What "color" are shadows?

[LadyDreamkiller]

This isn't meant to sound cheeky (ok, maybe it is Just a little bit lol) ... but...

Could someone please translate what Larry said into simpler terms for those of us who are dunces and/or new to trying to sort out light and shadow in the first place? Please???

[bigflea]

For the premise that the shadow color is the local color darkened to be correct, one would have to conclude that the shadow colour of an area of clean snow is grey or black. Further, one can observe that the local color of snow is not white in the sunlit areas, and that the concept of the local color as an absolute is not a reliable one in describing a visual experience.

Visually, the local color of any and all objects is changed by the quality and conditions of the lighting present at the time of observing them, including the shaded areas.

So a painter has to choose whether they are attempting to depict the visual experience, illusory though it may be, or are they attempting to be a scientist describing data, imo.

It is possible, I feel, to learn to observe illusory effects and gain an understanding of how to paint them in a visual way, without being visually "tricked" by them. Some of the examples given of contrast illusions are extremes because of the context of seeing them on a computer screen. Contrast effects in nature, in my experience anyway, are not nearly so deceptive and difficult to study/observe visually.
Ken

[WFMartin]

Originally Posted by LadyDreamkiller

This isn't meant to sound cheeky (ok, maybe it is Just a little bit lol) ... but...

Could someone please translate what Larry said into simpler terms for those of us who are dunces and/or new to trying to sort out light and shadow in the first place? Please???

One can detect subtle color more readily by looking in some direction other than at the object whose color you are attempting to determine. (Make use of your periferal vision.)

In the very simplest terms.......ya' wanna' see the color of somethin' better? Then, look at somethin' else.

Bill

[Einion]

Originally Posted by bigflea

For the premise that the shadow color is the local color darkened to be correct, one would have to conclude that the shadow colour of an area of clean snow is grey or black.

No, that's far too broad - you're going from a general statement to a specific condition and just like doing the reverse it's almost always a mistake.

To begin with this presupposes that there's one lightsource or lightsources of only one colour. If we imagine a clear sky and we compare it to a person standing within sight wearing a white shirt for example, one should not assume that the shadows would look exactly the same; for a start the shadows on the shirt and snow are likely oriented differently (facing downward and upward respectively) so the elephant in the room we're not talking about - the obvious, but unspoken, second lightsource works differently on each.

Even without this, on an overcast day, where the huge, coloured secondary light is gone, slight difference in basic colour can have a noticeable effect on the shadows. And on such a day the shadows of snow stop being blue and do become grey.

In a real-world scenario outside there's no assumption of a simple one-light setup except when there is solid cloud cover.

Originally Posted by bigflea

Further, one can observe that the local color of snow is not white in the sunlit areas, and that the concept of the local color as an absolute is not a reliable one in describing a visual experience.

Apart from the fact that the first part will vary with conditions the second part is an opinion Ken - many other people, including many other artists, don't agree with it. There is far more art made and being made that is, or is close to, what you've referred to as "local color literalism" than not among realist painters.

Originally Posted by bigflea

Visually, the local color of any and all objects is changed by the ...lighting present at the time of observing them, including the shaded areas.

I don't believe anyone here has said that they're not.

Originally Posted by bigflea

So a painter has to choose whether they are attempting to depict the visual experience, illusory though it may be, or are they attempting to be a scientist describing data, imo.

Anyone who's interested, see first link I provided above.

Originally Posted by bigflea

It is possible, I feel, to learn to observe illusory effects and gain an understanding of how to paint them in a visual way, without being visually "tricked" by them.

What does "paint them in a visual way" mean? How else would one paint them, with our eyes shut? LOL

Originally Posted by bigflea

Some of the examples given of contrast illusions are extremes because of the context of seeing them on a computer screen.

Yes, that is true. Not directly because they're on a monitor but because they're usually made to be pronounced to make the point strongly.

The illusory hue of caucasian skin shadows is a prime example and one that in reality and in good realist depictions works much the same - to make the point most strongly that painted illusions can closely match what is seen in the flesh for any doubters.

Einion

[Einion]

P.S. I don't agree that it's a good idea to use peripheral vision to try to 'see' the colour of shadows accurately myself because I personally don't think it works; and there's a strong physiological reason that we see colour best in the centre of the visual field since that's what the eyes were 'designed' to do

Einion

[Patrick1]

Originally Posted by Einion

...I don't agree that it's a good idea to use peripheral vision to try to 'see' the colour of shadows accurately myself because I personally don't think it works; and there's a strong physiological reason that we see colour best in the centre of the visual field since that's what the eyes were 'designed' to do

I can take this two ways:

1) you're saying that comparing a color with surrounding colors won't give you a more accurate read on the target color, or

2) comparing with surrounding colors will give a more accurate read on the target's relative color, but look at the surrounding colors directly rather than using peripheral vision

[Patrick1]

Anybody notice how contentious these shadow color threads are?

[bigflea]

Ladydreamkiller,
Larry's comments, to me, were about the visual difference in color perceptions, between a peripheral or glancing/scanning observation, and a fixed, or "staring into" visual habit. In the latter the subtle color effects of shadow in contrast to direct light planes,(or areas of color) can become "lost" or neutralized by the effect of staring as opposed to allowing your eye to glance across the whole visual field in a relaxed and natural way.

The debate over the peripheral visual glance vs. the fixed, or staring into vision, or over how does the eye best see color, revolves, imo., around the issue of the importance of the local color of an object vs. the importance of the light effect ( the light source and atmospheric conditions) on the local colors of all objects, and the shade areas near or on them.

On the local color enthusiast side, the argument seems to be mainly that the local object color is best recognized by the direct or staring into vision. From that pov, one may conclude that the dominate aspect of color is the local object color.

On the light effect enthusiast side, the argument I make is that the light effect on color is best seen in the peripheral glance across the whole composition. Shadow color from that pov is determined by the interaction of all colors together, which results from the quality of the light sources and the atmospheric conditions present at the time of observation.

Einion, you are right that the example of snow is broadly drawn. But that is exactly the point of it. It is to say that the concept of the local color as the most dominant aspect of a shade color, as promoted in earlier comments, is misleading and too broad, and over simplified. In visual observation in natural light conditions the painter is faced with the effects of light reflecting into and around shade areas, and also having to distinguish between optical visual effects, such as the differences between peripheral perceptions and focal vision perceptions.

While local object colors, or surface colors, are commonly taken to be the "color" of an object, that does not also mean these local colors are what the painter ought to use to paint them. The realist convention for using local color to depict an object's color is a convention only, and as such is an opinion about object color only. In that sense it is no more correct than the perceptions of an impressionist colorist. A realist portrayal of skin coloring may look convincingly real to a realist painter, and to others who agree with the conception of the local color, but to myself and others who see the light effect and changes to the local color differently, it will likely lack color qualities visually present in natural light conditions.

Just to clarify, I am not contending that local object colors are irrelavent to the understanding of solving a painting problem. What I am contending is that the light and atmospheric effect on all local colors is the most important aspect in a harmonic unity, and in the modeling of form as visual description.

Pictorial conceptions of form are not always visual descriptions, imo. That is why I keep using the phrase, "in a visual way". I am trying to speak to the difference between a pictorial conception vs. the way form is seen. I suspect most painters would agree with the statement, but disagree over which painted version is pictorial and which is more visual. And so the debate continues.
Ken

[LarrySeiler]

all of us are amused when from experience using an approach we hear others saying it can't be done. Its why I painted a painting using only sticks when insistence only the best brushes could make painting possible...and just painting a painting of a winter scene standing on location at 5 degrees arriving at the location only to discover I had no white. So I painted without white.

Wish I could say I invented the peripheral vision thing...I did not...I read it elsewhere then saw it reaffirmed by Kevin MacPherson in his book "Fill Your Oil Paintings With Light & Color"

Perhaps about five or six years ago...in a fall season, I was driving my truck down a dirt road having spent a wonderful afternoon hunting ruffed grouse, and doing a painting along a river area. It was near dusk.

I was in my own little world when I noticed intense orangish/red pine tops to the trees...and a greenish glow of the sky. I was looking down the road at the time, and saw/sensed these colors peripherally...that is to my side vision where the eyes do not see directly.

I pulled my truck over and got out to study this moment.

The tree tops were indeed reddish in appearance looking directly at the tops...but appeared more intense looking more at the ground or to the side. The sky had very very little hint of greenish color looking directly at it. At first I thought I was wrong to sense the color...but remembering this phenomena and not really making use of it before I looked down at the ground masses to judge the color of the sky indirectly or peripherally.

Sure enough...I could again sense the greenish color.

It was shortly after that I read MacPherson speaking about this phenomena and how he often uses it.

In my own book I spend time sharing this technique and I'm only sorry to say that if it doesn't appear to work for others, I can understand that. I would ask anyone not to presume because something fails to work for you to assume then it can't work for someone else. This is one reason why we read, why we tinker around, and we find what DOES work for us.

I found especially painting several weeks ago outdoors during a snow storm, sitting on the tailgate of my truck that I was able to sense much more color judging areas looking away or to the side. It would confirm especially if I thought I saw a color directly to look aside and sure enough sense it to an even greater degree.

here was that painting, a 9"x 12" painted in 1-1/2 hours...on location...



Here are closeups...and I arrived at judging the color present using this peripheral vision judgment not in substitution of looking directly, but in conjuction with...







now...my question philosophically and practically is...why minimize the degree by which we see? If the eye sees peripherally as well as directly, why say seeing counts only directly? The human experience has a broader existence. Thus...the painting is not holding less integrity to include color that is sensed to exist.

Truth is too...that if we make a habit of looking for something believing it to be there, (sometimes because someone may have told us to look for it) then we might be likely to discover it indeed is there.

for example...if you are at an amusement park responsible for a youth trip and 34 kids on the bus and all agree to show up at a certain time to leave, but one boy is missing...you have the ability to walk back into the amusement park despite its having 10,000 enjoying the rides and such. The mind has a way of blanking out everyone that is not going to meet with the understood features. Girls with long hair, guys with hair color not matching...too tall, whatever are glossed over and ignored. An amazing mechanism which the mind has.

If you believe shadows are dark and hold little color, then you are likely not to look for evidence to the contrary.

I can tell you...that if you approach nature with greater naivety believing to see as a child the wonders of the world for the first time with awe and without prejudice, nature can be a marvelous and capable teacher.

IF you approach nature having determined to believe only what you are going to believe...you are likely prejudiced against further seeing. When you do see something outside your expectations it is likely to come as a surprise or to be dismissed.

Peripheral vision works for many artists.

Just this summer in one of the workshops I taught...I had seven of the participants standing together and we looked across a field to a distant tree mass about 140 yards away. I pointed out how if you looked directly at the mass which was in shadow and backlit that one color was basic and determinable. I then asked them all to look up in the sky about a hands height above the tree tops and tell me what color in the shadow they saw...?

They all agreed they had seen a definite bluish-violet hint of color. Now...if that is the case, why not make use of it?

So, I will argue...a sensitive eye can see color in the shadows peripherally.

Here was one more example.



If I took a photograph and used a photo for reference where no peripheral vision is possible, thru the lens metering would turn the shadows of the distant mass simply to darks and the photo would argue (as if photos were authoritative evidence) that the shadows are dark. Standing on location, painting and observing directly from life...the shadows changed and appeared differently. If I looked directly...the area lit up by light looked differently, the shadow not feeling quite as intense, and I could see details of trees, trunks and so forth. When I looked at the lit up foliage the shadow's color appeared another way, the trunks barely discernible...AND...when I looked up at the sky the shadows revealed a hit of bluish-violets, and particular trunks and details not discernible at all....and knowing that cool colors would assist in pushing the illusion of the mass further back I opted for what THAT vision technique would provide other than what direct looking would have provided.

So...here's a question. Does this last painting work, or not work? If it works...then it works inclusively to the fact that it was painted by my making decisions, and it was peripheral viewing judgments that gave me a working decision to make. If someone wanted to know how I arrived at the decision to paint the background as I did...it would be wrong testimony for me to say I did not use peripheral measures to judge how to approach painting it. It is what I did. It is the reason it appears as it does. Perhaps it could be said I simply got lucky. Lucky....again.

Now...lest it be said perhaps this can only be true for plein air landscapes, I use peripheral judgments in viewing color or the full measure of color available to me when I paint my incidentals which I put up on my blogspot. My incidentals (for those that do not know) are everyday mundane simple objects I paint as postcard exercises to keep the eye sharp.

[LarrySeiler]

here's an example of one of my incidentals...smoked chubs...oils.



I could sense...a hint of the presence of other colors in the shadows. I struggled at first to immediately know their presence, and by directing my eyes to the area above the chubs sitting on my table (painted from life, not a photograph), I was able to more easily discern the presence of pinks, some hints of neutral blue in the shadows as you see here which I then painted...




again...the question is...does it work? Does the painting look like chubs though painted very quickly...does it aesthetically work? If so...one must account for the fact that I struggled to see and judge the color in the shadows, and used peripheral (or indirect seeing) to make final determinations of what color was indeed there. Is that a wrong way to see? What ought makes the right ought in seeing, being what it ought to be?

One last example so perhaps what I'm saying and arguing for is understood, and might help some...



In this detail of an incidental I painted of a 35mm film canister, I have added the graphics of the spheres and arrows indicating for example that I would have looked in those areas to judge the color for the areas the arrows point to...one develops this by habit, but I believe adamantly it has at least for me made my painting manner more proficient and easier. Please don't be offended, for it is not my intent...but I'm a bit aghast to hear what can't be done when it is part of my everyday practice. Can't help but strike me..well, I dunno...silly maybe? I can appreciate perhaps it not working for someone else or be their preferred way of working. It is, like many other things...one option to consider.

[Patrick1]

I have the feeling that Einion isn't disagreeing that using peripheral vision can/does help to emphasize color differences (simultaneous contrast effect) but that this shouldn't be mistaken as seeing more accurately. I'll wait to see what he says...I too was surprised by his comment.

Nice examples Larry. If you haven't already, don't sell those last two - keep 'em and hang them adjacent each other on the wall...such a striking warm/cool color contrast.

[JamieWG]

When artists ask about purple shadows, I like to show some pics I've taken. These are photos I took while out with some of my plein air friends one day. I did not imagine these colors.





and another, in winter, with the strongest blue-violets!



Jamie

[FriendCarol]

Well, first here's my take on the 'how to see color' issue, then I'll try to respond to the Lady's plea.

The anatomy of the eye definitely reveals the greatest concentration of receptors -- particularly color receptors -- are in the fovea, which is at the center of the eye. (Not the very, very center, where the optic nerve enters, but that's a tiny 'dead' spot we're really not aware of.)

So, to see color best, we should look straight at the object. BUT, if we do this for too long, another factor causes our color vision to become less accurate or less sensitive: The specific receptors that react to specific colors 'burn out' and need to be chemically replenished. This is why if you stare at a colored pattern for about 20-30 seconds, then look away to a white or gray surface, you will generally see the optical complements of that pattern -- the non-used color receptors are creating that illusory color.

Conclusion: To see color best, glance at the scene. Repeated glances give you the best data, physiologically speaking.

Peripheral vision is better for seeing light, however, because the rods are the receptors for light (not color, that's the cones), and they are not concentrated at all at the center of the eye. That's why it's easier to see stars without looking directly at them.

Okay, now to try to answer Lady: Shadows are conceptually easy to understand, but can be hard to get right in practice. Suppose we are outside, on a sunny day. We have a very large blue sky overhead, and that is actually producing a blue light. That indirect, diffused lighting pales into insignificance beside the light of the sun (usually a bit yellow). If something is shaded from the light of the sun, however, that blue light may still be illuminating it.

In addition, any light reflected from other objects -- which will tend to take on the color of those objects -- could also be illuminating a shaded area. Furthermore, occasionally objects are somewhat translucent, and if a translucent object (flower petal, perhaps) is blocking the sun's light, some of the petal's color will get through.

So the shadow color of the object is equal to the local color of that object (dulled, lower in chroma) plus any light that is reflected into it (such as blue skylight), or filtered onto it (such as a pink rosepetal's filtering). This is the true 'general rule.' Easy to understand, but in practice it's hard to locate all the sources of indirect, reflected, filtered, (etc.) light that is illuminating that area shadowed from the major source of illumination.

Little trivia for you: Since sunlight is basically yellow, and skylight happens to be blue, the Impressionists apparently helped give artists the general idea that shadows are complementary in color to the local color of their objects. This general statement is not true, it just happens to work outdoors on sunny days. Because of a completely different fact about color vision (called 'simultaneous contrast'), however, it often appears to be true. And this is what Bill (WFMartin) is emphasizing. Bill suggests we should paint shadows using complementary colors because simultaneous contrast makes it look like that, so we should paint it the way it looks. Einion (and I) say if we want to reproduce the illusion, the way to do that is to paint the shadows the way they really are, and then simultaneous contrast will make them look more complementary than they are. Bill suggests the effect doesn't work as strongly with painted objects as with real ones. That's why I'd like to see 3 versions, so we can decide which looks more 'accurate.'

In case you're wondering, around this forum, we are careful about the concept of 'color matching' -- find the color of paint to represent objects in the world, and use quotation marks around words like 'accurate.' That's because otherwise a very clever expert (handle 'drollere') might come along and tell us off and make us do experiments to prove that painted things never really look like the things they look like.

Give a yell if you want more detail on any of this.

[LadyDreamkiller]

Gentlemen (and ladies) ... while some of you might find the thread a bit contentious, and even downright heated in a few spots... I'm finding the whole thing EXTREMELY helpful. Watching and taking part in this discussion has had qualities for me of sitting in the back row of a classroom filled with professors - all of whom have something invaluable to teach me about my avocation.

Larry, while the comments concerning colors seen via periferal vision don't help much with the piece that has me obsessed with finishing it by Monday - it relies heavily on a reference photo of someone who is no longer living - I went out in the backyard earlier with the dog and actually DID end up noticing colors that I didn't realize were previously present in my surroundings. Once the weather warms up, I plan a trip out to one of the lakes not far from home to try some plien air experiments - and I'm currently hopeful that the periferal viewing will help me when I do that.

Einon, for the current portrait piece, what you've been arguing about shadows being simply a darker shade of the same color seems to be working out well. At least for most of it.

Having never attempted to lay in shadows in a painting before, and never having attempted a portrait before, there are areas of the painting that I'm having significant difficulty with getting to look Right to me. Below are a close up of the elbow/arm area of the WIP, and the photo it's being done from - some of the shadows just feel "odd" even though I can SEE them in the photo, and have tried with several variations of the base color, as well as other colors. Neither technique seems to be doing me much good for the shadows in the bend of the elbow. I've tried everything from grey with a thin glaze of the flesh tone color over it, to burnt sienna with a glaze, to simply darkening the base flesh tone a bit at a time. I've even tried walking away from it for a couple of hours and then going back to see if I can't get a fresh perspective on the problem.


The shadows where the sleeve lays, and along the back of the arm, took 4 attempts before I quit grimacing every time I glanced at them. I'm even mostly satisfied with the shadows around the neck and chin area. That Smudge in the crux of the inner elbow was the latest attempt to get the area looking like I want it, and it's Still driving me nuts. Any assistance that you folks can offer by way of suggestions would be IMMENSELY appreciated.

This is the photo that is being used as a reference for the piece -
Obviously I've taken some artistic liberty in interpretation for my painted piece, but it's very important to me for several reasons that it turn out well. I'm determined to do justice to the memory of this woman who was incredibly important in my life.

[JamieWG]

Originally Posted by WFMartin

Cameras won't see it that way. Film won't see it that way, Why? Because, as you say, it's an optical illusion, and that takes place in your eye/ brain. No, it's not actually blue--it isn't there!

Bill, I respectfully beg to differ. See my pics. They are really there. They are not always as dramatic as those photos display, but I can't deny their existence!

Jamie