Trompe l'Oeil painting class

I just figured out what I'll do for my Trompe l'Oeil painting class: The vase above, which I photographed outside the Vanderbilt mansion in Rhode Island. There's so much great detail going on within the vase itself that it's a perfect lesson in all the various forms of ornament, and grisaille painting. This photograph reminds me of the panel by De Lairesse which you've seen before on this blog. The balance of a large value range within the space of the vase, against the smaller value range of the rest of the niche is reminiscent of his trompe l'oeil masterpiece.

It could be combined with an Italianate landscape in the background as in the photograph below. Or maybe just painted as it is, in shades of grey.

I did something similar for the Kips Bay Show house in New York City in 2003, but I've always wanted to do a taller panel, with less stuff going on around the vase. I hadn't yet studied the De Lairesse piece and learned its lessons about value. Everything I painted in the panel below was treated with a similar range of values, and the result is a flattened painting with little sense of depth.

Kips Bay Show House mural, Alan Carroll 2003

Giovanni Battista Zelotti, Villa Godi

click on the image for much larger view

It's a little unclear whether Andrea Palladio enjoyed seeing the interiors of his buildings decorated with paint, but who cares?! This is awesome! Just look at the cast shadows thrown from the beaded molding around the door frame. Absolutely spectacular examples of wall painting.

The only thing that'd really rub old Palladio the wrong way would be the complete disregard for the architecture of the room. Zelotti's painted opening pays no attention whatsoever to the position of the actual door to the room. The trompe l'oeil columns just disappear into the door frame. It's as if Zelotti had a better idea as to where the door should have been placed than Palladio himself!

Giovanni Battista Zelotti added to the already overflowing decorations started by Gualtiero dell' Arzere in Villa Godi (Palladio's first villa), constructed in 1537. His trompe l'oeil in this, the large central salon, is more fluid and monumental than Gualtiero's, and takes on a popular theme of the time: the depiction of a human figure (presumed to be Palladio himself) in an opening or doorway.

These images are from Volume II of Antiques in Italian Interiors, which came out in 2010 from Verbavolant books. Both volumes are an incredible treasure trove of gorgeous photographs by Mario Ciampi, of interest to anyone curious to see the painted decoration of Italian villas. Highly recommended.

Natural History Museum dioramas, New York City

Source: Research Library, American Museum of Natural History

James Perry Wilson (above) painting the background for Western Pine Forest Group, Forestry Hall

The painted dioramas in The Natural History Museum in New York City are some of the most outstanding examples of mural painting that I have ever seen. There is a palpable sense of place in these dioramas, such that the viewer feels instantly transported. The feeling is not unlike vertigo. Though the hand of the artist is distant, the sense of joy and awe they clearly felt is there in abundance.

James Perry Wilson was one of a group of incredibly talented artists responsible for painting the dioramas. Yale Peabody Museum (for whom he also painted dioramas) presents an excellent and in depth study. The author says of him that "it is worth noting one of the differences between Wilson's plein air painting and his diorama work.  Accuracy was possibly more pronounced in Wilson's diorama work than in his plein air studies, but in both, he tried to paint himself out of the painting to allow the viewer to focus on the illusion of reality and not the artifice of its creation.  Wilson sought to make the artist transparent."

In fact, it takes close inspection to tell that there are different hands involved in the production of the paintings. Yet somehow they defy blandness of expression, and maintain a freshness that is still present many years after they were produced, despite aging and the damaging effects of ultra-violet light.

"Wilson's paintings have a clarity and lack of muddiness that would have been impossible to achieve had he fussed over mistakes, over-painted, or made numerous changes."

The article goes on to describe the fascinating characters and working conditions at the Natural History Museum in the '30s:

"In 1934 when Wilson started at the AMNH, many of the major figures producing dioramas were already at work: Francis Lee Jaques, William R. Leigh, Robert Rockwell, James  L. Clark.  Belmore Brown, Carl Rungius, and Hanson Puthuff would come a few years later.  Even though Carl Akeley had died eight years before, his influence was still strongly felt.  All of these men were not only extraordinary artists, but they were also outdoorsmen, hunters, and explorers.  They walked with a swagger, they had strong personalities and big egos.  Conrad Schwiering described working as an apprentice with these diorama painters in the summer of 1941:

Imagine my sitting there on a ladder in the middle of a wildlife setting listening to [Charles] Chapman and William R. Leigh debate points of perspective, with Belmore Brown finally coming over from where he had been working to tell them they were both nuts-then Brown giving his own views on the matter in eloquent style...."

The effect of staring into Wilson's sunset is almost blinding. It's that convincing. Largely self-taught as an artist, his immense painterly skill is matched by his technical mastery of perspective and composition. James Gurney, the Dinotopia creator, writes on his great blog Gurney Journey of Wilson's working method that;

"The full backdrop was often 35 feet wide or more, with a dome-shaped ceiling. This required large amounts of paint and careful mixtures. To make the perspective accurate across such a oddly curving surface, Wilson worked out a unique grid system that he called “the unsquare square” to compensate the foreshortening of the side sections. Every backdrop was carefully planned using maquettes, color sketches, and full-size charcoals drawn on the backdrop before the final paint was applied."

These artists were dedicated professionals and all went to extraordinary lengths to create their work. Carl Akeley (along with William R. Leigh) famously made an expedition to the Congo in 1924 to research the landscape for the Gorilla diorama, and wrote a book detailing his experience, called In Brightest Africa, available for free download. They were dangerous and physically taxing journeys into the blankest sections of the map, pith helmets and pistols in hand. In fact, Carl Akeley died on the Eastman-Pomeroy expedition to Africa in 1926, underscoring the demanding nature of the work.

James L. Clark, the man responsible for building the diorama halls, had a number of close scrapes himself, on one occasion falling ten feet into an elephant trap and miraculously missing sharp, poisoned bamboo stakes on the way down. He and Akeley, in a fit of impetuousness driven by boredom, once decided to ignore warnings of "hostile natives with poisoned arrows" to climb Mount Elgon (14,176').

I posted a large set of photographs of the dioramas on my Flickr page.

Jules Dalou

I like Dalou (1838-1902) for a couple of reasons. Not least of which was that he was technically superb, but also because of his politics. Though he worked on commissions for incredibly wealthy clients, he sided with the working-class politically and actually got himself into a lot of trouble by being a little too close to the Paris Commune. He was made to flee to England, and was given a life sentence (later repealed) in absentia.

Jules Dalou studied under Carpeaux (creator of the incredible sculpture Ugolino and his Sons), and created a number of fantastic public works including this rather raunchy depiction of bacchanalia.

I love the illusion of space that he created in a relatively shallow field, especially in the foreground, and the kneeling figure on the left. Look how his leg appears to recede. This work follows on from a post I made regarding a trompe l'oeil masterpiece by Gerard de Lairesse, and another on the work of Christen Kobke, who created a couple of masterful illusions of depth in grisaille.

How to Draw a Volute, or Acanthus Scroll

I don't think there is a more beautiful line in the world than the curve described above. Zeising went so far as to claim for the Golden ratio that;

"... [It is] the universal law in which is contained the ground-principle of all formative striving for beauty and completeness in the realms of both nature and art, and which permeates, as a paramount spiritual ideal, all structures, forms and proportions, whether cosmic or individual, organic or inorganic, acoustic or optical; which finds its fullest realization, however, in the human form."

Well actually the nautilus shell is not a true Phi-spiral, but it's pretty close so we'll go with Zeising's effusiveness for now. This is Part IV in the series How to Draw the Acanthus, and deals with the topic of how to draw a volute, or acanthus scroll.

I don't mean to mislead with the opening image, but in this instance we are going to look specifically at the Ionic volute (not the curve described by the logarithmic spiral above). I have linked out in this article to other mathematical scrolls if you're interested in further reading regarding Fibonacci numbers, Golden Spiral, or logarithmic spirals.

But first, a little background. The following quote is from Elements of Geometrical Drawing, by John Henry Spooner (1901), describing the Ionic volute.

Erechtheum capital from Stuart and Revett’s The Antiquities of Athens (1790)

James Stuart and Nicholas Revett’s The Antiquities of Athens (1790) was the book that popularized the Ionic order of the Erechtheum through it's beautiful illustrations. Historian Calder Loth points out that "The Erechtheum’s volutes are the most complex of all Greek Ionic capitals, having a series of tightly spaced creases or canals resembling fine drapery. The central swag in the canals, a characteristic feature of Greek Ionic capitals, emphasizes the drapery-like character." 

Corinthian Acanthus Scroll

The volute made it's way into the Corinthian order too. A Textbook on Ornamental Design (1901) introduces the Choragic monument of Lysicrates in Athens as "the principal structure of Greek origin designed in the Corinthian order", and describes the section of Acanthus scroll pictured above as showing...

"... not only the elaboration of ornament characteristic of this order, but also a principle of ornamental design that is of the utmost importance in its relation to the lack of invention of new forms and the restraining influences of certain art periods. It will be observed here that the various parts of the acanthus scroll grow out of one another in a continuous line. This is a principle of design that originated with the Greeks, and was continued by the Romans."

So we're beginning to see the importance of the scroll design, whether as described by the natural roll of a fern or shell, as evidenced in Greek, Roman or Islamic architecture, or as a mathematical design principle adopted by the likes of Le Corbusier. It seems we should probably learn how to draw one of these things then, right?

There is ample historic precedent for drawing the volute, but oddly the information seems to vary. A number of different methods are described in The New Metal Worker Pattern Book, by George Watson (1901). Spooner maintains that the ratio for describing it's curve is 7 : 6 : 5, and claims that the Greeks would agree. Another source follows a ratio of 6.5 : 5.5 : 4.5. [Though I am strictly a layman, these ratios seem like approximations to me; 7 : 6 is surely not the same as 6 : 5, nor is 6.5 : 5.5 as 5.5 : 4.5 (but please feel free to enlighten me). A more accurate Ionic ratio would be 7 : 6 : 5.143, which is what I will use here].

The point is that although the ratios may change, our method for constructing the volute remains the same. We could substitute 7 : 5.5 : 4.321 (which would describe a different curve) just as easily, but for now we will concentrate on drawing the distinctive ram's horn shape of the scrolls of an Ionic capital as described by Spooner's ratio.

Here are the steps you need to follow:

Fig. 1

Step 1: I decided arbitrarily to start with a rectangle 3.5cm wide and 6cm tall [Fig. 1], and to start the convolutions of the curve at A, 3.5cm from the base of the volute.

Fig. 2

Step 2: So, according to the traditional ratio of 7 : 6 : 5.143, we divide the length of AB by 1.1666666666 to give us the length of BC. [In the case of the ratio 6.5 : 5.5 : 4.5, we would divide by 1.1818181818].

Continuing in this manner, we determine the length of CD [Fig. 2], by dividing BC by 1.1666666666. This gives a length for CD of 2.57cm .

Fig. 3

Step 3: Next, draw lines through the points ABCD at 45º [Fig. 3] to where they intersect to form a rectangle in the middle.

Fig. 4

Step 4: See that little rectangle that you just made? You'll have to find the midpoints of each of those sides. Let's call them 1,2,3, and 4. Now draw lines perpendicularly from those midpoints to intersect AB, BC, and CD [Fig.4].

Fig. 5

Step 5: Using 1 as the center of the first arc, describe the curve connecting the top of the volute to AB [Fig.5]. Similarly; using 2 as the center of the second arc, we describe the curve connecting AB to BC, use 3 as the center to describe the curve from BC to CD, and 4 as the center to describe the curve from CD back around for the first full rotation of the spiral 

Fig. 6

Step 6: Let's now draw the second, inner rectangle EFGH which we will use to draw the second revolution of the spiral [Fig. 6].

According to our 7 : 6 ratio, DE = 2.2cm [CD ÷ 1.16666666].
EF = DE ÷ 1.16666666 = 1.9cm.
FG = EF ÷ 1.16666666 = 1.6cm.
GH = FG ÷ 1.16666666 = 1.37cm

Fig. 7

Step 7: As in Step 3, draw lines from points E,F,G,H, at 45º to where they intersect in the middle to form a rectangle [Fig. 7].

Fig. 8

Step 8: Draw lines perpendicularly from points 5,6,7 and 8 to connect with DE, EF, FG, and GH [Fig. 8].

Fig. 9

Step 9: Connect the dots (as in Step 5), using points 5,6, and 7 as the centers of our arcs [Fig. 9].

Fig. 10

Step 10: If you wished, you could repeat Step 6 through 9, to give you a third inner curve to the spiral, but I'm going to stop here. From the last point (8), instead of describing a short arc, we continue around and draw a full circle [Fig. 10], which is the central circle of our newly completed volute. So there you have it.

Fig. 11

Step 11: I could have just stopped at Step 10, but I'm going to add two more steps. I want to extend and soften the curve at the top (or the "thread", to borrow a term from a stair-makers lexicon). This will make it look more like the acanthus scroll. In order to do that, take the line that passed through Point 6 and Point 1 [Fig. 11] and extend it out to the left. Here's where a little creativity comes in. Draw a line at 45º that intersects with your horizontal line at point a'. You can position that line wherever you want.

Now draw an arc centered on Point a', that goes from Point 1 and stops where it intersects the 45º line.

Fig. 12

Step 12: Now we're going to finish it off. Using a point (b') further down the 45º line as a new center point for the last section of the arc, connect Point 2 and Point 3 [Fig. 12].

Here is the completed scroll without all the construction lines, and with one of the beautiful ornament drawings of the acanthus scroll from a previous blog post superimposed for comparison:

Ornament Drawings

The best drawing I have found of the structure of the acanthus scroll

These incredible details are from a huge (13 3/4" x 9") three volume set of prints engraved in 1872 by C. Claesen entitled Spécimens de la décoration et de l'ornementation au XIXe siècle. The amount of work involved In the drawings alone is staggering, let alone the time spent carving them out of stone or wood, casting them in iron, gilding them, painting them on panels, or sculpting them in clay and casting them out of plaster. These guys must have been just churning this stuff out. It's no wonder they got a little weird every now and then, like: What exactly are these two monkey people up to?

The quality of the draughtsmanship in every engraving is just superb. I'm giving you a close up view here, hoping that you'll enjoy these as much as I do. They are exquisite examples of how to draw ornament.

I posted a set of 187 images from the book here, on my Flickr page. Don't forget to click the largest size before you download.

How to Paint Trompe l'Oeil

Wouldn't we all like to be able to paint like this guy?

Gerard De Lairesse's trompe l'oeil grisaille masterpiece is an excellent lesson in how to paint trompe l'oeil. Taking a closer look at Allegory of the Sciences, we can discern his technique for creating a successful illusion of depth - particularly with reference to the value range he employed.

When I speak of 'value' here, I'm referring to the Munsell color system. I personally use the term interchangeably with 'tone' or 'tonal value', although that's probably not correct.

I will never forget speaking with my friend Dermot Power (concept artist for films such as Star Wars, Harry Potter, Batman Begins, Charlie and the Chocolate Factory and Alice In Wonderland), who told me that you have to be really careful with highlights: The sign of a novice is the tendency to touch everything with a dot of white on the highest point. "It looks like it's snowing" he would say. That really stuck with me.

Examining De Lairesse's painting, we can see that he reserved his brightest highlights and darkest lowlights for a section of the painting occupying no more than 30% of the overall composition, inside the lower right section above. I'm calling this his 'High Contrast' section.

The rest of the canvas is roughly broken in half by two areas of 'Medium Contrast' and Low Contrast'. It is important to recognize the sense of balance achieved by this: the painting neither jumps out at us, nor recedes too much. In my opinion, a general rule of thumb would be to divide the 'value sections' of your canvas into percentage divisions of 30/40/40, as above. If every area within the painting was given the same value range the trompe l'oeil effect would be flattened and unconvincing.

'High Contrast' sample, with tonal range

'Medium Contrast' sample, with tonal range

'Low Contrast' sample, with tonal range

Let's examine this further: Sampling from the High, Medium and Low Contrast areas above and isolating the value of the tonal ranges, we see that the ranges are represented within the graph below.

When plotted against the fullest possible value range (solid black through white), we notice some interesting things. To begin with; even the area that contains the darkest lowlights and lightest highlights (lower right section of the painting with the seated figure), doesn't come close to representing the full range of tones available. In other words, the outside boundaries of the value range De Lairesse employs represents no more than 75% grey at its darkest, and 25% grey at its lightest.

Of course, this is borne out through our own observations of nature: Shadows are never 100% black, nor are highlights 100% white. That last point is worth restating, because I have seen it done incorrectly many times: Highlights in trompe l'oeil painting are never 100% white. What? Oh alright, not "never", but let's just say they should be used extremely sparingly. Unless you're rendering a high gloss sheen on a white marble statue, pure white highlights should be a last resort.

The value of pure white (Value 10 according to Munsell) and the value of pure black (value 0) are actually unattainable with pigment, being theoretical limits.

The following illustration of spheres receding in space uses the same value range De Lairesse employs to differentiate between foreground and background objects. You can plainly see that the optical laws of chiaroscuro apply: If you want an object to appear to recede in space, limit the values to the middle of the spectrum.

Note that the whitest highlight in the foreground sphere is still not pure white

The point gets a little confused here due to the fact that we are dealing with a grisaille representation of carved statuary, not simply a painting of objects in space, as say, a trompe l'oeil landscape. Our objective is to convey a sense of how an object is carved in stone. So not only do we flatten the value range to represent how an object recedes from us, we also flatten the it when representing objects in low relief, as in the detail of the frieze, below.

I posted these before, but they are such perfect examples of trompe l'oeil low-relief cameos that I am adding them here again. They are by the artist Christen Kobke, and demonstrate precisely the technique of expanding the value range to imply high relief, and reducing the value range to imply low relief. And yes; these are paintings!

Note the reduced range of tonal values in the low relief of the wing

Generally speaking though (when painting trompe l'oeil molding for example), there is a simple rule of thumb that we should adhere to when dealing with middle distance objects, of medium relief: That is the rule of thirds. Roughly speaking, a sphere should be divided into thirds between highlight, mid-tone and shadow, as below.
Not only that, but an accent tone, light or dark, should be one third the width of the half-tone, as below: