When still occupied, old houses have almost invariably been modernised and modified, sometimes obviously and sometimes in ways which may mislead the investigator, whilst when deserted they are mere shells, melancholy reminders of a vanished way of life, where it is often impossible to determine the uses to which particular rooms were put and to visualise how they were once furnished and equipped.

In two respects at least the consideration of documentary and pictorial evidence may serve to reduce the possibility of error and enlarge our knowledge of the Welsh house. In the first place, it may result in an accession of knowledge leading to the modification of hypotheses based solely upon the study of existing structures. There is clearly every danger of distribution maps of ground plans and architectural features being misleading to some extent, because they reflect only what has survived. Take, for example, the present distribution of traditional timber-framed houses, most of which date from the sixteenth century and later and so are found today mainly in areas where timber was still plentiful in the early modern period. An examination of documentary sources makes it clear that huge areas in Wales were cleared of woodland between the twelfth and sixteenth centuries, and this in turn clearly implies the possibility that timber-framed buildings were once common in areas where very few are to be found today, such as within the confines of modern Dyfed. Indeed, this very fact leaves one wondering whether the level of craftsmanship in wood was not as high in pre-Conquest Wales as it was, say, in England or Norway, and whether the fine timberwork to be found in the late medieval houses of north and east Wales represents the dying flicker of a once-vigorous native tradition, the influence of immigrant English craftsmen, or a combination of both. Of more direct importance in the present context is the fact that documentary sources tell us what rooms were called in the various localities at specific periods, how they were furnished, and the uses to which they were put.

In this article an attempt will be made to indicate some of the possibilities of this approach by examining some non-archaeological sources of evidence relating to peasant houses in Stuart Pembrokeshire. 2

Before doing this it may be well to review some of the salient characteristics of those early farmhouses in the shire which survived into the modern period. Most, as Peter Smith has pointed out, were built on the three-unit plan with secondary rooms lying at each end of a hall, 3 and normally there was a cross-passage with the front door at one end and the back door at the other. In some cases the front doorway assumed the shape of a Gothic arch and was protected by an outshut porch. In the St. David's district most houses of this type had at least one lateral outshut or side-aisle outside the main walls of the house but under an extension of the roof, thereby, in the words of Romilly Allen, 'increasing the area of the ground floor without the necessity for making a roof of unduly wide span'. 4 Such outshuts could be used as bedrooms or sculleries, and sometimes large chimney stacks were built into them, though it was also common for huge external stacks to be built, with blatant disregard for symmetry, off centre to the ridge of the gable ends of houses. The principal chimney stack of an old Pembrokeshire house was frequently large and occasionally massive, containing several ovens and a cavernous open fireplace within. The chimney stack at Hodgeston farm, for example, is about 19 feet high, 10 feet 11inches long at the bottom, and 5feet wide and 6 feet 3 inches deep at the top, whilst a few miles away the fireplace within the great chimney stack at Bangeston near Stackpole is no less than 8 feet deep : the chimneys themselves were sometimes square or oblong and sometimes round. So solidly built were the stacks that they were frequently left standing when other parts of the house were demolished or rebuilt, and forlorn examples of isolated stacks marking the sites of vanished houses may be seen today in the villages of Lamphey and Carew. Writing in 1942, Sir Cyril Fox suggested that the external hearth may have been developed as a means of isolating the fire and the flue from contact with the gable of the building, 5 an argument which gains force when one remembers that until about the middle of the last century most houses and cottages in rural Pembrokeshire were thatched, the outshuts being roofed with slate. Their windows were small and frequently unopenable, and the floors paved with great slabs of slate. Apart from the outshuts, the ground-floor rooms varied between 7 and 8 feet in height and were ceiled by the floorboards of the room above, and to the beams were fixed hooks from which various articles were suspended and racks which frequently contained flitches of salted bacon and hams. The doors leading from one room to another were made of thick boarding, ledged, and fastened by wooden latches.

The farmhouses in the St. David's district visited by Romilly Allen in the early 1880s contained few factory-made products, and domestic arrangements preserved the essential features of a peasant way of life hundreds of years old. Cauldrons and other cooking utensils hung from cranes in the great ingle-nooks, and wooden bowls and platters were piled on wooden shelves supported by strong wooden pegs................

................... driven into the walls. Wooden spoons were kept in a rack fixed to the wall, and further storage space was provided by open recesses, about a foot square and nine inches deep, in the thickness of the wall and by wall cupboards with wooden doors. Larger articles such as tubs rested on slate slabs ranged along the walls and standing on low stone walls. There were few chairs, seating accommodation being provided by three-legged stools and by 'skews' or high-backed wooden settles whose seats comprised the lids of chests for storing such things as blankets. There was often, too, a stone bench adjacent to the main door in the entry porch. Almost everything was hand-made locally, preserving ancient craft traditions and the use of a whole range of artefacts which were shortly to disappear from everyday use.

As an indication of the possibilities of documentary investigation as an ancillary aid to the study of vernacular architecture, two sources will be considered, a survey of lands of Charles, Prince of Wales, within the hundred of Roose which was compiled in 1623, 6 and a considerable number of probate inventories. 7

The Roose survey of 1623.

This was carried out by Sir John Philipps, George Ellis and the county feodary, Sir Thomas Canon, in accordance with instructions issued by the prince's commissioners of revenue in a letter dated 7 October 1623. The following data relating to houses and farmbuildings have been abstracted from the survey, and presented in summary form.

STEYNTON.

1. Anable Pill, 64 acres. Tenant William Tasker. House 4 rooms on ground floor and loft at lower end of house, 21 couples. 8 Covered with thatch. A little outshut adjoining hall and covered with slate. Outbuildings A barn, stable and cowhouse of 10couples. Backhouse, 9 4 couples. Outhouse, 3 couples. Carthouse, 4 couples. Sheepcote, 8 couples. Coalhouse, 6 couples. Hayhouse, 4 couples. All covered with thatch.

2. Ford, 32 acres. Tenant Francis Thomas. House 4 rooms on ground floor and a loft, 11couples. Covered with thatch. Outbuildings Barn, 6 couples. Backhouse and chamber adjoining it, 8 couples. Sheepcote, 6 couples. All covered with thatch.

3. Ford, 32 ares. Tenant Philip Moyle. House 4 rooms on ground floor and a loft, 11couples. Covered with thatch. Outbuildings Barn, 6 couples. Backhouse and chamber adjoining it, 8 couples. Sheepcote, 6 couples. All covered with thatch.

4. The Fenn, near Pill, 64 acres. Tenant David Howell. House 6 rooms on ground floor and 2 lofts over 2 of the rooms, 20 couples. Covered with thatch. Outbuildings Malthouse, 5couples. Barn, stable and hayhouse, 10 couples. Carthouse, cowhouse and sheepcote, 20 couples. Backhouse, 3 couples. Swinecote, 3 couples. All covered with thatch.

5. Egebs Moor, 32 acres. Tenant Henry Codd. Houses A hall, barn and sheepcote of 20 couples all under on(sic) thatched roof. Also a 'house adjoining the hall and built cross upon it, divided into 2 rooms by a loft', 4 couples. Covered with thatch. Outbuilding One outroom used as a backhouse, 3 couples. Covered with thatch.

ST. ISHMAEL'S.

1. 64 acres. Tenant Griffith Reynold. House Hall, 6 couples. Loft, 2 couples. Outshut of 2 rooms. All thatched. Outbuildings A room near the dwelling house, 5 couples. Malthouse, 4 couples. Barn, 13 couples. Cowhouse, 4 couples. Carthouse and sheepcote, 12 couples. All thatched. Backhouse, 2 couples; covered with slate.

2. 32 acres. Tenant Philip Hire. House 6 couples; thatched. Outbuildings Barn, 3 couples. Sheepcote, 4 couples. Little outhouse, 4 couples. All thatched.

3. 32 acres. Tenant Philip Hire. House 6 couples; thatched. Outbuildings Barn, 3 Couples. Sheepcote, 4 couples. Little outhouse, 4 couples. All thatched.

4. 28 acres. Tenant Philip Price. House 12 couples; thatched. Outbuildings Backhouse, 4 couples. Cottage, 4 couples. Another cottage somewhat removed from the dwelling house, 8 couples. All thatched.

5. 64 acres. Tenant Lewis Cock. House 3 rooms, 12 couples. Thatched. Outbuildings Malthouse, 4 couples. Backhouse, 6 couples. Barn and stable, 6 couples. Sheepcote, 8 couples. A little cottage, 3 couples. All thatched.

6. 56 acres. Tenant John Skone. House 9 couples; thatched. Outbuildings Barn, 6 couples. Decayed sheepcote. Cottage somewhat removed from the dwelling house. All thatched.

7. 44 acres. Tenant David Prosser. House 3 rooms, 8 couples. Thatched. Outbuildings A backhouse and other outhouses containing 8 couples. All thatched.

ROCH.

1. 44 acres. Tenant Roger Webb. House 8 couples; thatched. Outbuildings Outchamber, 4 couples. Barn, 7 couples. Cowhouse, 8 couples. Thatched.

2. 44 acres. Tennant Roger Webb. House 13 couples; thatched. Outbuildings Outchamber, 4 couples. Malthouse, 6 couples. Another little chamber near the house, 3 couples. All thatched.

CAMROSE.

1. West Pelcomb, 96 acres. Tenant John Bennett. House 12 couples; thatched. Outbuildings Barn and stable, 8 couples. Cowhouse, 6 couples. Sheepcote, 3 couples. All thatched.

2. West Pelcomb, 32 acres. Tenant David Mathew. House 7 couples, thatched. Outbuildings Barn and sheepcote, 5 couples. Outstall, 4 couples. All thatched.

3. West Pelcomb, 36 acres. Tenants Philip Thomas and Watkin Cornock. House 5 couples, thatched. Outbuilding Barn, 4 couples. Cowhouse, 4 couples. Sheepcote, 5 couples. All thatched.

4 West Pelcomb, 32 acres. Tenant Sir Thomas Canon. House 8 couples, thatched. Outbuildings Barn, 3 couples. Thatched.

5. West Pelcomb, 64 acres. Tenant Sir Thomas Canon. Building 'The dwellinge house beinge iii roomes containing vii coples, one cowhowse v coples, one barne v coples, one sheepecott v coples, all under a roofe and covered with thatch'.

6. West Pelcomb, 64 acres. Tenant Sir Thomas Canon. Building 'The dwelling howse being iii roomes containing vii coples, one cowhowse v coples, one barne v coples, one sheepcott v coples, all under a roofe and covered with thatch'.
[5/6 seem to be duplicated]

7. Leweston, 96 acres. Tenant Sir Thomas Canon. House 3 rooms on ground floor, 13 couples. Thatched.

8. North Camrose, 72 acres. Tenant Jenkin Renish. House The hall, 5 couples. An outshut, 2 couples. An inner room and loft over it, 4 couples. A room below the hall door, 3 couples. All on the ground floor and thatched. Outbuildings Barn and stable, 8 couples. Cowhouse, 5 couples. Sheepcote, 16 couples. Carthouse, 6 couples. 'All lowe builte, open to the roofe, and covered with thatch'.

9. North Camrose, 24 acres. Tenant John Estmond. House 12 couples. All on ground floor and thatched. Outbuildings Sheepcote and stable, 8 couples.

10. North Camrose, 38 acres. Tenant John Estmond. House 3 rooms on ground floor, 10couples. Thatched. Outbuilding 'One barne being a sheepcott', 10 couples. Thatched.

11. North Camrose, 56 acres. House 3 rooms on ground floor, 12 couples. Thatched. Outbuilding One little room adjoining the house. Thatched.

12. Broughton's Land, North Camrose, 32 acres. Tenant John Codd. House 2 rooms on ground floor, 2 couples. Thatched. No outbuildings.

13. North Camrose, 32 acres. Tenant John Byrren. House 3 rooms at ground level, 12 couples. One offshut. 2 couples. Thatched. Outbuildings One little thatched building, 2 couples. Barn and stable, 6 couples. Outhouse, 3 couples. Sheepcote, 6 couples.

14. North Camrose, 24 acres. Tenant John Byrren, 24 acres. House 2 rooms, 6 couples. Thatched. Outbuildings Barn and stable, 5 couples.

Most of these farmhouses seem to have had three or four rooms, and in several cases pairs of farmsteads within the same parishes are described which appear to have been very similar in size and building facilities (e.g. those of Francis Thomas and Philip Moyle at Ford in Steynton, the two distinct tenements at St. Ishmael's held by Philip Hire, and the two 64-acre holdings of Sir Thomas Canon at West Pelcomb in Camrose). Like the houses of the St. David's district described by Romilly Allen, William Tasker's house at Pill was thatched apart from the outshut, which was roofed with slate. There are three references to what may have been long-houses. Henry Codd's house at Steynton had a hall (i.e. farmhouse), barn and sheepcote under one roof, and two identical 64-acre tenements at West Pelcomb in Camrose parish each had a three-roomed farmhouse, a cowhouse, a barn and a sheepcote under a single roof. There would appear to have been slight differences between the farmhouses of the various parishes in 1623, but perhaps it would be unwise to read much significance into this.

Probate inventories

Only a small minority of the probate inventories emanating from Stuart Pembrokeshire yield room-by-room descriptions of the contents of houses, and of these far more relate to the south than the north of the county. Even where such inventories do exist, one must be cautious in interpreting them, and especially in watching for errors of omission. Instances occur where rooms known to have ................

................... existed were not listed and, as might have been expected, features like the cross-passages or 'entries' which existed in the great majority of ancient farmhouses surviving into the late Victorian period were rarely mentioned simply because they contained no articles of pecuniary worth. It is clear, too, that the appraisers who drew up the inventories varied greatly in conscientiousness and attention to detail: on occasions such as these much must have depended on the lead given by local parsons, who so frequently acted as the advisers and scribes of their parishioners. Most peasant houses in Stuart Pembrokeshire had either two or three rooms on the ground floor. The names assigned to them varied, but one which was almost always mentioned was the hall. Where there were two other rooms they were usually known as the upper and lower chamber, the upper room and kitchen, or as the parlour and kitchen. Many and perhaps most of these farmhouses had first-floor rooms or lofts: sometimes all ground-floor rooms were ceiled, and sometimes only one or two of them. While it is impossible to reconstruct ground-floor plans precisely from inventory evidence, one repeatedly comes across descriptions suggesting a lay-out which may be expressed in diagrammatical form as follows:

Where small service rooms existed, these were normally located near the kitchen or lower chamber. Two-roomed houses almost invariably consisted of a hall and a kitchen. Beds were frequently to be found in the hall and, indeed, in almost any room within the house apart from the kitchen, 10 and farm servants often slept in outbuildings, a practice which was still common in the early part of this century.

The hall was the living-room and simply furnished with locally-made furniture. In 1652 Henry Leach's hall at Slade in Castlemartin contained a bed, a table, a chair, a form, four stools, two 'skews', four chests and a cupboard, whilst that of John Perrot of Carew, who died in 1691, had a bed, a table, a chair, a form, a settle, two coffers and a cupboard. The table usually consisted of a long board laid across trestles, but by the early eighteenth century round or oval tables were occasionally owned by more prosperous farmers. In the case of peasants occupying houses ........................

................................. with two ground-floor rooms, the hall was usually a combined living room-kitchen. Only a minority of farmhouses had a room called a parlour, but where it did exist it was usually a bedroom-sitting room containing the best furniture in the house. In the house of John Vaughan of English Redwalls (Henry's Moat), the parlour was furnished in 1650 with the best feather bed and the most valuable table, together with a chair, two forms, seven stools and two chests, whereas the hall had an old feather bed, a smaller table, a press or cupboard, a chair, six old stools and an old chest. In 1693 Thomas Mathias's parlour at Cresborough (Slebech) was the only room in the house with a carpet, and at Lamphey in 1705, the parlour of Henry Rowe of Cleggars contained twelve chairs, whereas in the hall there were only six stools and a form. The parlour was usually situated at the upper end of the hall, at the furthest end of the house from the service room or rooms.

There is nothing to indicate that the word 'chamber' meant anything more precise in Stuart Pembrokeshire than the word 'room'. Like most rooms apart from kitchens chambers usually contained beds, but they were not simply regarded as bedrooms. Occasionally the two chambers were described as being situated either above or below the hall, in which case the upper chamber would be at the end of the hall furthest from the main door and the service rooms, whilst the lower chamber would be in the vicinity of the latter, and sometimes in practice no more than a kitchen. Sleeping arrangements varied widely from one household to another, but on the whole there was a tendency for the farmer's family to sleep in the hall, upper chamber or parlour, or in the lofts above them, whereas servants slept in the service rooms and outhouses. Increasingly, from about the middle of the seventeenth century, chambers in large farmhouses were described in terms of colour. In 1659 the house of the rector of Penally, a farmer like many of his parishioners, had a green chamber (and the only example so far encountered of a children's chamber), 11 and at Brownslate (Monkton) in 1698 there were blue and white chambers, and in addition a new chamber, each being furnished with a close stool. 12

Houses with more than two rooms invariably contained kitchens, and a few large farmhouses like Mayeston (Cosheston) and East Moor (Manorbier) 13 had both inner and outer kitchens. There is nothing to suggest that at this date the kitchen was an outbuilding and not an integral part of the house. 14 It was usually furnished with a table and, more rarely, with a dresser. In almost every kitchen there were cauldrons and spits, a fact which suggests that the great chimney stacks so characteristic of the area were associated with the kitchen rather than with the hall (a number of inventories mention the existence of iron fire-bars in halls which may have been meant for use in comparatively small fireplaces rather than in great open hearths). Inventory evidence suggests that a well-equipped Pembrokeshire farmhouse kitchen of the period would contain, apart from a table, cauldrons, brass pots and pans, skillets, a frying pan, and a mortar and pestle; pewter dishes, porringers, plates, trenches, cups and tankards; spits, andirons, brandirons, pothooks, dripping pans, boiling irons, pot forks, chafing dishes and hand irons,......................

.................... together with barrels, tubs and bottles. Naturally, variations in wealth and status were reflected in levels of furnishing and equipment. The tables of the richer peasants boasted some silverware: their poorer counterparts made do with wooden trenches, vessels and spoons, and perhaps a few pieces of earthenware or pewter.

As might have been expected, dairies (sometimes termed 'milk houses') usually contained pans, churns, troughs, cheese vats and cheese presses, and sometimes cheese, butter, salted meat and other provisions were stored in them: cheeses were also kept in 'cheese-lofts' in some farmhouses. A few peasant houses had butteries, but these were mostly general storage rooms and not places where drink was kept. In those cases where rooms were used mainly for the storage of beer they were termed cellars, and they were normally above ground. Spinning wheels were to be found in many houses, suggesting that it was common practice for peasants to carry their own yarn to local weavers, who then worked on it. This would explain why, despite the decline which had occurred in the Pembrokeshire cloth industry during the second half of the sixteenth century, a surprising number of fulling mills remained in operation throughout the Stuart period. At Stackpole Thomas Gibbon of Rowston had a 'spinhouse' containing two tourns or spinning-wheels (as well as two beds for servants), 15 a quite exceptional arrangement: in most cases both long tourns and sitting tourns were put either in upstairs rooms, where they existed, or else in convenient service rooms, no matter what their primary functions might be.

The majority of peasant farms in Stuart Pembrokeshire were rented or leased, so most farmhouses were probably built at the instance of landlords, though evidence has been found to suggest that on occasion the responsibility might be assumed by a tenant in return for an appropriate financial concession. A house 24 feet high with interior dimensions of 39 by 18 feet and with walls 2 feet 3 inches thick was built in 1660 at Pembroke by a Mr Browne, the cost being about £250, 16 but apart from this isolated and perhaps untypical case no data concerning building costs have yet been found.

Each peasant household was an individual unit, with its own way of organising its domestic life, but there is a considerable weight of documentary evidence to suggest that in respect of nomenclature, room use and furnishings, norms did exist and, what is more, that they did not vary greatly throughout the whole Stuart period. The same cannot be said of gentry households: after all, one obvious way of flaunting gentility was to show that one was abreast of fashion in such matters as dress, furniture and household arrangements, and rich yeomen with social aspirations were not slow to get the message. At the other end of the social scale, the rural poor lived in hovels described by the person who answered Edward Lhwyd's parochial queries about Walwyn's Castle as 'a sort of extempore erections of dirt and clay called clam ( recte clom), and very uncertain being tho' but lately built, [and] suffered after soon to fall into ruins ... ' 17 The peasantry occupied a broad band of the social spectrum, the richer yeomen being able, if they chose, to ape the life-style of the lesser gentry, and the poorest husbandmen struggling................

.................to maintain themselves at subsistence level, but for the most part peasants in Stuart Pembrokeshire occupied houses which though modest in scale, conservative in design and furnished with a limited range of locally-produced artefacts, were capable of offering, by the undemanding standards of the day, a reasonable degree of domestic comfort.

BRIAN HOWELLS

JOHN HOWELLS

Lampeter

Notes;

1. For a useful guide to the literature of this subject see R. de Zouche Hall, A Bibliography on Vernacular Architecture (Newton Abbot, 1972).

2. In this study the term 'peasant' is used to denote farmers who rented or owned tenements but who were not regarded by their peers as belonging either to the landed gentry or to the professional classes.

3. P. Smith, Houses of the Welsh Countryside (London, 1975), 20.

4. A. R. Allen, 'Old Farmhouses with round chimneys near St. David's', Archaeologia Cambrensis, series VI, ii (1902), 2.

5. C. Fox, 'Some south Pembrokeshire cottages', Antiquity, 16, 1942, 307-19, provides description of a number of old farmhouses which formerly stood within the Castlemartin tank range: most of theme are now destroyed.

6. P.R.O. L.R.2/206, fos. 1-39. In many cases the farms were occupied not by tenants but by sub-tenants. See B. E. and K. A. Howells, The Extent of Cemais 1594 (Haverfordwest, 1977) for much detailed evidence on this score.

7. These are all in the National Library of Wales. Of the many probate inventories relating to the goods and chattels of Pembrokeshire peasants of the Stuart period, about 120 were selected for detailed analysis because of the detail contained in room-by-room descriptions, and these form the basis of the generalisations made concerning room use and furnishings.

8. One of the definitions of a couple in The Oxford English Dictionary is 'one of a pair of individual rafters or beams that meet at the top and are fixed at the bottom by a tie, and form the principal support of a roof'.Clearly the surveyors felt that the distance between couples provided a rough and ready indication of the size of a building.

9. The form found in the original document is 'backehowse'. This may be interpreted either as backhouse or as bake-house, and it is only the fact that bread-ovens were usually built into chimney stacks which inclines us to favour the former form. For discussion of these terms see M. Barley, 'A glossary of names for rooms in houses of the sixteenth and seventeenth centuries' in I. Foster and L. Alcock (eds.), Culture and Environment: Essays in Honour of Sir Cyril Fox (London, 1963).

10. In the Stackpole area and probably elsewhere in the south of the county cupboard beds were in common use in the early eighteenth century. See B. E. and K. A. Howells, Pembrokeshire Life: 1572-1843 (Haverfordwest, 1972), 67.

11. Probate inventory of Francis Hudson of Penally, 1659.

12. Probate inventory of William Holcombe of Monkton, 1698.

13. Probate inventories of Thomas Philipps of Cosheston, 1704, and Thomas Lort of Manorbier, 1687. Both belonged to the lower gentry but occupied farms held in the earlier part of the century by yeomen.

14. See P. Smith, op. cit., 231.

15. Probate inventory of Thomas Gibbon of Stackpole, 1703.

16. Glansevern MS. 14,096.

17. R. H. Morris (ed.), Parochialia ..., iii, (London 1911), 24.