Consider the source.

Canaletto did not use a camera obscura. People repeat that because they’re uncomfortable with the fact that they can’t draw.
Westminster Abbey with a procession of Knights of the Bath, 1749, Canaletto
It has long been held that Canaletto achieved the amazing accuracy in his vedute through the use of the camera obscura. This is not a modern thesis, although it is widely repeated as fact. It came down to us from Canaletto’s earliest biography, written in 1771, but it’s convenient for our modern sensibilities. After all, Canaletto’s landscapes are so perfect, they could not have been rendered from life—could they?
The Royal Collection Trust has released a report that seems to prove, conclusively, that this theory is wrong. While infrared technology is often used to examine what’s under the surface in oil paintings, it’s not commonly used on drawings. The Trust applied this technique to their collection of Canaletto’s works on paper. This is significant; they own a third of known Canaletto drawings.
They discovered the ruler edges, pencil markings and other traces of the drawing process under the finished surfaces. It was enough for the curators to state “categorically” that the stories of Canaletto’s use of the camera obscura were mythical.
Architectural Capriccio, drawing, Canaletto. “Capriccio” means it’s a fantasy landscape.
There is nothing inherently wrong with the camera obscura or any other mechanical aid to drawing. Nor was David Hockney revolutionizing the art world when he proposed that our ancestors used it. Leonardo da Vincidescribed its workings in 1502, and a similar pinhole drawing device was illustrated in Albrecht Dürer’s Four Books on Measurement. For Canaletto, born simultaneously with the Age of Reason, the temptation to try the camera obscura would have been overwhelming. But he would have quit for the same reason many mature artists stop working directly from photos:
The results are boring.
The Stonemason’s Yard, 1726–29, is considered Canaletto’s early masterpiece.
What is seen by the human eye, with its pronounced center pole, is so much more interesting than the flattened line of ‘real’ optics. That is why our photographs so often disappoint us, and why photography really is a lot more complicated that simply pointing a camera and firing away.
The popularity of the Hockney thesis lies in an uncomfortable fact: by and large, moderns don’t draw well. We haven’t put in the hours with ruler, pencil and paper. We rely on viewfinders, photographs, and other devices for our underpaintings. Rather than face up to that deficiency, it’s easier to imagine that drawing is impossible.
Of course, it’s not, and nobody can really paint until they master the elements of drawing. Too often, modern landscape painting is about fragment and impression. Is that because fragments are so interesting, or because we’ve given up on drawing?

The genius isn’t inside the camera

An artist drawing a seated man onto a plane of glass through a sight-vane, from the fourth book of Albrecht Dürer’s Four Books on Measurement.
Last month, a number of people sent me Vanity Fair’s pieceon engineer Tim Jenison’s painstakingly-complex recreation of Johannes Vermeer’s The Music Lesson. Jenison faithfully built the room and objects depicted in the painting, and then repainted the scene using a version of a camera obscura. In the end, he discovered what any freshman art history student could have told him: yes, it’s possible that Vermeer used a camera obscura. He’d have hardly been alone.
Sir Robert Hooke’s portable drawing machine should be familiar (in concept) to any of my plein air painting students. There is nothing new under the sun.
The peculiar properties of the pinhole camera were known and described in antiquity from Greece to China. In the west, the principles behind it were analyzed and described by the eleventh century Arab scientist, Ibn al-Haytham. In the 13th century, Roger Bacon described the use of a camera obscura to watch solar eclipses. Leonardo da Vinci puttered with one, and even came up with a proto-telescope based on it:
…in order to observe the nature of the planets, open the roof and bring the image of a single planet onto the base of a concave mirror. The image of the planet reflected by the base will show the surface of the planet much magnified.
Above, A man drawing a can, and below, A man drawing a recumbent woman, in foreshortening through a frame with a network of squares on to a paper also with squares, in order to be able to reduce or enlarge proportionally, both from the fourth book of Albrecht DĂźrer’s Four Books on Measurement.
Filippo Brunelleschi (1377-1446), sculptor, architect, artist and engineer, is generally (and perhaps falsely) credited with the invention of linear perspective in art. He also created a device to demonstrate that perspective, which we call Brunelleschi’s peepshow, but which is in fact a version of a camera obscura:
“[He] had made a hole in the panel on which there was this painting; … which hole was as small as a lentil on the painting side of the panel, and on the back it opened pyramidally, like a woman’s straw hat, to the size of a ducat or a little more. And he wished the eye to be placed at the back, where it was large, by whoever had it to see, with the one hand bringing it close to the eye, and with the other holding a mirror opposite, so that there the painting came to be reflected back; … which on being seen, … it seemed as if the real thing was seen: I have had the painting in my hand and have seen it many times in these days, so I can give testimony. (Antonio di Tuccio Manetti)
A Man Drawing a Lute, from the fourth book of Albrecht DĂźrer’s Four Books on Measurement. 
Albrecht DĂźrer was not a visionary in the manner of Leonardo, but he was peerless in investigating and recording his experiments. Several woodcuts of drawing aids come down to us from him, so we know he used them. Nevertheless, using a camera obscurawas not the only thing he could do; the man could draw brilliantly. Very few living artists today could duplicate his Young Hare or Large Piece of Turf, neither of which relied on optical tools.
Courtyard of the Former Castle in Innsbruck without Clouds, 1494, Albrecht Durer. Was it done with a drawing device? Perhaps. Does that make it less brilliant? I don’t think so.
Ultimately, it doesn’t matter whether Vermeer used an optical device. That’s the one thing Tim Jenison’s experiment proves. It’s ultimately nothing more than a fair copy of a masterpiece, such as art students churn out every day. Vermeer’s incalculable genius is safely his own.
Artists used the camera obscura until the development of photography made it obsolete. Here are four camera obscura drawings by Canaletto from the early 18th century.


Let me know if you’re interested in painting with me in Maine in 2014 or Rochester at any time. Click here for more information on my Maine workshops!