Monday, August 26, 2013

A Dado from the Alhambra's Hall of Justice, via Owen Jones and Eric Broug

This art project, on the doors of my garden shed, began with a pattern in Owen Jones' 1856 Grammar of Ornament.

Jones' Plate XLIII, No. 11, as shown in the online version of the Grammar Of Ornament (rotated 90°)
Jones notes that this pattern is a dado from the Hall of Justice in the Alhambra, Granada, Spain. I had to look this up: a dado is the lower half of a wall, below a dividing rib called the dado rail. This is a horizontal feature sticking out of the middle of a wall, something your grandmother might have called the "chair rail."

My project was to paint this pattern on a wall. I wasn't going to paint it on as a dado, however: I would cover the whole the wall of my garden shed with it, using the rotated version shown above.

Although we have wonderful computer software now that allows us to make fabulous patterns and to experiment with colouring them in different ways, and we have copiers and scanners to facilitate this, and indeed we have a cheap and bountiful supply of paper, markers, rulers and compasses, none of this equipment allows us to get a design anywhere other than on paper. What are the challenges involved with getting it onto a wall?

Construction method

If you look at this dado pattern, you will see that it is centred on 12-pointed stars, around which radiate 12 pairs of parallel lines. I had Eric Broug's book Islamic Geometric Patterns, so I looked through it to see if he reproduced something like this. As luck would have it, I found the same pattern, without the colouring.
The same pattern in Broug, Notice the hexagonal symmetry.
Broug gives an elegant method of construction for this pattern using only a compass and straightedge. In essence, one constructs hexagons within a circle, and then connects various intersection points. Once the host of construction lines are drawn, a few segments (only a few!) are selected to make the pattern.
Illustration from Broug showing the construction lines (faint) and the pattern (red). The heavier black line indicates the bounds of the basic hexagonal unit

Paper Test

Doing these things on paper is relatively easy. Following Broug's instructions, I made the construction lines and then inked in the pattern on a small sheet of paper. The radius of the master circle was 8.5cm.
Broug's design realized on paper with compass and straightedge. Construction lines are in pencil, with inked-in segments showing the actual pattern.
By tracing, I transferred this hexagonal cell to another sheet, and then traced in a couple of other copies at the edges. I experimented with colouring it.

Repeated pattern traced onto other paper and coloured with felt tip pen.

One of the advantages of doing this small-scale proof before going to the wall is that you become familiar with the various shapes that made up the pattern. There are the blue stars (A), and connecting them are blue or grey darts (B), pointed shapes with convex backs. Between the lines of darts we have petals (C), pointed shapes with concave backs, coloured in trios of orange or green. All other space, including the hexagons (D), the rays (E), and the winged darts (F), is yellow.

The basic repeating unit in both Jones' pattern and Broug's pattern is the hexagonal cell. However, in Jones' pattern the hexagons are packed in horizontal rows, whereas in Broug's they are packing in vertical rows.

Repetition in Jones' version uses vertical packing (black hexagons), whereas Broug's version uses horizontal packing (red hexagons).
So, to build Jones' pattern using Broug's construction method I'd have to rotate it 30°.

Paper master

My basic strategy here was to work on the wall using a stencil. The stencil would be made from a paper version that I would draw by hand.

In this full-size pattern, the master circle was 40 cm in radius. The hexagon within it was 35 cm from centre to mid-side, a measurement that would become important later.

I drew in horizontal and vertical axis lines to remind me how this master hexagon would need to be placed on the wall to create the vertical packing in Jones' pattern.  
Large pattern on paper: vertical axis in green, horizontal axis in blue.

To channel or not to channel

Before making a stencil, I had to make a decision about the white channels.

Are they there between the blocks of colour in Jones's reproduction? They are definitely there, although whoever drew the illustration for the Grammar of Ornament wasn't consistent about the width of these channels. Broug suggests that in Islamic design you generally do want channels separating polygons (as opposed to polygons touching one another), and that you strive to have channels of consistent width.

I chose a channel width of 16 mm, or 8 mm on either side of each pattern line. The 8 mm was not chosen because of an eye for design, or as the result of a mathematical calculation: it was simply that I had a clear plastic ruler with a second line 8 mm from the edge.


I transferred the vertices of the pattern lines from paper to foam core using a pin; then I drew between pin holes to reproduce the pattern on the foam core.  I marked channels out around the centrelines with the clear plastic ruler, and then cut out the stencil.

This stencil covered only part of the design, one quarter of the figure. Although in many ways it would be ideal to cut a stencil of the entire hexagonal cell, it would have taken a long time and required a bigger piece of foam core than I had. Practical considerations!

By clipping the channels at the outer hexagon and the horizontal and vertical axes running through the figure, I made a stencil that could abut itself and repeat in all directions.  It included three full rays, a dart, a petal and three points of the central star. The hexagon and the winged dart were there as partial edge figures.  I could flip it over and work with either side, and four repeats should cover an entire figure.

The one-quarter pattern stencil. Yellow lines indicate where stencil was trimmed to both the outer hexagon and the quartering lines.
I also cut a flexible stencil out of cardstock. This is useful where walls meet ceilings and floors. You can press this kind of stencil into corners.

Notations on the stencil

A stencil is a remarkable tool, and it can carry all sorts of extra information. Before I got to drawing on the wall, I added two kinds of notations to my stencil: matching zones, and alignment marks.

Matching zones were places where the stencil would abut against another repeat of itself. When drawing along the edge of the stencil, you can skip these zones.  I flagged them with felt tip pen on both sides of the stencil.

Alignment marks showed where I could expect the stencil to align with horizontal and vertical lines drawn through the centre of central star.

A's indicates matching zones; line B-B connects alignment marks

Picking Colours

At this point I was curious about the colours in the original pattern, not in Jones but in the Alhambra. I was suspicious about Jones' reproduction because all the patterns on that plate had the same four or five colours. As well, the image from his book online was somewhat different in colour from the Dorling Kindersley reprint of the Grammar of Ornament that I owned. I went searching on the web for pictures of the Hall of Justice at the Alhambra, pictures that might show the dado.

It turns out this room is known by several different names: the Hall of Justice, the Hall of Kings, Sala de Justicia, and Sala de Reyes. Detailed large images of the lower walls are not very common, because people are mostly photographing the ceiling. However on Wikimedia Commons I found a nice large high-res image showing the dado, taken by José Luiz Ribeiro in 2013 and released under a Creative Commons attrubution share-alike license.
Hall of Kings, Alhambra. © José Luiz Bernardes Ribeiro
Let's zoom in on the dado on the left side of the photograph and compare it with the image in Jones.
Photograph of the dado from the Alhambra (left) and Plate XLIII #11 from the present Dorling Kindersley edition of Jones's Grammar of Ornament (right)

Although the pattern is recognizably the one that Jones reproduced, the differences in colour are appalling. Where the original has black stars and darts, Jones's were blue. Where light blue darts ran across the pattern, Jones used a warm grey. Dark, jade-green petals in the original became a leaf green in Jones. The pale orange petals in the Alhambra were changed into rich orange petals. A white background in the original became gold.

And importantly, here are no channels in the original. There are simply tiles placed next to each other. (Although it's not clear what the material is: it could be semi-precious stones or it could be tiles. I haven't been there.)

So my wall painting was not going to be a reproduction of this dado in any sense. It would be a pattern inspired (distantly) by a dado in the Alhambra, recoloured and presented to the English-speaking world by Owen Jones in the nineteenth century, and then further altered by that idea that I got from Eric Broug's book, of having channels between the shapes.

Final result

Using the stencil I sketched the outlines of all shapes on the wall. The flexible stencil was indeed handy where the wall met the ceiling and floor.

I painted it using leftover house paint we had, but trying to approximate the colours in Jones's plate.

It is in fact even difficult to see where the doors are now.