What it is, where these hands came from, and which of the old books still hold.
Paleography is the study of old handwriting: identifying it, dating it, localising it, and — first of all — reading it. That last is the part this site is for, and it is worth saying plainly that it is a separate skill from knowing the language.
A reader who knows Latin brings a powerful advantage to a manuscript: they know which words are likely, so a half-legible run of letters can be resolved by sense. What they do not yet have is the other half — the trained eye that maps ink to letters in a script it has never seen. The two halves work together, and each can mislead. Fluent Latin will confidently supply a word the scribe did not write; a sharp eye with no Latin will read letters faithfully and produce nonsense. Learning to read a hand is learning to hold both at once.
Until about the year 800, Greek books were written in majuscule — capitals throughout, no spaces between words, no accents or breathings as a matter of course. This is the writing of the great fourth-century Bibles. Reading it is an exercise in dividing a continuous stream of letters into words, and ancient readers did it out loud, which helped.
Then, within a few generations, Byzantine scribes adopted minuscule — a small, connected book hand developed out of everyday cursive. The change was not merely cosmetic. It brought word division, accents and breathings written as standard, and punctuation. A tenth-century Greek page is far closer to a printed page than a fifth-century one is, and that is why this site teaches minuscule: it is both the more approachable hand and the one in which nearly all classical and patristic Greek actually survives, because it is what the Byzantine scriptoria copied.
What replaces the old difficulty is a new one. A scribe writing at speed fuses letters into ligatures — compound shapes that may resemble neither of their parts — and one letter can take several forms in a single line depending on what stands next to it. That, rather than the alphabet, is what a reader of printed Greek has to learn.
The Latin story runs the other way round: from a formal, monumental hand towards a workaday one, with a deliberate reform in the middle.
Roman books were written in capitals, then in uncial — a rounder hand suited to pen and parchment rather than chisel and stone. The word itself survives from a complaint. Jerome, in his preface to the Book of Job, grumbles at luxury manuscripts:
Habeant qui volunt veteres libros vel in membranis purpureis auro argentoque descriptos vel uncialibus, ut vulgo aiunt, litteris, onera magis exarata quam codices.— quoted in Thompson, Introduction (1912)
“…or in uncial letters, as they commonly call them — burdens rather than books.” Nobody is sure what Jerome meant by the word. The name stuck to the script anyway, on the strength of a fourth-century aside about other people's extravagance.
After Rome the western kingdoms each developed their own minuscule out of Roman cursive, and by the eighth century a scribe from one region could struggle badly with a book from another. The Carolingian reform ended that: the schools of Charlemagne's empire standardised a clear, round, well-spaced minuscule that could be read anywhere. It is the easiest medieval Latin hand, and it is where a beginner should start.
It also has a long tail. When fifteenth-century humanists went looking for classical texts, what they found were Carolingian copies — which they took for genuinely ancient writing and imitated. Their imitation became roman type. The letters you are reading now are its descendants, which is why a modern reader meeting Caroline minuscule for the first time often feels an odd flicker of recognition.
Scribes abbreviated constantly, to save parchment and labour. Thompson draws the distinction that still organises the subject:
Abbreviation is the shortening of a word by the omission or suspension, as it is called, of the end…; contraction is the shortening of a word by omitting letters from the body and leaving the beginning and end. The system of contraction is superior to that of suspension, in that it affords a key to the inflections.— Thompson, Introduction (1912), ch. VII
That last clause is a practical fact rather than a taxonomy. A contraction keeps the word's ending, so its case and number survive and the sentence can still be parsed; a suspension cuts the ending off and destroys exactly that information. It is why one abbreviation can be resolved from the grammar and another cannot.
The surprise is where contraction came from. The ancient Greek system, Thompson says, was suspension only. Contraction arrives later, and:
The contraction system was a Christian system, an innovation brought in through the Hellenistic Jews who translated from the Hebrew, particularly the authors of the Septuagint translation of the Old Testament.— Thompson, Introduction (1912)
The nomina sacra — DS for deus, DNS for dominus, XPS for Christus under a stroke — are the visible residue of that. They pass from Greek into Latin practice and then into the whole medieval system of abbreviation. A habit of reverence about divine names became, by degrees, a general technology for saving parchment.
Two are public domain and both are used here directly.
E. M. Thompson, An Introduction to Greek and Latin Palaeography (Oxford, 1912). His own expanded replacement for the 1893 Handbook — same architecture, roughly twice the text. Prefer the later book; the earlier one is not wrong, only thinner.
A. Cappelli, Lexicon Abbreviaturarum (1929). The standard dictionary of Latin manuscript abbreviations, some fourteen thousand forms. Where a sign's expansion is marked verified in the exercises, it means Cappelli's or Thompson's page was read as an image and says so.
What has not held. A century of work has redrawn much of Thompson: the dating and localising of individual hands above all, and the tidy national-hands scheme of his chapter on Visigothic, Lombardic and Merovingian writing.
He half-expected it. A footnote in the 1912 edition points the reader to a work by E. A. Lowe, Scriptura Latina Minuscula Antiquior, “announced, but not yet published” — the project that became the Codices Latini Antiquiores and superseded a good deal of what surrounds the footnote. For current guidance the standard replacements are Bischoff's Latin Palaeography and, for Greek, Reynolds and Wilson's Scribes and Scholars; neither is public domain.
It teaches reading, and only reading. It will not teach you Latin or Greek, and it does not try to date or localise a hand — that is the graduate seminar, and it needs a teacher.
Everything shown is cut from photographs of real manuscripts along transcribers' own baselines, on openly licensed ground truth. Nothing is redrawn. Where an expansion has not been checked against a printed authority, the tool says so on the face of the gloss, because a confident wrong expansion is worse than none.