The Vasari Graph

In 1568 Giorgio Vasari published the second, expanded edition of Le Vite de’ più eccellenti pittori, scultori e architettori, the “Lives of the Most Excellent Painters, Sculptors and Architects.” It is usually called the first work of art history. It is also, if you squint, the first art-history dataset: a few hundred artists, sorted, dated, praised, blamed, and linked to one another by who studied under whom.

So I stopped reading it the usual way and counted it instead. I pulled the full 1568 text, split it into 162 individual Lives, matched the artists to Wikidata for dates and birthplaces, and built a graph. What follows are five things the numbers say. Some confirm what every survey course tells you. One or two do not.

A note on method up front, because it matters for the last chart: Vasari did not just describe the canon, he helped build it. Any modern measure of fame has his fingerprints on it. I have tried to be honest about that rather than pretend otherwise.

The graph: schools that draw themselves

Start with the relationships, since that is the whole promise of a book organized around lineage. I took every “student of” link recorded in Wikidata for Vasari’s artists and drew the network: each artist is a dot, each master-to-pupil connection is a line, dot size scales with how widely the artist is known today.

Then I let a community-detection algorithm colour the dots, with no knowledge of art history at all. It only sees who connects to whom.

The schools fall out on their own. The algorithm finds Venice (the Bellini clan, Giorgione, Titian, Lotto), the Umbrian and central-Italian group around Raphael and Perugino, the Verrocchio workshop that produced Leonardo, the Ghirlandaio shop where Michelangelo trained, the founding generation of sculptor-architects around Brunelleschi and Ghiberti, and two distinct waves of Florentine painting. Nobody labelled these clusters. They are just what happens when you follow the teaching lines.

That is the quiet argument of the whole book made visible: Renaissance art was a family business, and style traveled through workshops. Vasari knew this. The graph is his thesis, drawn.

Does he play favorites? Not the way you think

Vasari was a Florentine writing largely for Florentines, and he has a reputation for it. The usual charge is that he inflates Tuscan artists and is stingy with everyone else.

Half of that is true. Roughly two thirds of the Lives go to Tuscan-born artists, and about a third to Florentines alone. The bias in who gets in is real and large.

But bias in the writing itself is a different question, so I measured it. For every Life I counted the density of praise words, the superlatives Vasari cannot help reaching for: divino, eccellentissimo, raro, perfetto, mirabile, and the rest. Praise per thousand words, Tuscans against everyone else.

The cliche does not survive contact with the count. Word for word, Vasari is if anything slightly warmer about outsiders (about 7.9 praise words per thousand) than about his fellow Tuscans (about 5.9). His favoritism lives in the table of contents, not in the adjectives. Once an artist is in the book, a Venetian or a Lombard gets at least as generous a vocabulary as a man from Arezzo.

Geography, and the story he wanted to tell

Vasari organized art into three “ages,” each better than the last, marching toward perfection. That is a claim about time. It is also, quietly, a claim about place. Here are the birthplaces of his artists, coloured by which age they belong to, next to a test of the progress story.

On the map, the centre of gravity drifts north across the three ages. The first age is tightly Tuscan. By the third it has pulled in Venice, Ferrara, Parma, the whole northern arc, as Florence’s monopoly loosens and the field goes national.

On the right is the teleology, caught in his own prose. His praise vocabulary climbs step by step across the ages, from about 4.8 words per thousand in the first, to 6.4 in the second, to 7.1 in the third. Vasari believed art got better over time, and he wrote as if it did. The progress narrative is not just in his framing chapters. It is baked into the temperature of the sentences.

“Disegno” against “colorito”

There is an old fault line in Italian painting between disegno, drawing and design, the Florentine faith that everything begins with the line, and colorito, the Venetian conviction that painting is built out of colour. Vasari is not a neutral party here. He founded the Accademia del Disegno; the word is practically his religion.

You can watch the argument happen in his word choice. For each Life I counted how often he reaches for disegno against colorito and colore.

Vasari talks about drawing everywhere, so disegno wins across the board. That is the house style. But the relative weight of colour rises exactly where you would predict: the Venetians sit highest on the colour axis, Antonello, Giorgione, Correggio and the rest, while the central Italians cluster down among the draughtsmen. The great theoretical divide of Renaissance painting shows up as a measurable tilt in one man’s vocabulary. He was taking a side, and the counts catch him doing it.

The long shadow

Now the hard question, and the reason for the caution up top. Vasari gave each artist a certain amount of room, some a few hundred words, Michelangelo nearly fifty thousand. Does that attention predict who we still care about?

I plotted the length of each Life against a proxy for modern fame, the number of Wikipedia language editions with an article on that artist, on log scales, and fit a line.

The line is weak. How much space Vasari gave an artist explains only about a tenth of their modern presence. The canon reshuffled itself after him.

The interesting part is who sits far off the line. Below it are the long Lives that time quietly forgot, and they are almost all Vasari’s own circle: his collaborator Cristofano Gherardi, called Doceno; Aristotile da Sangallo; the sculptor Tribolo; Simone Mosca; his friend Lappoli. He gave the people he knew and worked with pages that posterity did not ratify. Above the line are the artists he treated briefly who tower over us now: Leonardo, on a fairly short Life; Botticelli, whom the nineteenth century rediscovered and Vasari rather underrated; Giorgione, Duccio, Masaccio.

Here is the catch, and it is why this chart is a mirror as much as a measurement. Wikipedia is not an independent judge. It inherits centuries of art history that begins with Vasari. So the surprise is not that his attention correlates with modern fame, it is how weakly it does, even when the fame is measured with a ruler he helped forge. The divergences are the real signal. They mark the places where later taste, connoisseurship and rediscovery overruled him, and where his personal loyalties failed to carry.

Conclusions

None of this replaces reading Vasari. The counting is not smarter than the book; it just asks different questions and answers them without getting tired. It confirms the thesis of lineage, complicates the charge of favoritism, catches the progress story in the act, and shows both the reach and the limits of one Tuscan’s opinion four and a half centuries on.

The first art historian was also, it turns out, a fairly good data source. He would probably have hated the charts. He might have liked being at the centre of one.

The corpus is the 1568 Giuntina edition, from Wikisource. Artists were matched to Wikidata for dates, birthplaces and teaching relationships. Fame is proxied by Wikipedia language-edition counts, a measure that, as noted, partly descends from Vasari himself. Code and data available on request.