Ink and Vellum: A Night in the Medieval Scriptorium

Duration: 13 minutes 59 seconds Format: Solo narration Topic: Medieval scriptorium manuscript production

Chapters

#	Title	Timestamp
1	Opening Hook	0:06
2	What Is a Scriptorium	0:37
3	Preparing the Vellum	1:07
4	Making the Ink	2:35
5	The Quill	3:59
6	Layout and Script	4:41
7	The Workshop	5:57
8	Color and Illumination	7:07
9	Binding	9:13
10	Scale and Time	10:08
11	Decline of the Monastic Scriptorium	11:24
12	Closing Reflection	12:33

Full Transcript

Somewhere in northern France, around the year 1150, a monk scraped a dried goat-skin stretched tight on a wooden frame. He had been doing this for two hours. He would do it for another two.

Welcome. Tonight we are going to talk about medieval manuscripts. Specifically, how they were made — from the animal skin to the finished page. This is, uh, a slow process. It was slow then too.

A scriptorium was a room, usually attached to a monastery. It was where monks copied books. Sometimes they copied scripture. Sometimes laws, or medical texts, or old Roman poetry they had no particular use for.

The room was cold. That was intentional, kind of. Cold air kept the vellum stable. Cold fingers, well — those were just a consequence.

Before any writing happened, you needed a surface to write on. And in medieval Europe, that surface was vellum. Which is — you know — treated animal skin. Usually calf. Sometimes sheep or goat.

The skin was soaked in a solution of water and lime. Calcium hydroxide, essentially. It loosened the hair follicles and began to break down the fat. The monks called this liming. It took several days.

After liming, the skin was pulled over a wooden frame called a stretching frame. Wooden pegs held it under tension while it dried. Tension is the key word here. If you stretched too far, the skin would split. Too loose, and it would wrinkle as it dried.

While it was still damp and stretched, a monk would scrape it with a curved blade called a lunellum. Circular strokes. Very light. Removing the last of the flesh and the surface irregularities.

When it finally dried, the result was not leather. It was something different. Translucent, almost. Smooth on both sides. It had a faint, warm smell — like very old paper, but slightly animal. Scholars still notice it, in the archives, centuries later.

Once you had your vellum, you needed ink. The most common medieval ink was iron gall ink. It sounds complicated but the basic recipe is — oak galls, iron sulfate, water, and a little gum arabic.

Oak galls are those little round growths you see on oak trees. They form when a gall wasp lays eggs in the bark. The gall is essentially the tree's — well — overreaction. Full of tannic acid.

You crush the dried galls and soak them in water, usually rainwater. Then you add ferrous sulfate — iron salt — and the tannic acid and the iron react to form a dark, dark compound. Almost black.

The gum arabic comes from acacia trees. It is a binder. It keeps the ink particles in suspension, and it gives the ink a slight viscosity — a kind of slow, thick quality — that lets it sit on the quill without just running off.

Freshly made iron gall ink was actually brown. It darkened as it oxidized. Left on vellum for a few years, it turned the deep brown-black you see in old manuscripts. And, uh — it also ate into the skin very slightly. Which is why some old pages have these tiny holes where the ink sat heaviest.

The writing tool was almost always a quill. Usually from a goose. The outer feathers of the left wing were preferred, actually — they curved away from the right-handed scribe's line of sight.

Before use, a quill was soaked in water to soften it. Then the tip was trimmed and split with a small knife — the penknife. That is, quite literally, where we get the word penknife.

The split created a small channel that drew ink up by capillary action and released it evenly onto the page. A practiced scribe could keep a nib writing for — maybe half an hour before it needed re-trimming.

Now. The actual writing. Before a scribe wrote a single letter, the page had to be ruled. You used a blunt stylus — or sometimes a lead point — to press faint, horizontal lines across the vellum. These guided the script so lines stayed even.

Vertical lines marked the column boundaries. Prick marks — tiny holes made with an awl — kept the ruling consistent from page to page. There is something, I think, quite pleasing about this. A system of holes, guiding the words that would last a thousand years.

Different monasteries used different script styles. The Carolingian minuscule, developed in the eighth century under Charlemagne's scriptorium reforms, was one of the most widespread. Clear, regular, open. You can still read it without much training.

By the twelfth and thirteenth centuries, scripts grew more compressed. Gothic textualis — the angular, dense hand we usually picture on medieval documents — became dominant in northern Europe. It fit more words on a page, which mattered. Vellum was expensive.

A scriptorium was rarely one monk working alone. Typically it was organized. One monk, the armarius, oversaw the process. He assigned texts, maintained the books of the library, and set the copying schedule.

Scribes worked at individual desks called carrels. Sometimes these were fixed, with a slanted writing surface. A model book — a copy of the text being reproduced — would be propped nearby or laid flat while the scribe copied from it.

In larger operations, a lector would read aloud while multiple scribes wrote simultaneously. This was faster. It also introduced errors, because speech is ambiguous and scribes occasionally misheard. Some of those mishearings survived in manuscripts to this day.

Corrections were made with a small scraping knife. You could — carefully — scrape the iron gall ink off the vellum surface. Not perfectly. Under ultraviolet light, the ghost of the original text almost always reappears.

After the text was complete, the page would go to the rubricator. This was sometimes the same monk, sometimes a specialist. The rubricator added the red text — rubrics. Chapter headings, saints' feast days, section markers. The word rubric comes from the Latin for red: ruber.

For decorated manuscripts — the ones that survive in museums now — there was also the illuminator. This was a skilled, specialized role. The illuminator added the painted initial letters, the borders, the miniature scenes.

The gold in illuminated manuscripts is real gold. Applied as gold leaf — sheets thinner than this sentence — or as powdered gold mixed with gum arabic into a kind of paint. The process of laying gold leaf began with gesso. A raised ground made of gypsum, white lead, sugar, and egg white. Stiff but slightly sticky.

The illuminator would breathe onto the gesso to reactivate it — just slightly — before pressing the gold leaf down. Then burnish it with a smooth stone, often agate. The gold adhered. It caught the light. It still does, nine hundred years later.

The pigments were extraordinary. Lapis lazuli from Afghanistan — ground to a powder, washed repeatedly to remove impurities, producing the deep blue called ultramarine. The word means beyond the sea. It describes where the stone came from.

Vermilion, made from ground cinnabar. Malachite for green. Orpiment, a yellow arsenic sulfide, which was beautiful and extremely toxic. Scribes and illuminators handling orpiment regularly likely absorbed low levels of arsenic through their skin. Most would not have known.

When all the pages were finished, they were gathered in order. Folded into quires — groups of four, six, or eight folded sheets. A quire of four sheets gave you eight leaves, sixteen pages. Standard format.

The quires were sewn together along the spine using linen thread. Loops of thread caught onto raised cords or thongs stretched across a sewing frame. This is the structural core of a medieval book. Those cords, when the cover was applied, often showed as the ridges on the spine.

The cover was wooden boards — oak, beech, or lime — attached to the sewn text block. Covered in leather. Clasps of metal kept the book closed, which mattered, because the vellum inside would expand and contract with humidity if left open.

A large Bible or psalter — the kind with full illumination — might take a single monk several years to produce. Working through all seasons. Through cold winters with poor light. Through feast days when work stopped. Through illness.

Some scribes left notes in the margins. These are called colophons and marginalia, and they are, uh — remarkably candid. One scribe in ninth-century Ireland wrote: as the cold wind comes, I cannot hold the pen. Another wrote: three fingers hold the pen, but the whole body toils.

There was one scribe, probably in a monastery in Germany, who finished a long text and wrote: now I've written the whole thing. For the love of God, give me a drink.

Somewhere between twelve and forty sheepskins were needed to produce a single copy of the Bible. For a large, full-format illuminated Bible — the kind kept chained to a lectern in a cathedral library — it might be upward of a hundred and sixty calf skins. The economics of that alone shaped what got copied and what did not.

By the late fourteenth century, urban commercial scriptoria had begun to appear. Book production moved outside the monastery. Lay craftspeople — booksellers, illuminators, parchment-makers — organized into guilds in Paris, in Bruges, in Bologna.

The demand for books had grown beyond what monasteries could supply. Universities were opening. Lawyers needed law books. Students needed glossed copies of Aristotle. The market existed, and the market, as it tends to do, found a way.

Gutenberg's press changed everything after 1450. But the monastic tradition did not vanish overnight. Some monasteries continued copying books well into the sixteenth century. For liturgical texts, in particular — missals, antiphonaries, large choir books — manuscript production persisted because the printed page was simply too small to prop on a choir stand.

The oldest surviving European manuscript on vellum dates to around the fourth century. The newest monastic manuscript copy was probably made sometime in the nineteenth century. That is — roughly fifteen hundred years of continuous practice.

What strikes me — and I find this quite settling, actually — is how much the process was about patience. About doing the same thing again tomorrow, and the day after. Soaking the skin. Scraping the skin. Ruling the page. Dipping the quill.

No part of it was fast. There was no shortcut that did not eventually ruin the work. A monk who rushed the liming would get vellum that cracked. A scribe who pressed too hard would cut through the page. The process demanded attention, and then — once you had given it that attention — it rewarded you with something that lasted.

That is, I think, enough for tonight. The scriptorium is quiet now. The fire has died down. The skin on the frame has dried, and tomorrow someone will take it down and begin to rule the lines. But not yet. For now, it rests.

Sleep well.

Sources

De Hamel, Christopher. Scribes and Illuminators. British Museum Press, 1992.
Brown, Michelle P. The Lindisfarne Gospels: Society, Spirituality and the Scribe. British Library, 2003.
Clemens, Raymond and Timothy Graham. Introduction to Manuscript Studies. Cornell University Press, 2007.
Stirnemann, Patricia. «Quelques bibliothèques princières et la production hors scriptorium au XIIe siècle.» Bulletin archéologique du Comité des travaux historiques et scientifiques, 1984.
Bischoff, Bernhard. Latin Palaeography: Antiquity and the Middle Ages. Cambridge University Press, 1990.

Audio and Music Notes

Narrator voice: MiniMax system voice English_ManWithDeepVoice — deep, mature, unhurried, documentary register. Speed set to 0.85× for sleep-inducing cadence. No pitch adjustment applied (voice already sits at the correct register for the channel persona).

Background music: Generated instrumental, title「Scriptorium Night — Sleep Bed」. Produced via fal.ai MiniMax Music v2.6, instrumental mode. Sparse, slow (~50 BPM) ambient piece with solo acoustic character, no percussion, monastic late-night quality. Duration: 178 seconds. The same track serves as intro clip (first 6 seconds, faded out over 2.5 s), outro clip (first 8 seconds, faded out over 4 s), and full-session background bed (looped, mixed at −27 dB below speech — at least 12 dB below the −18 LUFS speech target). This is AI-generated music; no third-party rights apply.

Loudness: Final mix normalized to −18 LUFS target. No sudden volume jumps.