Medici Porcelain Ewer

A small blue-and-white ewer, 20.5 centimetres tall, made in Florence in the last quarter of the sixteenth century at the world's first successful European porcelain workshop, a furnace built inside Grand Duke Francesco I de' Medici's private laboratory-palace, the Casino di San Marco. It is one of perhaps sixty to seventy such pieces to survive anywhere, and one of the smaller number that still carries the workshop's own painted mark, the dome of Florence Cathedral over a capital F.

The Louvre holds it, and three related but distinct pieces, in the Departement des Objets d'art. This one was bought by the French state in 1885 for 3,500 francs from an otherwise undocumented owner; the other three came in two years earlier as part of the bequest of Baron Jean-Charles Davillier, a historian who had written one of the first serious studies of the factory's output. The recipe the workshop used was never written down in a form that survived Francesco's death in 1587, and was only pieced back together through chemical analysis of the surviving pieces themselves, most recently by Raman spectroscopy carried out directly on museum objects in the early 2000s.

Where the piece stood before 1885, and who "Pennelli" was, is not recorded in any source consulted for this record. That gap is honest rather than filled in, in keeping with how little is documented about almost anything Medici porcelain touched between the 1772 dispersal of the Tuscan grand-ducal stock and its slow re-collection by nineteenth-century connoisseurs like Davillier.

What it is

A small ewer, or chevrette, in white porcelain painted in monochrome blue: 20.5 centimetres tall and 11.3 centimetres across at its widest, with a single scrolled handle that rises into an open heart-shaped loop above the rim, a short pouring spout set into the neck, and an ovoid body covered edge to edge in flowering sprigs picked out in cobalt underglaze. Turned upside down, the foot carries a small painted mark, the dome of Florence Cathedral above a capital F, the closest thing this workshop had to a signature.

Close inspection of the glaze shows what a fully successful porcelain body would not: small dark specks where impurities in the frit fired unevenly, and tiny pinholes and bubbles trapped in the surface where the glaze did not settle perfectly smooth. These are not damage. They are exactly what a soft-paste, frit-based body pushed to the edge of what a sixteenth-century Florentine kiln could achieve looks like, and they are part of why this object matters more than its size suggests.

A furnace inside a palace

The piece was made at a workshop that did not resemble a commercial pottery in any way. It operated inside the Casino di San Marco, a palace built for Francesco I de' Medici, Grand Duke of Tuscany, by his court architect Bernardo Buontalenti between 1568 and 1574, on a site near Piazza San Marco in Florence. Francesco intended the building as both a residence and, in his own words as later summarised by scholars, a laboratory of alchemical and physical experiments, and he moved the Medici grand-ducal foundry there, converting part of the complex into rooms for glassmaking, pharmacy, and porcelain.

This is not incidental dressing. The porcelain furnace shared a roof, and very likely shared workers, with an alchemical laboratory and a glass workshop that later produced Antonio Neri's L'arte vetraria of 1612, the first serious European treatise on glass technology. Francesco's court biography consistently describes him as more interested in scientific and mechanical pursuits than in the business of ruling, and the porcelain project reads as one output, among several, of that interest rather than a stand-alone commercial venture.

Why porcelain, and why then

By the middle of the sixteenth century, Chinese blue-and-white porcelain was reaching Florence through Medici trade contacts and diplomatic exchange in small but conspicuous quantities, prized for a combination of qualities no European ceramic could match: translucency, hardness, and extreme thinness fired without collapsing. Ottoman and Iznik blue-and-white wares, closer at hand and more available, offered a related but distinct model that Florentine painters drew on for some of the same floral vocabulary seen on this piece.

The Venetian ambassador Andrea Gussoni reported to his own government in 1575 that Francesco had found a way to make what he called "the porcelain of India," adding that it had taken ten years of work and that, in the end, a traveller from the Levant had shown Francesco the way to succeed. That single sentence is the closest thing to a contemporary account of how the secret was actually obtained, and it is treated in the scholarship as a documented, if frustratingly brief, record rather than legend: an eyewitness diplomatic report, not a later retelling.

The craftsmen

Vasari's own writing already places Buontalenti at work trying to discover porcelain as early as 1568, before any success is documented, which makes him the earliest named figure associated with the project, in his capacity as court architect and engineer rather than as a potter. The maiolica painter Flaminio Fontana, from a family of ceramicists with roots in Urbino, is separately documented as having been paid for twenty-five to thirty pieces delivered to the Medici factory in 1578, which places a specific, named decorator inside the workshop during its active years.

A third figure, the Delft-born goldsmith and gem-cutter Jacques Bylivelt, who had joined the Medici court workshops in 1573 to head the hardstone and jewellery workshop, is documented in 1590, three years after Francesco's death, drawing up an inventory as he oversaw the transport of 154 pieces of porcelain, 64 of them identified specifically as Medici production, from the Villa Medici in Rome back to Florence. Bylivelt's inventory description of one vessel's neck, shaped, in his words, in the manner of a reed cut down the middle, is detailed enough that it has been matched by curators to the distinctive split-necked form still held today as the Louvre's own OA 3101. Beyond these three names, the great majority of hands that actually worked the clay, mixed the frit, and fired the kiln are not identified in any source consulted for this record.

How it was made

Medici porcelain is not porcelain in the sense the word describes Chinese or Japanese hard-paste wares, fired from kaolin at very high temperature. It is a soft-paste, or frit, body: white Vicenza clay mixed with ground rock crystal or quartz sand, feldspar, and calcium compounds, fired at a lower temperature and covered with a lead glaze deliberately opacified, in the manner of Islamic and Iznik ceramics, using calcined bone or a related calcium-phosphate source rather than the tin oxide used in ordinary European tin-glazed maiolica. Decoration was painted directly onto this glaze in cobalt blue, sometimes outlined in manganese, and fired again, producing the grey-toned blues visible on this piece rather than the purer blue of Chinese originals.

The process pushed contemporary Florentine kiln technology to its limit. Achieving a translucent, thin-walled body at these temperatures without warping or collapse required firing conditions that period sources describe as producing very high costs and, on the evidence of the surviving pieces themselves, a high failure rate: nearly every known example carries some combination of firing cracks, glaze bubbles, or the kind of dark impurity specks visible on this ewer's surface under close inspection.

How production ended

Francesco I de' Medici died on 19 October 1587. No source consulted for this record documents any secure evidence of porcelain manufacture at the Casino di San Marco after his death, and the standard scholarly position, reflected in the reference works cited here, is that production ceased at or very close to that date. An inventory drawn up at Francesco's death recorded 310 pieces of porcelain in the grand-ducal collection, a figure worth holding onto for what it says about scale: several hundred pieces existed at the moment production stopped, against the sixty to seventy, by differing modern counts, that are known to survive today.

Francesco's brother and successor, Cardinal Ferdinando de' Medici, inherited the collection and, on becoming Grand Duke, had part of it moved from the Villa Medici in Rome to Florence; the 1590 transport overseen by Jacques Bylivelt, discussed above, is the documented record of that move. A large portion of what remained in Tuscan grand-ducal storage was later dispersed at an auction held at the Palazzo Vecchio in 1772, the point after which individual pieces begin to scatter into private and, eventually, museum hands across Europe with little continuous paper trail.

How many survive, and where

Modern scholarship gives different counts depending on the cut-off date and the criteria used. The Metropolitan Museum of Art's own published figure is 59 surviving examples; older surveys, including the standard illustrated catalogues by Giuseppe Liverani (1936) and by Giuseppe Cora and Angiolo Fanfani (1986), are more often summarised in secondary literature as putting the number nearer sixty to seventy. This record treats the exact figure as unsettled rather than picking one estimate to state as fact.

Whichever count is used, the surviving pieces are concentrated in a small number of institutions: the Musee du Louvre, the Metropolitan Museum of Art, the Victoria and Albert Museum, the National Gallery of Art in Washington, the J. Paul Getty Museum, and the Sevres - Cite de la ceramique, which alone holds about ten examples, along with the Museo Nazionale del Bargello and other Florentine collections closer to the object's origin. The Louvre's own four Medici porcelain pieces, described further below, are a small but genuine part of that total.

What the rest of the type has sold for

This ewer itself was bought by the French state in 1885, not at public auction, so it carries no hammer price of its own to report. Other pieces of Medici porcelain have appeared on the open market since, and they establish what this small class of object is worth to collectors when it does trade, which is useful context for a piece the Louvre bought outright. A Medici porcelain vase dated to 1575 sold in Paris in 1995 for a reported price of around 870,000 pounds sterling, quoted in secondary reporting as approximately 6,430,000 French francs. A Medici porcelain cruet of about the same date sold at Christie's in 1999 for 287,000 pounds.

Neither sale involved this object, and this record does not treat those figures as evidence of this ewer's value. They are recorded here only because they show what the wider market has been willing to pay for authenticated pieces of the same rare, twelve-year production run, which is the only real comparison available for an object that has not itself been back on the market since 1885.

Authenticating a piece with almost no signature

Establishing that any individual object is genuine Medici production, rather than a later or contemporary imitation, is a live problem for this material precisely because so few pieces are marked, and the marks that do exist, the domed-F device or the Medici palle, appear inconsistently even among pieces otherwise agreed to be authentic. In practice, museums and scholars build a case from several kinds of evidence together: documented descent from the Medici grand-ducal collections where a paper trail survives, comparison of a piece's decorative vocabulary and potting against securely marked or documented examples, and, increasingly, direct chemical and microscopic analysis of the body and glaze against the now well-characterised composition of confirmed Medici porcelain.

This particular ewer sits on the more secure end of that spectrum for a specific reason: it is marked. The painted dome-and-F device on its underside removes one layer of the usual uncertainty that surrounds unmarked pieces, although the Louvre's own broad cataloguing date range of 1585-1600, which extends past the 1587 end of documented production, is a reminder that even a marked piece is not dated with full precision.

The recipe that had to be rediscovered

No workshop formula for Medici porcelain survived in a form later chemists could simply read and reproduce. What contemporaries wrote down, in the Gussoni report and scattered court records, describes the fact of success and something of its cost, not the technical recipe itself, and once the Casino di San Marco furnace stopped operating after 1587, the specific combination of clay, frit and glaze chemistry that had been worked out over roughly a decade of experimentation was not carried forward by any documented successor workshop in Italy.

Understanding how the body and glaze were actually put together has therefore depended entirely on later scientific analysis of the objects that survive. Non-invasive Raman spectroscopy carried out directly on museum pieces by Philippe Colomban, Veronique Milande and Helene Lucas, published in the Journal of Raman Spectroscopy in 2004, identified the body's characteristic components, including alpha- and beta-wollastonite, quartz, feldspar and cassiterite, and showed the glaze's opacification relied on calcium phosphate consistent with a calcined-bone source rather than the tin oxide of ordinary maiolica, a technique with roots in Islamic ceramic practice. That analysis, done more than four centuries after this ewer left the kiln, is the closest anyone has come to reading the recipe the Casino's craftsmen never wrote down for posterity.

Three siblings in the same room

This ewer is not the Louvre's only piece of Medici porcelain. The museum holds three further examples, all bequeathed together in 1883 by Baron Jean-Charles Davillier, a French collector and one of the first serious historians of the ware, whose 1882 study Les origines de la porcelaine en Europe remains a cited reference in the Louvre's own catalogue entries for these objects. They are a two-handled gourd-shaped bottle painted with flowers and birds (OA 3102), a ribbed, pear-shaped vase with rosette medallions (OA 3104), and an ovoid drinking vase with two scroll handles and a distinctive split-reed neck (OA 3101), the same neck form Jacques Bylivelt described in his 1590 transport inventory.

Photographs of the Davillier gourd and the Davillier ribbed vase, taken by the same Wikimedia contributor in the same Louvre gallery in 2009, are included with this record as clearly labelled comparanda from the same small holding, not as this object. They are worth including precisely because they show how varied Medici porcelain's surviving forms are even within a single museum's four pieces: an ewer, a gourd, a ribbed pear-shaped vase, and a two-handled drinking vase, all from a workshop that operated for barely twelve years.

From an unrecorded owner to the French state

Against the depth of documentation available for the object's manufacture and technical composition, its ownership history before 1885 is close to a blank. The Louvre's own record names a previous owner or vendor only as "Pennelli," with no further identification, immediately before the French state's purchase of the piece on 26 February 1885 for 3,500 francs, a transaction recorded under archival reference AN 20144787/9. Nothing in the sources consulted for this record establishes how or when the ewer left Florence, whether it passed through the 1772 Palazzo Vecchio dispersal of grand-ducal porcelain stock, or where it was in the roughly three centuries between the workshop's closure and its appearance in French hands.

That gap is stated here rather than smoothed over. A collector or curator working from this record should treat the object's history as beginning, in documented terms, only in 1885, with everything earlier resting on the general history of the workshop rather than on any paper trail specific to this piece.

What is still unresolved

Two kinds of open question run through this object: how it is dated, and where it was before the paper trail begins. The Louvre's own catalogue bracket of 1585-1600 sits awkwardly against a production history that every other source consulted dates as ending in 1587, and no source resolves that tension directly. Separately, the identity of "Pennelli," the piece's last owner before the state's purchase, and the object's whereabouts for roughly three centuries before that, are not recovered in any source available for this record.

Both gaps are recorded as gaps. Neither is bridged with a guess.