It all begins with a sheep.

The first stage of medieval manuscript production was, of course, the provision of raw materials for the pages; papyrus had passed out of fashion in the 4th century or so, leaving parchment and, from the 13th century onwards, paper. Depending on the size and planned dimensions of the manuscript, sheep (common on the mainland), calfs (common in insular manuscripts), or even goats (rarely, and mostly in Italy) were slaughtered and skinned. For manuscripts like the famous Codex Amiatinus, the lives of up to 500 sheep were sacrificed for the project. The skin was not tanned, but soaked in a solution of calcium lye, to dissolve and loosen any remaining hair and fat, then scraped clean with a sickle; for the parchment of higher-quality manuscripts, this process may have been repeated. A famous woodcut series from the 15th century provides us with pictorial evidence for the next step: the skin was stretched on racks and let dry for several days. For Carolingian, Byzantine, and early Armenian parchment, the process ended here, leaving a clear distinction between the flesh side (smooth, with few marks and small pores) and hair side (rougher, with larger pores and individual follicles occasionally still embedded, not to mention the scars of any diseases or wounds the animal may have had during its lifetime).

Italian parchment was the first to be treated additionally with chalk dust to whiten the pages. Insular parchment was further rubbed, after drying, with pumice stones, leaving a rough surface comparable to modern cold-press paper, particularly suitable for bonding with the heavier, brighter pigments of colour painting. This very thin, rough material, usually made from calfskin, is known as vellum, thus distinguished from parchment. After it was thus dried and treated, the parchment was trimmed (usually from the main part, cutting off the protruding sections of the limbs, neck and tail) and ready for use

Paper first came into production in Europe in the mid-13th century. Traditionally, the movement of the paper-making process from China has been assigned to the presence of Chinese captives in the city of Samarkand in the mid-8th century, which then spread to the Muslim world and into the west through the crusader contact and the Arab colonies in Spain. Whatever the case, by the 12th century there was already a paper-mill in Arab-controlled Valencia; in 1390, Ulman Stromer founded the first paper mill in Germany, and the use of paper in manuscripts gradually grew, in no small part for economic reasons; it was ultimately much cheaper to harvest a single grove than to slaughter a herd.

Usually the next step was to take the individual leaves and copy the text, before binding. Most texts of the medieval world were copied by whole scriptoria of monastic scribes. It is easy to take pity on their working conditions; we picture the lone scribe, trapped in a drafty stone cell in front of a small wooden table, sloped towards him with a single sheet pinned down and a dimly-lit text at his side. Thanks to the common motif-paintings attached to many gospel-books depicting the pensive Evangelist awaiting divine inspiration in his scriptorium, we are left with many scenes of the scribe's basic equipment. In the early period, the pen, or calamus (Latin, "reed") was usually a trimmed reed or wooden pen, a rod with a triangular shape and tapered at the end. Ever more common from the 10th century was the quill, usually a goose feather. Beside him, the scribe kept a knife for trimming the pages, two ink-horns, a pair of razors for erasing, a pumice-stone for evening the texture of the parchment, chalk for whitening the surface, a straight edge, and an awl, used to prick guidelines in the page to keep the text straight. The oldest and cheapest inks are made from soot, water, and gum, though some examples are made from sepia or gall gum and iron vitriol. It became common starting in the 6th century to make the ink from boiled whitehorn or blackthorn mixed with wine and blackened with soot. Usually this produced a thin, dark brown or yellow colour. Only in the British Isles do we see a true black. Some of these inks, especially those made from vitriol, were so acidic that all we are left with today are "little windows" in the page where the letters once were. In the rarest cases, for royal manuscripts, the inks are made from a metallic base, a deep, rich gold or silver meant to cover manuscripts dyed with purple murex. From the end of the 13th century, the scribe might even have a pair of spectacles to aid his wretched eyesight.

Usually, copying and illumination were done by separate persons, and indeed involved entirely different skills. Some of the best evidence for the process comes from Armenia, from which we have the most vibrant paintings of the medieval world, as well as numerous example and pattern-books, containing charcoal sketches which provided the artistic models for the painter. Inks varied depending on the scriptorium's resources as well as the period and age. Vegetable-bases were common in the Byzantine empire and Italy. Richer, thicker pigments were made from minerals. The design was usually etched in charcoal or sepia-inks before it was filled.

The pages of the finished parchment were let dry and then collected to form the manuscript. The basic form is the codex (Latin caudex, "tree bark"), already mentioned by the first century poet Martial and common by the beginning of the fifth century. The trimmed pages are stacked and folded along the center once to form a gathering. Usually a gathering consists of 4 folded pages, 5 in insular manuscripts, more in the later middle ages, though there certainly exist Gospel books and smaller works with gatherings of up to 50 pages. Pages folded twice, once across and once over, form a quarto; anbody who has done yearbook in highschool will still recognize a basic gathering of 4 quartos, 16 pages.

These gatherings were then pressed, punctured with an awl and sewn vertically across the fold to a cloth or parchment backing. Often these narrow strips of parchment are recycled from discarded manuscripts, and in dissecting or repairing these manuscripts today we often find important texts scribbled on them, occasionally plugging fragments or supporting disputed readings. The parchment is often doubled, sometimes coated in a thick glue to stiffen the backing. Reinforcing strips known as headbands are sometimes added to the topmost and bottommost edge of the backing to reinforce the structure and provide additional support when the book is stored standing. Alternately, the book may be sewn instead to a series of cords laid perpendicularly to the spine.

An additional layer of backing is then attached, this time leaving additional material flowing over the edges of the spine. The next step is, of course, the cover, and this depends on the form. There are three basic cover-types used in the middle ages:

  • de luxe bindings, made from precious metals or ivory and inlaid with gems and stones, representing the richest and most complex metalwork available at the time. Often these are hammered coverings attached with nails or rivets to a thin wooden board. These covers are attached directly to the overflowing strips and then inlaid with a solid piece of parchment which connects the inside cover and the first page of the first gathering.
  • The next type are the eponymous wooden boards from which the codex gets its name. These are attached on the front and back to the headbands of the spine, then covered with a sheet of leather riveted to the boards and folded onto the inside of the cover. This is then covered with the two endsheets, as before. This is certainly the most common type of cover, providing excellent protection while not requiring nearly the resources the de luxe bindings demanded.
  • The third and last type is a simple, soft leather or cloth cover, attached to the headbands and endsheets; this provides the most basic protection for the book, but was relatively cheap.

The life of a manuscript naturally didn't end there. Texts which were commonly used often wore through their covers, in which case they were rebound again and again; the oldest Armenian text, dating from the 7th century, is currently contained in a 15th century binding. Librarians were thoroughly practical about their books, though. Texts which were considered heretical were sometimes destroyed, taken apart and used as material for endsheets and backings. More common is the palimpsest (from the Greek, palin psao, "I wipe over again"). When a text was no longer useful, or, in the case of many Hebrew, Greek, and Gothic manuscripts, could no longer be read, it was often simply erased, the ink washed off with water and scraped off with a strigil. If the paper and binding were still of good quality, it was then re-used and written over. Depending on the inks used and the degree of erasure, it is still sometimes possible to read these hidden texts.

One of the most important conclusions drawn from all this information is that a book in the middle ages was, above all, a commodity, requiring enormous resources and labour to complete. Only the wealthiest communities could create large works like the Book of Kells, which demanded not only vast, expendable herds but the available man-power dedicated entirely to the creation of a document. A single book required at the very least a herdsman, a butcher, a tanner, a metal-worker, a scribe, and a painter, working for several years to produce a finished product. The slow and gradual pace of the dissemination of literacy and knowledge throughout the middle ages had a very material basis. Even so, these books are still the most beautiful artifacts of their age remaining to us.