One of the strangest and most beautiful forms of art is also one that is almost universally ignored: the medieval pavement mosaics that decorate the floors of many churches in Italy. In some ways, it is not surprising that when visiting churches in order to appreciate art, people look up rather than down. Many Italian churches are full of glorious frescoes and statues which are more than enough to absorb your attention: when you are in the Sistine Chapel, cricking your neck as you stare at Michelangelo’s miraculous ceiling, the last thing you think of doing is casting your eyes down to the floor. Yet the floor of the Sistine Chapel is covered by a marvellous floor mosaic which, while not the equal of Michelangelo’s masterpiece, is certainly an exhilaratingly complex and absorbing work of art.
Most floor mosaics aren’t in locations that possess art as stupendous as the Sistine Chapel — although the room in the Vatican with Raphael’s fresco of the School of Athens also has a very fine floor mosaic that no one ever looks at. But there is another reason why floor mosaics tend to be ignored, even in churches graced with only the most mediocre statues and frescoes: floor mosaics are usually very hard to see in their entirety, since they generally have people and pews perched on them, and both get in the way of a clear view. To see the patterns made by floor mosaics properly, you need to look down on an empty church from a position high above the floor. This is usually not possible, and it is often only from photographs that you can see a floor mosaic as a complete composition. And then the smallness of the photographic reproduction diminishes the effect of the original work: the size and scale of the real thing are critical to its impact.
We also like works of art to “mean something”, or at least to tell a comprehensible story. Medieval floor mosaics do not tell a story, and there are no familiar human figures: often there are no figures of any kind. Floor mosaics were frequently conceived as an entirely abstract form of art. A series of interlocking geometric patterns may, in the Middle Ages, have helped those who looked at them contemplate the relation between heaven and earth. But for most people today, they are not easy to imbue with any meaning at all, and they do not fit with the usual idea of what religious art from the Middle Ages is supposed to be like: there are no images of the sufferings of Christ or of his mother Mary, and no gory depictions of the martyrdom of saints.
But even given all those drawbacks, the extent to which art historians have passed over medieval floor mosaics in almost total silence is still surprising. It is true that the mosaics cannot be bought and sold easily, since they are cemented to the floors of the structures for which they were originally intended, so there is no market in floor mosaics, and no auctions with record-breaking prices — two practices which are always a potent stimulus to academic art-historical endeavour. But frescoes are stuck to walls, and can’t easily be bought and sold either — yet that hasn’t stopped there being a huge scholarly literature on fresco-painting. I could only find two books in English on medieval floor mosaics. The most recent was published in 1980. There are a few more books on the topic in Italian, but even in Italy, the amount of published work on floor mosaics is pitifully small.
But what does it mean? The chickens have caught the fox in one of the figurative mosaics of Saints Mary and Donatus, Murano (VASSIL CC 0 1.0) Part of the reason for the lack of scholarship may be that there is so little evidence for art historians to investigate. The artists who designed and made medieval floor mosaics have left very little deposit in the surviving written records. Nothing has come down to us that indicates how much they were paid, or how they were commissioned, or how many of them were involved in each project, still less what the message in their abstract patterns was supposed to be. But the artists were obviously proud of their work, because some of them signed it — although since the men who designed the floor mosaics often also designed the churches in which those floors appeared, it is not always clear whether the signature refers to the floor, or the building, or both.
Four families seem to have monopolised the production of floor mosaics in the two centuries between 1100 and 1300 in Rome and the area north of the city, up as far as Tuscany and Umbria. Each of those four families produced an astonishing four generations of artists who worked as architects and designers of floor mosaics. And they didn’t only design mosaics for floors: they often decorated pulpits, tombs and other church ornaments with their patterns.
While they don’t seem to have been related to each other, those four familes are collectively referred to as “Cosmati”, which is strange, because only two of the 27 artists whose names we know were actually called “Cosma”, while there were three Pietros and four Jacopos. But the floor mosaics those families produced have been called “Cosmatesque” since the 19th century, when the first systematic study of them was made, even though there is no particular evidence that the men called Cosma were more active or creative than their fellow-workers.
Cosmatesque mosaics have a very distinctive style that usually involves porphyry circles connected by bands of coloured marble decorated with geometric figures, often triangles or hexagons. One of their distinctive designs — it often forms the centrepiece of the whole pattern — is the “quincunx”: four circles symmetrically arranged around a fifth, central circle. The Cosmati did not invent the quincunx, but they did develop and elaborate novel forms of it. Their designs have a subtlety that is not immediately obvious. For instance, a quincunx in one of their floor mosaics will typically look symmetrical along both horizontal and vertical axes, but in fact will not be when examined more closely.
Designing and constructing a floor mosaic requires a number of technical skills, including very precise cutting of marble. Some of the restorations over the last 100 years have demonstrated how difficult it can be to get those techniques right: restorers have reset and relaid large sections of marble floor mosaics, only to find that their constituent pieces quickly started to come apart, so that elements of a mosaic that had fitted together perfectly for close to a millennium started to separate from each other and from the background in which they were embedded within less than 50 years, once the restorers had “improved” them.
How did the Cosmati craftsmen learn their remarkable skills? At the end of the 11th century, Leo of Ostia wrote a history of the abbey of Monte Cassino. Leo’s history records how Desiderius, who was abbot of Monte Cassino when Leo was a monk, completely rebuilt the abbey. Leo wrote that Desiderius “sent envoys to Constantinople to hire men who were experts in the art of laying mosaics and pavements”. Among other tasks, they were to “lay the pavement of the whole church with various kinds of stones”. The effect of the Greek artists’ work was duly stupendous. “One would believe that in the marble of the pavement flowers of every colour bloomed in wonderful variety,” Leo enthused. That floor cannot be seen today: it was unfortunately destroyed even before the Allies flattened Monte Cassino abbey in World War Two. But an engraving made while the floor was still extant gives an impression of what it must have been like, even if it cannot give any idea of those spectacular colours.
Leo adds another very striking claim. He says that “since these arts [of making mosaics and laying pavements] had been left uncultivated for more than five hundred years . . . the abbot in his wisdom decided that a great number of young monks in the monastery should be thoroughly initiated in these arts in order that their knowledge might not again be lost to Italy.” It has certainly been frequently claimed by art historians that craftsmen from Byzantium had a critical role in educating Italians in artistic techniques that had been lost after the fall of Rome. But as far as I know, Leo of Ostia provides the first and only record to indicate that it actually happened.
It is plausible that Greeks from the Byzantine empire, where the art of creating and laying floor mosaics had been maintained, taught the monks of Monte Cassino how to make mosaic floors. And it is very easy to imagine the Monte Cassino monks spreading the techniques to craftsmen in Rome, including to the earliest of the “Cosmati” dynasties. But is Leo of Ostia right when he says it was those Greek mosaicists who brought the art of floor mosaics back to Italy?
The difficulty is that Monte Cassino was not the first church to be decorated with a mosaic pavement since the Lombard invasions of the late 6th century and the subsequent loss of nearly all of the technologies that had been developed during antiquity. Desiderius’s new church of Monte Cassino was consecrated on October 1, 1071. The abbey of Pomposa near the mouth of the river Po has a magnificent floor mosaic which is based around circles and geometrical figures, and which even has a quincunx on its central axis. It was not built by marble workers from Rome: the small size of the “tesserae” or mosaic tiles, together with the techniques used, are both different from those employed by any of the Cosmati families. Pomposa’s floor can be dated to 1026, which makes it nearly 50 years earlier than the floor mosaic at Monte Cassino.
Perhaps Greek workmen were involved in Pomposa’s floor mosaics too. We know they worked on the mosaics from the 11th century in Saint Mark’s Basilica in Venice: the inscriptions they left in Greek demonstrate their presence. Pomposa, although not part of the Veneto and never directly controlled by Venice, is not far from that city, and it is perfectly possible that Greek mosaicists played a central role in creating the floor mosaic at Pomposa. Many of the designs used at Pomposa have antecedents in earlier Byzantine churches in the Middle East — in just the same way as do many of the pattens in floor mosaics executed by the Cosmati families in Rome and its environs. So even if Leo is wrong that Italy’s first floor mosaic since antiquity was constructed at Monte Cassino, he may be right that it was Greek artists who made it possible for floor mosaics to be revived in Italy in the Middle Ages.
Subtlety, precision and skill: The mosaic floor of San Clemente, Rome, features “slices” of porphyry taken from Roman columns (©Godong/Universal Images Group via Getty Images)
The floor mosaics of northern Italy have one characteristic that those of Rome and central Italy do not: they frequently contain representations of people and animals. The splendid floor mosaic of 1141 in the church of Saints Mary and Donatus on the island of Murano has several such figures. They have defeated all attempts to make sense of them. There is, for instance, a depiction of two insects, one on top of the other. This has been variously interpreted as “two copulating crickets”, symbolising “the birth of religion and the growth of the church”; as “insects in a state of nature”, symbolising “humanity singing the praises of the Lord”; and as “two ants sniffing flowers”, symbolising God knows what. The truth is, no one knows what the animals mean. They may not have had any specifically Christian significance at all. The church of San Miniato al Monte, just outside Florence, has a series of astonishing black-and-white floor mosaics with various figures in a circle enclosed within a square. They seem to depict the signs of the zodiac — but what are astrological symbols doing on the floor of a Christian church? The medieval Catholic Church seems to have had no difficulty in combining astrology with divine providence: signs of the zodiac pop up, not just on floor mosaics, but in relief sculpture around church doors.
Some of the divergences between floor mosaics in churches in northern and central Italy are the result of the different availability of materials for making them. In Rome and central Italy, there was a ready supply of porphyry columns left over from antiquity: these could be sliced like salami, and the resulting circles used to provide roundels for floor mosaics. There was an active market in taking porphyry columns and other remains from ruined Roman buildings. It was so vigorous that the popes, who claimed the exclusive right to such rare and precious materials, had repeatedly to threaten with excommunication anyone who took them without authorisation.
Floor mosaics were popular, and can be found all the way down the Italian peninsula and into Sicily, where the influence of Arab designs on the patterns used is very obvious. Floor mosaics even made it to England. Odoricus, a Roman marble worker, travelled to London where he worked on the construction of a very remarkable Cosmati pavement at Westminster Abbey. He was commissioned by Richard of Ware, abbot of Westminster Abbey, who went to Rome at least twice, in 1258 and 1268, when he would obviously have seen a lot of Cosmati floor mosaics. It is thought that it was on his second trip that he hired Odoricus. Supplies of coloured marble stones were brought from Rome, but no porphyry: the popes kept that for themselves. The colour purple was indelibly associated with Roman imperial power, and many of the floor mosaics in Rome were commissioned at the height of the contest between Italian popes and German emperors as to who was the rightful inheritor of imperial authority. But it doesn’t really help us understand what the patterns were supposed to mean: beyond the idea that floor mosaics may have played a role in emphasising the superiority of sacerdotum over regnum, no one has anything very plausible to say.
The Cosmati pavement at Westminster is the only one in England that has survived intact, although there is evidence that there was also one at Canterbury Cathedral, and another in St Paul’s before Christopher Wren demolished the remains of the old church on that site that had burned down in 1666. The Westminster Abbey pavement is spectacularly beautiful but until very recently it was hidden underneath a thick carpet. A very careful and effective restoration has made it possible to see it in full — and unlike floor mosaics in Italian churches, it does not cover the whole church, but is restricted to a relatively small square in front of the altar.
Looking at any of the beautiful floor mosaics that have survived, it is as impossible to work out what the interconnecting circles and the geometrical shapes were supposed to mean as it is not to wonder what their creators were trying to get at. You cannot avoid the sense that they were trying to communicate something important. To contemporary eyes, the designs of medieval floor mosaics, with their iterations and their emphasis on mathematical proportions, seem to reflect a very modern preoccupation with the role of numbers in the world. Nearly four centuries ago, at the dawn of the scientific age in which we still live, Galileo wrote: “The book of nature is written in the language of mathematics, and its letters are triangles, circles and other geometrical figures.” Staring at Cosmati mosaics, I think I know what he meant.