colors in the Conver and CBD Tarot de Marseille

in this essay i explain the color aspects in the reproduction of the Conver (1760) tarot deck, which i recently published as "CBD Tarot de Marseille": how and why i chose this coloring scheme and this color palette. a fuller discussion of the CBD deck is in this thread
http://www.tarotforum.net/showthread.php?t=163515

also a separate essay on the face expressions is here
http://www.tarotforum.net/showthread.php?t=168219

as this is long, i separate it ino 4 consecutive posts.
------------------------------------------------------------

colors in the Conver and CBD Tarot de Marseille

1. lines and colors

the colors of the card images have a great impact on the emotional experience of the reader. in my way of using tarot this is an essential factor in the interpretation, so in working on the CBD deck i had to pay a special attention to the coloring. actually this involved two sets of decisions: (1) the coloring scheme, i.e. in which color to do each area in the image, and (2) the color palette, i.e. once i decided, for example, that certain surfaces should be colored red, which shade of red i should use.

for both sets i first had to understand how the original colors in the Conver decks were produced. for this i relied on several sources, including the professional knowledge of Leela Ganin who drew the black lines for the cards, hints from internet sites and forum posts, and especially the historical studies in "Cartes a jouer & Tarots de Marseille, la donation Camoin" published by Musees de Marseille (2004).

because of the technique of production, the marseille decks have a set of basic and uniform colors, not like the gradual shades of modern deck. in fact the black lines and the colors represent two different levels of production and authenticity. the lines were printed from woodcut blocks made under the supervision of the master (Conver) himself. Possibly they were copied from Conver's design but done by a professional specializing in woodcutting (it is also possible that the letters were cut by another specialist). when i look at the decks which i used for the restoration (the Conver editions published by heron and by lo scarabeo, the bicentennial (1960) camoin printing, plus copies of specific cards i found here and there on the internet), they are all printed from the same original woodcuts. Thus, what i see in the lines is the actual work of the master and/or the persons working under his direction.

in contrast, the colors were hand painted with a set of stencil sheets ("pochoire"), one for each color, in which holes were cut in the appropriate places. they were put on the printed sheets and painted over with a brush. so the actual painting would be the work of a "simple" worker doing a batch of decks one after another. the stencils were not always put exactly so you can see that each color has some offset relative to the lines, which would be different from one copy to another. also the painter could miss a small hole here and there, or have too much paint on the brush that would smear. this means that i could not always be sure about the original coloring of small areas - whether the red should end on this or on that line, etc.

also, the stencils were much less durable than the woodcut blocks, as they were made from heavy paper coated on both sides with several layers of paint. in decades of use they were worn out and had to be replaced. this is probably why the scarabeo copy (second half of the 19th century) has many coloring variations relative to the heron (early 19th) deck. as can be expected, the scarabeo coloring is less precise and less elaborated than in the heron deck - it was copied from former sheets which were already worn out, and not under the caring eye of the original creator.

2. the coloring scheme

a main principle of my restoration was that the interpretation should follow from the image details, not the image details from the interpretation. for the coloring scheme, this meant not to rely on symbolic color meanings (e.g. "yellow=intelligence, so this detail should be colored yellow to symbolize intelligence working on matter"). such symbolic considerations would depend too much on my own idiosyncratic vision of the cards, and anyway they would come from a tradition which historically developed much after the cards themselves. so, i just tried to keep as close as possible to the original coloring scheme as i can see it, and not make considerations of "what it should symbolize".

in tracing the lines i relied on all the original printings available to me. but for the coloring scheme i took as a main source the oldest (heron) printing, which for all i know may have been colored with the original stencils made under Conver's supervision. still, there were many places were the heron coloring is not clear (e.g. the difference between the diluted skin color and the original off-white paper is sometimes almost invisible). in such cases i also consulted the scarabeo copy, but generally i took it as less reliable.

still i had to take into consideration the effect of the different color palette i was using, and also the different perception of color by people living today, as i explained in the forum post about the face expressions. so my consideration was not only "how is it in the original" but also "how does it appear to the eye". thus, sometimes i made deliberate modifications in the exact limits of specific areas to keep what i felt as "the spirit of the original" in the much stronger and sharper color shades of the modern printing techniques.

also, i had to consider the fact that in the stencil technique, it is impossible to produce a colored area with an uncolored hole in it (there is nothing to hold the covering stencil sheet piece in place). in the different cards made by Conver there are different solutions to this constraint. sometimes a darker color is printed over in the middle of an area with a lighter color, so that the top hides the bottom color (e.g. a red circle overprinted in the middle of yellow coins). by the way, in some other cards a similar technique is used to produce an additional color effect - light blue printed over yellow to produce a distinct yellowish green, for example in the ace of cups. sometimes the middle area keeps the color of the surrounding even though it does not make sense from a realistic point of view (for example, skin color under the right arm of the girl in the star card, which is made less conspicuous by the over-printed green). and sometimes an uncolored band appears in what should be a continuously colored encircling shape (e.g. the hands of the children in the sun card, which should have a blank space between them). in such cases i tried to figure out Conver's motivation to choose this or that solution, and to choose my solution in a similar spirit but also considering the visual effect of the new color palette.



i tried to be more or less consistent, but allowed myself to change the rules for specific problems in specific cards. for example, when i did the bottom of the tower card, some people felt that it appeared strange and dull when painted only in yellow as it is in the heron copy, and i could see their point. so here i looked at the scarabeo, and saw that unlike in other places, in this card the ground is done in a more elaborate coloring scheme than in the heron. this may reflect the evolving nature of the tarot as a living work of art - i can imagine that the person who designed the new stencil sheets felt a similar difficulty with the ground color, and wanted to make the composition more vivid. i did want to allow for such an evolution - after all, it also gave me a certain latitude to make my own decisions and mistakes. so, in the ground of the tower card i tried to integrate additional coloring elements from the scarabeo into the basic heron scheme, and played with the exact color borders until it looked right.



in similar places i usually tried to choose the coloring scheme which appears more dynamic and interesting, but also more open to different visual interpretations. so, in case of a doubt whether a certain area should be colored at all, i preferred to leave it blank unless it became sore to the eye. also i usually kept blank areas where they were blank in the original, for example the small triangle to the right of the top of the middle table leg, which other restorations either removed, or colored to make it a continuation of the magician's leg. when left uncolored you can also interpret it as a part of the table, or maybe as something else standing for itself. you can also more easily deconstruct the surfaces to see the female bosom shape under the table (intentional? Freudian? i know it is not only my own dirty mind because others can also see it..). you can also notice the white spaces on both sides of the table, which make the ground feel open and unreal (maybe it is a part of the illusion show?).



one more thing was about the faces. in the original, some faces appear in skin color, while others (and sometimes also hand palms) are left white (uncolored). but when i test printed the cards, as the contrast between skin color and white paper is much stronger than in the old printing techniques, the white faces appeared eerie and lifeless. other restorations solved this by doing all the faces in skin color, except for the popess which is sometimes left blank. but i felt that this is a distinction in the original that i don't want to lose: which faces are blank, and which ones are colored. my solution here was to break the rules of the coloring scheme that i kept elsewhere, and do the originally blank faces in a lighter shade of skin color (same hue, but with partially transparent white on top). if you look it up in the heron copy, note that the white faces were re-colored with a redder shade of skin color. but this was added only in the reprinted copy, probably for similar reasons.

3. the impact of colors

for the color palette i knew i had to start from scratch - there was no way to restore the original palette of hues and shades used by Conver. the old heron copy was made about 50 years later, by which time pigment availability and prices may have been very different, especially with all the changes in the commercial conditions after the French revolution and napoleon. so the colors shades are not necessarily Conver's originals. also, the colors in this copy faded and changed during 200+ years. and even if i could somehow reconstruct the original pigments and mixtures of 1760, i would still have to account for the huge difference in printed color perception of people living today. so any attempt to reproduce the original palette was pointless.

i considered the possibility of imitating an old handmade look, like an offwhite color for the paper and handmade-like textures in the color surfaces. but this did not feel right to me: there is an essential difference between doing such things because this is the available technology, and doing them as an artificial imitation. so here i thought that my task is not that of a restorator, but more of a translator: to re-create in the language of a new printing technique what i perceive as the essential content of the original. so i decided to use white paper and uniform color surfaces which feel natural in today's color printing context. what i needed was a color palette which would be internally harmonious, convey distinct and appropriate feelings, and also be usable for card readers who interpret the colors symbolically.

as i said, i did not want to rely on a specific symbolic system of my own. but i had to have a clear idea of the emotional impact that i want each color to suggest. so i put together some ideas that i got from my own experience and from things i read here and there on the meaning and impact of colors. for the basic concepts i took elements from Wassily Kandinsky's theory of colors and their emotional impact, presented in his book "Concerning the Spiritual In Art" (1910). i thought that in a way he was dealing with similar questions as an abstract painter who expresses himself in pure form and color. in his study Kandinsky characterizes differnt colors as warm or cold, dynamic or calm, outward or inward-going etc. this is a language which is precise about color effect without going to abstract symbolism.

using such terms, i could also understand something about the original cards. in the Conver deck, the different colors appear in different measures. i did not do exact statistics, but looking on the whole deck i can see that the order of colors by prevalence is: yellow (most prevalent), red, light blue, skin, green and blue (least prevalent). thus, the hot and dynamic colors yellow and red are the most dominant, while the cold and calm colors green and blue are the least present. the interim colors light blue and skin are different from the other four - they are not pure spectral colors, and both are done in diluted pigments so they appear much lighter. the greater prevalence of the cooler light blue over the warmer skin color is contrary to the domination of hot and active
in the four spectral colors, but this i can see as some counter-balance. so, generally, i can understand that Conver intended his cards to have an warm, active and outreaching feel, not cool and meditatively absorbing like some new-age decks of today. but as with other things, he also tried to balance this preference and not to make it too extreme. this is the spirit i wanted to maintain.

4. the new color palette

finally, i got more or less this scheme for the visual feel of the colors:

yellow - hot and bright, gives light and warmness to the cards. i did it in pure yellow with almost full saturation. in the CMYK 4-color printing this is a pure color ink, so it prints clear and bright.

red - active and outgoing, but considering the bright yellow i thought i should make the red a little cooler and quieter, otherwise the combination of these two strong colors would make the cards too flashy and disquieting.

still, red is a strong color and often it gives a skeleton to the card composition. it is interesting to see this with the following experiment: look at a card from the CBD deck and try to separate only the reds, as if to lift them out from the other colors. often you can see in this way the compositional structure of the card or some part of it. to understand what i mean you can see sample compositions in the following video:

http://www.youtube.com/watch?v=LYkK5BKI5mo

once you get to it, you can add other colors like green and blue and do a similar exercise with their combination. or look at each of them separately to see how Conver distributed it in the card. when i do it, i am strongly impressed and fascinated by this work of a genius with his sense of dynamic harmony and right balance.

light blue - to have its own character i wanted it to be airy, suggestive of the sky. i used 30% pure cyan so that again it has the clear quality of pure ink in CMYK printing (in contrast, red is a mixture of yellow and magenta, so as dots of different ink color overlap, it may get a slight brownish dullness).

skin - obviously i wanted it to look alive, but not too pink as would make the figures "cute".

green - similarly, it should be natural to suggest plants, vegetation and growth, and be harmonious with the red.

blue - quiet, deep, absorbing, but in a right balance with the green and red so that none of them looks off-palette and sticks out from the rest. also, as it is a darker color, i had to take care that it is not too dark so as to hide black lines inside blue areas. finally it proved to be the most difficult color to get right.

i also needed a starting-point palette to play with and modify until i reach the right shades. a friend of mine who is a long-time researcher and participant in the digital culture scene pointed out that there is a good example of a color palette which is pure basic colors in a combination which looks updated and appealing to people today: the google logo. so i took the google logo color shades, and started to change them one by one until the test prints looked like what i was looking for.

still, even high-quality test prints differ from the actual machine printing. the difference may be significant when the emotional effect of the color is important, which is exactly the case with tarot cards. so, as i was advised by several people with printing experience, i traveled to belgium for the actual printing so that i could be present at the final calibration of the printing machine. this proved justified - i could not expect, for example, the professional printers to consider the "bright and warm, but not too much" quality of the yellow, or the "airy, not opaque" feeling of the light blue.

you can get some impressions of the CBD deck colors from the card images in the galleries on my website or on flickr. however, note that these are not the actual color shades as printed on paper, but their conversions into the RGB color space of computer screens.

http://www.bendov.info/tarot/cbd/gallery/index.html
http://www.flickr.com/photos/48485995@N00/sets

or see the short video tour here
http://www.youtube.com/watch?v=qKhj3WFybv0

for other info, please go to my site and choose english
http://www.bendov.info

Thank you for posting this. Sorry it took so long to express my appreciation. It's interesting to see your thought process. There seems to be a fad for "restoring" old decks lately and I don't know of anyone who's documented their decisions in such detail. I often wonder if their choices are systematic or more a matter of impulse. At any rate, a quick comment on this:

Yoav Ben-Dov said:

3. the impact of colors

...

in the Conver deck, the different colors appear in different measures. i did not do exact statistics, but looking on the whole deck i can see that the order of colors by prevalence is: yellow (most prevalent), red, light blue, skin, green and blue (least prevalent). thus, the hot and dynamic colors yellow and red are the most dominant, while the cold and calm colors green and blue are the least present. ...

so, generally, i can understand that Conver intended his cards to have an warm, active and outreaching feel, not cool and meditatively absorbing like some new-age decks of today. but as with other things, he also tried to balance this preference and not to make it too extreme. this is the spirit i wanted to maintain.

Click to expand...

I wonder if the cost and usability of pigments for paint had something to do with it. This would take more research than I have time for now--but I was just reading about how Vermeer used unusual amounts of blue paint compared with his peers, blue being very expensive. Anyway, here's the site I started to read and then...reality called. http://www.paintmaking.com/historic_pigments.htm

I love the Google logo story