PMdB > Reconstructions > Interpreting Colours

Interpreting Colours

Up

B148

Composition 1922

In 1922 Mondrian sent Van Doesburg "two photos which give some impression of my work". One of the photographs was of B148 and it was published in De Stijl. It is the only information there is on the work: no indication of colours or size. B148 will therefore be used to explore reconstruction process.

26 rue du Depart
26, rue du Départ

Mondrian was working at a time of significant advances in the chemistry of photography as the blue- and green-sensitive orthochromatic emulsions gave way to the spectrum-wide panchromatic. The subject is explored in some detail in Postma’s 26, rue du Départ [pp.55-63], in which Mondrian’s studio is painstakingly reconstructed from black-and-white photographs and contemporary descriptions.

B104
B104

The key to the reconstruction is the inclusion of B104, Composition No. II, 1920, now at the Tate in one of the three photographs taken of the studio by Delbo in March 1926. The research almost foundered when the shades of grey in the photographs could not be related rationally to the known colours of the painting. Despite using and replicating a variety of photographic emulsions, the rendering of yellow as black could not be explained until descriptions were discovered of an obscure and short-lived Russian portrait film with those characteristics which Delbo could plausibly have used.



The problems of reconstructing individual paintings are similar, but on a much smaller scale and can be illustrated with the example of B148, Composition, 1922 which exists only in De Stijl, no.8, 1924 fig. 1. The work’s title offers no direction and the dimensions are unknown. While there are clearly at least two colour blocks in the piece, the reproduction is of little help in determining what those colours might be.

B148
fig. 1 B148, De Stijl

The extant 1921 painting B129 Tableau II with Red, Black, Yellow, Blue and Light Blue also appears in De Stijl and fig. 2 shows how the reproduction, showing only black and grey, fails to represent colour values.

B129
fig. 2 B129, De Stijl
B129
fig. 3 B129

The reconstruction of B148 rests on observing other similarly-structured pictures of the time which all had more than one colour with red and blue usually predominant: most had all three primary colours, but with no suggestion of where any yellow might be, fig 4 shows the reconstruction with red and blue, arbitrarily assigned.

B148 reconstruction
fig. 4 B148 reconstruction

This is a good time for a reminder of the scope of this exercise,

"An important factor in pursuing this plan is the knowledge that, to Mondrian, structure was more important than colour: in confirmation of this, Blotkamp [p.205] quotes a letter Mondrian wrote to Alfred Roth, a prospective purchaser, in September 1929, ‘Let me know whether you prefer blue and yellow, white and grey, or perhaps red, a bit of blue and yellow and white and grey. The latter works with red in them are more “real”, the others more spiritual, more or less.’ While it is important to get the colours in the reconstructions as close as possible to the originals, the structure is key."

One of the most useful exercises in practicing reconstructions has been to work on paintings which still exist but for which only black-and-white images are currently available. This presents the same problems as working on the Missing Mondrians but with the possibility of a true evaluation of the reconstruction when a colour image is found. Results have been mixed and a couple of the comparisons provide an insight into the process.

Figs. 5-7 illustrate the process with B199, Composition with Yellow, Red and Blue, 1927, now in the The Menil Collection, Houston. The black-and-white photograph (fig. 5) shows two colour blocks but the title clearly requires at least three. Colours were assigned to the middle image by analysing the shades of the grey blocks. Comparing the facsimile (fig. 6) to the actual painting (fig. 7), shows a reasonable degree of success, but there is simply no way of determining with any certainty that there was a second yellow block.

B199
fig. 5 B199
B199 facsimile
fig. 6 B199 facsimile
B199
fig. 7 B199

By contrast, figs. 8-10 shows how easy it is to make a fundamental mistake in this process. B208 Composition No.I with Yellow and Blue, 1929 appears at first to be a straightforward task from black-and-white with the ‘blue’ area easily identified and the area above it clearly differentiated from the other light blocks. The middle image (fig. 9) can be asserted with confidence, but when the postcard from the Kyoto National Museum of Modern Art arrived (fig. 10, kindly posted by a correspondent on the day of the 2011 earthquake), it was a surprise to learn that the yellow block was in one of the three other indistinguishable grey areas. The actual painting is more satisfying than the attempted facsimile, but the translation from grey to yellow remains perplexing.

B208
fig. 8 B208
B208 facsimile
fig. 9 B208 facsimile
B208
fig. 10

Sources
Snap Dragon