How to date a manuscript by eye

1. The support

The cheapest signal, and the bluntest. Papyrus means antiquity: Greek and Egyptian material, mostly recovered from Egypt, and in this corpus almost all of it is earlier than anything written on skin.

Parchment is the Latin Middle Ages. It tells you very little on its own, because it spans a thousand years, but how it was prepared can: a hard, white, evenly thin leaf is more likely to be late and Italian than early and insular.

Paper is the useful one, because it is a terminus post quem. It reaches Europe through Islamic Spain and Sicily, is being used in Italy from the thirteenth century, and becomes ordinary north of the Alps across the fourteenth and fifteenth. A paper codex is a late codex. In this corpus 223 manuscripts are paper against 1,815 parchment, and they cluster exactly where you would expect.

2. The script

The strongest signal, and the one that takes the longest to learn. The very broad sequence for Latin books:

• Late antique and early medieval — capitals, uncial, half-uncial; then the regional hands, of which insular is the most recognisable, with its wedge-topped ascenders and its own abbreviations.
• Caroline minuscule, from the late eighth century — round, separated, legible, the ancestor of the letters you are reading now. If a page looks unexpectedly easy, this is often why.
• Gothic textualis, from the twelfth and thirteenth centuries — compressed laterally, angular, the curves flattened and adjacent bows “biting” into each other. Densely abbreviated. Two columns, in a large book.
• Later gothic cursives and hybrids, fourteenth and fifteenth centuries — written faster, with looped ascenders and a single-compartment a. In England, anglicana and secretary, often mixed in one book.

The trap. Humanistic minuscule, developed in Florence around 1400, is a deliberate revival of Caroline — the humanists took eleventh-century copies for classical ones and imitated the letters. So the most round, open, readable page in a set of five may be the youngest thing on the table by three hundred years. The game will punish you for this, which is the point of including it.

3. The ruling and the layout

How the scribe laid out the page before writing on it is often easier to read than the writing. Ruling with a hard point, scored into the skin so it stands up on the other side, is the earlier technique; ruling in plummet — lead, a faint grey line — becomes usual in the eleventh and twelfth centuries; ink ruling is later still. Look for the prick-marks in the margins that guided it.

Two columns suggest a large book meant to be read from a lectern, which in practice often means a gothic bible, a law book or a commentary. Long lines suggest something smaller and more private. Counting the lines in the written space is worth more than it sounds: a thirteenth-century pocket bible may have fifty and a ninth-century gospel book fewer than twenty.

4. The decoration

Decoration dates a book by fashion rather than by technology, so it is the least reliable of the four and the most fun. Gold leaf, painted borders that grow out of an initial and run the length of the margin, and framed miniatures with modelled faces all point later rather than earlier. A plain red initial points nowhere in particular: scribes did that for a thousand years.

Be careful about how much this tells you. 1,940 of the 2,201 manuscripts here carry some decoration, so "decorated" is almost the default state of a surviving book — which is itself a fact about survival, not about production. Handsome books were kept.

What the eye cannot do

It cannot give you a year. Cataloguers write "14th century, first quarter" or "c. 1320–30" because that is the honest resolution of the evidence, and a page that tells you 1323 is telling you about a colophon, a dated document, or a guess.

It cannot separate a script from the place it was written: regional hands overlap in time, and a Bolognese and a Parisian gothic book of the same decade look very different. It cannot see a scribe copying an exemplar faithfully enough to reproduce its archaisms. And it cannot see a composite — a volume whose quires were written a century apart and bound together later, which in this collection is common enough that two records sometimes share a shelfmark.

Where to read properly

The standard introductions are Raymond Clemens and Timothy Graham, Introduction to Manuscript Studies, and Michelle P. Brown, A Guide to Western Historical Scripts from Antiquity to 1600, which reproduces dated specimens you can compare against. For gothic book hands specifically, Albert Derolez, The Palaeography of Gothic Manuscript Books. All three will disagree with the simplifications above, and they are right to.