Saturday, April 2, 2011

Color abstraction

I have been mulling over the idea that...

basic geometric forms give specific identity to objects in reality that have none

I was exploring this idea, using "nose" as an example. We all have a simple concept of the nose as something sticking out from the face with holes in it for smelling things. Like all concepts, it unifies all the noses we have ever seen, while leaving out the details. The absence of details is oddly empowering. I say "odd", because we usually associate "knowledge of reality" as being knowledge of actual things in reality. But reality is so overwhelming, that if we tried to contain all of it's specific things in our minds, we would fail. Consciousness cannot contain reality, it can only represent reality, and these representations cannot presume to be the thing itself. Concepts give us power over reality precisely because they give us distance. Power (or usefulness) is relationship too, just like knowledge, and so it must be detached from reality in this way. One has to detach from literal reality in order to gain conceptual reality.

So then I shifted focus from the ordinary concept of a nose, to a "visual concept" of a nose. So I drew three views of a nose (line drawings)... one in profile, one in 3/4 view, and one in frontal view. Immediately these present a problem. First of all, for any of the three, it is not obvious where the nose should begin and end. This relates to the "continuity problem" inherent in the continuous shapes in the reality of the face. The second problem, is that the frontal view of the nose (for instance) cannot be "rotated in the mind" to produce the 3/4 or profile view, because the frontal view doesn't abstract anything general about the nose. The same goes for the other two views, which cannot be rotated to produce other views. These three views of the nose are NOT abstractions of the nose, they are simply icons for (or symbols, or signs) that tell us "nose thing" in different position, but they do not function as visual abstractions. Because they strive to "look like" a nose, they represent the nose using specific details, which makes them simply "another nose" in reality.  Interestingly, this is the exact point of one of Aristotle's criticisms of Plato's forms... that if they literally exist in another realm, then they must be something in particular, which means they couldn't be any thing in general, which means they can't be forms.
Formalizing this drawing issue of drawing made it clear in my head, and I made an interesting connection to the issue of color. I have been studying and considering the issues of color for a long time now (well, since 2004). Color is addressed in art school both directly and indirectly, and I have also read a lot on the subject, etc, etc. This studying has had the effect of stuffing my head full of various details, facts, observations, and pragmatic suggestions, as well as any number of conceptual organizations of paint on a palette. Obviously, that kind of stuff is not satisfying for me, as I cannot rest until I reach the magic, transcendent moment of a unified theory and practice. I think I just achieved this in the past few days.

I bought this book called "Still Life Painting Atelier" last week, and began reading it. I was hoping it would be on the level of the "Classical Drawing Atelier" and "Classical Painting Atelier" books by Juliette Aristedes,  in terms of weaving deep thought about art into the subject matter of Still Life Painting. Unfortunately, it does not do that. Instead, it is simply a simple step by step book that describes some basic palettes and approaches to still life painting. But a funny thing happened as I sat there reading the chapter on the Blue/Orange two color palette. A light went off in my head. The author showed the two color palette in a photo, along with a range of mixes that could be produced by it. The author mixed the the orange and blue to get a neutral, and then between orange and neutral he mixed a dull orange, and between blue and neutral, he mixed a dull blue. Then he tinted those 5 little piles of paint  (which were dark) to get a middle value, and then tinted some more to get a light valued mix. In all, there were 15 little piles of paint.

It was all very simple. But within those 15 mixes existed all the fundamental elements of painting.

HUE: the tube colors of orange (burnt sienna actually) and blue (ultramarine blue).
CHROMA: The mixes between the orange and the blue.
VALUE: The tinting of the tube orange, tube blue, and the 3 mixes in between them.
TEMPERATURE: between the orange and the blue.

With that simple palette on could model form, have a range of neutrals, and have contrasts of hue, value, and temperature.

The breakthrough thought here was this... that THAT is exactly what painting is. It is representing reality via (first drawing), but by modeling form with value... of creating depth with neutrals... of creating interesting with contrasts of value, temperature, hue. And so on. Those are the ESSENTIAL aspects of painting.

It's interesting that HUE, CHROMA, VALUE, and TEMPERATURE are often described as the essential aspects of color... BUT what I have found is that these are ALSO the essential aspects of painting. This last point is not taught very well from my experience, and I think I know what the reason is. The reason is that we presume that the purpose of painting is to match reality. In attempting to match what we see we are drawn so close to the phenomenon that we cannot see it for what it is, i.e., we cannot see it for it's essence. We have not distance from it, and therefore cannot form useful conceptions of it.

The essence of artistic vision is related to value, chroma, temperature, and hue. Because value, chroma, and temperature are all aspects of hue, we can (in a sense) drop hue from the equation. Hue is the least important aspect of color in art, though in our minds it seems the most important. This is not to say that hue doesn't matter, but simply that hue becomes the omitted measure of color, relative to the other aspects of color.

Given this, it is clear how misleading much color theory and painting classes are. These classes put undue emphasis on hue. And this reinforces the mistaken notion that matching the specific hues in reality is the objective of painting. I myself have been totally sidetracked into considering hue as the most important thing. For instance, wondering how many reds could be mixed from a collection of tube reds. I have wondered how many reds could be mixed by combining various combinations of burnt sienna, cadmium red, indian red, iron oxide, etc. I have actually felt guilty that i didn't spend hours mixing up the endless permutations that could be created from them. I thought that the answer must certain lie in that. And the same logic applies to the yellows, and blues, etc.. to all the colors. And once you have mixed 100,000 variations on all those colors, you can turn around and start combining the reds with the yellows, and yellows with blues, etc. And this could (and would) go on endlessly, as you attempt to reconstruct every possible combination of every color that can possible be mixed. And then you start in with tinting and shading them, and so on for forever. This endless exercise is a literal analogy with the mindlessness that is inherent in trying to match reality. Such an approach implies that painting is simply the matching of a infinite spectrum of colors.

It is the reality trap. Thinking you have to match natures hues is the reality trap, and allows for no conceptualization or control.

Just like the forms of the face, the hues in the world are infinitely nuanced and continuous, such that no order can be found in them other than what we bring. The means of ordering form is geometric abstraction. The means of ordering color are the abstractions of value, chroma, temperature. To restate the idea from above...

value, chroma, and temperature give specific identity to hues in reality that have none

Once hue is dropped from the equation, simple palettes suddenly seem relevant, and not simply as training exercises. This is another point of failure in art teaching... that limited palettes are presented as training grounds for full palettes, where the superior virtue of the full palette is the range of hues that can be mixed from it. This reinforces the idea that matching nature's colors is the goal of painting.

So, I am really excited by this breakthrough idea. It settles in my mind much of what was confusing.

As a side note, I was thinking that the obsession of with specific hues might be related to impressionism, with it's emphasis on impressions of light (hue) rather than on form and space, which is more an issue of value, chroma, and temperature.

1 comment:

  1. Mike,

    Hue and temperature are the same thing. One cannot shift temperature without shifting hue and the reverse.

    Temperature is typically detached from hue, in academic settings, as it provides an extra exercise after the learning of value, hue and chroma. And, they should be in that order for the sake of clarity to the widest audience.

    You’re right, too many novices try to (or are trained to) “think” of the local color. Or, some systematic “color theory” organizational response to what they see.

    What is (mostly) taught is very different from how our human brains respond to color. It’s a long story that was resolved by Chevreul and Rood over a hundred years ago.

    Perceptual psychology has come a long way in the past decade to help reconcile the success of the great colorists (including some Impressionists) and how we perceive them.