UP
Essential Vermeer 3.0
Looking for a painting by Vermeer? Find it with QUICK SEARCH!
WHICH IS YOUR FAVORITE PAINTING BY VERMEER?

Vermeer's Studio

The Painter's Studio in the Netherlands

Although the artist's studio had been occasionally portrayed during the Renaissance—the place was not actually called a studio until the nineteenth century—it became a popular motif in the seventeenth-century Netherlands. Particularly numerous were pictures of the workplaces of destitute artists. Certainly, not all Dutch painters were paupers nor their studios dilapidated, but the idea that a painter's working space was somehow different from those of other craftsmen was ubiquitous throughout Europe.1

Click here to access 50 images of sixteenth- and seventeenth-century paintings of artists' studios.

he Artist's Studio, Frans van Mieris fig. 1 The Artist's Studio
Frans van Mieris
c. 1657
Oil on panel
formerly Gemäldegalerie, Dresden (lost in WWII)

In the seventeenth century, most painters kept their studios in their homes, often in the upper story where dust and noise might be less intrusive. But not all artist's studios were laid out in the same way. The landscape painter had different requirements from the still-life and portrait painter (fig. 1), these were again different from those who painted interiors only. The first always worked from drawings and sketches he had made on his trips by field and path, and therefore a space with only one window was sufficient. If he got enough light on his work he was satisfied; it was not necessary to group anything but a few drawings on the floor nearby his easel. The still-life painter collected the objects of his choice and arranged them on a small table, painting them directly and from close range. He too had need of no more space. The portrait painter, instead, required more space and a larger window to cast light on both his sitter and his easel while remaining at a comfortable distance from the subject, as many representations clearly demonstrate.2

The painter of interiors needed more space still. If he had the opportunity, he liked to work in an environment with at least two windows, which afforded him the possibility to distance himself from the subject as well as the opportunity to open and close the windows to create compositional alternatives and interesting atmospheres. History painters often found a separate studio (schilderhuys) necessary.

The most important points to consider for an ideal studio were the size of the room and the lighting. The German art historian and painter Joachim von Sandrart (1606–1688) wrote that studios ought to be about 9 meters square and illuminated sufficiently, although direct rays of the sun should be avoided. He wrote:

Many of our forefathers ... have erred in that too small chambers, everywhere filled with light and sunshine, served them for their work: whereby lacking the space and the necessary distance to walk far enough back and forth from their model or panel, to inspect their work from a distance ... as well as the strength of the light so that all colors become diminished and confused to them. They would, if they had been in a proper studio, have given their excellent works far more life, force and veracity?

The most advantageous system of illumination was to be obtained in studios with large windows divided into four squares, called cross-windows. Nonetheless, there are various paintings that show light coming from a window built high up on the wall, for example in The Painter's Studio, by Joost van Craesbeeck (c. 1605/1606–c. 1660; fig. 2). A workshop scene by Jan Miense Molenaer (1610–1668; fig. 3) shows plainly another high-set window of this kind.

The Painter's Studio, Joos van Craesbeeck fig. 2 The Painter's Studio
Joos van Craesbeeck
1655
Oil on panel, 48.5 x 66 cm.
Frits Lugt Collection, Paris
Painter in His Studio, Painting a Musical Company, Jan Miense Molenaer fig. 3 Painter in His Studio, Painting a Musical Company
Jan Miense Molenaer
1631
Oil on canvas, 86 x 127 cm.
Staatliche Museen, Berlin

If the incoming light was too bright, it could be mitigated with an additional wooden frame fitted over the window covered with cheesecloth or oiled paper (blaffeturen). One of the few, if not the only, representations of such an article is found in painting by Jacob van Spreeuwen (1611–1658) in a private collection in Sweden (fig. 4). The screen is clearly visible in the window, and can by no means be mistaken for a pane of glass, as no panes of the size existed at that period. At best, the Dutch painter had up to fifteen hours of light a day in the summer but only about three or four hours in the winter. Candles and oil lamps could be used as a source of illumination if daylight was unavailable although both create light that is ill-adapted for painting, with the consequence that colors applied with the artificial light appeared radically different when viewed in natural light. However, artificial light was ideal for drawing because the shadows they project are more marked than those produced by filtered studio light. Dust was one of the most insidious enemies of the fine painter. In a painting by Adriaen van Ostade (1610–1685) a piece of linen is suspended under the roof to keep the dust from coming through. A work by Gerrit Dou (1613–1675) show a Japanese umbrella fastened to the top of the easel for the same purpose. It is said that when commencing to paint Dou sat in front of his easel for a half and hour to allow the dust settle.

>A Painter in His Studio, Jacob van Spreeuwen fig. 4 A Painter in His Studio
Jacob van Spreeuwen
Collection Braunerhielm, Djursholm
An Artist with a Model in his Studio, Cornelis Dusart fig. 5 An Artist with a Model in his Studio
Cornelis Dusart
1690–1700
Paper, 24 x 19 cm.
Rijksmuseum, Amsterdam

The many painting studios (schilder-camer, painting-room) pictured in the Golden Age range in size and decor but often portray common elements. In those of the vast army of artist/artisans who were barely able to make ends meet, no luxury items can be found; nothing but the most elementary utensils and props are visible (fig. 5 & 6). The floor is made of wide wooden planks or paved with simple stones. The walls are whitewashed or constructed with crude wooden planks. Ceilings are supported by precarious-looking wooden beams. In a background room an apprentice or journeyman is often pictured grinding paint or preparing palettes. Occasionally, the apprentice is seen working in the same room as the master. A large grinding stone is never lacking. Not only do such pictures show us this indispensable article repeatedly, but written statements also mention it—one such object was described in a posthumous inventory of movable goods in of Vermeer as "a stone table for grinding paints." The most famous rendition of this inglorious but indispensable tool is featured in an early self-portrait by Rembrandt (Artist in his Studio, c. 1628: fig. 7). From what we can observe, Dutch studios were always lit from the side. Top-light, which appears in many scenes of artist's studios of the eighteenth century, was never used. Only Rubens seems to have painted with a toplight.

An Artist in his Studio, Rembrandtfig.6 An Artist in his Studio (detail)
Rembrandt van Rijn
c. 1628
Oil on panel, 24.8 cm × 31.7 cm.
Museum of Fine Arts, Boston
An Artist in his Studio, Rembrandt van Rijnfig.7 An Artist in his Studio (detail)
Rembrandt van Rijn
c. 1628
Oil on panel, 24.8 x 31.7 cm.
Museum of Fine Arts, Boston

The contents of the studio depended on the character of the painter's work but also on his means, although certain pieces of furniture and equipment were indispensable, such as an easel, a stool, a cabinet for storing brushes, lumps of raw pigment, ground pigments, drying oils and essences. Plaster casts of classical sculpture were extensively used for instruction and can be seen in almost any portrayal of an artist's studio of the time (fig. 9). Skulls too are constantly pictured and not only in the studios of painters who devoted themselves to Vanitas still-life genre. The artists who were most successful in painting ducks are known to have kept them in their gardens. Otto Marseus (c. 1613–1678), famous as a painter of insects and reptiles, kept his "models" in a building behind his house to study them at his convenience. On the other hand, flower painters worked extensively from prints or sketches, usually in watercolor, that they had done during the blooming seasons, allowing them to create rich but completely impossible flower arrangements throughout the entire year. If the could afford it, figure painters, especially interior painters, usually had a mannequin on which they could arrange fancy costumes. Some painters, like Rembrandt (1606–1669), collected exotic costumes, natural oddities, weapons etc. The small army of Rembrandt's followers depicted books, globes (fig. 10), musical instruments, swords and any number of exotic props haphazardly heaped up in a foreground corner of the composition. In Caspar Netscher's (1639–1684) studio various pieces of fine silk and satin fabric were inventoried after the artist's death. Some painters owned the Turkish carpets which appear repeatedly in their works, and occasionally lent them to their colleagues. Inexpensive musical instruments such as flutes, fiddles, citterns and lutes were also common features in artist's studios. However, luxury items such keyboard instruments, tapestries and chandeliers would have been borrowed for the occasion. Although it is not exactly clear why, artists seem to have had a particular liking for convex mirrors and crystal balls, perhaps, in order to study the bizarre manners in which the world is reflected in them. A few Dutch painters portrayed themselves in the reflections of a glass ball, such as a Vanitas still life by Pieter Claesz. (1628; fig. 8).

Still Life with Crystal Ball, Pieter Claesz. fig. 8 Still Life with Crystal Ball (detail)
Pieter Claesz.
c. 1628
Oil on oak, 35.9 x 59 cm.
Germanisches Nationalmuseum, Nuremberg

At any rate, some of the more unusual items may have no doubt had the additional advantage of creating exotic ambiance to charm visiting connoisseurs and clients.

We can assume that the most commonplace objects that we see in Vermeer's interiors, such the assorted chairs, mirrors, tables, wine jugs, drinking glasses, cushions and tablecloths, were his own. Others may have been borrowed for the sake of variety. Various inventories suggest that a few of the more expensive items, such as the large silver ewer, the Cupid and bordello paintings that hang on the back walls, were those in which he traded or property of his well-to-do mother-in-law, who owned a discreet collection of paintings. However, the most expensive objects, such as the gilt chandelier, woven tapestries, virginals, harpsichord and Turkish carpets, were probably above the living standards of the artist or his mother-in-law, and were lent to him by supportive patrons or friendly connoisseurs who were happy to collaborate in the master's next masterpiece. The citterns and lutes, both fairly affordable objects, were property of the painter, perhaps inherited from his grandfather, who was a professional musician.

Art Class, Jan ter Borch fig. 9 Art Class (Drawing Lesson)
Jan ter Borch
1634
Oil on canvas, 120 x 159 cm.
Rijksmuseum, Amsterdam
x fig. 10 An Artist in his Studio, Possible a Self Portrait next to a Violin, A Globe, A Sculpture and a Sea Shell on a Draped Table
Pieter Corneisz van Egmondt
Oil on panel, 34.6 x 30.6 cm.
Private collection

Vermeer's studio

The Art of Painting, Johannes Vermeer fig. 11 The Art of Painting
Johannes Vermeer
c. 1662–1668
Oil on canvas, 120 x 100 cm.
Kunsthistorisches Museum, Vienna

How did Vermeer paint? If The Art of Painting (fig. 11) were a literal representation of the artist's studio, we would have a fairly good picture of how he worked. The painting, however, is neither a autobiographical statement nor a "painted photograph" of Vermeer at his easel. It is an ideal representation made explicitly to trumpet the painter's elevated social position and learning, as well as to showcase his technical prowess and exceptional pictorial intellect. There exist, in fact, numerous discrepancies between the real working habits of seventeenth-century painters and those illustrated in The Art of Painting. The fact that Vermeer represented the painter beginning with Clio's laurel wreath alone should probably be taken as a measure of the iconographic significance of the wreath rather than literal evidence that, for example, Vermeer painted figures from the top down.3 While the drawing on the canvas is made with white paint or chalk, microscopic examination of the painting itself proves that the artist used dark brown lines to define the outline the arms of Clio, the chandelier and the map. The costume worn by the painter is anachronistic; few similar can be found in paintings of the same period. In almost every other representation of artists' studios, a small cabinet for the painter's equipment stands close to painter's easel. It would be hard to imagine that the dandified artist in Vermeer's picture would have found it practical to walk back and forth countless times each day just to clean a dirty brush or lay a bit more paint on his palette. Nonetheless, The Art of Painting provides evidence of one of the most significant aspects of the Vermeer's working habits, and one that differed from those of his contemporaries: he worked largely in the same environment represented in his paintings. Vermeer painted from deliberately and carefully constructed scenes, which we might call mise-en-scènes, although they were not always furnished completely or exactly as we see in his pictures.

Portrait of an Artist in His Studio, Michiel van Musscher fig. 12 Portrait of an Artist in His Studio (traditionally identified as Willem van de Velde II)
Michiel van Musscher
Oil on panel, 47.5 x 37 cm.
Dickinson Gallery, London and New York

In the seventeenth century, in fact, painting directly from nature was not a normal working method. As it has been pointed out, history painters and landscape painters generally created sets of drawings of the figures, background and props, and then combined them into complete compositions. An illustration of this practice can be seen in Michiel van Musscher's (1645–1705) The Artist in His Studio (1679; fig. 12), which portrays a young artist at work on a landscape as he mixes his paints and studies a group of sketches strewn out on the floor. Certainly, the highest-tier interior painters, such as Dou, Gerrit er Borch (1617–168) and Frans v an Mieris (1635–1681, also portrayed complete environments. But in many case their works betray discrepancies in scale of the objects, lighting and perspective that suggest they were not painted from a single composite scene but artfully composed from objects viewed in different circumstances. Only Vermeer's paintings and Pieter de Hooch's (1629– 1684) finest Delft pieces, despite the sometimes awkward drawing of the latter, give the impression of the spatial unity and coherence of lighting that can be understood only from a concerted study of a single, extant environment from a unique point of view.

Informed specialists maintain that Vermeer probably worked in two or three different studios during his twenty-year career. This hypothesis is based on the differing structures and patterns of the side windows, although it cannot be ruled out that he painted in fewer rooms but introduced variants to make them look different, a practice which is particularly evident in the work of De Hooch. Henk Zantkuijl, former head of the Amsterdam Office of Monuments, suggested that the windows shown in some of Vermeer's earlier paintings might be rooms in Maria Thins' house; but according to Philip Steadman these are pictures dated to the period before Vermeer and his wife moved to Thins' house some time in the late 1650s. It has been speculated that the windows shown in The Astronomer and The Geographer belonged to the house of the Delft scientist Antony van Leeuwenhoek, although there is no objective evidence in regards.

The locations of only two of the three proposed studios can be identified with any degree plausibility. The first is the upper floor of his father's inn, called Mechelen. This conjecture is based on the architectural features of the early The Little Street, which supposedly gives a view of the façades of early Delft houses as seen from an upper floor window at the back of Mechelen across the canal on Voldersgracht. The houses were torn down to make way for the new Guild of Saint Luke soon after the artist depicted them. Some believe that the rustic-looking windows of the Officer and Laughing Girl and The Milkmaid suggest that both works were painted there as well. Tim Jenison has recently speculated that the patterned reflections of the window in the Officer and Laughing Girl reflect the some features of the arched entrance way of the Nieuwe Kerk as seen from a window of Vermeer's studio facing south to the Markt at Mechelen. The second candidate, more credible than the first, is the Oude Langendijk studio, which is probably represented in The Music Lesson and other paintings with similarly-constructed window fittings (fig. 13).

Vermeer's studio, drawn from three drawings of Philip Steadman fig. 13 Vermeer's studio, drawn from three drawings of Philip Steadman, Vermeer's Camera: Uncovering the Truth behind the Masterpieces (2002) courtesy of Philip Steadman

According to Steadman's estimates this studio was about 6.6 meters deep, 4 meters wide and about 3 meters high. The raw brick walls were layered with thick coats of white-wash to create a smooth, hygienic and light-reflective surface. Since it was located on the second floor, the brownish coloring of the glass panes of the window nearest to the background wall of The Music Lesson may possibly represent a part of the tower of the Nieuwe Kerk or other buildings nearby.4 Vermeer's chief biographer, John Michael Montias, speculated that Vermeer might have maintained a studio in the family inn after this move to Oude Langendijk. In any case, archival evidence proves that Vermeer had a studio in a "front room" of his mother-in-law's house at one time or another although the year in which he transferred his family to Oude Langendijk is not known precisely.

As is apparent from the information given above, it is generally assumed that Vermeer worked in a studio that was located within his private living quarters. Interestingly, however, it has recently been questioned if two paintings may have been executed outside this context. Huib Zuidervaart, of the History of Science Department of the Huygens Institute for the History of the Netherlands, has advanced an interesting theory on the origin of Vermeer's Glass of Wine and the The Girl with a Wine Glass, both of which represent the same ceramic floor tiles and an open window that features an identical coat of arms. Zuidervaart asserts that the two works were commissioned as wedding presents for two of the grandchildren of Moijses van Nederveen and Janetge de Vogel. Moijses van Nederveen came from of a prominent Delft family which was one of the four local producers of gunpowder, delivered to the Dutch army. Although historians had previously identified the families to which the coat or arms belonged—the heraldic emblem is a combination of those belonging to the Van Nederveen and De Vogel families—no one had as of yet explained why Vermeer would have chose to include such a conspicuous elements into two ambitious compositions. Zuidervaart points out that dates given to both paintings by Vermeer experts coincide with the dates of the weddings in question. Moreover, the symbolic implications of the coat of arms (figure of female virtue holding a crest) would appear in agreement with the idea as a wedding gift as well as being consistent with the background of Vermeer's presumed clients. Given that both paintings present not only the same window but the same set of ceramic tiles, which Vermeer never painted again, and having reasonably established the circumstances in which two paintings originated, Zuidervaart has cautiously advanced that Vermeer executed both works on the premises of the Van Nederveen/De Vogel residence in Delft, rather than in his private studio. This idea deserves proper consideration in that had the painter not executed his painting in the room where the window was originally located, he would have had to detach it and somehow fit it into the window frame of his own studio. In this event, either the Nederveen/De Vogel window had the same precise measurements of the window frame of Vermeer's own studio, or it would have been necessary to alter it to make it fit. While both solutions are possible, the simplest explanation, to which there seems to be no counter evidence, is that Vermeer painted both works at Van Nederveen/De Vogel residence as Zuidervaart suggested.

Windows

An Interior, with a Woman drinking with Two Men, and a Maidservant, Pieter de Hooch fig. 14 An Interior, with a Woman drinking with Two Men, and a Maidservant (detail)
Pieter de Hooch
c. 1658
Oil on canvas, 73.7 x 64.6 cm.
National Gallery, London

The windows that appear in Vermeer's interiors are typical of Delft houses of the period. This type of window is called cross-window (kruisvenster in Dutch). In most cases a lintel, the horizontal separation between two compartments, was placed in the middle of the window so that light and air compartments are the same size. Each window is divided into four sections, called lights. On the lower half of the frame there are two swinging windows, called casement windows (kloosterkozyn), attached to the frame by two hinges. The two upper lights are fixed. The casement window was divided into small square pieces of not fully transparent glass, held together by strips of lead or other metal. This kind of window was usually equipped with external shutters over the lower casements only, which can be seen in Vermeer's The Little Street. In order to look outside one had to open the shutters, which was not comfortable to do in the colder months of the year, although they were necessary for the purpose of security. Some paintings by De Hooch (fig. 14) show internal shutters on the upper lights but they do not appear in any of Vermeer's paintings. Almost all the casements display decorative patterns formed from the shapes of their panes. The individual panes were held together by lead glazing bars/muntins (roede in Dutch). Three types of patterns can be easily distinguished, which, for convenience, Steadman dubbed "lozenge," "hourglass" and "squares and circles." The first type are apparent in the early Girl Reading a Letter by an Open Window, Officer and Laughing Girl and The Milkmaid. The hourglass type is confined to two paintings; The Astronomer and The Geographer. The squares and circles type is by far the most complicated and can be seen in 8 different paintings. A fourth design of window is to be found uniquely in Lady Standing at the Virginals, which is is similar to the lozenge type. In four paintings there are stained-glass panes. The structure and molding of the frames vary within types. Three sets of windows allowed Vermeer to improvise a variety of compositional solutions. Different parts of the windows could be opened and shut to create the interesting patterns of light and shadow, in which the painter obviously relished.

The mysterious glass globe (fig. 15) which hangs from the ceiling in the late Allegory of Faith confirms the three-window hypothesis. Although the brushmarks that stand for the globe's reflections are rather abstracted, it is nonetheless possible to identify some features of the studio in which this picture was painted. The wobbly rectangular gray patch near the center of the globe represents a portion of the dimly lit back wall. According to Steadman, the playful "blurbs of light paint at the bottom of the sphere are most likely reflections of the woman's face, white skirt, and sleeves, and of the Bible on the table. At the center left we see reflections of the three windows of the room, none of which are portrayed in the painting itself. The middle window shows three clearly visible lights which describe an inverted L-shape. The leftmost window, which is the nearest to the background wall, is almost entirely curtained except for the tops of the lights. To the right there is another rectangular patch of light which may be upper light or lights of a third window, at the back of the room, adjacent to the back wall. This third window cannot be seen in any other painting, since it is behind the viewpoint of the painter in every case."

Allegory of Faith, Johannes Vermeer fig. 15Allegory of Faith (detail)
c. 1670–1674
Oil on canvas
114.3 x 88.9 cm.
Metropolitan Museum of Art, New York

The most practical way to paint was to position the easel directly below the second or third window. The broader, multi-figure scenes would have required the painter to sit as far as possible for background wall directly under the third window, while single figure arrangements were best viewed from closer up, under the second window. That the artist sometimes worked under the third window is comforted by Steadman's analysis of the objects and architectural features of Vermeer's interiors. By means of a procedure called reverse perspective, or reverse geometry, Steadman was able to calculate with sufficient precision the actual view point—the position where the camera obscurity's lens or the painter's eye was placed—of 6 paintings by Vermeer (The Girl with a Wineglass, The Glass of Wine, Lady Writing a Letter with her Maid, Lady Standing at the Virginals, The Music Lesson and The Concert). It happens that the view point of each of these works was located directly under the third window. Steadman utilized this information as a partial confirmation that Vermeer used the camera obscura as an aid to his painting. The perspectives of the Officer and Laughing Girl and The Milkmaid suggest they were painted beneath the second window.

Floors

In Vermeer's 26 domestic interiors there are 11 paintings with black and white tile floors, 2 with ceramic tiles and 5 with simple unadorned pavements. Never once do we see the large wooden planks represented in so many Dutch paintings of domestic interiors. Differently from most every other interior painter of the time (fig. 16) except for De Hooch, Vermeer always set the tiles of his scenes diagonally to the picture plane. The imaginative De Hooch experimented with many different patterns, with some scenes featuring both perpendicularly and diagonally set tiles (fig. 17). In any case, Vermeer's foray into tile painting was not without difficulties. The spatial recession created by the small blue and yellow ceramic tiles in the two early shoe-box interiors (The Glass of Wine and The Girl with a Glass of Wine) is problematic, most probably due to the intricacies of three-point perspective and the difficulty of hand painting the perimeters of each tile without losing the underlying scheme, especially those in proximity of the background wall which are very small and drastically deformed. The artist quickly remedied this problem by moving his easel farther from the scene but above all by using much larger tiles and by adopting broader, cross-like patterns in which the white tiles are partially isolated from one another. Had the tiles of The Music Lesson (fig. 17) been painted with conventionally alternating white and black tiles, the perspective thrust would have interfered with the overall compositional equilibrium. Jacob Ochtervelt (1634–1682) seems to have understood Vermeer's lesson and painted tiles set in broadly spaced, diagonal patterns. Various measurements would seem to demonstrate that the dimensions of Vermeer's marble floor tiles are relatively consistent across every work which represents them.

x fig. 16 Merry Company Making Music
Jan Miense Molenaer
1631
Oil on panel, 56 x 79 cm.
Private collection
Woman with a Child in a Pantry, Pieter de Hooch fig. 17 Woman with a Child in a Pantry
Pieter de Hooch
c. 1656–c. 1660
Oil on canvas, 65 x 60.5 cm.
Rijksmuseum, Amsterdam
The Music Lesson, Johannes Vermeer fig. 18 The Music Lesson with different tile pattern

Tiled floor patterns are represented in painting as early as in Roman times. Early Renaissance artists and designers harked back to classical precedents and routinely employed the motif as a means of creating spatial depth and providing sumptuous decor for religious scenes (fig. 19). These early floors, however, appear to tip awkwardly forward creating the sensation of accelerated perspective. Many experiments were necessary to harmonize the perspectives of the floors, architectural features and figures, which had previously been treated as separate problems. With the diffusion of mathematical, one-point perspective the slipping and sliding of tiled floors was eventually tamed, and intricately patterned tile floors became a means for the advanced artist to showcase his technical prowess, unrelated to the painting's narrative content.

Frontispiece of the Chroniques de Hainaut, Rogier van der Weyden fig. 19 Frontispiece of the Chroniques de Hainaut
Rogier van der Weyden
1447
Opaque paint, gold, pen and ink on parchment
Royal Library of Belgium
Town Hall, Delftfig. 20 Town Hall, Delft

Black and white floor tiles are one of the most characteristic features of Vermeer's interior scenes, although Dutch genre painting offers many chances to delight in similar motifs. Marble tiles make their debut in the Netherlands during first decades of the seventeenth century. Willemijn Fock, a historian of the decorative arts of the Netherlands, maintains that it is highly improbable that the marble floors that appear so often in Dutch interior painting were painted directly from life. Such a luxury items could be found only in the homes of the rich and so were beyond reach of both Vermeer and his mother-in-law, with whom he lived and in whose house he kept his studio. Moreover, there is no historical evidence that marble floors were used above the ground floor (remembering that Vermeer's studio was on the second floor). Period inventories reveal that marble floors in domestic settings were generally restricted to one room, the voorhuis (the main entrance), where they would have most impressed visitors. To see real marble floors Vermeer could have visited the Delft Town Hall (fig. 20) or the palatial Huis ter Nieuwburg at nearby Rijswijk. The painting of marble tiles must have had a three-fold propose for Dutch interior artists: to intensify the illusion of spatial depth, to showcase the artist's command of perspective and to create richly decorated environments that would appeal to the tastes of discerning upper-class clients.

There are two types of tiles in Vermeer's painting: ceramic and marble. According to Steadman, the minute cracks and chips of the ceramic tiles in The Glass of Wine and The Girl with a Glass of Wine suggest that they were observed and therefore painted from life, even though creating the such minor effects would have been easy for any modestly trained painter. More importantly, Steadman discovered that the side measurement of these tiles is exactly half of that of the larger black and white marble tiles, allowing four ceramic tiles to fit into a single marble tile. Steadman suggests that Vermeer may have painted the marble floors in a room that in reality contained cheaper ceramic tiles, exploiting their underlying geometric grid to project the larger tiles with different patterns according to the compositional exigencies of each work.

There are essentially three patterns of marble tiles in Vermeer's paintings. "In The Music Lesson alone, separate white tiles are framed in a lattice of black stripes. In Allegory of the Faith alone, a pattern of what can be read as white Maltese crosses, each made from five tiles, is set on a black background. In the six remaining pictures the colors are reversed, to make a pattern of black crosses on a white ground.5 In one painting only, Woman with a Lute, do the tiles meet the base of the background wall at an arbitrary, intermediate point rather than cutting them in half into two neat triangles. The fact that Vermeer's tiles exhibit no reflections—in reality they would have been polished, as can be seen in other paintings of the time—would suggest that they were invented, although a skilled painter such as Vermeer could have easily eliminated the reflections for aesthetic reasons.

Art historians have unanimously discounted the possibility that Vermeer's studio was actually fitted with real black and white marble floor tiles. Imported marble flooring was far too expensive for the artist's income—there is no marble native to the Netherlands. Nonetheless, Chris Eckersley, a freelance designer and sculptor, has suggested that Vermeer may have possessed a stock of real marble tiles and laid them out on the floor in these varying patterns for each of his paintings.6 After Eckersley had read Steadman's calculation that the floor tiles in Vermeer's paintings measured 29.3 cm. (derived by a process of adjusting the entire size of the studio and all its contents to arrive at the best fit to the known sizes of many objects: chairs, tables, virginals, painted paintings, maps, as well as the Delftware skirting tiles), he realized that this size is extremely close to half of a Florentine brachia (29.18 cm.), which was the standard size of seventeenth-century Italian marble tiles (presumably from Carrara). However, it is hardly out of the question that the artist used faux tiles made of colored wood which could be easily arranged in any pattern desired. With the collaboration of the industrial designer Allan Kuiper (based upon reconstruction drawings by Zantkuijl) the Dutch art historian Kees Kaldenbach developed a digital reconstruction of Vermeer's home at the Oude Langendijk.7

Adelson's Checker-Shadow Illusion fig. 21 Adelson's Checker-Shadow Illusion

Representing the variations of brightness on a checkered floor tile is a particularly complex undertaking due to the fact that the two differently colored tiles must be modified simultaneously, in both tone and, to a lesser degree, hue, with respect to the the amount of light they receive. For every change in tone of a white tile, the adjacent black tiles must be proportionately modified. Owing to the sharp contrast between the black and white tiles and the perceptual mechanism called brightness constancy it becomes exceptionally problematic for the painter to "see through" the local contrasts of adjacent tiles to the broader falloff of light on the floor. The well-known diagram by Ted Adelson (MIT) (fig. 21) illustrates the vexations of evaluating the relative values of tone when applied to the checkered floor motif. Even when the viewer is informed that squares "A" and "B" of Aldeson's floor are precisely the same value, the perceptual system "corrects" them to make them look as if they were differently colored tiles, with the result that white tile labeled "A" strikes the viewer as intrinsically darker than tile labeled "B," although in actuality it is not (if the reader is not convinced he may see Aldeson's by clicking here). This is because from a biological standpoint there is nothing to be gained by understanding the absolute tones of each separate tiles. Instead, understanding brightness in relative terms allows us construe a plausible picture of a cylinder which projects a shadow on a checkered floor. The perceptual forces at play in brightness constancy are complex and are probably elaborated on different levels of the optical system. The artist, then, must find ways to undo this correction in order to render the tiled floor realistically, otherwise, his floor will seem to have no particular light. A phenomenon similar to Adelson's occurs in Vermeer's The Music Lesson and the late Lady Writing a Letter with her Maid. A detail of this painting (fig. 22), which is rendered in black and white to facilitate the reading of tonal value, shows that the value of one of the white tiles near the side wall is only slightly lighter than that of a black tile near the foot of the carpet, and yet within context we unquestionably understand that the left-hand tile is intrinsically white and the right-hand tile is intrinsically black.

Lady Writing a Letter with her Maid, Johannes Vermeer fig. 22 Lady Writing a Letter with her Maid (detail)
Johannes Vermeer
c. 1670–1671
Oil on canvas, 71.1 x 58.4 cm.
National Gallery of Ireland, Dublin

Location & weather

The studio in which Vermeer painted some his finest pieces was a noisy place overlooking the bustling hub of the town's civic life (Markt), only one story above street level. Upstairs was an attic where the artist ground his paints and stored extra equipment. In the streets below carts full of fruit and vegetables began rolling in hours before the artist had enough light to work by. Throughout the day street peddlers hawked their wares, traders shouted and children played kolf. Street musicians sang and played the latest tunes on droning bagpipes and screeching hurdy-gurdies.

The back side faced a narrow alley which connected Oude Langendijk to the noisy cattle market. In front, it was only a stone's throw away from the mastodonic tower of the Nieuwe Kerk, which was armed with 32 bronze bells that clanged throughout the day at regular intervals keeping time and lifting the spirit of the town's hard-working citizens. Some foreigners enjoyed the frequent carillon concerts but some were annoyed by their "incessant jangling." Moreover, Vermeer's resilient wife, strong-willed mother-in-law, domestic maid and a slew of children milled around just beneath the timber floor which separated the artist's studio from the living quarters below. Factoring in the already noisy environment and the fact that children are children—it is well know that foreigners thought Dutch parents were intolerably tolerant to their children's misconduct—the silence which exudes from the artist's perfect constructs is all the more perplexing.

Meteorology also played an important role in Vermeer's studio activities, a fact that is generally ignored by many who look at his paintings. Dutch winters, already inimical to the fine painter who must remain seated dead-still at his easel for hours on end, was even more merciless in Vermeer's time owing to the so-called Little Ice Age that started during the mid-sixteenth century and ended about 1850. In this period, the average temperature was significantly lower, winters were very much colder and expanses of water remained frozen over longer periods (fig. 23). Vermeer's biographer, John Montias, discovered that the artist may have bought an ice sled (fig. 24) equipped with a sail for his children for eighty guilders to make the best of the stern conditions.8

Oostpoort, Delft (photo courtesy of Pieter Haringsma)fig. 23 Oostpoort, Delft (photo courtesy of Pieter Haringsma)
Winter Landscape with Ice Skaters, Hendrick Avercampfig. 24 Winter Landscape with Ice Skaters (detail)
Hendrick Avercamp
c. 1608
Oil on panel, 77.3 x 131.9 cm., Rijksmuseum, Amsterdam

"In Dutch houses an open fire was the only source of warmth, with peat the usual fuel in most dwellings. An enormous amount of peat was consumed in an effort to keep warm: an estimated 1.6 tons annually for each city resident during the seventeenth century. Even so, people wore several layers of clothing most of the time."9 And so did Vermeer while he was painting. Despite the fact that upper-tier painters enjoyed portraying themselves draped in imported satin Japonese rock (fig. 25) or clothed in historical costumes, there is little documentary evidence about how painters really dressed when at work. Leonardo da Vinci (1452–1519) once commented that painting was superior to sculpture also because painters could work in their finest clothes. But it would seem that "the seventeenth century had no specific occupational costumes in the sense of clothing specifically designed and made for various activities. Most trades involved nothing more than protective clothing like aprons over one's every-day dress. In other words, painters were likely to have worn their everyday dress with floor-length, loose-fitting protective gown, which were known as a tabbaard or a nacttabbaard (nightgown). Such garments were very practical because they gave freedom of movement and could be lined with fur for to keep them warm in the winter. There is frequent mention of these items in the inventories of Amsterdam painters. Rembrandt once made a drawing of himself with a tabbaard (fig. 26) which was inscribed with the words. ‘Drawn by Rembrandt van Rijn after his own image/as he was attired in his studio.'"10 The situation must have been different for the artist's models. Since is difficult to imagine that the young woman who posed in the Woman with a Pearl Necklace could have stood motionless with arms bared to the bitter cold of the studio, he must have painted from dressed mannequins during the winter months, or during warmer weather.

Self-Portrait, Michiel van Musscherfig. 25 Self-Portrait (detail)
Michiel van Musscher
1679
Oil on panel, 57 x 46.5 cm.
Museum Rotterdam, Rotterdam
Self Portrait in Studio Attire, Full Length, Rembrandt van Rijn fig. 26 Self Portrait in Studio Attire, Full Length
Rembrandt van Rijn
c. 1655
Pen and bistre, 20.3 x 13.4  cm.
Rembrandt Huis, Amsterdam
View of Haarlem from the Northwest, with the Bleaching Fields in the Foreground, Jacob Isaacksz. van Ruisdael fig. 27 View of Haarlem from the Northwest, with the Bleaching Fields in the Foreground
Jacob Isaacksz. van Ruisdael
c. 1650–1682
Oil on canvas, 43 x 38 cm.
Rijksmuseum, Amsterdam

Studio light

Studio lighting is a most important factor for a painter, and it is narrowly connected with the architecture of the studio, more especially with the source of light. Whether a window faces north, east, south or west makes a noteworthy difference in the type of light it receives. If one prefers morning sunlight to spray across the breakfast table, the breakfast room window should face east, but if one wants to paint, the window must face north. The direction of direct rays of the sun coming from the east, west and south shift angles from one minute to the next creating fascinating but hopelessly complex patterns. Every half hour the scene needs to be repainted. Moreover, for the great part of painting styles, direct sunlight produces such a wide range of lights and darks that they are impossible to capture with the artist's poor reflective paints. For these reasons painters have always preferred a northern exposition because northern light is cooler than southern light, but above all because it is diffused and relatively constant throughout the great part of the average working day. Thus, the light that falls on any object or model will almost be always the same. Leonardo seems to have been the first to have written about the advantages of northern light for the painter.

On the average, Vermeer had to factor into his working schedule more than 200 days of rain per year with less than two hours a day of sunshine in the winter months. In the Netherlands, winter days can be exceptionally gloomy even at midday; rain, intermittent drizzle and cloudy skies were, and still are, the norm. Light is so scarce that windows could fill from two-thirds to three-quarters of an external wall. Various combinations of curtains, shutters, and half-shutters were used to retain warmth or exclude sunlight on bright days.12

Although the three large windows of the Vermeer's studio let in the cool, constant light of the north adapted for painting, its intensity was often unpredictable owing to the incessant march of low-flying clouds that without warning plunged the studio into a deep penumbra and released it back to brilliance in a matter of seconds (fig. 27). In such conditions, the rendering of light and shadow, which plays such a crucial role in establishing atmosphere and compositional unity in Vermeer's art, would have been almost maddening. On heavily overcast or rainy days the artist may have found more productive things to do than paint his brightly-lit pictures. Perhaps, the brighter works were programmed for the summer months when light was more abundant. In any case, Vermeer's studio was well-lit. Approximately 40% of the area of the wall visible to the left was occupied by windows.

Studio equipment

>An Artist in his Studio, Jacob van Spreeuwen fig. 28 An Artist in his Studio (detail)
Jacob van Spreeuwen
After 1650
Oil on panel, 30.5 x 36.8 cm.
Private collection, New York

Easel. Easels (a Dutch word for "donkey") were invariable equipped with three legs and made from wood (fig. 28). The forward two legs are angled towards each other at the top, and lean against the third leg hinged near the top, similar to those used in former times for black-boards in schools. The front legs were studded with a series of round holes at regular intervals. Wooden pegs could be inserted in the holes at varying heights to support a horizontal wooden slat on which the painting itself could rest. The canvas was generally positioned at a height so that its middle (where the most important parts of the painting are generally located) could be reached without any effort. When it became necessary to work on the higher or lower parts of the painting, the wooden pegs could be quickly adjusted. The easel was tilted backwards to insure stability, avoid reflections on the canvas and facilitate a more natural posture for the painter. Many easels were reinforced with a horizontal slat which connected the two front legs near the bottom, which was also used to rest the painter's forward foot (some depictions show this slat worn down by continual use).

Mannequin ( Please note that the following information regarding mannequins below is drawn largely from the excellent Silent Partners: Artist and Mannequin from Function to Fetish by Jane Munro (Yale University Press, 2014, to accompany an exhibition organized by the Fitzwilliam Museum, Cambridge)

This term is derived from mannekijn, an old Dutch word for "little person." It was absorbed into English usage at about the same time that English speakers took from the Dutch the words "easel" and "landscape." It was sometimes referred to as "the boy" or a "lay figure."

Dummies, called "dobbles " were used by knights to hang their suits of armor.13 In 1464, the Florentine architect, writer and sculptor Filarete (Antonio di Pietro Averlino or Averulino, c. 1400-c. 1469) bragged he invented the mannequin when writing about the need to dress sculpted or painted figure in costumes of suitable circumstances, advising artists to draw draperies di naturali by using a figuretta di legname, a "little wooden figure" with "jointed arms, legs and neck." Filarete's mannequins must have been much smaller than life. If the fabric was not of a suitable weight or pliability to hang well, he recommended the mannequin and its drapery be "bathe[d]...well" in melted animal skin glue in order to stiffen the folds into the desired shapes as they dried. This process could be undone by dipping into warm water and reshaping. Most early Italian mannequins were small, typically the height of a braccio (an arm) or "half an ell" (ulna or forearm), measures which in reality could range between around 18 and 70 centimeters in length. Although small-scale mannequins were convenient for plotting out a composition, a larger, life-size model was considered necessary to obtain the most convincingly naturalistic effects, and more especially when it came to painting clothing or drapery, since most cloth is too bulky to fold correctly on smaller mannequins. In his Vite, Giorgio Vasari (1511–1574) first records the use of a large wooden mannequin by Fra Bartolommeo (1472–1517) —one of his drawings portrays a draped mannequin (fig. 29)—who worked not only with small-scale figures, but also used a jointed wooden model, grande quanto il vivo (life size), specifically in order to paint and draw the folds of drapery, armor "and such like things." By the end of the fifteenth century, it is clear that the practice was also widespread among painters, such as Piero della Francesca (c. 1415–1492), and particularly those who like Leonardo, who had studied in the workshop of sculptors, or painter-sculptors, such as Andrea Verrocchio (c. 1435–1488)." Leonardo's exquisite drapery studies (fig. 30) were most likely created with the aid of life-scale mannequins.

Drapery Study for figure of Christ in the Last Judgment, Fra Bartolommeo fig. 29 Drapery Study for figure of Christ in the Last Judgment
Fra Bartolommeo
1499–1500
Drawing
British Museum, London
Drapery Study, Leonardo da Vinci fig. 30 Drapery Study
Leonardo da Vinci
c. 147–1484 Drawing
Louvre, Paris
La prima-[quinta] parte della luce del dipingere et disegnare, ..., Crispijn van de Passe fig. 31La prima-[quinta] parte della luce del dipingere et disegnare, ...
Crispijn van de Passe
Publication date 1643
Publisher Ghedruckt t' Amsterdam : Ende men vintse te koop by Ian Iantsz. ... als mede by den Getty Research Institute, Los Angeles

In the seventeenth-century, mannequins began to be noted in Dutch painters' studios. Simon Kick, Michael Sweerts (1618–1664), Van Spreeuwen (fig. 10), Adriaen van Ostade (fig. 6) and Wallerand Vaillant depicted mannequins in paintings (fig. 33) and Ter Borch is known to have owned a relatively large lay figure. Ter Borch's father wrote to his son in London, "Dear child, I am sending you the mannequin, but without a stand because it is too large and too heavy to be put into the trunk. For a small amount of money you can have the stand made there. Use the mannequin and do not let it stand idle, as it has done here, draw a lot: large dynamic compositions." From this letter it appears that mannequins were unknown or unavailable in London. Gabriel Metsu (1629–1667) once exchanged a portrait for a mannequin. Typical examples of these devices appear in engravings by Crispijn van de Passe (1597–c. 1637; fig. 31). De Passe, however, warned they be used only for the study of draping the body, writing that "The Painter ought to avoid all manner of stiffness and hardness of folds, and be careful that they don't smell of the layman (mahhewuin)." Life-size mannequins are featured in studio scenes by Simon Kick (1603–1652; fig. 32) and an obscure Dutch painter D. Witting (active in the second quarter of the seventeenth century). It is very likely Vermeer most likely possessed at least one life-size mannequin.

Artist in his Studio, Simon Kick fig.32 Artist in his Studio (detail)
Simon Kick
c. 1645–1650
Oil on panel, 92 x 69.5 cm.
National Gallery of Ireland, Dublin

Over the course of the nineteenth century, the mannequin gradually emerged from the studio to become the artist's subject, at first humorously, then in more complicated ways, playing on the unnerving psychological presence of a figure that was realistic, yet unreal—lifelike, yet lifeless.

Mahlstick. The maulstick, or mahlstick (painter's stick), is a long, round stick that is primarily used for painting detail (fig. 34). The word mahlstick derives from the Dutch maalstok from malen, "to paint." The mahlstick is held (by the right-handed painter) with the left hand and leaned against the upper or right edge of the canvas to steady the hand that holds the brush, preventing contact with the wet paint. It is the difference between writing with the heel of your palm on the table or writing with your whole hand in the air.14 The mahlstick can also be used to paint straight lines by slowly dragging the lower part of the brush (ferrule) along its length. Mahlsticks are represented almost every time in portrayals of European painters at work. In most representations one end of the mahlstick is topped with a leather cover—one with a red cap is proudly displayed in Vermeer's The Art of Painting (fig. 35).

A Young Boy Copying a Painting, Attributed to Wallerand Vaillant fig. 33 A Young Boy Copying a Painting Attributed to Wallerand Vaillant
c. 1660
Oil on panel, 31.1 x 39.5 cm.
Guildhall (Samuel Collection), London
The Painter's Studio, Adriaen van Ostade fig. 34 The Painter's Studio (detail)
Adriaen van Ostade
c. 1670–1675
Oil on panel, 37 x 36 cm.
Rijksmuseum, Amsterdam
The Art of Painting, Johannes Vermeer fig. 35 The Art of Painting (detail)
Johannes Vermeer
c. 1662–1668
Oil on canvas, 120 x 100 cm.
Kunsthistorisches Museum, Vienna

Oriental Carpets. By the thirteenth century, merchant travelers like Marco Polo had remarked on the beauty of the Oriental carpets they encountered in their journeys. Soon after, carpets were imported by Venice and then distributed throughout Europe. While early carpets of this kind are rarely preserved, painted versions provide a good idea of how they looked. European painters, such as Giotto, Ghirlandaio (1449–1494), Hans Holbein (c. 1460–1524), Jan van Eyck (c before 1390–1441) and Lorenzo Lotto (c. 1480–1556/57), frequently depicted carpets originating from Turkey and Iran. Turkish carpets were one of the many exotic imports that appealed to seventeenth-century Dutchmen and can be seen in extraordinary numbers of Dutch genre paintings of the time. However, they were rarely appear on the floor, but always on tables, most probably because they were too precious to be soiled. For interior painter their colorful arabesque patterns and plush textures provided a perfect opportunity to show off their abilities and simultaneously create a sense of exotic luxury which would appeal to upper-class clients. The fact that so many turkse and persiche tapijten appear in Dutch interiors of the time might suggest that they were an integral part of the life of the upper classes. However, they do not occur so frequently in inventories nor are they documented in appreciable quantities on the cargo of Dutch merchant ships. It is known that some painters supplied clients with their own carpets and that the same carpet might be used for generations of artists. Vermeer seems to have used at least one carpet more than once. Most carpet commonly seen in Dutch interior paintings of the time and can be identified with extant types.

It is generally assumed that the carpets featured in Vermeer's paintings were painted from real models, although some liberties were taken with coloring. The level of detail varies considerably with the late renditions reduced to abstract pattern, without a single knot visible. Carpet expert Onno Ydema has shown that the carpet in The Music Lesson (fig. 36) is a sixteenth-century Medallion Ushak type, from Turkey, and is faithfully described by the artist.15 The carpet of The Procuress is of a similar type. The carpet that appears in the Christ in the House of Martha and Mary seems to be almost identical with the one represented in A Maid Asleep painted a few years later. Both have a broad, light orange border ending in a fringe, but the medallion in the former is yellow, in the latter green, which suggests the artist used one rug as a model and painted imaginary variations of it. Observing the bold but uninspired facture of Vermeer's early carpets (Martha in the House of Martha and Mary, The Procuress and A Maid Asleep), the vitality of those in his later interiors is surprising. Another carpet which appears in Vermeer's painting is the so-called "Lotto" type, whose name is drawn from the carpets portrayed in the works of Lotto. Differently from Dou and the most meticulous Dutch fine painters, Vermeer was not obsessed by the individual knots of the carpets, but the broad patches of color. The patterns were then punctuated with a series or small, roundish splotches (pointillés), which should not be mistaken for literal representations of physical knots but painterly renditions of camera obscura's disks of confusion.

Medallion Ushak type carpet (Courtesy Tim Jenison) fig. 36 Medallion Ushak type carpet (Courtesy Tim Jenison)
The Concert, Johannes Vermeer fig. 37 The Concert (detail)
Gerrit van Honthorst
1623
Oil on canvas, 123.5 x 205 cm.
National Gallery of Art, Washington D.C.

Repoussoir Curtains. The term repoussoir is derived from the French verb meaning to push back. In painting, a repoussoir is a means for achieving (i.e., forcing) spatial depth by placing a large figure or object, generally dark, in the immediate foreground of the composition. The repoussoir device tends to physically involve the spectator in the pictured scene in as much as his instinctive reaction is to "look around" it in order to observe what is taking place behind it. Repoussoir motifs set a dramatic entrance for the painting and help determine a sense of scale, and by tonal contrast accentuate the sensation of light. Caravaggio (1571/2–1610) had become famous for his use of ordinary people as repoussoir devices. In the classical landscapes of Claude Lorraine (1600–1682) the repoussoir element was usually a large, dimly-lit tree placed in the left-hand foreground. The Utrecht Caravaggist School frequently used human figures as repoussoir (fig. 37). Dutch landscapists used repoussoir trees, windmills, animals and buildings to enliven the uneventful Dutch countryside and create innovative compositions. Chairs were frequently employed as repoussoir devices by interior painters, but most of all hanging textiles.

A number of Vermeer's interiors feature repoussoir curtains. Some are quite large, and some occupy only a small portion of the canvas. The larger ones, usually elaborately designed tapestries, are pushed to the left and gathered up at the bottom so that they give the impression of a stage with the curtain drawn back.16 Such tapestries are present in The Art of Painting, Allegory of Faith, The Love Letter and Lady Writing a Letter with her Maid. In A Lady Seated at a Virginal and The Geographer they occupy only a small part of the upper left-hand corner of the composition, probably in order to alleviate the predominance of the rectangular layout rather to create spatial depth.

A large, green carpet once occupied a large part of the right-hand background of Mistress and Maid. The green pigment used have dramatically degrade over time. Viewed from life or in good reproductions the carpet's sweeping folds are visible to the naked eye.

Prints. Prints were to be found in every studio of the time. No painter, however poor, was without a little collection. The most popular prints, often arranged in large folio form, contained engravings and woodcuts by Dürer, or prints after the Italian masters—particularly Raphael (1483–1520), Mantegna (c. 1431–1506) and Michelangelo (1475–1564)—which served as compositional models. So many prints and copies were made by earnest pupils that it became impossible to separate the originals from the copies, a problem which would occupy a small army of art historians centuries later. Various "folios" were reported in a posthumous inventory of movable good in Vermeer's studio. The large volume which lies flat on the table of The Art of Painting is probably an example (fig. 38).

The Art of Painting, Johannes Vermeer fig. 38 The Art of Painting (detail)
Johannes Vermeer
c. 1662–1668
Oil on canvas, 120 x 100 cm
Kunsthistorisches Museum, Vienna

Plaster casts. "Plaster casts are simultaneously copies and originals. Once they leave the mold that makes them a replica they take on their own distinctive lives. And yet, cast collections of ancient, mostly Greek and Roman, or European sculpture share a common, if not global history.

Plaster casts have been formative for many artistic movements such as Renaissance, Baroque or Neo-Classicism. From the fifteenth century on casts were part of private collections of scholars, artists, aristocrats and royals in Europe. The replicas constituted an early canon of what were considered masterworks of the ancient Greek and Roman world. The latter formed the basis for the establishment the royal or courtly academies of arts, the first of which were several precursors of the later Académie des Beaux-Arts founded in Paris in the second half of the seventeenth century. Casts enabled artists to study human anatomy and to learn 'the idea of beauty.'"17

In depictions of drawing lessons or painters in their studios plaster casts are portrayed literally countless times in depictions of drawing lessons or painters in their studios (fig. 39 & 40). Willem Goeree, a Dutch art theorist who wrote an extensive treatise on drawing in 1668, wrote "that one make drawings after copies in the round [i.e., from a solid 'round' object rather than a 'flat' drawing] and plaster casts of good masters; such as one can come across very easily, some being very common and familiar; many of these can be bought for a modest sum and used to great advantage in the practice of art."

The Drawing Lesson, Jan Steen fig. 39 The Drawing Lesson (detail)
Jan Steen
c. 1665
Oil on panel, 49.2 × 41.2 cm.
John Paul Getty Collection, Los Angeles
>A Young Woman Seated Drawing, Gabriel Metsu fig. 40 A Young Woman Seated Drawing (detail)
Gabriel Metsu
c. 1655–1660
Oil on oak, 36.3 x 30.7 cm.
National Gallery, London

Given their near ubiquity it is reasonable to suspect that the term "accessories," found in the catalogue description of a now lost self portrait by Vermeer ("a portrait of Vermeer in a room with various accessories, uncommonly beautiful"), might have indicated, along with other common objects, a plaster cast of one kind or another. In any case, The Art of Painting features a oversize plaster cast of of a male head turned upward which in the past has been mistakenly called a "mask." In an informative essay dedicated to this picture,18 the art historian Sabine Pènot has speculated that the particular features of Vermeer's cast—in her words "a headband runs above the eyebrows, which blends into a diadem-like element, such that the top of the cast is pointed"—suggest it may be a portrayal of Apollo (fig. 41), the Greek god of light and patron of the arts. She believes that such an association is strengthened by the fact that the head is turned away from the viewer toward the unseen window that floods the studio with light, as well as by the most likely iconographic readings given to this most ambitious of Vermeer's paintings, which refer to the role of the art of painting among the liberal arts, il paragone, and the unification of poetry and painting embodied through Apollo and Apelles, a pervasive seventeenth-century topos in the Netherlands.

An Italian, eighteenth-century plaster cast of Apollo Belvedere fig. 41 An Italian, eighteenth-century plaster cast of Apollo Belvedere (left) and a detail of Vermeer's The Art of Painting (turned 45 degrees clockwise to facilitate comparison

Grinding Stone. It is generally assumed that the grinding of pigment and drying oil to make paint was done in the studio of the artist or his apprentices, although in depictions of artists' workshops one sees as many older men, perhaps journeymen (fig. 42), as young men bent over the grinding stone. An amusing pair of drawings by the Dutch painter Andries Both contrast the "successful" painter and the "poor" painter by the fact that the paint of the successful painter is being ground by a gleeful apprentice while the pigment of the poor painter is being ground by his distraught wife. It is known that paint was being mulled on large scale and commercially prepared in the seventeenth century but painters who cared particularly about the quality and consistency of the paint choose to grind their paints themselves. The inventory taken after the death of the Vermeer listed a grinding stone in the attic space above his studio.

Grinding is done with two elements the muller and the grinding slab (fig. 43). The muller was generally a stone rounded at the top and completely flattened at the bottom so it will be fully in contact with the grinding slab. Sometimes mullers were made of flint or whetstone although both elements were often made of the same hard stone, such as porphyry, so that the would wear evenly. Today, both are made of solid glass.

Paint Making in a Painter's Studio, David Ryckaert III fig. 42 Paint Making in a Painter's Studio
David Ryckaert III
c. 1635–1638
Oil on panel, 42.8 x 31.7 cm
Stadtgeschichtliches Museum "Schabbelhaus," Wismar
The Workshop of Evangelist Luke, Karel Slabbaert fig. 43 The Workshop of Evangelist Luke (detail)
Karel Slabbaert
c. 1645–1654
Oak wood, 64.5 x 50.1 cm.
Gemäldegalerie, Berlin

 

aint Luke painting the Virgin and Child, Workshop of Quinten Massys fig. 44 Saint Luke painting the Virgin and Child (detail) Workshop of Quinten Massys c. 1520 Oil on oak, 114.9 × 35.4 cm. National Gallery London

Stools. Differently than how painters work today, almost every early depiction of painters in their workshops show the artist seated. The kinds of chairs the used varies, from formal arm chairs to short-legged chairs with backs, four-legged stools and the box-type type of stool. A fancy four-legged stool is proudly featured in Vermeer's The Art of Painting (fig. 45) while a more humble version is featured in an informal studio scene by Joos van Craesbeeck (fig. 46) . However, the most common was the humble three -legged stool (fig. 44), which can be seen in a great variety of settings.

The Art of Painting, Johannes Vermeer fig. 45 The Art of Painting (detail)
Johannes Vermeer
c. 1662–1668
Oil on canvas, 120 x 100 cm
Kunsthistorisches Museum, Vienna
The Painter’s Studio, Joos van Craesbeeck fig. 46 The Painter’s Studio (detail)
Joos van Craesbeeck
Oil on panel, 48.5 x 66 cm.
Frits Lugt Collection, Paris

Painting in Vermeer's studio: A subjective account

Nothing is more distant from Vermeer's studio environment than that of the hackneyed image of the brooding twentieth-century artist boldly postured in the midst of his white loft strewn with litter, coagulated paint pots, huge unpainted canvases and assorted paint-encrusted palettes. More than a solitary lair where pent up emotions are vented, Vermeer's work space looked like a laboratory.

Although we know nothing of the artist's personality, it is nearly certain that he worked in an orderly manner. This is not because we assume that the subject matter and aesthetic qualities of an artist's work must conform to his persona, for while it is true that the lives and works of painters often appear to relate to each other very closely they sometimes are desperately remote. On one hand, the behavior of Spaniard Salvador Dalí is almost as bizarre as his surrealist landscapes. On the other, it is difficult imagine that De Hooch, who fashioned among the most amenable portrayals of domestic peace ever painted, could have died in an Amsterdam insane asylum. However, when creating such meticulously crafted works like those of Vermeer the demands imposed by the painting process and rebellious materials are nothing short crushing. It may not be coincidence that some of the most skilled fine painters in the Netherlands hailed from families of jewel cutters or silverworkers. Vermeer's own father worked in fine caffa (satin). In effect, painting in the fine manner necessitates a rational and systematic approach which virtually excludes spontaneous expression of the kind associated with the twentieth-century art production. Complicated compositions must be planned with precision in order to avoid time-consuming repainting. Rare and costly materials must be meticulously processed and properly conserved. Essences must be filtered, and brushes must be scrupulously cleaned each evening lest they turn hard become unusable the next morning. A slip of the hand or an unchecked cough at the easel transforms a one centimeter smile into a leer. And dust, one the most insidious enemies of sticky oil paint, must somehow be kept at bay, for even a speck becomes immediately visible on the glass-like surfaces of small paintings. In order to let the dust settle when entering his studio, it is said that Dou sat immobile in front of his easel for half an hour before picking up his brush. The order, method and mental discipline indispensable for creating fine paintings like those of Vermeer and his colleagues are confirmed by the habits of expert fine realist painters of today.

Vermeer's studio was divided in two functionally distinct but physically communicating spaces. On the farthest end, the west end, he staged his mise-en-scène in proximity of the left-hand window. Here, the model, dressed more or less appropriately, and all the movable furnishings and props were provisionally assembled according the initial picture that had formed in the artist's mind. In a second area, near the third window a scant few yards from the scene, Vermeer conducted, as it were, his painted experiments in optics, illusionism and poetry. Here, the tools of the artist's trade were carefully grouped together: a three-legged easel, a painting cabinet, a small stool, a camera obscura and a large curtain for controlling the fall of light.

It is probable that the initial planning stage Vermeer employed a rather large booth-type camera (as proposed by Steadman) big enough to contain the seated painter from which he could study the distribution of the elements of his composition and the patterns of light and dark they cause. When closed, the booth shielded the camera's image from the light of the studio's windows so that it would be as clear and bright as possible. The back wall of the booth was set flush against the wall opposite to that of the mise-en-scène. Once the artist had finalized the composition he would attach a sheet of paper to the back wall and trace the most telling contours of the camera's projection.

Before achieving the final compositional unity of his more complicated settings, Vermeer must taken considerable pains to optimize the reciprocally influential three-dimensional and two-dimensional orders of the painting. Paintings were taken down from the walls and replaced by others, or by maps. Different chairs with different ornamental material were brought in and moved around. Windows were shut or opened, curtains pulled forward, then again taken away or replaced by those with other colors. The figure's position and pose must have been considered with the utmost attention and adjusted and readjusted many times, as made evident by numerous pentimenti in the artist's work. When all the elements finally congealed into a indissoluble unity, the real figure was substituted with a mannequin, dressed accordingly and repositioned where it would remain perfectly immobile during the months to come.

In order to actually paint, the cumbersome booth camera had to be dismantled or moved to make space for the artist and his painting equipment, which could have been easily done by equipping it with a simple set of wheels. Once the device was removed, the painter set up his easel, sat on a stool and set his eye at the same point which had been previously occupied by the camera's lens. In this manner any discrepancy between the outline drawing on his canvas and what he was seeing and was minimized. Some authors have proposed that for further optical consultations, a small portable box-type camera would have imposed fewer mechanical restrictions than the bulky booth-type camera. But the optics of the small camera produces an image that is somewhat less suitable for painting purposes. Although he could have set up his scenes at the opposite end of his studio, like all painters Vermeer preferred light which originated from the left.

Vermeer painted most of his pictures while seated. His bodily movements were confined to the forearm and the fingers of his right hand, his head bobbing back and forth within a foot or two of the canvas. His handiwork was slow but nimble, deliberate but never schematic. The only sign of sloppiness that has ever been detected in the artist's oeuvre is a pair of barely discernible fingerprints embedded in the gray paint along the uppermost edge of The Guitar Player. Rather than painting all over the whole canvas at once with a full range of colors, Vermeer completed his composition in a jigsaw-puzzle fashion, piece by piece. He moved along to the successive passage only when the former was reasonably finished or properly worked up for successive refinement or glazing. His graceful, but firm, brushstrokes mimed the flow of cloth or the glint of light as it grazed across the carved relief of a gilt French frame. At all times he was attentive to the quality of his brushmarks: no two must be alike.

The time that Vermeer spent gazing at the actual mise-en-scène was probably much less than the time he spent looking at the painting that slowly arose from his handiwork. He occasionally stood up to ease the pain in his back or to delicately reshape a graceless fold of a satin gown with his hand. In his self-contained world, which was tainted with the odor of turpentine and noise coming in from every side except from the attic above, neither of which he noticed, nothing he did was haphazard or compulsive. His movements were deliberate and measured. When he depicted the finest details he held his breath as all painters instinctively do but never notice. He could spend hours depicting the slate-blue hills and valleys of an anonymous table cloth as it cascades from the top of a massive oak table: the same cloth that three-hundred and fifty years later even the most meticulous art historians would never once find worthy of a comment. Vermeer, instead, knew that it was absolutely crucial to the compositional balance because it concealed the complicated patterns of the floor tiles and the ornate features of the table in the lower part of the picture that would have diverted the viewer's eye from the poetic simplicity of the figure above without minimally upsetting the verisimilitude of the scene.

While painting, Vermeer kept his mind's eye open for possible analogies between visual phenomenon and the painting process, which he knew would create a far more convincing illusion of reality than any literal description, however precise it might be. In fact, the only subject matter he could transcribe literally were the occasional strands of text and the lines of the topographical designs on the wall maps. But even these features had to be processed, progressively darkened from left to right in order to confirm the sensation of the gradual falloff of light. Everything else, be it a white Italian wine jug, a pretty Dutch face or a Turkish carpet, had to be attentively observed, broken down into essential structural elements, and then reformulated into a specific hue, tone and shape according to the light it received and the distance from the painter. During a typical work session, he kept only a few paints on his tiny, oval-shaped palette, sometimes as few as three. A mahlstick and various kinds of brushes issued from its thumb-hole. In his other hand, he kept a clean rag to wipe excess paint from his brushes.

Vermeer occasionally leaned back from his easel curious to see how his current labor related to the whole. Sometimes he was annoyed by the contrast between a patch of bright color and the drab gray ground, or by the glare of freshly applied paint against the matte underpainting. Here and there the more highly finished passages popped out unnaturally against the vagary of the underpainting. Details were lost amid the overall unevenness of the work. But the artist had taught himself to be patient and even-tempered—he understood that his labor would bear full fruit only in the final stages of the painting process, perhaps during the final week or days. In fact, until the work had reached its mid-point, only a well-practiced painter familiar with fine painting methods would have been be able intuit in this disharmonious ensemble an exceptional work of art. For the moment, the painting had to executed on faith, according to plan, with patience and unwavering concentration.

Had a twentieth-century visitor been lucky enough to observe Vermeer at work for a few hours, he would have probably gotten the impression that the great artist had forgotten who he was, and, at moments, even what he was painting. The drag of paint, the spring of the brush and the constant ballet of oil and pigment seemed to monopolize his attention. Brushes were scrupulously cleaned with turpentine and wiped incessantly. It is likely that he used different brushes for different colors, because certain pigments have such a high tinctorial strength that even repeated cleansing will not keep them from contaminating the lightest paints. He carefully calculated the precise amount of paint to be applied to the surface of the canvas because some passages had to be worked up with grainy impasto while others scumbled on delicately with a dry brush barely tinged with paint, according to the visual effect sought by the artist. The painting was sometimes turned on its side or upside down on the easel to facilitate certain movements of the hand. Like many painters today he occasionally examined his work reflected in a mirror, a trick which allowed him to discern the most defective parts of the composition. If inspiration lurked behind this artisanal process, it never seemed to gush or spurt. At the end of the day our virtual visitor would have been assailed by the unsettling sensation that what the great Vermeer did, after all, looked much like ordinary work.

But the visitor's eyes tells him only what surfaces reveal. Like the child, who for the first time scribbles on a piece of paper and discovers something that was not there before, the Vermeer who paints willfully abandons himself to the painting process. The only difference between the two is that the artist conserves the original excitement throughout his adult life, even as he learns that his gestures must be in great part controlled in deference to the mimetic function of his activity. The picture-making experience is sometimes so strong that it is kinesthetic. The artist not only conjures an image on his canvas what caused him to paint, he takes possession of it. As the American writer and comic book artist Frank Miller somewhat unglamorously stated, "When you have a brush in your hand, inking a beautiful woman is a lot like running your hands over her." The extraordinary aspect of Vermeer's painting is that these intense picture-making pleasures are caused not by things beautiful in themselves; they can be intuited even in the knots of an Oriental carpet, the mundane head of a nail protruding from in the an uneven surface of a wall or the uneventful edge of a window sill.

† FOOTNOTES †

  1. SWILLENS, P.T.A. Vermeer. Utrecht. 1955. 114.
  2. In seventeenth-century Holland, the studio picture flourished as never before. Painters, whose numbers reached unprecedented proportions, were acutely aware that to succeed in an oversaturated market they needed to both protect their hard-earned intellectual status and to promote the art of painting. From H. Perry CHAPMAN. "The Imagined Studios of Rembrandt and Vermeer". Inventions of the Studio. Renaissance to Romanticism. Eds. Michael COLE and Mary PARDO. 2005. 127.
  3. GIFFORD, Melanie. "Painting Light: Recent Observations on Vermeer's Technique." In Gaskell and Jonkers, eds., Vermeer Studies. 185–199.1998. 198
  4. Personal communication with Tim Jenison.
  5. STEADMAN, Philip. Vermeer's Camera: Uncovering the Truth Behind the Masterpieces. Oxford University Press, 2001.
  6. ECKERSLEY, Chris. "Vermeer's tiles." Chris Eckersley: Making Things website <http://www.chriseckersley.co.uk/section745248.html>
  7. Kees Kaldenback, in Website, Johannes Vermeer. < http://kalden.home.xs4all.nl/>
  8. The purchase was recorded a year or two before 1660. Unfortunately, we cannot say for sure the buyer of ice sled was the artist Vermeer rather than his virtual namesake, the apothecary Johannes van der Meer.
  9. MUIZELAAR, Klaske and PHILLIPS, Derek Phillips. Picturing Men and Women in the Dutch Golden Age: Paintings and People in Historical Perspective. Yale University Press, 2003. 56.
  10. de WINKEL, Marieke. "Costume in Rembrandt's Self Portraits." Rembrandt by Himself. Ed. Christopher White and Quetin Buvelot. Zwolle. 1999. 65–66.
  11. MUIZELAAR, Klaske and PHILLIPS. 2003. 26
  12. See Vermeer's The Little Street.
  13. A pair of lay-figures from the sixteenth century, now in Innsbruck, are among the oldest that have survived.
  14. CAVES, Julie. "A classic tool: the Mahl stick." Website: Jackson's. http://www.jacksonsart.com/blog/2012/05/29/a-classic-tool-the-mahl-stick/
  15. YDEMA, Onno. Carpets and their Datings in Netherlandish Paintings, 1540–1700. Woodbridge: Walburg Pers. 1991
  16. An exception in Vermeer's oeuvre is the green satin curtain in Girl Reading a Letter at an Open Window, which is placed on the right-hand side. Vermeer had altered once more the original composition (with a "roemer" and a Venetian "winged" glass, similar to those known from the still-lifes by Willem Kalf, standing on the right-hand side of the table) by covering the glasses with the large repoussoir curtain. In any case, this curtain was not made to represent a repoussoir curtain but rather one which has commonly attached to particularly precious Dutch pictures to protect them from dust.
  17. "A Short History of Plaster Casts." Cornell Collections of Antiquities: Originals and Reproductions. Cornell University Library. <https://antiquities.library.cornell.edu/casts/a-short-history>
  18. PENÒT, Sabine, "Johannes Vermeer's 'The Art of Painting,' A Picture Marked with Light: Questions on Pictorial Invention," in Vermeer: Die Malkunst, 2010.