Obscura Knowledge

Part One: A use for a magnifying glass (neither close looking nor fire starting)

With some gentle encouragement from my college’s Associate Dean, I agreed to take part in a public discussion about David Hockney’s book Secret Knowledge. It would be a debate format. On the pro side, a colleague from the School of Architecture would describe the merits of Hockney’s ideas; then I, the art historian, would roll up my sleeves and systematically demolish them.

Such was the initial idea. But since I’d previously only dipped into Hockney’s book, I was a little reluctant to be cast as the spoiler. And upon making my way through the text, I actually found myself feeling fairly sympathetic to Hockney’s enterprise if not–as will become clear in the second part of this post–entirely convinced by his arguments.

Part of this sympathy I attribute to my being a fellow believer in the mesmerizing power of the projected image. (Surely art historians can agree on this: the primitive attraction of the darkened room with, at its center, a strangely magnetic cone of light?) I share Hockney’s fascination with the camera obscura, and can trace the origins of  my interest to the one in Edinburgh’s old town, which I visited in the summer of 1992. It’s the strangest combination of a tawdry tourist spot (holograms anyone?) and, once you’ve climbed the turret and finally entered the camera itself, pure magic. There, projected downwards onto a white disk, is a luminescent slice of Edinburgh. The attendant slowly rotates the periscope of a lens and takes you on a 360-degree tour of the oddly flattened and rounded city.

Camera Obscura, Edinburgh (exterior and interior)

Although I should have expected it, the fact that the image of the city was truly alive came as a genuine surprise. The motion was captivating. Streets and buildings shifted as the light fell around them and then retreated into shadows; twisting wreaths of smoke rose up to the clouds, catching the eye as they ascended; cars and pedestrians flitted back and forth. I felt instantly converted to this relatively simple form of image production, even though I’d grown up surrounded by superficially more impressive technology: television, videos, films, computer games and, yes, holograms.

While I was reading Hockney’s book, and thinking about the appealing simplicity of the camera obscura, I began to play around a little with my own seldom used magnifying glass. Catching the light from my office window, I found I could project a fairly good image onto a piece of white paper. Then one morning I persuaded Megan, my photographer wife, to help me repeat the experiment at the doors of the University’s chapel. (Romantic, I know!) Megan had been with me all those years ago in Edinburgh and together we had subsequently visited seaside camera obscuras in Santa Monica and San Francisco. As Hockney discovered, the bright Californian light is well suited to optical experimentation.

Camera Obscura (and Holograph Gallery!), San Francisco, 1949

Here are a couple of the images Megan shot that morning. One shows the view of the student union from the chapel doors; the other shows the image of the same view projected back onto a sheet of paper using my magnifying glass. As you can see, the projection is fainter, smaller, upside down and laterally reversed.

Part Two: Hockney’s History

I was attracted to the chapel on campus for a couple of reasons. The view was pleasant and the doorway offered the necessary contrast between a dark interior and a bright exterior. But I was also looking for a scene that approximated the subject of one of art history’s great lost works. Around 1415 (the exact date is uncertain), Brunelleschi made two demonstration panels, one of which depicted the baptistery in Florence, as seen from the west doors of the cathedral. Hockney brings the subject up in his book. “Brunelleschi demonstrated perspective”, he writes, “by painting a small panel (half a braccia square). To paint this, he stationed himself just inside (some three braccia inside) the central portal of Santa Maria del Fiore, in short, in a dark room looking out to the light” (286). It’s a view that the photographer Abelardo Morell has replicated using the camera obscura of his “tent camera.”

According to his biographer, Brunelleschi’s demonstration piece used materials with reflective properties (a mirror and polished silver) but, as an aside, Hockney entertains the tantalizing possibility that his knowledge of a “mirror-lens” might also have aided him in his rediscovery of linear perspective. “Did Brunelleschi”, Hockney asks, “devise the rules of perspective to make the picture bigger than those the mirror-lens could produce?”

The central point of Hockney’s book is now well known: using a set-up not unlike the one Megan and I orchestrated at the chapel, it’s not too difficult to “fix” a scene by tracing elements of the image onto the paper. This type of procedure, he proposes, led to the emergence of a new “optical” style of painting in the fifteenth century. Although he seems less excited by it, Hockney occasionally falls back on a weaker version of his argument, which may actually have greater explanatory power. This proposes that, having seen projected images and/or art made using such projections, artists could then represent the world in a more “optical” way. Thus even if an artist didn’t actually use a camera obscura in the production of a particular work, their art could still partake in this broader style. Let’s call these, respectively, the strong and the weak versions of Hockney’s argument.

Hockney is making an argument about artistic change. How can we explain, he asks, the radical change in style that occurred in European art around 1430? He believes that this quantum leap can only be accounted for by the fact that many artists used optical devices (the camera obscura in its various manifestations and in combination with lenses and mirrors). But art history is surely a lot messier than this and at least two other major changes occurred at the same time: the development and use of single-point linear perspective, and the sudden increase in the popularity of the oil medium. Single-point perspective is found in Florentine relief sculpture and frescoes of the 1420s, and although oil painting had been around for a long time, it only became commonplace in Netherlandish painting in the 1420s.

By comparison, the chronology of the camera obscura (in its various manifestations) is much harder to establish, although its basic optical principles were known to Aristotle. But lenses complicate things. The image in a camera obscura can be greatly enhanced using either a concave or a biconvex lens (as in my magnifying glass), and according to Vincent Ilardi, spectacles were invented “in Tuscany between 1280 and 1295″. Convex lenses, that is to say, had been available for many decades before Brunelleschi’s experiments. By comparison, spectacles with concave lenses probably weren’t manufactured in Florence until the “middle of the fifteenth century.” In short, it’s very hard to know (1) what kind of lens-based camera obscuras would have been available to artists, (2) when they would have become available, and (3) what quality of image they might have produced.

At any rate, the camera obscura’s possible impact surely shouldn’t be considered alone, but in combination with these other factors, as well as others we might also mention (regional styles and patronage patterns, for example). The importance of oil painting, in particular, emerges as an oddly underdeveloped theme in Secret Knowledge. Mediums, after all, have their own particular optical qualities, and oil paint is particularly well suited to depicting (and reenacting) light’s complicated interface with the material world. As artists became increasingly aware of the possibilities of oil paint, they would surely have paid more attend to the type of phenomena that might be successfully depicted using it. Consider the kind of eye-catching effects seen in Van Eyck’s work: the glint of jewels and metals, the infinitely subtle way light interacts with skin and fabrics, and even the image lying on the sheen of a mirror’s surface.

Hockney’s visual argument is made primarily through his bravura use of comparisons. For him, these reveal a shift from an earlier “eye-balled” style of painting to a later, more “optical” style. But consider the following table, which is based on the same comparisons (please click on the table for a larger view):

It is particularly noticeable that, whatever other factors might help to explain Hockney’s shift, oil paints are a common denominator behind all the works identified with the new “optical” style.

Finally, I want to consider another type of painting that emerged during the fifteenth century, the self-portrait. Typically, making a self-portrait has involved the use of a relatively simple optical device, the mirror. But it’s hard to grasp how an artist would be able to paint a self-portrait using a camera obscura. How could one pose outside of the camera while also drawing or painting inside of it? The self-portrait therefore seems to provide us with a kind of limit-case, indicating what it was possible to paint without the assistance of a lens or camera obscura. In the Netherlands, where Hockney finds most of his early examples of “optical” painting, and where oil painting first became popular, there are also some early examples of self portraiture. Notably, Van Eyck’s Man in a Red Turban (1433)  is often thought to be a self-portrait. But Charles Falco, one of Hockney’s collaborators and interlocutors, writes that “if you agree that the realism of this portrait is the result of a lens, the implication is that this cannot be a self-portrait of van Eyck” (269). Hockney readily agrees with this conclusion. His commitment to the strong version of his argument compels him to do so.

Jan van Eyck, Portrait of a Man in a Red Turban, oil on panel, 1433

So what about a work that is undeniably a self-portrait, like some of the remarkable images made by Albrecht Dürer? His self-portrait of 1500 would, at first glance, seem to fit all the qualities of Hockney’s “optical” style, but there’s little doubt that is was made using a reflected image, not a projected one. A work like this–an oil painting like this–makes me believe that humans can achieve astonishing feats of veristic painting without recourse to any technologies beyond the obvious ones: a mirror, a panel support, brushes, oil paints, and thinner. In the terms of the two versions of Hockney’s argument, Dürer’s self-portrait shows just what an artist can do without using image-projecting technology. It’s as though the weak version of the argument suddenly threatens to swallow up the strong.

Durer, Self-Portrait, 1500, oil on panel

Hockney does make a convincing case that at least some fifteenth- and sixteenth-century artists must must have used camera obscuras in order to produce their works. Thanks to Secret Knowledge it will certainly be hard for future art historians to avoid considering this possibility. Now almost ten years old, the book still reads as a welcome wake-up call, a reminder that art historians should be attempting to reconstruct studio practices as thoroughly as possible. We might, so to speak, benefit from dusting off our magnifying glasses every now and then, not just to read texts but in order to make images. And for followers of contemporary art, the book is also fascinating as a self-portrait of Hockney himself–of his energy, ambition, sociability, and occasional impatience. It’s a self-portrait of the artist fixed in an unlikely place, in the camera obscura of his own historical imagination.

6 Responses to “Obscura Knowledge”

  1. H Niyazi says:

    Great piece Ben!

    It’s interesting you haven’t mentioned Vermeer yet – saving that for part 2?!!

    Have you read Vermeer’s Camera? http://www.vermeerscamera.co.uk/bookhome.htm

    It’s quite convincing IMO. I also have a feeling some of Caravaggio’s work used some similar type of design aid – but saying that out loud in some circles is quite contentious!

    Kind Regards

  2. Benjamin Harvey says:


    Part two is already part of this blog… Would love to chime in on Vermeer, but don’t have time to catch up on all that bibliography! Isn’t it generally accepted that later artists like Vermeer and Canaletto were using camera obscuras–at the very least as ways of pre-visualizing their final paintings? I found his arguments about Ingres and the camera lucida fairly persuasive, too. Wasn’t so convinced about Caravaggio.
    One thing I didn’t include, and found completely fascinating, was an image of Joshua Reynold’s camera obscura. It’s disguised (poorly) as a book! & Reynold’s speaks fairly negatively about the CO in his Discourses–perhaps that can be the subject of another post.

    See: http://tinyurl.com/36andex and http://tinyurl.com/2vrez36


  3. H Niyazi says:

    Cheers for those pics – how fascinating! I found another article by Steadman on Vermeer at the BBC, very interesting stuff



  4. tents says:

    I found his arguments about Ingres and the camera lucida fairly persuasive, too. Wasn’t so convinced about Caravaggio.

  5. Larissa says:

    Hello, I am currently a senior in high school, and I am doing a project over perspective, the impact of optics and the photograph specifically. Do you have any advice for me on this? Suggested articles? It is my final measure to graduate from high school in white, which is a great accomplishment, and I am so close to achieving it. I would really appreciate this. Thank you.

  6. Benjamin Harvey says:

    Hi Larissa: a good place to start would be the work of Martin Kemp. Take a look at his book The Science of Art: Optical Themes in Western Art from Brunelleschi to Seurat. See: http://www.amazon.com/Science-Art-Optical-Western-Brunelleschi/dp/0300052413
    And good luck with your project!

Leave a Reply