Through more than one hundred and fifty years of rather painstaking study since 1850, scholars have identified thirty-four, perhaps thirty-five, paintings they now safely attribute to Johannes Vermeer. Their task was made difficult for a variety of reasons: Vermeer's varied and changeable painting style; the range of his choices of subject matter; the fact that he signed less than half of those works which yet survive and dated only one; and that, for several hundred years after his death in 1675, no one knew the true extent of his oeuvre. In addition, his contemporary reputation probably did not extend much beyond Holland, in all likelihood because only a small number of local connoisseurs collected his relatively few paintings. In a professional career that spanned twenty-two years, Vermeer completed perhaps no more than forty works, and he left no drawings or preliminary paintings behind.
When so little is known about an artist, the science of artistic attribution becomes a weaving of a few threads of hard historical data with the fabric of informed but subjective interpretive analysis based upon a shared sense of the artist's style, technique, composition and and subject matter. An attribution's authenticity is greatly strengthened if it can establish direct links over time to the artist himself or to an ownership during the artist's lifetime or fairly soon after his death. And this is precisely what Vermeer scholars have attempted to do. In examining relevant records of art and estate auctions of the late seventeenth and early eighteenth centuries, they have rather confidently connected those documents with about two dozen extant Vermeer paintings. There also appears to be another nine (or maybe eleven) paintings which have survived for which no contemporary corroboration in Vermeer's time has yet been found. Conversely, there seems to be at least six, and perhaps eight or ten, Vermeer paintings identified by historical records which today either remain hidden or have not survived. This latter group is known as the "missing Vermeers."
The most important source of information for twenty-one of the artist's paintings was the 1696 Amsterdam auction of the estate of Jacob Dissius, who had inherited by marriage the art collection of Pieter van Ruijven, Vermeer's principal patron. Scholars have gleaned from their descriptions in the auction catalog the identity of fifteen existing Vermeer's. Three other catalog descriptions from the Dissius ledger—the various tronies—are so vague that an assured connection to present day Vermeer pictures of faces is impossible to make. However, no one can account for three of the twenty-one works, items 3, 5 and either 32 or 33, because the descriptions of those paintings don't fit the subject matter of any known works; they are therefore presumed missing:
Item 3 – "the portrait of Vermeer in a room with various accessories, uncommonly beautiful, painted by him," could be the exquisite Art of Painting, still in Vermeer's possession at the time of his death. However, the painting at the auction fetched only forty guilders, a price so low that it discredits any notion such a sum could have purchased this large and magnificent painting. The only other conclusion is that item 3 was another, and very direct, portrait of Vermeer, now lost or missing.
Item 5 – "in which a gentleman is washing his hands in a see-through room with sculptures, artful and rare, by ditto [Vermeer]." This painting is clearly lost or missing, but Vermeer may have used a similar compositional device here to that which he used in painting The Love Letter. And the subject of a man washing his hands finds a parallel in Gerrit ter Borch's A Woman Washing Her Hands; the two artists knew each other, probably very well, as documents attending Vermeer's marriage indicate.
Item 32 – "a view of a house standing in Delft, by the same [Vermeer]" or item 33 – "a view of some houses by ditto [Vermeer]." Since only Vermeer's Little Street fits one or the other description, the remaining painting is now lost or missing.(For a discussion of other surviving Vermeer paintings identified through eighteenth century transactions, see Vermeer's Clients and Patrons.)
In addition to the Dissius auction, the estate sales of three other collectors mention Vermeer paintings with descriptions that don't match known works today. First, in 1657, the death inventory of the Amsterdam art dealer and collector, Johannes Renialme, listed a painting as "The Visit to the Tomb" ascribed to "Van der Meer." It is unclear whether this description indicates a biblical subject or a more contemporary theme, but in any event no such painting seems to have survived.
Another lost or missing painting by "J. ver Meer," mentioned in the mid-eighteenth century auction catalog of the collector Willem van Berkel, was entitled "Jupiter, Venus and Mercury" (although this descriptor was likely a misnomer, for this kind of painting almost always placed Virtue or Psyche in the picture rather than a Venus). A work with this sort of mythological theme would likely have been painted early in Vermeer's career as a complement to Diana and Her Companions.
Finally, there was also a "face by Vermeer" which The Hague/London sculptor, Jean or Johan Larson, had acquired as early as 1660, perhaps from Vermeer himself. The early date of this painting makes it unlikely to have been the Girl with the Pearl Earring, or the Girl with the Red Hat, or the Portrait of a Young Woman, or even the vexatious Young Girl with a Flute – all of which seem to have been painted later than 1664. This particular "face by Vermeer" now appears to be lost or missing, too.
Along with these six paintings, two others may be "missing" Vermeer's as well. A Delft innkeeper, Cornelisz de Helt, had a Vermeer painting "in a black frame" hanging in the front hall of his Young Prince Inn in 1661. Black was a preferred color for frames at that time and perhaps most artists of that era displayed their work in a black frame. Nothing more is known about this work.
In 1663 a visiting Frenchman, the Baron de Monconys, visited a well-to-do baker, Hendrich van Buyten, who had a painting of a single figure by Vermeer. Given this date, it is possible that the painting could have been Vermeer's Woman in Blue Reading a Letter or the Woman with a Water Pitcher, both relatively small, similarly-sized works, painted at about the same time (1662–1665). It could also have been the large Girl Reading a Letter at an Open Window, executed relatively early in Vermeer's career. Indeed, it is not out of the question that it could have been the Girl with a Pearl Earring, for the price which the baker claimed to have paid for it—six hundred gilders—could have been justified by the beauty of this work.
Neither of these two works was mentioned in seventeenth-century documents. It is entirely possible this group is not among the missing Vermeer's and that all passed forward to the present day intact; but their vague descriptions stymie accurate attribution.
Altogether, there seem to be at least six Vermeer paintings that still remain hidden or are forever lost. Additionally, two more works were described so vaguely that no one today can pinpoint their attribution with certainty.
Ten of Vermeer's surviving paintings have not been accounted for using contemporary seventeenth- early eighteenth-century documentation. Nonetheless, scholars believe they have garnered sufficient evidence to support credible attributions for them. These paintings are: Diana and Her Companions, Christ in the House of Mary and Martha, The Procuress, The Glass of Wine, The Concert, The Love Letter, The Geographer, The Astronomer, and the Woman with a Water Pitcher. In 1712, there was another Amsterdam auction for the estate of Pieter van der Lip, with a painting in it described as "A woman reading in a room, by vander Meer of Delft." This surely referred either to the Woman in Blue or the Girl Reading a Letter at an Open Window, though it is not clear which one. But the listing would account for one of them. Finally, in 1703, at the estate of the Amersfoot paper merchant, Paulus Gijsbertsz, another painting was listed, "A piece by Van der Meer, being a lady in her room." However, this work fits the description of at least twelve preserved Vermeer's, many of which were mentioned in earlier documents. This is not a candidate for the missing in Vermeer's inventory.
The last three items of the Dissius inventory – number 38, "a tronie in antique dress, uncommonly artful;" number 39, "Another by ditto;" and number 40, "A pendant by same"—shed helpful but ultimately too dim light for an assured identification of these works. A Dutch "tronie" was not necessarily a portrait of a particular individual but rather a character study depicting a head and shoulders dressed in colorful, often highly textured clothing, especially fancy hats and collars. Such clothing was often distinctive, even theatrical, out of fashion and exotic costumes for this purpose were frequently termed "antique." With a tronie, an artist could showcase his painterly skills while demonstrating how well he might do a commissioned portrait. Though portrait painting in Vermeer's day was not considered the highest form of art, it was in relatively high demand and it did help pay the bills. A tronie, therefore, was often an artist's business card, a very commonplace form of advertising.
Vermeer seems to have painted four such tronies that have survived (if one accepts that he at least began Young Girl with a Flute, though did not finish it himself). By modern eyes, his Girl with the Pearl Earring fits the description in Dissius number 38 perfectly. It is uncommonly artful with an exotic ("antique") turban. The Dissius accountant who listed this painting may have been so taken with its beauty that he used the rare accolade "artful" in his description of a tronie. The painting's obvious pendant is the Portrait of a Young Woman in New York, for it is the same size and depicts the same general demeanor and torsion as the Girl with the Pearl Earring.
But note that Dissius number 39 describes only "another" tronie by Vermeer, not a pendant. Dissius number 40 does specify a pendant tronie, however, ostensibly referring back to item number 39. Since there are no other extant Vermeer's large enough to be a true pendant for the Portrait of a Young Woman, perhaps number 39 refers instead to the very small Girl with the Red Hat, which is most certainly a tronie. Item number 40 then could be the similarly-sized Young Girl with a Flute, almost certainly (if it is genuine) a pendant for the Girl with the Red Hat. Both pictures seem to portray the same girl or closely related girls posed in very similar, even complementary, aspects. And both are painted on wood panels (though nearly all surviving Vermeer's were painted on canvas, six wood panels were found in Vermeer's studio after his death).
The social historian, John Michael Montias, is the scholar most responsible for piecing together many of the documents that augment a proper consideration for the provenance of Vermeer's oeuvre. In addition to many other references, he also found that Vermeer had at least four of his own paintings in his mother-in-law's house along the Oude Langendijk in Delft at the time of his death: his showcase Art of Painting (which Catharina, probably unsuccessfully, attempted to save from her husband's creditors), a "Woman Wearing a Necklace" found in a basement room, most probably today's Woman with a Pearl Necklace that, by 1696, was sold at the Dissius auction, and "two tronies in Turkish fashion" found in Catharina's interior kitchen. These tronies may have found their way to become part of the Dissius auction or they may now be lost. It is tempting to speculate, however, since they remained in Vermeer's household, that they indeed are depictions of Vermeer's daughters: look-alike tomboys, one wearing a hat of blazing red, and the other a most mysterious, somber, Chinese millinery.
Such speculation is a joyful part of Vermeer studies. But it has no real explanatory power. The historical record firmly indicates there are six Vermeer paintings for which modern attribution cannot account—the "missing" Vermeer's. Another half dozen historical references for other paintings are too vague to make certain attribution, although it is quite possible some or all of these form part of the known works today. Perhaps scholars will discover new evidence to clarify these references and even confirm new additions one day to Vermeer's slender canon.