Historic thatch (Ireland) study: an interim report
J B Letts reports on an important project which it is hoped will produce a detailed picture of Irish thatching practice.
Flax thatch, County Londonderry
In the early spring of 1996, a research project on the history of thatching in Ireland funded jointly by the DoE (N. Ireland) and the DACG (Ireland) began with a tour of 40 thatched buildings in Ulster province. The study is being conducted by John Letts of Reading University, an archaeobotanist who has pioneered the analysis of ancient thatch over the last four years in research projects commissioned by English Heritage and the governments of Northern Ireland and the Republic of Ireland. In an initial study of 'smokeblackened' thatch recovered from the base coats of late medieval open-hall buildings, and in Thatching in England from 1790 to 1940 undertaken jointly with building historian James Moir, Mr. Letts has integrated the in situ evidence from excavations of ancient multilayered thatched roofs with ethnographic, historical and technical data to provide an overview of thatching over the past five centuries. This provides the academic and historical basis for legislative action which English Heritage which has been under pres~ sure to adopt to protect historically significant roofs, materials and thatching techniques.

Research on thatching in Ireland follows roughly the same approach and has broadly similar goals, although with a greater emphasis on recording the in situ evidence which survives from every historical period and highlighting technical issues which could improve the performance - and hence the economic viability - of thatch vis a vis common alternatives.

Turf and marram grass, County Donegal
The research is propitious as traditional Irish thatch is in an advanced stage of decline, but evidence of the techniques and materials used in the past survives either under corrugated iron or on semi-derelict houses scattered throughout the countryside. The new 'Euro-thatch' that is emerging to take its place relies on imported materials and methods which are far removed from Irish tradition and are in many ways unsuited to Irish climatic conditions.

An integrated study will provide a basis for legislation which could help preserved at least a portion of ancient Irish thatched roofs, as well as elucidate some of the underlying factors that should be taken into account by owners, thatchers, architects and Conservation Officers as 'lrish thatch' adapts to the economic and technical realities of the new millennium.

Project Brief
The core of the field work involves surveying a selection of multi-layered thatched roofs in each of the four provinces of the island of Ireland. Each building is excavated' as thoroughly as possible in order to establish a chronological sequence in terms of the materials and techniques used at each rethatching. Essential constructional aspects of the building and roof structure are also recorded, along with details on the local environment and agriculture that will help situate the building in its historical and economic context. Wherever possible, local informants are interviewed to obtain more detailed information on the building and thatching materials and practices once used in the local area.

A photographic and video record is also made of the building, the roof and the excavation. Samples are taken of every layer of thatch revealed, and these are later analysed in the laboratory in order to identify the species used and the way in which they were harvested and processed for use as thatch.

An immense amount of reliable ethnographic data on thatching in Ireland survives in a unique collection of replies to a questionnaire on thatching that was circulated to thatchers in most counties by the Irish Folklore Commission in the 1940s. Approximately a third of the c. 500 replies in the Department of Irish Folklore at University College Dublin are in the Irish language and have been translated into English for the purposes of this study. This collection is without doubt the richest source of information on 'traditional' thatching practices available anywhere in Europe, and will greatly enhance the data recovered by in situ sampling.

Eel grass and straw, County Down, Northern Ireland

Work completed
For the purposes of this study, buildings surveyed fall into three categories:

  1. derelict buildings that are sampled and recorded in detail;
  2. inhabited dwellings that are surveyed, but which provided limited sampling or recording opportunities; and
  3. buildings with modern thatched roofs (i.e. replaced within the last 50 years).

As much time as possible is being devoted to the recording of level (a) buildings as these will soon be lost due to general dereliction. Unfortunately, these buildings are not always easy to locate. Level (b) buildings can also provide data of technical, historical or botanical interest.

Recording of the in situ evidence began in the spring of 1996 with a tour of Ulster Province, and has been followed by additional sampling visits to Leinster, E. Munster and N. Connaught. The success of the first Ulster tour was due to the fact that buildings suitable for analysis had already been identified by the Environment Service of the DoE.

Sampling in the South has been less methodical, but nevertheless successful because of the sheer number of derelict thatched buildings that exist in most southern counties. As of September, 1997, 24 level (a) buildings have been sampled and recorded in Ulster Province along with c. 25 in S. Leinster and E. Munster. A further c. 20 have been surveyed to level (b) in each province. A dozen buildings have been surveyed in North Connaught and a half dozen in the Dublin region. Further sampling visits in October and November will focus on Munster and West Connaught, and the sampling component of the project will be completed by the end of November.

In general, sampling has gone more slowly than originally planned due to logistical difficulties and the need to develop local contacts. Locals are invariably aware of buildings suitable for surveying, whether derelict or inhabited, and home owners are almost always happy to provide access, but these contacts take time to develop and are difficult to pursue during short sampling visits. A number of contacts were provided by Mr. Peter Brockett, a thatcher and thatch instructor, who unfortunately passed away in the summer of 1996. Many buildings have also been located through historians and individuals with a strong interest in vernacular architecture. In contrast, and with a few notable exceptions, the response from county councils for assistance in locating suitable buildings has been less than enthusiastic.

On occupied buildings, sampling is constrained by the type of rethatching that is occurring. Traditionally, a new top coat or patch was simply 'scolloped' or 'stuffed' into a weathered coat when required. Older thatchers are always eager to talk about the ,coating' methods they are using, but such roofs do not provide sequential series of samples as they offer access only to surface layers. Contemporary practice often involves stripping roofs completely for rethatching with either water reed or combed wheat reed, and numerous complete series of samples have been obtained from roofs being rethatched by a recently trained younger generation of thatchers. Semi derelict buildings provide the ideal sampling opportunity as they can usually be partially dismantled, recorded and sampled as methodically as time permits.

Some modern Irish thatchers still use methods and materials that disappeared long ago in England, Scotland, Wales and the Continent. An effort is being made to interview, and when possible record, these practices as most are unlikely to outlive the generation of thatchers now approaching retirement. The older generation's intuitive understanding of thatch and the link between structure, materials and performance is crucial to the success of thatch in the future whether a roof is thatched with rushes, oat straw or imported water reed. The technical basis of thatching has already been sketched in considerable detail by Mr. Letts in his work on thatching in England.

The historical literature contains abundant information on thatching in Ireland and related topics, although more recent writings are often anecdotal and not nearly as botanically or technically 'rigorous' as one might wish. These sources provide the basis for sketching a broad history of thatching in Ireland taking into account many of the social, economic and environmental changes that have occurred in recent centuries. A great deal of information can also be obtained from photographic archives, particularly in reference to the appearance and techniques used for ridges, caves and gables.
Flax thatch, County Armagh.
Waterford, Wexford, Kildare and Kilkenny counties have been particularly rich in such 'sampling opportunities', and the additional effort devoted to this region is justified by the variety of practices and materials discovered and the general applicability of this data to other regions. In contrast, derelict thatched dwellings are rare in adjacent Co. Wicklow, although the two that have been recorded revealed unusual materials and methods that will contribute significantly to the overall story. No building can be ignored, as almost every building visited to date has provided evidence for idiosyncratic detail and a considerable degree of regional and local diversity in the principal materials and fixings employed.
Netted thatch, County Donegal.
Materials research
Modern Irish thatch is largely derived from oat and rye straw, whereas many of the older roofs excavated so far have a basal coat of wheat straw. This suggests that the shift to oat for thatching is a relatively recent one in many areas. Contemporary thatchers would certainly use straw of 'tougher', longerlasting varieties if these were available, but unfortunately almost no stocks of ancient Irish wheat or oat varieties survive in Ireland. One must turn to gene banks in Wales, Scotland and England for small samples of varieties similar to traditional Irish varieties that must then be multiplied over several years to reach production level. A very rare example of an ancient, genetically-diverse' land race 'populations of bristle oat (Avena strigosa) was recently discovered by Mr. Letts in a remote part of Donegal and is currently being multiplied for experimental purposes.

Some oat varieties undoubtedly produced high quality thatching straw in the past, but a short life expectancy for a thatched roof now seems to be accepted as inevitable - perhaps because of growers' attempts to produce marketable grain as well as straw from the same crop. The evidence suggests that a considerable increase in the longevity of Irish thatched roofs could be achieved by growing older varieties better suited for thatching, reducing the use of chemical fertilisers and harvesting the straw while a little unripe. Experimental plots of a few older varieties of wheat and oats were sown for experimental purposes in Oxford and Co. Kildare in late 1996 to provide a basis for discussing the potential for varietal research in Irish thatch. The preliminary results of a screening trial of c.250 old wheat varieties being undertaken by Mr. Letts at the University of Reading in Berkshire, England, will also be available soon, providing a basis for selecting suitable wheat varieties for more detailed analysis under Irish growing conditions.

Collection and analysis of ethnographic and historical data
The translation of the Irish language replies to questionnaires and associated sources in the Department of Folklore at UCD is nearing completion. This work is being carried out by Diarmuid O'Gruagain under the supervision of Criostoir Mac Carthaigh, and will eventually be made available to interested researchers through the Folklore Department. The unanticipated richness of this resource has encouraged a greater emphasis to be placed on the ethnographic data compared with studies of thatching in England.

Straw and water reed, Ardmore, County Waterford.
Preliminary results
The in situ record of thatching in Ireland has turned out to be far richer than had been expected, both technically and materially, and Irish roofs clearly hold treasure troves of data useful to historians of agriculture, the landscape and vernacular buildings. Equivalent diversity in England and the Continent appears to have been stripped away long ago, and Irish roofs are thus unique in a wider context in having preserved methods and materi~ als once much more widely used throughout Western Europe.

The evidence suggests that wheat straw was more commonly used than oat straw in many areas in the last century, along with barley and rye straw. Somewhat surprisingly, combined barley straw was still used in some areas until very recently. Truly appalling examples of straw thatching are not difficult to find in Ireland. Cases exist of newly applied thatch that could not withstand a winter without major repair, and the oat straw currently used rarely lasts for more than five to seven years. In contrast, straw roofs in Wales and Western England - districts with 'Irish' rainfall and shallow-pitched roofs - routinely last for 20 years with minimal maintenance. Ethnographic records suggest that roofs were expected to last from 10 to 15 years at the beginning of this century, and the reduced longevity is at least partially linked to the introduction of modern hybrid varieties with straw too short for use as thatch.

Unusual main coat and fixing materials survive in profusion; corrugated iron shields older base coats of heather in Wicklow and Antrim; decay-resistant eel grass (Zostera maritimus), collected in bulk on the shore after a storm tide, was used in a similar way in coastal districts in the North; the almost ubiquitous turf 'scraw'- whether exposed or underlain by a thin flecking of straw - provided a firm base for scollops throughout much of the country; marram grass cut on a sustainable basis from coastal dunes was used effectively on rope thatched dwellings in many coastal areas; and flax was widely used in parts of Ulster when the harvest could not be sold for the production of linen (and continues to be popular in some regions).

Water reed has been cut from beds along the River Suir in Waterford and Kilkenny for generations, but in this area it is applied in much the same manner of straw and is not dressed into position as in Norfolk-style water reed. In contrast to England and the Continent, however, native reed was generally considered to be inferior to straw for reasons rooted in population genetics as much as technique. Rapid improvements in the quality of domestic Irish reed could probably be achieved with relatively minor effort.

In general, the in situ, ethnographic and historical evidence suggests that 'traditional' lrish thatch once performed better than is commonly believed, and a relatively small number of material and technical changes could increase significantly the average longevity of such roofs while preserving the visual, technical and material continuity with ancient Irish thatching traditions.

Context is grateful to Richard Oram and Lawrence Manogue of the Environment and Heritage Service, DoE (Northern Ireland) for permission to reproduce these interim findings. A final detailed report is to be published in due course.

J. B. Letts is with the Department of Agricultural Botany, The University of Reading

Context 56 December 1997