-4-
Two characteristic elements are seen in paintings deploying perspective. First, objects are smaller as their distance to the viewing increases—so objects that are farther away from the viewer will be smaller, objects closer will be larger. Second, objects are “foreshortened”—the size of an object’s dimensions are shorter along the line of sight relative to its dimensions across the line of sight. Foreshortening mimics the distorted appearance of a form when it is not perpendicular to the line of sight; for instance, a hand held out directly toward the viewer will look shorter, and larger, than a hand hanging straight down by a figure’s side.
Both of these elements help artists depict depth on a two-dimensional canvas. In both cases, the use of perspective requires the artist to paint what the eye actually sees, not what we expect to see. We know that two people are the same size, even if one is farther away, and that two hands are the same size, even if one is held out toward the viewer and the other is hanging down at the figure’s side. Prior to the development of perspective theory, artists would tend to paint figures the way we know them to be. But what we see is something different—that the person farther away looks smaller and that the hand held out toward the viewer looks larger. The painter now had to try to depict figures this way.
Vermeer’s most evocative use of foreshortening in his entire oeuvre is visible in The Procuress. We see the prostitute’s hand open, palm up, to receive the gold coin her client is about to toss into it. Her four fingers are, as one writer has put it, “so intensely foreshortened that only the genius of Vermeer makes her gesture appear so absolutely natural.” The thumb of her client’s hand holding the gold coin is so foreshortened as to become almost invisible, “and yet we almost sense its built-up energy ready to cause the coins to fall.” This mastery of one aspect of perspective is in contrast to what we find in Vermeer’s earlier picture Diana and Her Companions, in which it appears that the golden plate at Diana’s feet has been placed on an incline and is about to slip from the frame of the picture. Although the use of foreshortening in The Procuress is adroit, the figures still appear to exist in a split space, with the group of three figures existing in a separate plane from that of the laughing man. Vermeer would continue to struggle with the problem of representing a logically consistent space in his paintings, a problem that he would seem to solve only after 1657.
How did Vermeer achieve the remarkable effect of the foreshortened thumb? He most likely used a mirror. Painters had been using mirrors to help them portray foreshortening from the time that Alberti published On Painting in the fifteenth century. The Florentine sculptor and architect Antonio di Pietro Averlino, known as Filarete, had advised painters in the 1460s,
If you should desire to portray something in an easier way, take a mirror and hold it in front of the thing you want to do. Look in it and you will see the outlines of the thing more easily. Whatever is closer or farther will appear foreshortened to you.
Filarete was describing a characteristic of mirrors of the time. In an imperfect mirror—like the flat glass mirrors that had just begun to be available in Venice—the reflection of an object will appear flattened because of how the light reflects off the surface and slightly merges before reaching the two eyes of the observer. Because of this effect, in the mirror image the artist could see more clearly how to depict the object onto his or her flat canvas in order to correctly represent the perspective and foreshortening of that object from the frontal viewpoint.*2
Leonardo da Vinci pointed to two additional reasons for painters to use mirrors. By looking at a mirror image of the subject, a painter is forced to attend to aspects of his visual perception that he would otherwise not typically notice. Within a mirror’s frame (for they were mostly framed in those days), the scene is brought to the viewer’s focused attention. Like the tube Aristotle suggested for seeing “farther,” and the lensless telescope used by Tycho Brahe, a mirror could be used to concentrate attention on a small scene by blocking out what is around it. When we look at a scene directly, for example, we do not notice that it appears in perspective, that is, with the lines converging to a point in the horizon. But when we view that same scene in a mirror we tend to notice its perspective more. This is partly because of the mirror’s flattening effect, but also because the use of an optical device disrupts our visual habits and forces us to look at the scene in a new way. “Since you can see that the mirror,” Leonardo explained, “by means of outlines, shadow, and lights, makes objects appear in relief, you, who have in your colours far stronger lights and shades than those in a mirror, can certainly … make your picture look like a natural scene reflected in a large mirror.” The mirror cast nature into a new light for artists, by showing nature reversed, framed, and with its visual characteristics intensified.
Leonardo also noted that by comparing a picture to a mirror image of the same subject, the artist could judge whether he had successfully painted the subject. “When you want to see if your painting corresponds throughout with the objects you have drawn from nature,” he advised, “take a mirror and look in that at the reflection of the real things, and compare the reflected image with your picture.” Earlier, Alberti had called the mirror the iudex optimus, the optimal judge, of paintings, because it intensifies the picture’s properties. By the seventeenth century, a mirror was a common part of the painter’s toolkit.
The supposition that Vermeer used a mirror to depict foreshortening is strengthened by the realization that the figure to the left—Vermeer himself, we think—looks like nothing so much as an imperfect mirror reflection, with its forms flattened, and its outlines slightly blurred. If Vermeer was using the common method of painting a self-portrait from a mirror reflection, it would have been a simple matter to turn the mirror to the model of the prostitute and see how the fingers of the outstretched hand were reflected there as well. He could have learned this technique from Fabritius, whose self-portrait of 1648–50 also vividly resembles a mirror reflection—in its subject, pose, and glance as well as the way light is depicted. The use of a mirror is evidence of Vermeer’s increasing preoccupation with the way things actually appear to our vision, shown in this picture by his depiction of the green glass and the ceramic jug, the ribbon and feather on the customer’s hat, the texture of the lace on the white scarf.
Besides the mirror, Vermeer also availed himself of another instrument in painting this picture. At the far right of the painting (in danger of being toppled over by the prostitute’s left hand) is an elaborately painted white-and-blue wine jug. It is clear that Vermeer used a pair of compasses—such as those used by surveyors—in order to achieve the incredible accuracy of the vessel’s contours and decorations; the point at which the compass pierced the canvas, and the scratches made to define the contours and designs on the jug, can still be seen in the paint layers. Even at this early stage of his career, Vermeer was showing a willingness to use devices—mirrors, compasses—that could help him achieve the effects he sought.
In this willingness to deploy aids in his painting process, Vermeer was taking part in an artistic tradition of long standing. From about the time that perspective theory was devised by Brunelleschi, various instruments were contrived to help artists instantiate these geometrical principles in their paintings and drawings.
-5-
The person who had first laid out the laws of perspective, Alberti, had invented a simple device called an intersection—also known as an Alberti veil—in order to help painters depict three-dimensional objects on two-dimensional paper or canvas. The Alberti veil was a sheer piece of fabric, loosely woven of fine thread, divided by thicker threads into parallel squares, stretched on a wooden frame. The artist looking at his scene through the sheer veil could transcribe the arrangement of forms onto a drawing or painting surface that had been divided into corresponding squares. Alberti described this velo in his discussion of perspective in Della pittura: “Nothing can be found, so I think, which is more useful than that veil which amon
g my friends I call an intersection.… This veil I place between the eye and the thing seen, so the visual pyramid penetrates through the thinness of the veil.”
In his widely translated and distributed treatise Instructions for Measuring (1525), Albrecht Dürer described a version of the Alberti veil that could be stretched over a glass pane placed perpendicular to a table; the artist would sit at the table, look through the glass at the object being painted, and draw the image onto a sheet of paper before him. Numerous depictions and descriptions of the Alberti veil would later appear in works by Italian writers on perspective theory, as well as in writings by the seventeenth-century Dutch mathematician and engineer Samuel Marolois, whose book on perspective, originally published in 1614, was probably the most widely read volume on perspective theory by a Dutch writer.
A related method was soon developed as well: by inserting a pin, with an attached string, through the canvas at the desired horizon point, the artist could use the string to show him where the orthogonals—the straight lines that meet in the vanishing point—should go. The painter could sketch the lines in chalk (which would dissolve under the oil paint) or could even chalk up the strings, pull them tight, and let go so that the chalk would be transferred onto the canvas. This method was commonly employed in the seventeenth century in Delft by the architectural painters Houckgeest and De Witte, as well as by Fabritius, Vermeer, and others. Tiny pinholes at the vantage points have been found, on x-ray analysis, in some paintings, including a number of Vermeer’s canvases.
More elaborate devices were also described in books on perspective during the sixteenth and seventeenth centuries. Undoubtedly some of these were fantastical in their conception and were never built, intended more as demonstrations of the author’s imaginative prowess than as useful tools for artists. Some inventions were basically mechanized versions of the Alberti veil, such as one described by the architect Jacopo Barozzi da Vignola in his Le due regole della prospettiva pratica (Two rules for practical perspective) of 1583. This large-scale machine used a rope-and-pulley system to move an eyepiece up and down and side to side along a vertical and horizontal ruler, so that the artist could first measure, then transcribe, an image bit by bit onto a gridded surface, perhaps with the help of an assistant to whom he would dictate instructions.
Not long after Vignola’s book was published, an innovation was introduced by the painter and natural philosopher Lodovico Cigoli. Cigoli was a friend of Galileo’s and such a supporter that, even before the ink had fully dried on the pages of Galileo’s Sidereus nuncius, Cigoli had painted a ceiling in the church of Santa Maria Maggiore in Rome showing Mary standing on a pockmarked Galilean moon rather than the conventional smooth Aristotelian crescent. At the same time that he painted the church’s ceiling, Cigoli was conducting his own study of sunspots from the church’s roof. His perspective machine mechanized both the viewing of a scene and the recording of it.
Using this device required some coordination and effort, however. With his left hand, the draftsman would twist a pulley clockwise or counterclockwise to make a vertical pole move left or right over a flat base covered by paper. With his right hand, he would move a pencil connected to a sighting bead. With the motion of the pencil, the bead is moved up and down the shaft of the vertical pole as the pole is moved back and forth by the twisting motion of the left hand. The sighting bead can in this way trace the outlines of the object. At the same time the pencil is recording the resulting configuration directly onto the paper.
Unlike Vignola’s device, which would have been the height of a person and nearly as wide, Cigoli’s was a tabletop model and fairly portable. We know that Cigoli’s machine was produced in his time. One was in the collection of Louis Hesselin, a counselor to Louis XIV of France. The device was still being produced in the midnineteenth century. But its use seems to have been mainly as a kind of “perspectival robotics” that demonstrated how perspective drawings could copy reality, rather than a practical instrument used by working artists.
Another natural philosopher, the astronomer Christoph Scheiner, invented the “pantograph,” which was designed for copying designs or maps, or for enlarging or reducing drawings. Four rods hinged together to form a parallelogram connected two pens in such a way that the movement of one pen, drawing or tracing an image, produced an identical motion in a second pen. These instruments were, essentially, copying machines. But mounted vertically, half on a canvas or paper-covered board and half on an open frame, a pantograph could be used as a device to draw an object in proper perspective. Looking at the object through the frame, the artist would hold the pen end and move it so that the stylus attached to the hinged rod would “trace” the outline of what he sees. This was difficult to do because, in effect, he would need to “predict” the path of the tracing stylus as he drew. A complex version of the pantograph, using a system of pulleys and counterweights, was devised in 1669 by Christopher Wren, the English natural philosopher and architect. When Wren demonstrated his device to the Royal Society, he introduced it as an “instrument for drawing in perspective.”
Artists had various instruments available to help them depict perspective properly. But at the same time, new studies of vision began to expose cracks in the edifice of classical perspective theory. And artists—especially in Delft—were paying attention to this development. Eventually, these fissures would lead Delft’s artists to experiment with different kinds of devices—optical instruments—to help them achieve the effects they sought.
-6-
Johannes Kepler, the mathematician, astronomer, and astrologer (those last two being noncontradictory in his time), had shown in his Ad Vitellionem paralipomena (1604) and in his Dioptrice (1611) that the actual character of the physiological nature of vision was not reconcilable with the geometrical assumptions about optics made by Euclid or with the anatomical claims made by Galen and his followers. Kepler, following discoveries described by the anatomist Felix Platter in his De corporis humani structura et usu (1583), claimed that the crystalline substance of the eye was not the seat of vision but was, rather, a lens that magnified images before they passed to the retina. Platter had shown that the crystalline humor is not connected to the optic nerve, as would be necessary on the view that it was the humor that transmitted its perceptions of an object through the nerves running to the brain. Instead, it was the retina that was connected to the optic nerve (more precisely, as we know today, the optic nerve begins in the cerebrum and culminates in the retina). Kepler also disagreed with the earlier writers on vision, who claimed that the crystalline humor had a flat posterior surface; this hypothesis was necessary in order to ensure that the rays from the visible object would not intersect at the center of the eye, which would result in an inverted visual image—something these thinkers wished to avoid, because it seemed so counter to our experience (we do not see everything upside down). Kepler, however, fearlessly pointed out that the posterior surface of the crystalline humor is rounded—and that the image projected on the retina is, contrary to common sense, inverted.
Kepler also showed that the geometrical assumptions of the earlier theorists were flawed. On the view of the medieval writers, since every point within the eye receives rays emanating from every point in the visual field, there will be pure visual confusion unless some of these rays are ignored. (Seeing everything equally, the eye would see nothing.) This confusion was avoided by the claim that only the rays that are not refracted—that is, the perpendicular rays—are responsible for vision. Kepler pointed out the absurdity in claiming that only perpendicular rays are responsible for vision. What about rays that are only slightly oblique? While it could make sense to say that rays that are nearly perpendicular do not act on sight as strongly as perpendicular rays, it is absurd to say that the nearly perpendicular rays cannot stimulate the visual power at all. Kepler was then faced with the problem of explaining how an infinite number of rays from each point in the visual field could be drawn into a coherent, point-to-point corresp
ondence in the eye. Against the medieval tradition, Kepler argued that the crystalline lens refocused intromitted rays upon the retina, where vision was thus made possible. Significantly, Kepler called this retinal image a pictura. In this way he drew a distinction between visual pictures (projected onto the retina, or onto walls or screens) and visual images (produced by the mind, and in the mind, out of the information received through the nerves from the senses).
One consequence of Kepler’s new theory of vision was that the single “pyramid of vision” of the medievalists, and of classical perspective theory, was split into numerous cones or pyramids being refracted onto the retina from various points of the visual field; he used the term pencilli, or “pencils,” for these rays—a word that was used by artists of the time to refer to their brushes. It became unclear how the actual way we see could be made consistent with how the classical theories of perspective said artists should depict what we see. No perspective theorist of the time really attempted to come to grips with Kepler’s optical theory. Indeed, even several years after Kepler published his new theory of vision, the Dutch theorist Samuel Marolois explicitly abandoned efforts to accommodate perspective theory to the up-to-date optics, reverting to older optical theories in his perspective treatise of 1614. Marolois emphasized that understanding the nature of vision was irrelevant to his concerns, and he exhorted artists to follow established laws of perspective—and explicitly told them they should not paint according to their visual impressions. However, this clashed with the sensibility of mid-seventeenth-century Dutch painters, who were resolved to record exactly these visual impressions. A kind of uneasy uncertainty about the relation between vision and perspective theory led Dutch painters of the time to feel that they were free to experiment with perspective rules.
Eye of the Beholder: Johannes Vermeer, Antoni van Leeuwenhoek, and the Reinvention of Seeing Page 10
