LAST SEPTEMBER, an earthquake shook Italy's Umbria region, sending a Cimabue fresco on the ceiling of Assisi's St. Francis Basilica crashing to the floor. Two friars assessing the damage were killed in an aftershock. In the wake of the disaster, Italians were heartened to learn that the cathedral's most treasured artwork--the St. Francis Legend frescoes, attributed to Giotto--was spared. The radiant panels were obscured only briefly, by a plume of dust that floated up from the Cimabue's shattered remains.

A recent book by Italian art restorer Bruno Zanardi, however, has sent another shock wave through Assisi. The Florentine master Giotto didn't paint the 1296 St. Francis Legend, claims Zanardi, who cleaned the frescoes between 1978 and 1982. In fact, he writes in Il cantiere di Giotto (Skira, Milan), three unheralded Roman artists painted the series. In Giotto's stead, Zanardi places Pietro Cavallini, the member of the trio responsible for the cycle's finest panels, such as "St. Francis Preaching to the Birds."

Does the history of art need a rewrite? Ever since Vasari attributed the frescoes to Giotto in 1568, the St. Francis cycle--which chronicles the life of the peace-loving monk--has been celebrated as the creation of a daring Florentine who single-handedly pumped blood into the lifeless poses and frozen surfaces of Byzantine painting. Ernst Gombrich's Story of Art claims that Giotto "rediscovered the art of creating the illusion of depth on a flat surface." Before Giotto brandished his brush, Gombrich writes, "Nothing like this had been done for a thousand years." As a result, Giotto's Florence has long been seen as the wellspring of the Renaissance.

Zanardi's study is by far the most comprehensive look at the twenty-eight frescoes. He agrees that the cycle is groundbreaking: "For the first time, we see completely the naturalistic style that will dominate Western art until our own century." But he disagrees that Florentine know-how was responsible for the breakthrough. To make his case that Cavallini was the fresco's principal author, he points to the habits of medieval fresco workshops. A workshop's methods, reports the restorer, "remained consistent even in different works"; therefore, today's scholars can detect the signature flourishes of these traveling troupes of artists. And, Zanardi says, the St. Francis Legend frescoes have "Rome" written all over them.

The argument is complicated. First, for all its supple realism, Zanardi reveals, the St. Francis cycle was created with a technique that evokes not medievalists but Matisse: paper cutouts. In order to standardize output and ensure stylistic unity in the hastily painted figures, the leader of a workshop fashioned numerous patroni, or tracing patterns on waxed paper. While no patroni have survived to modern times--making their existence a source of dispute--Zanardi says the evidence was right before his eyes: "I simply traced the heads in the Franciscan cycle on transparent paper and then checked how well all these outlines fit onto each other." They fit like a glove.

What does all this have to do with determining authorship? Zanardi concludes that, in fact, three discrete sets of patroni appear in the course of the St. Francis series--suggesting that different workshops executed the cycle. Consider how the figure of Francis gently morphs: In the early panels, Zanardi reports, he is "1.8 meters tall with wide eyes and mouth and a pink complexion." In panels 7 through 22, the monk shrinks, becoming "1.7 meters tall with long mouth and eyes and a yellowish complexion." The final six panels display a sickly saint "1.9 meters tall with tight mouth and eyes and a greenish complexion." The master who created the second set of panels, then, produced the lion's share of the work--as well as the most elegant compositions.

Zanardi next scrutinizes how the workshops created such vibrant flesh tones. He claims that the paint recipes used in the St. Francis Legend are all Roman in flavor--and therefore unlikely to be the product of Giotto's workshop. The three cooks in the Assisi kitchen started with a generous sage-green priming layer, a particularly Roman preference. (In contrast, the complexions of Florentine figures are decidedly less Mediterranean in tone.) These verdant layers were then covered by a succession of pink and rose shades. The second Assisi workshop, Zanardi notes, had a curious touch: "Very thin but nonetheless visible threads of color" were placed on top of each other to enhance modeling. Although these striations blend together when viewed at a distance, from the scaffold Zanardi could distinguish the juxtaposed strands of green, ocher, and pink. (Up close, in fact, the effect gives the faces a colorful-yet-wrinkled appearance.)

Here is the key to Zanardi's argument. "Pietro Cavallini's frescoes in the Roman church of Santa Lucia," he says, "display precisely the same mode of applying color--the spreading of thread-thin brushstrokes to create a subtle weaving effect." Giotto's technique, Zanardi continues, was wholly different. Take Giotto's famed Arena Chapel frescoes in Padua, created ten years after the Assisi frescoes, in 1305. In lieu of a woven texture, the Florentine preferred velvety-smooth, translucent coats of color. Other Giotto frescoes suggest his style remained consistent throughout his career; as a result, concludes Zanardi, "we must exclude Giotto as the author of the St. Francis Legend."

So who was this Pietro Cavallini, then? He lived sometime between 1240 and 1330. A letter written by Cavallini's son around 1340 claims that his father "lived to be 100 years old, and always went about with an uncovered head, even in winter." A few commission statements have been found. That's about it. (In fact, so little is known that some Italian scholars have countered that Cavallini is a composite of various trecento artists.)

Making matters worse, most of Cavallini's works no longer exist. In 1823 his masterpiece, a huge fresco cycle in the Basilica of San Paolo outside Rome, was consumed in a fire. The expansion of St. Peter's Cathedral in the Vatican probably led to the destruction of other works. The Cavallinis that survive are primarily mosaics--in particular, a lush series at the church of Santa Maria, in the Trastevere section of Rome. These mosaics, completed around 1290, display a precocious sense of perspective that resonates with the St. Francis series: The knees of seated figures amply protrude from beneath drapery, and the faces blush with life. But mosaics, the signature medium of Byzantine art, have never been seen as a vanguard form. As a result, Cavallini's early stab at naturalism has frequently been glossed over by art historians.

While Zanardi's reattribution has unsettled many Italians--the newspaper La Stampa huffed that the judgment was "hasty"--it won't surprise many American scholars. Studies on Giotto have long been divided between Italians and Anglo-Saxons, with the latter being far more skeptical of Vasari's vaunting of all things Florentine. Several art historians have wondered aloud if it were really plausible for Giotto, born around 1270, to have been granted a grand papal commission in his mid-twenties. Nowadays, many English-language textbooks ascribe the St. Francis Legend to "Anonymous."

In a new book, Painting in the Age of Giotto (Penn State), Canadian art historian Hayden Maginnis concludes on stylistic grounds that the St. Francis Legend cannot be by Giotto. In fact, he believes Giotto consciously broke away from the naturalistic mode developed at Assisi. Compared to the scrupulously detailed architecture of the Assisi cycle, Giotto's frescoes are radically spare in design, with compositions that are less mimetic than symbolic. "Verisimilitude was sacrificed to narrative purpose," he writes. "Pure naturalism takes second place." In Giotto's Padua frescoes, he observes, stark blue backgrounds and a flattened perspective focus attention on the expressions of the figures.

For his part, Zanardi says that his discoveries will hardly crumble Giotto's marbled reputation: They simply underscore that Giotto was not an "isolated miracle" but the beneficiary of a creative ferment that had bubbled up from Rome. Indeed, Zanardi--who's from Pontremoli, in Tuscany, not Rome--thinks that the traditional notion that Florence spawned the Renaissance is awfully quaint. "Without question," he says, "in the thirteenth century, Rome was the most important city in the world. It just makes sense that the Renaissance began there."

DANIEL ZALEWSKI