Romanesque bindings

A post by Assistant Curator Katie Brew

I have recently been fortunate to attend a five day Romanesque binding course with experienced bookbinder and conservator Arthur Green. It was a largely practical course, attended to gain a deeper understanding of this particular binding style and its construction – to improve my own skills and directly inform the conservation of the early medieval bindings we have here in Special Collections.

Romanesque bindings were predominantly produced c1000-1250 and are sometimes called monastic bindings, as they were produced in monasteries for religious study and as the product of the pious discipline of hand-copying manuscripts. The components of a Romanesque binding all have practical purposes. The strong oak boards give protection to the textblock and the herringbone sewing is self-reinforcing. Joints are mechanical rather than adhesive and the covering material and textblock were made from parchment, readily available animal skin. Metal clasps had the purpose of compressing the parchment textblock and keeping it flat and covers were left largely undecorated.

As you can imagine, Romanesque bindings in their original state are now very rare. They were not produced on anything like the scale of modern methods, and remaining books have had to endure a millennium of wear and tear, various religious reforms and the changing fashions of collectors. Even bindings which have remained in good condition have had their endband tabs cut off, in order that they could stand upright on the newly conceived bookshelf. We are lucky in Durham to have access to a wonderful collection of early medieval manuscripts. Many were created here, gifted to the city and the bishops of Durham were often great manuscript collectors. Many remain in their original home as part of the Durham Cathedral collection and we even have one from Ushaw College with endband tabs! The course was therefore strongly relevant to the care of the collections here.

The course began with a morning at Hereford cathedral and a look at their chained library, one of the last remaining of its kind. Getting a feel for the bindings in context was a great way to start the week. The librarian then took us for a closer study of three of their Romanesque bindings, one of which was the binding O.ii.1, which had been selected for us to recreate in the coming days. This focus on a specific object, rather than a general Romanesque style, was valuable in several ways. It enabled us to interpret a process from the outcome and experience that process for ourselves. Focussing on one binding also avoided generalising too strongly about what to expect from a certain style, as although there are common characteristics such as lacing channels and sewing patterns, changes were organic and varied by location and binder. Surviving bindings are not enough in number to be truly representative of the period.

The rest of the week was spent at Green’s studio in the Malverns, constructing the model binding. It was an intensely practical few days, with lots of opportunities to learn new techniques. We began by constructing the textblock. Although we mimicked the original where possible, we used paper folios rather than parchment as the binding would otherwise have been ludicrously expensive! We used linen thread to sew the folios onto alum-tawed sewing supports, in a herringbone pattern. This linking style imparts reinforcement to the sewing. We then prepared the oak boards. They were planed to size and the outer edges bevelled. The lacing channels for the sewing supports were then chiselled and hand-drilled through. The alum-tawed slips enter through the side of the board through to the front, and drilling the correct angle for these first time was a tense and challenging part of the process.

We adhered the endband tabs to keep them in place while sewing the endbands. Although the tabs had been cut off from the original, you could still see some of the purple and green textile that had been used. It seems this was one area which the binder may choose to add decoration. The endbands were sewn with natural and dyed-indigo thread in a chevron pattern around a split alum-tawed core. They were ‘tied down’ through every section of paper which makes them very strong. Endbands aided the opening of the book, protected the ends of the textblock and further attached the wooden boards.

The endband cores and sewing supports were then laced through the channels on the boards and fixed in position. Many early medieval binders fixed these slips in position by hammering in small wooden pegs. Our binding O.ii.1 had the slips secured by pulling alum-tawed wedges through, which we were grateful for as this method is less likely to split the board.

It was then time to prepare the parchment for covering the book. The skin was pasted several times with wheat starch paste to relax it out, and then pulled tightly around the boards. Parchment is a very strong material which stretches while wet and contracts tightly on drying, so it was quite the battle to mould it into position and the book needed to be placed under heavy weights to dry slowly overnight. We had initially been apprehensive to cover up all of the visible details we had just worked so hard on, but were very pleased with the outcome.

On the final day we had the relaxing job of sewing the perimeter. This intricate task brought together the loose layers of the tabs and covering parchment and both mechanically and visually completed the binding.

Throughout the course we stopped at various points to discuss what we were doing and why, and what variations there may have been. We were also given reading to accompany the practical side and this combination really cemented my understanding of the process.

Whilst understanding Romanesque bindings is directly applicable to those we have in the collection here, more broadly the early medieval style demonstrates a much greater understanding of the materials used across the period than later interventions. Rebinding of parchment textblocks by later bookbinders for reasons of restoration or fashion used techniques designed for paper, which has very different qualities. For example, paper is more flexible, so a tight binding with smaller margins is less of an issue. Parchment also needs more room to open and respond to fluctuations in humidity, so using excess adhesive on the spine is a poor decision. New conservation treatments of bound parchment are returning to this earlier mechanical format.

The course has been a genuinely useful experience which will positively affect both my work and the work of my department as a whole. It is wonderful to have the chance to deepen my knowledge of our collection here at Durham, and I am enjoying sharing this with my colleagues at the university.

With great thanks to both the York Foundation for Conservation and Craftsmanship and The Anna Plowden Trust/Clothworkers’ Foundation for sponsoring my place on the course and without whom it would not have been possible to attend.