The Musicians

Caravaggio's first picture for his first great patron, painted in 1597 for Cardinal Francesco Maria del Monte, and the earliest work in which he built a composition out of separately posed models rather than a scene. It is an allegory of Music and Love, not a concert: the cornetto player is a self-portrait, the lutenist may be a singer in the cardinal's employ, and the winged boy gathering grapes is the only conventional thing in it.

Its provenance runs from Del Monte through Cardinal Antonio Barberini, the French ambassador Charles de Crequy, Cardinal Richelieu and the duchesse d'Aiguillon, and then stops dead for two hundred and forty-five years. It reappears in a house in Whitehaven in Cumberland in the 1920s, inventoried in 1933 as a musical party signed Helang da Caravagio, is sold for a hundred pounds in 1947, and is recognised in 1952 by the dealer David Carritt and published within weeks by Denis Mahon. The Met bought it that year. It is the worst-preserved great Caravaggio in any museum, and the damage is itself the reason the x-ray is so informative.

What it is

Oil on canvas, 92.1 by 118.4 centimetres in its present state, which is smaller than Caravaggio left it. Four youths fill the picture. One tunes a lute in the centre and looks out of the canvas with wet eyes; behind him a boy in a white shirt holds a cornetto and turns his head towards us; a third figure bends over a part-book; at the left a winged cupid reaches into a bunch of grapes. A violin and an upturned page of music lie in the foreground.

It is not a picture of a concert. Nobody in it is playing. The lute is being tuned, the music is being read, the cornetto is being held. Caravaggio has painted the moment before performance, which is the moment in which the instruments are objects and the players are models.

Made to order, and the first thing he made to order

Giovanni Baglione, a contemporary painter who wrote one of the two primary lives of Caravaggio, records that immediately after entering the household of Cardinal Francesco Maria del Monte he painted a concert with some youths portrayed from nature very well. The Met's picture is universally identified with that sentence.

The sequence matters. Del Monte already owned two Caravaggios, the Fortune Teller now in the Capitoline and the Cardsharps now at the Kimbell, and he had bought both on the market through the dealer Costantino Speti. This is the first one he ordered. A painter who had been supplying the trade acquired, in a single transaction, a patron with a palace and a programme.

The date of that transaction has moved. It used to be given as 1595. Lothar Sickel's archival work published in 2009-10 put Caravaggio's entry into the Del Monte household in 1597, which is why the Met's tombstone now reads 1597 and why every dating of this picture before about 2010 is out by two years or more.

An allegory of Music and Love, built out of separate sittings

The picture is an allegory, not a genre scene. The cupid gathering grapes carries the meaning: Love, and wine. Cesare Ripa's Iconologia, the standard handbook of allegorical figures, had been published in Rome in 1593 with a dedication to Del Monte himself, and it recommends including wine with Music because music was invented to keep spirits cheerful as wine does. Caravaggio took the wine and refused everything else Ripa offered. Where Ripa would have given a single idealised female figure as Music, Caravaggio gave four boys, an untuned lute and a room.

Scholars agree that the cornetto player is a self-portrait. The lutenist may be a specific person, possibly a male soprano in the cardinal's employ, an argument developed by Franca Trinchieri Camiz in 1988 from the costume and the repertoire. He wears a classicising pastoral costume, as though dressed for a performance rather than living in one.

Giovanni Pietro Bellori, annotating his own copy of Baglione, complained that the composition is flat. He was right about the cause: Caravaggio built it piece by piece from individually posed models, and the pieces do not recede. What Bellori recorded as a fault is the earliest description of the working method that produced everything Caravaggio did afterwards.

A cardinal, an ambassador and a first minister

Del Monte died in 1626 and his collection was sold at the Giardino di Ripetta on 8 May 1628. The Musicians went in a single lot with a Caravaggio of a carafe of flowers, now lost, and a music piece by Licinio. Prospero Fagnani paid for the lot five days later, almost certainly on behalf of Cardinal Antonio Barberini.

An avviso of 8 July 1634 records that Cardinal Antonio gave a Caravaggio, bought from the Vigna Ludovisi, to Charles de Blanchefort, Marechal de Crequy, the French ambassador in Rome. The Vigna Ludovisi was Del Monte's own casino near the Porta Pinciana, which he had sold to Cardinal Ludovisi in 1621. A Roman avviso reporting the provenance of a diplomatic gift got the address right and the history wrong, and it has taken modern archival work to reconcile the two.

Crequy died in 1638 and the picture appears as no. 119 in his posthumous inventory. It then belonged to Cardinal Richelieu, appears as no. 996 in the 1643 inventory of his estate, and was sold as the same number in the estate sale of January and February 1650 for 1,030 livres tournois to his niece, the duchesse d'Aiguillon. Her inventory of 1675 lists it as no. 48 and adds a sentence with physical consequences: the canvas had been glued to a wooden support.

Two hundred and forty-five years missing

After 1675 the record stops. The picture is next heard of in the 1920s in a house called Fernacre at Whitehaven in Cumberland, belonging to David Burns.

This is not a small gap that better research will close in a footnote. It covers the whole of the eighteenth century, the French Revolution and the Napoleonic dispersals, and it is the period in which most French cabinet pictures either went to England or went to the fire. Nothing in the sources consulted names an owner, a sale or a port. The gap is recorded here as a gap and this record refuses to bridge it.

Signed Helang da Caravagio

In August 1933 D. Graham Burns had an inventory made of the pottery, porcelain and pictures at Fernacre. Item 846 reads: a musical party signed Helang da Caravagio, 36 x 46.

Everything in that line is nearly right and nothing in it is quite right. A musical party is what the picture looks like. Helang da Caravagio is a garbled reading of an inscription; the Met's catalogue entry records that a yellow inscription in capitals giving Caravaggio's name was formerly visible in the lower left corner and is now painted over. And 36 by 46 inches is this canvas, within an inch on each side.

The document is therefore both the proof that the picture was at Whitehaven in 1933 and a demonstration of how a masterpiece can sit in a private inventory, correctly measured and approximately attributed, for a generation without anyone acting on it. When Benedict Nicolson wrote his editorial note in the Burlington in April 1952 discussing the provenance, the 1933 inventory was the document he leaned on.

A hundred pounds in 1947

Mrs Burns sold the picture shortly before her death in 1947 to Joe Cookson, a dealer in Kendal. Cookson sold it to Surgeon Captain W. G. Thwaytes of Maulds Meaburn, near Penrith, for 100 pounds. The Times printed the figure on 29 January 1952.

Five years separate that hundred pounds from the Metropolitan Museum. In between, the dealer David Carritt saw the picture in Thwaytes's house and recognised it. Carritt was twenty-five at the time and would spend the rest of his career finding things; this was among the first.

What the Met paid has not been published in any source consulted. The Whitehaven News of 31 January 1952 ran a letter under the heading That 50,000 Pound Picture, which records what the town was saying about the value and is not evidence of a price. This record does not convert it into one.

Mahon, Longhi, and a dispute about a decade

Denis Mahon published the picture twice in January 1952, in the Burlington Magazine and in Paragone. He reported its discovery by David Carritt in an English private collection, identified it with the concert mentioned by Baglione and Bellori in Del Monte's collection, described its condition, noted that it had been cut down slightly, and observed that Caravaggio had begun to give wings to the figure at the far left and abandoned them. He dated it 1594-95.

Roberto Longhi disagreed at once and in two directions. In Paragone in the same month he attacked the homoerotic reading of Caravaggio's early pictures, in a piece titled with a sneer about the inverted Caravaggio. In his book of the same year he argued that this picture precedes the Del Monte association altogether and that the concert Baglione described must have been a different work with figures in contemporary dress. Fritz Baumgart dated it 1589. Costantino Baroni dated it about 1590.

Mahon replied in the Metropolitan Museum of Art Bulletin in October 1953, refuting Longhi in a postscript, and again in the Burlington in 1953 in an article about art-historical method. Nobody ever doubted the hand. What was fought over for a decade was whether the picture came before or after the patron, and the answer arrived not from connoisseurship but from a Roman archive, fifty-seven years later.

The worst-preserved great Caravaggio, and why that is useful

Theodore Rousseau set out the condition in the Met Bulletin in October 1953. The picture has been relined at least twice. It has been cut down slightly at the top and bottom, about two inches on the left and about one inch on the right. The duchesse d'Aiguillon's inventory of 1675 records that it had been glued to a wooden support, which the museum's curators believe may have contributed to the damage.

The losses are severe and specific. The lute is scarcely more than a shape, its strings obliterated and the shadow of the player's right hand reduced to smudges. None of the musical scores is legible. The violin and the upturned page of music in the foreground were reconstructed by restorers on the basis of a later copy. Even the best-preserved passages, the lutenist's head and the still life, have lost their subtlety.

The compensation is that a damaged surface is a transparent one. The Met's curators argue that the picture's compromised state makes it possible to watch Caravaggio revising: the lutenist's shirt and sash were reworked with more rhythmic drapery, the right arm was painted first and the red mantle laid over it, each change pulling the figure further from the everyday and closer to allegory. The x-ray makes the same point more sharply. This was an experiment in a kind of painting that was new to him, and the experiment is legible because the finish is gone.

The copies, and one that was made to order and refused

A copy of this composition exists and it is reproduced in the Met's own catalogue entry. Its relationship to the copy documented in 1614-15 is uncertain, and the documentation of that one is unusually complete.

A series of letters running from 27 December 1614 to 13 March 1615 shows that Giulio Mancini, later papal physician and the author of a treatise on painting, arranged to have copies made of three Caravaggios then in Del Monte's collection: the Fortune Teller, the Cardsharps and the Musicians. He paid a professional copyist 15 scudi. The intended buyer was Agostino Chigi. The copy of the Musicians was finished on 13 March 1615 and sent to Siena on 26 June, because Chigi had decided against buying it.

The surviving copy introduces a number of changes, and the Met's curators note that a connoisseur as demanding as Mancini would have been unlikely to accept it. So the documented copy and the surviving copy are probably two different objects, and a reproduction of this composition may be either of them, or a third. Any photograph of the Musicians has to be checked against the Met's canvas before it is attached to this record.

What was read into it, and by whom

A biographical and homoerotic reading has been attached to this picture for most of a century. The Met's own catalogue entry takes a careful position on it and this record follows that position rather than improving on it. The sensual quality of the image is real, particularly the lutenist's wet eyes meeting the viewer's. Caravaggio's bisexuality can be established with reasonable confidence from documentary work. Del Monte's own sexual character is not known. What bearing, if any, the second has on the picture is therefore conjecture built on the first.

Longhi's 1952 objection to the reading was polemical rather than evidential, and the modern position is neither his nor his opponents'. The picture was commissioned by a cardinal who had just been dedicated the standard handbook of allegory, and it is an allegory that departs from that handbook. That is a documented fact about its making. The rest is interpretation, and the honest thing is to name it as such.