View Full Version : shadow rules invalid

06-25-2002, 01:51 PM
Have always heard 'shadows cool, highlights warm', but have been watching maple tree branches up close for months and can see that backlit and side lit temp can be reversed. Is anyone willing to address this?

Scott Methvin
06-26-2002, 12:56 PM
Shadows are complicated. They are the absence of light, but all kinds of reflections come into play.

I read somewhere that there are about 10 different kinds of shadows and 10 different kinds of highlights. Doing the math on all those permutations will make you pause.

Shadows done well aren't noticed. They make the highlight the highlights.

Like the copy in a well designed magazine ad.

06-26-2002, 03:07 PM
This truism is a classic example of over-simplification Linarty; we all need to keep in mind that just because there's a saying for it, it ain't necessarily so! Lighting conditions are infinitely variable and have to be assessed for what is there, not what we expect; or as it's more commonly put, paint what you see, not what you think you see :)

Let's look at this one for a start. If you think about a barn lit by a low sun, with a clear sky, then this will hold, since the lit side would be warm and the shadows full of reflected light from the blue sky. Now imagine a still life setup indoors, lit by a low-power incandescent bulb, say a 40W (or candles for that matter). Well here, without an additional lightsource or a cool object to throw off reflected lights, the shadows will also be warm. This is a truth most of us can see every evening in our homes but which we don't actually notice - in short, warm lighting does not automatically equal cool shadows, or vice versa.

Backlighting is a very special circumstance that can create most unusual effects, not least if there is any transmitted light which often is a very different colour to reflected light for the same object.


06-27-2002, 12:18 AM
Here is another conflict I've experienced...in a winter scene....I often see winter light as cool, and winter shadows as warm. It makes sence to me, but I've been told I'm nuts. ????

06-27-2002, 06:49 AM
Originally posted by Linarty
Have always heard 'shadows cool, highlights warm',
no. light (NOT HILITES) warm, shadow cool. this is for outdoors. indoors, it's more like light warm, shadows neutral. this is all VERY generalized, btw

but have been watching maple tree branches up close for months and can see that backlit and side lit temp can be reversed.
what do you mean by this,,,backlit and sidelit? backlit means the object(trunk, branch) is lit from behind. this is NOT the reflected light you see in shadows. sidelit means the object is lit from the side.

if this is what your meaning is, then i REALLY don't understand the question because the one concept, backlit/sidelit, has nothing to do with the other concept, light/shadow.

06-27-2002, 08:49 PM
Linarty, I also don't understand your first question; please specify what you mean. By 'backlit' do you mean the tree's base/shadow color (in other words, what you have before you put the highlight on)? And by 'side lit' do you mean the highlight?

I don't know if you're talking about snow, but in winter, I find snow shadows purplish blue under blue sky; more purplish than any of the sky (still don't know why). Around early sunrise/late sunset, with clear sky, snow shadows are sometimes a beautiful magenta. Even without snow, the winter sun is generally pretty low, so this might also explain warmer-than-usual shadows (perhaps more purplish than usual), although I haven't looked out for this specifically.

But I can't understand sunlight being cooler than the shadows...I never noticed that :confused:

cobalt fingers
06-29-2002, 04:26 PM
well, you may well be nuts but that's not for us here to say. Regarding shadows and temps these are general rules. For even these general rules to work, (by the way, the shadows generally define the light)...it sounds like maybe what you see with the maples is light coming thru the trees-thus they would be made transparent. There are tons of variables that can mess with the general "rules".

With snow it's really easy to read since the snow is "locally" white. White things read and define the temp color very nicely. The extrmes are the easiest to read naturally.

07-04-2002, 06:26 PM
Have been thinking about this...and here is my (quasi) scientific attempt to explain.....I'm sure part of it depends on where you are, but I think that maybe it depends on the literal temperature. When the humidity in the air is 'liquid', it refracts light, when it is frozen, it 'reflects'. So instead of breaking into prizum, it just reflects cool blue sky, leaving shadows relatively warmer

07-05-2002, 09:06 PM
In her book "Painting beautiful Skin Tones with Color and Light", Chris Saper discusses this very topic. Apparently at varying times of the day the sun's Kelvin Temperature bends the light between the red and blue hues of the spectrum. For instance at mid-day the sun would be cooler (more blue), causing the shadows to be warmer (more red). The opposite is true of morning and evening.

I've seen her post her before. If I've misunderstood what she's saying maybe she'll stop by to shed some "light" on the matter.:D

07-07-2002, 11:49 PM
Originally posted by Guy
.... For instance at mid-day the sun would be cooler (more blue), causing the shadows to be warmer (more red).

er,,,just think about what you just said,,,,or what SHE said, i mean.....{M}:rolleyes

07-08-2002, 01:13 AM
Heh, I meant the sun's hue not the blasted surface :p

cobalt fingers
07-08-2002, 12:40 PM
We should mention here that this is complex. At mid-day the sun is very bright. This effects much for a painter. The intensity changes probably more than the temp. of the light source say from 3:00 pm. The color shifts are more apparent when the skys cloudy-Schmid knows this and uses this knowledge-see his lovely warm shadows? (outside cloudy days)

07-08-2002, 08:09 PM
Tim, I think a clear distinction should be made between observable lighting effect and how painters elect to portray it, especially with someone like Schmid with such a pronounced style. I like Schmid's painting on the whole but the likes of Estes, Bateman or Chard would be far better pointers to accurate depiction - believe me I have far too much experience studying lighting on cloudy days and shadows ain't warm :D


cobalt fingers
07-11-2002, 07:32 PM
This need not be a disagreement, if you look at something white that has some single light source on it (inside outside, warm or cool) the shadow on that object will have a temperature. This does not consider reflected light and colors bouncing into this shadow. This is observable. Anyone reading this can pause and walk about a room or look outside and see this. The temp (color) of the shadow will identify the temp or color of the light source...it will be opposite: warm light -color shadow and vice/versa. Look at the "good" paintings by the Taos painters of horses in the sunshine.

A greater concern and something that ought to be talked about is that light in reality and painting are different in the we must imply light with pigments. The backlit maple are examples of this. We must depict the look of the light.

I really think Schmid's remarks are right-when I look about I see what he says everywhere. I see it in paintings for 3 centuries before he said it.

07-12-2002, 10:03 PM
Try this; get some colored light bulbs (I used aquarium bulbs) and some fairly large bones of some sort (I had a duck's keel or something like that.) If you light the bones on one side with a blue light, the "shadows" are yellow. Same with other colors of light. It's quite amazing and entertaining...

cobalt fingers
07-13-2002, 12:32 PM
It's clear and obvious-good example

07-14-2002, 08:41 PM
Originally posted by cobalt fingers
The temp (color) of the shadow... will be opposite...
No, I'm sorry it isn't. If this effect was true the shadows in candlelight would be cool, which we all know is not the case. In a less extreme example the shadows under incandescent light would be violet-tinged, which of course they are not.

Check for yourself. Take an egg, put it on white dropsheet. Light it with a single cool lightsource and, in the absence of a warm reflector nearby, the shadows will be COOL. Same setup, but with warm lighting shadows will be WARM.

In a bright outdoor setting such as in the Southwest, with a clear vault of sky, it's no wonder the shadows are cool - the primary lightsource for the shadowed areas is the blue sky. The effect is magnified in the morning and evening because of the "warmer" sunlight providing a larger hue contrast, which makes the shadows look ever cooler (lower value also tends to be equated to cooler hue in human vision). I believe it is an (unwarranted) extrapolation from this that results in this misconception.

Jaelle, you must be doing this with some white incident light as it is this that is providing the opposing light effect. If, on the other hand, these coloured lights are the ONLY lightsource you will see something very different - as you can see quite clearly under neon lighting if one keeps one's eyes open. Again, don't take my word for it, check for yourself.


07-15-2002, 12:16 AM
My 2 cents worth. Out of curiosity I tried Einion's egg idea...on a pure white sheet. I used 100W flood lights (one a slighly magenta-ish red, the other a cyan). There were no other light sources. Here it is:


In both cases the shadows seemed a darker, greyed version of the light source. So, with the egg under blue illumination, the shadow seemed darker and greyer. Since grey is warmer than blue, the shadow did look darker and slighly warmer here. I think this is where the notion of "cool light -> warm shadows" comes from; there seems to be some validity to it.

But I didn't notice the complement of the light source in the shadow; I was expecting to. But I forgot to look out for that...maybe if I looked for it I would've seen it. Then again, the pictures above should be close enough and I still don't see the complemt in the shadows, even though color theory often says you should see the complemnt of a color around it.

07-15-2002, 12:38 AM
I need a pair of 3D gasses.

cobalt fingers
07-15-2002, 11:02 PM
The eggs don't do it for me because we must see better than a camera and then there's the issue of the camera and or film blah blah blah..

I really admire and respect the disagreeing a parties here and I think I'll just live with my disagreemnt. Let's talk about religon or politics now...

07-17-2002, 02:49 PM

I can see additional research is needed; I definitely saw the effect, but your right, I can no longer remember whether I did it during the day with light coming in a window or at night. It's logical that with a single source, part would be lit with that source, and the other side would be simply dark. But we have air, and some incident scattering, I'm sure. If there was white light in addition, it would seem that the shadows would be a bit complementary by contrast; one side would be, say, bluer (if using blue light) than the other, because the blue would be added to the white light on that side. The thing that puzzles me was that the effect was quite spectacular; the opposite side was distinctly yellow, as though the blue had been removed from whatever light was there. Green light gave the shadows a red tint, yellow made the shadows very blue. Go figure. I'll have to try it again and take notes.

07-17-2002, 11:41 PM
Jaelle, I tried this again, but this time with an incandescent bulb in the background, on the opposite side of the colored, main light source. And it worked:


The shadows were quite a bit greener than shows in this photo.

07-17-2002, 11:45 PM
Egg under cyan main light source with incandescent bulb as secondary light source at opposite side:


Here the shadows clearly look orange-red. This is about how it actually looked. The white area on the left is because the colored coating on the cyan spotlight is chipped away in a few places; just ignore it.

07-17-2002, 11:49 PM
Ugg under incandescent bulb only:


Even though this picture doesn't show it much, the shadows looked a little bluer (darker and greyer) than the lit side.

07-17-2002, 11:58 PM
Egg under candlelight only:


The shadows are a beautiful dark brown...kind of like dark burnt umber.

cobalt fingers
07-19-2002, 02:44 PM
Domer a couple of thoughts

1 what kind of film and camera are you using? Our eyes always see better-truer color than camers at least for a few more months.

2 Natural light outdoor without added fill will show the effect. You might try the egg at sundown 15 mintutes before the sun sets and on a cloudy day at noonish outside these delicate effects should prove the points.

I can't wait

07-19-2002, 05:46 PM
Tim, all the pictures were taken with a digital camera. Yes...it doesn't show color very accurately (greyer and less contrast than eyes see).

If I have time I'll try the egg thing outdoors like you suggested.

Ruth Brigitte
07-26-2002, 09:01 AM
I've read these principles: Warm light, cool shadows. Cool light, warm shadows.
As I interpret this and as everybody knows, a good painting always needs contrast. This principle gives you the needed contrast. I've tried this in some value sketches, and it does work. It depends on what you want to focus on. If you have a scene that you want to depict accurately but put your artistic interpretation on, you follow the above. A summer's day can be depicted as hot by using the first one, and the same scene can be depicted as cool (like after a rain shower, or when the observer is supposed to feel the cold of loneliness, etc) by using the second one.
I think it's nice to observe the physical truths, but we as artists should use them only as guidelines and then use our paints in a way to depict a scene as we want the observer to see it. It's not just about realistic rendering, but about artistic interpretation.

cobalt fingers
07-26-2002, 09:00 PM
I don't understand. Values are in NO way way related to temperature-not directly anyway

08-04-2002, 11:07 PM
As I try to notice these things more in-depth to improve my own work, it has occured to me that...please correct me if I'm wrong here.........temperature doesn't really exist.

It's all just very subtle hue or saturation changes. Shadows SEEM, in a limited number of cases, to contain the opposite of the light source since the shaded area is so reduced in saturation. If the color is high in chroma, the shadow iwll seem to contain the opposite just by virtue of the high contrast between high and low chroma.

Put a neutral grey dot in the middle of a bright red swath of paint and it will look greenish, etc.

If a color is 'cooler' than the one on your pallette, your pile of paint can be cooled by adding blue........ i.e. creating a slight hue change, and possibly altering the saturation depending on how much orange is in the color you're using.

So, it seems to me that hue and saturation are the only real parameters.

Perhaps someone could educate me further on the topic if this isn't so?


cobalt fingers
08-12-2002, 09:32 PM
The intensity of the objects color will prove or display the shadow colors best (white for example) strong colored objects are hard to read.

08-15-2002, 11:45 PM

is there an example of 2 colors with the exact same hue, exact same value, and a different temperature?

It seems to me that in order to influence temperature you must alter the hue and/or intensity, and that therefore, "temperature" is just a term for subtle changes in these two properties.

I suppose it's a nearly irrelevant terminology question.


08-16-2002, 12:52 AM
Originally posted by nam26b:

is there an example of 2 colors with the exact same hue, exact same value, and a different temperature?

How about perinone orange and a thin application of burnt umber, thin enough to give it the same value. Same hue, same value, but the b. umber is cooler (closer to neutral, or its compliment, blue).

It seems to me that in order to influence temperature you must alter the hue and/or intensity, and that therefore, "temperature" is just a term for subtle changes in these two properties.

Yes, but value too. Supposedly a darker value of the same color looks cooler.

Virgil Elliott
08-17-2002, 10:28 PM
The color in shadows is influenced more by the color of the secondary light registering there than by the primary light, which does not register in shadows. On a sunny day, the sky may be considered a secondary light source. Reflected light is another, and carries with it the color of the reflecting surface, which registers in nearby shadow planes facing that way. This is readily observable if we place more trust in our eyes than in someone else's theory. Secondary light in shadow can be any color, depending on the color of the reflecting surface or on the color of the secondary light source, regardless of the color temperature of the primary light source.

Virgil Elliott

08-18-2002, 01:30 AM
Very true, Virgil. I'm working on a painting now with several colors of reflected light, which illustrates your point quite nicely.

I suppose I've digressed into a discussion of what exactly color temperature is, and whether it is quantifiable or definable in terms other than hue and intensity.


I've never seen this orange you mentioned, but I imagine any tube color would be of much higher chroma than what is essentially a glaze of umber on a white background (the thin layer would be of reduced intensity).

Orange being a warm color, a reduced intensity orange would naturally seem 'cooler' even at the same hue and value. Thus, this case also seems to be explainable in terms of hue and intensity. If I'm wrong about this orange, maybe you could post a poto? (I'm not looking to argue here, it's just that the subject interests me).


Drew Davis
08-18-2002, 12:43 PM
Not to pick nits, but it seems like the argument is purely a matter of terminology.

People generally describe color with three components, which I'll call hue, value (lightness), saturation (chroma, intensity).

The original question:

> is there an example of 2 colors with the exact same hue, exact
>same value, and a different temperature?

(Domer responds with an example in orange.)

>Orange being a warm color, a reduced intensity orange would
>naturally seem 'cooler' even at the same hue and value. Thus,
>this case also seems to be explainable in terms of hue and

Note the switch in terms from the first to second question. The first asks for a "temperature" difference with constant hue and value. Domer's example seems reasonable to me in that context. But then, we switch the question to hue and "intensity" (saturation, chroma). The earth tone will be less saturated than the tube orange, and I'd agree that that is indeed why it's cooler. But hue and saturation wasn't the original question.

"Temperature" is an imprecise and subjective term. You can move orange toward "cooler" either straight across the color wheel, lowering saturation (and perhaps keeping other attributes constant). You can also move toward "cooler" by moving around the color wheel, changing the hue while keeping saturation high. And one person's "cooler" might be another person's "warmer". Most people seem to think "coolest" is somewhere the blues, but I've heard some people that think the coolest color is in the greens. You can imagine the argument there as, say, cerulean moves toward green. Even if you agree on the approximate location of "cool", you still have the problem that there's two ways around the wheel to get warmer. Purple is warmer than cobalt blue in a different way than aqua is; but is aqua "warmer" than purple? And what about that desaturated path across the wheel? The word "temperature" is a handy quick reference to describe the effect of a color, but it's not sharp enough to split many hairs.

Some people are similarly sloppy with the term "value", just to add to the confusion. I've seen it used, particularly as part of a phrase like "color value" or "hue value", to refer to attributes other than light versus dark.

So, to return to the second question, if you have two colors of the exact same hue and saturation (varying only the value), is there a perceived temperature difference? My theoretical inclination from picturing such a thing in my mind's eye is that you could invent such a pair of colors. But then, my mind's eye is happily generating imaginary colors without a temperature difference, because that's what I'm thinking about. So that's a bit circular.

And if I try to make some test swatches, I'm going to run headlong into the limitations of mixing actual pigments. I'm not sure I can mix two real colors, varying only the value between them, to a high degree of precision. My white or black is going to have a neutralizing effect at best (there's some of the complement in it, right?), and in practice the "neutral" is actually going to have some slight color of its own. ("Warm" whites, "cool" whites, blue or brown "blacks".) Then, there's the masstone/undertone shift a lot of pigments have; they'll change hue as I lighten them. And some pigments (like the phthalos) aren't at their most saturated in their their most concentrated state anyway. If I try to correct for all those effects, then I could probably find a pair of colors where I feel no temperature difference, and the hue and saturation are similar. But then I'll just bump back into the problem that I've fine-tuned the color for my own subjective perception of "temperature", much like imagining colors in my mind's eye, and someone else may disagree. And if you whip out the photospectrometer, you'll probably find that I screwed up the hue angle or saturation a bit while tweaking. At some point the color-theoretical hair becomes too fine to support the burden of the limitations of humans and pigments in the real world.

I suspect at least some people will find lighter values "warmer" if for no other reason than the experience that hot things glow, and glowing things are often hot. This supposition falls into the "cool colors recede" category of generalizations that depend on perceptual experience, and which can be overridden by other differences in color and context.

08-18-2002, 07:30 PM
Yes, I'd agree with you , Drew......to quote myself.....

I suppose it's a nearly irrelevant terminology question.

I think that's been confirmed.

The first asks for a "temperature" difference with constant hue and value.

Yes, I should have used hue, chroma (intensity), and value in the original question. My apologies.

"temperature" is a handy quick reference to describe the effect of a color, but it's not sharp enough to split many hairs.

You say one person's cool can be another person's warm......i agree here, too.....this clears it up considerably in my mind.

Temperature isn't a real quanitfiable aspect of a color, or at least there's considerable "gray area" :rolleyes:

Thanks for taking the time for the tedious explanation. It's hard to believe I didn't put it together from all I have read about the subject.



08-18-2002, 10:03 PM
Delighted to see you here Virgil, I doubt I'm the first to say it but welcome to WetCanvas! I look forward to your input on a number of issues.


08-19-2002, 01:03 AM
Nathan, perinone orange (PO 43) is an organic, reddish orange (a carrot-colored orange). It's more reddish than cadmium orange, but not as reddish as scarlet. Perinone orange is very close in hue to pyrrole orange, another organic orange, although pyrrole is of higher chroma. They're both less opaque than cad. orange. The handprint site has info about almost all pigments. Click below, then click on 'orange':


I chose perinone orange and burnt umber because on the color wheel I look at most, the Handprint visual color wheel, these two pigments are identical in relative hue. But this color wheel is updated occasionally; in the next revision (if there is one), the author might have these two colors not perfectly lined up.


Here's a visual of taking an orange color and desaturating it, in varying steps, to neutral grey, keeping its hue and value constant. So the only thing changing is its intensity/saturation.


Bear in mind I did this visually, so the hues and values are not exact; just close enough. The dulled orange that is just next to the neutral grey is roughly what a thinned burnt umber would look like.

To my eyes, the orange does look less 'hot' as it's desaturated.

Drew Davis
08-19-2002, 01:45 AM
Sorry if I got too tedious. I have a bad habit of overexplaining things at times.

Love the egg pictures, Domer.

08-19-2002, 10:41 AM
Yes, thanks domer. Another helpful demonstration.


Don't worry.......I'm just dense enough to need tedious explanations at times.

Usually I can find the answers by reading, but not so on this one.

Thanks agin,


08-27-2002, 07:09 PM

Keith Russell
08-28-2002, 11:14 AM

Keep in mind, that the shadow of an object blocking the sun, is 'lit' by the blue of the sky. Shadows are where the stronger light source is blocked, but often there are other, secondary light sources, that cause the shadows to be 'lit'.

On a partially cloudy day, with clouds moving periodically across the sun, shadows don't 'appear', highlights do. When direct sunlight is blocked, everything is instead lit only by the sky.

If you have a bright cool light shining on an object, and a warmer, dimmer light shining on the same object, the shadow will be cast by the brighter light, but 'lit' by the dimmer light.

The shadow will be 'warm'.

(True, you won't encounter most of this stuff when painting en plein aire, but in urban landscapes especially at dust or at night, or indoors under artificial light, or a combination of natural and artificial, these things can happen.)


08-28-2002, 05:33 PM
I would like to encourage everyone here to purchase a blue light bulb and an orange light bulb or anyother color for that matter. Im not talking about bulbs that have a slight coolness or slight warmness. There is definately a complementary relationship between the light bulb's color and the shadow. Why? I dont know, maybe simaltaneos contrast? Who knows, Im only reporting what I have seen. Using only ONE light source, A blue bulb will cast an extemely warm shadow. Likewise a red-orange bulb will produce a blue shadow. The shadow colors are mutted but very very obvious.

08-30-2002, 03:07 PM
I have discovered something interesting.

When lighting an object with a colored bulb, the cool/warm relationships dont happen if the light is TOO close to the object. Im not sure why. If you pull the light away from the eggs maybe you can get the complementary relationship. I had to pull my light back three feet or more.


08-30-2002, 03:16 PM
I know this horse has been beaten up pretty bad but Im very intersted in this. I was reading some of the posts on page two and was wondering why the nuetral secondary light would make the shadow green in one instance and redish in the other. The obvious answer is simaltaneous contrast. I think Im going to punch a hole through some really thick card board for a peep hole. I can then look through the hole to isolate the shadows to get an idea of what hue thay are without all the surrounding color. Im sure everybody has been amazed at the illusions color interaction can create. It wouldnt surprise me if this is what was happening. It would go with the idea someone posted earlier about shadow not being able to dictate a color since it is only a relativity issue. Ill try and post my results.

08-30-2002, 07:34 PM
I have been checking out the colored lights again and came to an understanding that at least some of my statements earlier were wrong. The first time I did this experiment I must not have been careful enough. This time I made sure there was no chance of a secondary light and discovered that the shadows were the same color as the light only darker and grayer! But Im baffled and maybe we can figure this out. I don't understand why a neutral secondary light can pull the shadow towards the complement of the light. I tried isolating the shadows through a peep hole but cannot control the room enough to tell if its only simultaneous contrast. Anyone have any ideas? Maybe if you have two spot lights , one containing a colored bulb and the other a neutral, you can then build a big blind with a hole in it so that light can only be seen through the hole. You could then inspect the shadow color with and without a secondary light to see if it changes or remains the same. Like I said though some of you are correct in saying that WITHOUT a secondary light you cannot see a complement in the shadows. Interested in hearing what you think.


08-30-2002, 10:16 PM
. I don't understand why a neutral secondary light can pull the shadow towards the complement of the light.

Your eyes and brain do this. It's a real perceptual effect, but it's not in the physics of the light. It's a kind of image enhancement that happens in the early layers of image processing in the brain.

Your brain also enhances edges, and turns surface texture and light interaction into an understanding of surface orientation, using some very elegant cross wiring in many layers of neurons. The path from your eyeball to your aware mind is more like an assembly line than a pipe ;-)

There a many such effects - hellishly subtle ones - to which we don't have conscious access, and that are very hard to understand. Also, because of them, the physical models of light and color (notions like temperature and hue) are always incomplete - the other half is what your body does with all that stuff. And your body has a very intricate, complicated, organically-grown and not at all simple way of seeing.

08-31-2002, 01:44 AM
Originally posted by MikeN
This time I made sure there was no chance of a secondary light and discovered that the shadows were the same color as the light only darker and grayer!
See, didn't I tell ya? ;)

Originally posted by puzzlinon
Your eyes and brain do this. It's a real perceptual effect, but it's not in the physics of the light.
Good post puzzlinon, but this is a true effect. This is why peephole views of a scene work - if it was a relationship-only phenomenon the appearance of the shadows would alter markedly when viewed in isolation. Our visual system might amplify the effect but it is actually present in the photons.

I haven't done as much research on this sort of thing as I'd like but I believe there is a relationship to other effects like how headlights appear dim during daytime - photon interference of some kind. Quantum physics is beginning to explain this sort of thing in terms that are not rooted in 18th- or 19th-century science and it's really fascinating stuff, if a bit hard to digest at times.


08-31-2002, 02:12 AM
Yes, the physics is pretty subtle also, but, we were talking about complementarity showing up in a shadow, absent other effects. I really don't think that quantum effects can produce that.

In a test, there can be various reasons for it that come from the environment. For instance, someone mentioned that the effect seems stronger when the light source is further. When that's the case, the contrast is weaker, so the amount of ambient light in the shadow is comparatively high. If the room around you is a different color than the light source, then the shadows will have a different hue than the light source; you'll notice it more, because your color sense needs more light than just sorting out lights and darks.

Probably where the shift is coming from, when you try this at home, is simple: all the light in the shadows is coming from reflectance in the room around you. So its color will be different from the incident light color unless you have a completely neutral room. Your eye will try to let you know that the color is different, but (especially at low light levels) it has to amplify the difference to make you aware of it. It does that by shifting the color 'across' it's color space, raising the weaker signals and lowering the dominant one (so it emphasizes a difference from red, by raising the green and blue signals a bit and damping the red one a bit.)

A true test of the peephole theory (that objectively, shadows tend to the complement without ambient light or perceptual influences) would need:
a neutral room
a colored light
a neutral object
a neutral surface
an instrument measurement rather than eyes (eyes have things like color-temperature latency - they adjust slowly to diffences in ambient tone, and that effects your perceptions.)

In that 'pure' case, you won't find shadows tending to complements. But you might still see the effect if you look with your eyes, unless you're outside the room for a while so you de-acclimatize from the colored light.

To back that up with intuition about why the physics doesn't support it... complementary and the RGB/CMY color spaces are a product of the interaction of your eyes with the environment and the evolutionary accident that gave us three kinds of cones instead of two or four to distinguish colors with. The usual visual spectrum doesn't quite span an octave of actual frequencies, so it's not like harmonics and music; what makes a color complementary to another is a product of the machinery by which we perceive color.

Sorry for the long post. This stuff fascinates me.

cobalt fingers
09-03-2002, 08:14 AM
Keith's on the right track.

Virgil Elliott
09-03-2002, 12:46 PM
What is harmful is placing so much credence in a theory that we disregard whatever our own eyes and judgment tell us that contradicts the theory, and end up painting by formula rather than understanding. Most people think of "cool shadows" as meaning blue or purple, so if they accept the "warm light-cool shadows" theory, this is what they'll most likely put in their shadows if the primary light is yellow or orangish. Generally they'll exaggerate the chroma in the shadows out of proportion to what is in the lights and middletones, and overlook reflected color or color from secondary light sources, and the resulting picture will not read as convincingly realistic. It may exhibit a pleasing decorative effect in some instances, but it betrays a lack of understanding of visual phenomena on the part of the painter.

Absent secondary light, the shadow is unreadable. It is blackness. If we can see anything in the shadow, that means there is secondary light, either from a weaker or more distant source than the primary light source, or reflected from some other surface. If there is color in the secondary light source or in the reflecting surface, that color will register in the shadows illuminated by it, no matter what color or temperature the primary light source is. Light carries color.

Theories can be misleading, and are not a good substitute for perceptivity. Trust what your eyes tell you, and to hell with the theories.

Virgil Elliott

09-03-2002, 01:41 PM
You shouldnt put too much faith in either


cobalt fingers
09-03-2002, 06:57 PM
That was Virgil man,

he's saying it's not about faith at all that's the point-it's about observation!

Keith Russell
09-04-2002, 02:39 PM

If you aren't playing with your orange and blue lights in an otherwise absolutely dark environment, remember, you have three light sources--

--not two.


09-04-2002, 05:59 PM
Ok here is a question for you.

I have a blue light, an orange light, AND a nuetral secondary light. If the secondary light determines the shadow color how can a blue light cast a warm shadow and a orange light cast a cool shadow if its the SAME secondary light used in both cases?
Im stumped. If it was determined by the secondary light only, the shadows should be the same color in BOTH the orange and bue lights. Has anyone else witnessed this or is my observation flawed.

Virgil your comments are solid and nobody can deny the importance of observation. Of course copying what is seen in not necessarily knowlege. There are people who can accurately copy shapes and values even though they have no understanding of light and form. Repeated observation may lead us to understand patterns in nature which is very much a knowlege. If this is what you meant then i agree completely. People should not just blindly trust in theories without first hand experience or observation. On the other hand relying only on observation can have its disadvantages as well. Perceptions are sometimes only shadows of the real.

This particular situation is a bit baffling. Some people in this thread are correct in saying that in a controlled enviornment, with only an orange light, the shadow will NOT be cool. But! with the introduction of the secondary light source, some unusual things happen which is why I posted the question in my first paragraph. I think this is the question I would like the members to address. I think there is more to this then what has been mentioned on the thread.

Input or data anyone?


cobalt fingers
09-04-2002, 06:44 PM
Here's the thing--

light is complex as are edges color and numerous other things about painting representational art-so complex that trying to figure out many matters in your head will get you into trouble.

I once had a math teacher working with me on my house. He estimated the length of a board using trig and geometry. I then simply measured the distance. he was close but wrong. Color theory can be like that. Use your eyes and see what's there. It's the best way.

The "cool light-- warm shadows" refers to normal "natural" settings only.

Keith Russell
09-04-2002, 10:14 PM
Mike said:
"I have a blue light, an orange light, AND a nuetral secondary light. If the secondary light determines the shadow color how can a blue light cast a warm shadow and a orange light cast a cool shadow if its the SAME secondary light used in both cases?
Im stumped. If it was determined by the secondary light only, the shadows should be the same color in BOTH the orange and bue lights. Has anyone else witnessed this or is my observation flawed."

Mike, in comparison to a 'cool' blue light, your neutral shadow 'looks' warm--just as, in comparison to a 'warm' light, the neutral shadow looks 'cool'.


Virgil Elliott
09-05-2002, 12:32 AM

We build our knowledge of visual phenomena by observing and by contemplating what we observe. Drawing and painting are the most intense ways of observing, and of recording visual information in our memory banks, where it is filed for future reference as part of our stored knowledge.

If we trust some theory more than our own powers of observation, we're more susceptible to being misled. What do we do when faced with contradictory theories? Figure things out for ourselves, right?

I demonstrate what I was talking about to my students quite often, in this way: I place a strong light on one side of my white drawing cast, so there is light on one side and shadow on the other, then I hold an orange card up a few inches from the shadow side, and the shadows read orange in all planes facing that card. I then replace the orange card with something blue, and blue appears where there was orange before. I go through this with several other colors, and each tiime, the strongest color in the shadow is the color of the reflecting object, no matter whether the primary light is warm or cool.

Now, one might argue that it is the core shadow, aka the shadow accent, where less secondary light registers, where this theory of complementary colors to the primary light manifesting itself in the shadows applies.

In somewhere around fifty years of painting, whenever I've seen cool colors in shadows, there has been a secondary light source that was that color. Ditto warm colors in shadows. In low-light situations, such as indoors, the chroma of the colors in shadows diminishes, i.e., they become more grey. Outdoors in sunny weather, more color can be read in shadows, because there is more secondary light, from the sky and reflecting from surfaces where the sun strikes directly. The color of the reflecting surfaces influences the colors in shadows illuminated by this reflected light.

In your experiment, you are using brightly-colored light sources in an otherwise neutral environment, and your eye reads the neutral as the visual complement of the strongly-colored bright light. This is the same optical effect that gives us after-images of the complement color to a bright, colored light after we look away from it or close our eyes. It is similar to the point Josef Albers makes in his book, "Interaction of Color," where he shows how a neutral surrounded by a single color reads as the complement of the surrounding color. This is why a neutral grey reads as blue in a predominantly warm picture, and the same neutral grey reads brown in a predominantly cool picture. It is a matter of relativity.

In normal situations, however, our light sources tend to be less strongly colored, and reflected color or the color of a secondary light source is most often the strongest influence on the colors we see in the shadows. If we overlook that, because we think the shadows must be complementary to the primary light source, then the realistic aspect of our picture will suffer.


Originally posted by MikeN
Ok here is a question for you.

Virgil your comments are solid and nobody can deny the importance of observation. Of course copying what is seen in not necessarily knowlege. There are people who can accurately copy shapes and values even though they have no understanding of light and form. Repeated observation may lead us to understand patterns in nature which is very much a knowlege. If this is what you meant then i agree completely. People should not just blindly trust in theories without first hand experience or observation. On the other hand relying only on observation can have its disadvantages as well. Perceptions are sometimes only shadows of the real.

This particular situation is a bit baffling. Some people in this thread are correct in saying that in a controlled enviornment, with only an orange light, the shadow will NOT be cool. But! with the introduction of the secondary light source, some unusual things happen which is why I posted the question in my first paragraph. I think this is the question I would like the members to address. I think there is more to this then what has been mentioned on the thread.

Input or data anyone?


09-05-2002, 02:02 AM
Thanks for you response Virgil,

The term that I am familiar with in regards to your explaination is called Simultaneous Contrast. I agree that this is the most logical answer.


Virgil Elliott
09-05-2002, 02:17 AM
Originally posted by MikeN
Thanks for you response Virgil,

The term that I am familiar with in regards to your explaination is called Simultaneous Contrast. I agree that this is the most logical answer.


Mike, yes, but remember that your experiment is an extreme situation quite unlike what we see under any other circumstances. You required very brightly colored lights and a neutral setting to allow you to see this effect at all. In any other situation, the effect is negligible, and the stronger influences on the color in shadows will be secondary light.


09-14-2002, 12:07 AM
I have to say I agree with Cobalt Fingers: Warm light source = cool cast shadows. Cool light source = warm light source.

I used to think this was hogwash, until I had the following experience: At my work place we used to have a room illuminated entirely with GREEN LIGHT. What color do you guess the cast shadows were....you got it--PINK, at least dark with a PINK cast. Logic tells us the shadows can't be pink. Why? Because there is absolutely no other light in the room but GREEN. Could these cast shadows be influenced by other objects or the nearby wall or ceiling reflecting other colors of light into the cast shadow? Not a chance, because there's only green light in the whole room! Would a camera/film record pink cast shadows? Absolutely not! Why. Because they ARE NOT PINK.

It is an OPTICAL ILLUSION! And are optical illusions part of the way that we humans see things? I'll say. Therefore, I think I'll continue to paint cast shadows as (at least SLIGHTLY) complementary in color to the light source.

Just my humble opinion. BTW, Cobalt, You paint GOOD STUFF!
Recently caught your article in one of my art mags.


09-14-2002, 04:06 PM
Sorry for the typo/misstatement on my first line. It should have read: Warm light source = cool cast shadows
Cool light source = WARM CAST SHADOWS

10-15-2002, 09:52 PM
On the subject, check out Scott Burdick's article on page 24 in the November issue of The Artist magazine.

cobalt fingers
10-17-2002, 08:54 AM
...probably read this thread, and said hey I should write an article!

Glad to see he choose the right side of this thing to be on.:evil:

10-23-2002, 03:01 PM
Hey guys, I think the warm light = cool shadows and cool light = warm shadows is a nice formula for creating very interesting paintings by creating a nice contrast in hue. However, I think it's a rather bogus formula to *always* apply if you are aiming for reality. There's just as much chance of a warm light creating warm shadows if not more in most circumstances.

In a perfectly controlled environment with one source light, the shadows will always be BLACK. However, this is practically impossible to simulate since light bounces around everywhere, including the shadow area, and indirectly illuminates it to some degree.

If you have one warm light illuminating a controlled scene: it's more likely that shadows will be warm if you have just a white object such as an egg. The only case in which the shadows will be cool is if the warm light contains some blue in it and there's also an object in the scene which reflects blue light and absorbs most of the red/green from the warm light (perhaps if you are wearing a blue shirt). If the light consists only of warm hues (which in themselves are debateable), such as a *perfectly* red light, there is no way that the shadows will contain any blue since there's no blue light of any sort in the light source to get reflected into the shadow. Even if there are objects which reflect blue light in the scene, they won't have any blue light to reflect (to work with you might say) - they'll simply absorb all light and appear perfectly black (this rarely happens in reality since every light, even if it appears perfectly red, might contain a tiny hint of green or blue).

The same holds true for cool lights - cool shadows will generally be the result in a controlled environment and can only become warm if there is some hint of undetectable warmth in the cool light and there exist warm objects in the environment. Either that or there must exist a refractive object in the environment which can disperse the light and separate the cooler components of it and cast/bounce that into the shadow region.

Even so, in more realistic circumstances, there's often more than one light source and there can exist multiple objects which reflect certain hues and absorb others to create an inifinitely variable number of combinations. Everything from the color of your shirt to your skin tone affects the hue of objects and shadows all around you in some form (though perhaps indiscernible by the naked eye in many subtle cases since even a trained eye cannot detect extremely small changes in frequency). If you don't believe this, look in a mirror and you'll see that everything in the room that is within the mirror's line of sight will be bouncing light to it (in other words, you'll be able to see the reflections of these objects in the mirror). I've never seen a case where I took an object I could see and put it in front of the mirror and didn't see it in the mirror(with the exception of vampires, of course). This is pretty obvious, but when the light being reflected off of things is not being transmitted to such reflective surfaces as mirror, it may not be so visible - if visible at all to the human eye.

When you add objects that can refract light into the scene, the situation becomes even more complicated as light can be dispersed into its separate frequencies.

I can say this with confidence: 100% blue light sources = blue shadows or neutral (black). 100% red light = red shadows or neutral. 100% green light = green shadows or neutral. However, we usually don't find 100% and even a slight amount of cool in a warm light or warm in a cool light can be enough for a cool/warm object to reflect it into a shadow area and perhaps a keen eye will be able to detect the slight changes. Light doesn't magically appear out of nowhere - if you have a truly cool light with no warmth of any sort (say a 100% blue light), no warm colors are going to ever get reflected (unless you consider certain *values* of solid blue to be warm - note, I don't mean variations in hue/saturation since the solid blue light won't permit any such variations). Only blue will be what you see.

A shadow is simply an area which is not illuminated directly by a source light. However, it can be illuminated indirectly by the same source light by bouncing off of objects: that's why we hardly ever see a perfectly black shadow even when there's only one light source. As a result, there are good chances that the shadow area will have a fairly different hue in a single light source environment since the shadow has to be illuminated by light which is bounced off of objects from the source light in order to reach the shadow area.

When you introduce secondary light sources, the shadow areas can appear to take on the colors of the secondary light sources at times far more than the areas illuminated by the primary light source because the primary light is not reaching the shadow region but the secondary sources are: creating the greatest hue contrast in the shadow areas between the primary and secondary light sources.

I think the easiest way to even begin to try determining the hue offset of a shadow from the local color of the object the shadow is casted on is to observe what secondary light sources might be affecting the shadow region and also take into consideration the reflective color of the objects that are nearest to that shadow area. Note that this isn't easy - there could be some hidden object way in the corner of a room or even a slight opening in a wall or a single ray of light shining in from a window which can be affecting the region.

Perhaps the best way of all is simply to free yourself from all these conceptions which will inevitably bias your perception.

BTW - there can be plenty of pink in green light. Pink is simply desaturated red light: green can contain plenty of it and still appear very much green.

Look at this green formed by combining red light and green light - the green at the top left and in the center of those two swatches are the result of the mix of roughly 80% green light and 48% red light (more than half the light consisting of red for exagerration to show how a green can contain plenty of red in it and still appear green - not even half that amount of red is necessary). The green formed is RGB(123, 206, 0) on the computer which is basically 48% red and 80% green.


Note that I'm working with the additive primaries of light which is red, green, and blue: not the subtractive pigment primaries that artists typically work with. If you were to take spotlights of the same color and intensity as the red and green boxes in my picture and shone the two spotlights on the same spot so that they are overlapping in a similar fashion as my picture, the overlapping area would be the same green color you see at the top left and in the overlapping region of my picture. If you were to increase the intensity of red to match the green, you would get yellow.

Don't believe me? Prove me wrong :) This particular intensity of green and red is difficult to simulate through real lighting, but simply take any red spotllight that's not too intense and combine it with an intense green spotlight and you'll see a pretty clear green in the overlapping areas even though it contains a good amount of red. Flashlights with filters also work.

This green formed contains far more than enough red light to create a slightly pink illumination in the dark shadow areas from a series of reflections off of objects (they don't even necessarily have to be red - the combinations of lights being cast could cast a pink through a myriad of reflected light being mixed together).

While modern optics is a highly developing branch of physics, there is plenty of research which proves this is how it works.

10-23-2002, 03:41 PM
Here's another example if you aren't convinced:


The green in the middle (RGB 148, 255, 82) was formed by mixing the green at the left (RGB 0, 255, 82) which also contains a small amount of blue with the red (RGB 148, 0, 0) at the right. When mixing with light, all you have to do is add the amount of red, green, and blue together. It's much simpler than mixing pigments, which is a subtractive process which is further complicated by chemical properties. If you want to see a clear demonstration of additive light mixing, shine a red, green, and blue spotlight at a wall so that each light is overlapping and you'll get white in the overlapping area.

The dark pinkish color was formed simply by taking red, green, and blue light of the green formed from the mix at the center and taking away some of the green and blue in it along with a small amount of red (RGB 123, 82, 82). This simulates the act of objects reflecting light where they will absorb some of the light and reflect the rest. Indeed a dark pink color can be formed - and this isn't necessarily a rare case either.

So, let's assume we have a green light that matches the mixed green in the middle of the 'purer' green on the left and red on the right. The light is cast out in the room, bouncing off all sorts of objects. Some of the objects that aren't green will absorb some of the green in the light, reflecting back red/blue light. This creates a situation in which pink shadows can be formed from the combinations of reflected light if there simply comes a situation where red becomes slightly more prominant.

It's hard to believe, I know, but take any digital paint/photo-editing program and check out the rgb values of the colors in my pictures with a color dropper tool.

Oh, I should also mention that if you ever have a true 100% green light and have no other lights in the room, you will never be able to tell if anything is blue or red at all without knowing ahead of time what color the object is. Everything MUST appear green, dark green, or black since there's no blue or red light to be reflected back - you won't see even a hint of blue or red in anything.

If you want to perform a test to see if your light is really 100% green or close enough to 100% so that you won't be able to detect any other colors reflected from objects, have a friend buy an object that one can't associate color with (like a sphere or something - not like an orange which we can usually assume to be orange) that's the brightest, reddest object or brightest, bluest object that he/she can find and don't let him tell you what color it is. Have him put the object in a box. Now, in your room illuminated by only this green light, open up the box and see if you can tell what color it is. If you can detect even the slightest hint of blue or red and the object doesn't appear jet black, then your green light is imperfect enough for objects to reflect different colors.

Another test to see if your light source is pure green that's more accurate than trying to see if we can perceive different colors is to take a red color filter and shine your green light through that filter. If it is truely green, no light should shine past that filter since it will filter out everything but red and there must exist no red light in the filter if your green light is pure. You can do the same thing with a blue filter and the same thing should happen. If it's a spotlight and you put the filter over it, it shouldn't even show a hint of light (now you can see this is practically impossible for almost all real-world lighting).

cobalt fingers
10-23-2002, 03:53 PM
I disagree with nearly everything you just said, respectfully

10-23-2002, 04:02 PM
Now to acknowledge the warm light cool shadows, cool light warm shadows rule, here's a case in which it works:


The orange light can be formed by mixing the three lights at the bottom: note that it does contain a small amount of blue light from the bottom left - enough to be very clearly blue if it was to be reflected into a shadow area. Thus, in such a case, even warm light in a controlled environment can produce cool shadows.

I hope I'm not annoying anybody with all these examples but I feel the necessity to demonstrate these circumstances as proof of why we might see things that don't necessarily agree with the lighting. Orange lights can contain blues, green lights can contain reds and blues, blue lights can contain red and green, etc. and all without affecting the hue strongly enough to detect when the light is mixed. However, when these slight amounts of foreign colors become separated from the light through reflective and sorbefacient processes, they can be seen as clearly as the three separate colors in the above picture where, for example, the blue on its own is very obviously blue :)

Who would have known that there was blue in that orange light above? There was, nevertheless, and therefore any object which absorbs red and green light and reflects blue light will appear blue in a room lit by this orange light because the blue color is being reflected to our eyes. If the object can reflect blue to our eyes, it can also reflect that blue light into the shadow region which is not directly illuminated by that orange light and hence we can get a cool shadow region. :evil:

10-23-2002, 04:43 PM
Originally posted by cobalt fingers
I disagree with nearly everything you just said, respectfully

I should mention that none of the things I've mentioned are original. They are all things I've learned from physics papers and textbooks.

However, I don't think physics is of any use to an artist if an artist, such as yourself, can paint such lovely paintings while completely disbelieving nearly all the basic properties of light as proposed by physicists throughout the world. :D

I believe I can, however, prove many of the things I've said. Perhaps I can get the equipment some time and demonstrate some of these principles.

Here is a very basic introduction to ray optics targeted at a high school level which is rather simple but explains the concepts very clearly:


Perhaps we can start with just one of my statements and debate that. That'll prove a lot simpler than trying to dispute everything I've said. How about the proposition that in a controlled room where no other light can enter with a perfectly solid red, green, blue light source, the only hue you will be able to see will be the same as the light source (depending on what light you picked - solid red or solid green or solid blue) of varying intensities but same level of saturation? That'll give me something to start with.

I'm open to people disproving these statements. I am ready to learn the truth, provided there is some solid evidence and research to go along with it.

After all, those scientists can be plain silly sometimes. For example, there are scientists who still think that frogs with no legs can retain auditory senses!

I know for a fact that this is not true, having performed the experiment myself once where I chopped one leg off and told the frog to jump. That it did. I very carefully measured the jumping distance, then chopped another off leg and told it to jump. It did so again. Then, I chopped another off leg and told it to jump. This time it didn't jump. I shouted very loudly "JUMMPPPP" and finally the frog jumped halfheartedly, not certain whether I said jump or take a dump. Finally, I cut the last leg off and told it to jump. It didn't budge. I yelled to the top of my lungs "JUMP JUMP JUMP" and nothing happened. Therefore, I knew then for a fact that frogs with no legs are deaf.

10-23-2002, 08:02 PM
On the note of optical illusions, there does exist one where the eyes can see complementary colors of light through an afterimage - especially when gazing into dark areas or closing the eyes after staring at a brightly lit area. I remember reading about it from my psychology classes. Even so, I don't think it applies to these cases as it requires you to stare at a brightly illuminated source for a fairly long duration then immediately look at a different area for the afterimage to be seen. It's the only case, however, in which I think a person will be able to see other colors from a pure source of green light.

cobalt fingers
10-23-2002, 08:53 PM
1. brevity is the soul of wit

2. In representational painting, if it's wrong and it looks right-it's right. If it's right and it looks wrong-it's wrong. Few painters that don't teach art in college will agree that shadows are made with black.

10-23-2002, 09:38 PM
With the first statement, I should say that I don't think it applies to my propositions. My statements refer simply to the properties of light (that the complete absence of light creates what we call black, and in an ideal, controlled situation where no light can bounce into a shadow region, the shadow will be black). They have nothing to do with the mixing of pigments. I have not read any scientific research to prove or disprove anything about the use of black pigment when painting shadows.

I personally think that black is necessary if you want to accurately capture certain situations in reality, but that's just my personal belief and should be detached from all my other propositions which are based on information which I learned from my science courses.

An example in which I think black pigment might be necessary: go to a room that's sealed shut and does not allow any light to enter. Turn off all the lights in the room. Now, try to find a mixture of paint without using any black in the mix to accurately capture what you see.

As for the second statement: I'm not very witty, unfortunately, and feel the necessity to explain my reasoning for every statement in excruciating detail under such circumstances, along with the use of repetition to reinforce what I'm saying (due to the lack of brevity in which people might forget what I said earlier).

Virgil Elliott
10-24-2002, 03:25 AM
The dark side of the moon is a good example of a shadow with no secondary light. Its blackness is visually indistinguishable from the blackness of space beyond it. Black is a useful mixing color if one knows how to use it, as the greatest painters have been doing for centuries.

cobalt fingers
10-24-2002, 07:00 PM
I think places w/o some reflected secondary light source (absolute black/dark) is about as commom on this earth as absolute zero. All this conversation 9at least my part refer to natural and mostly outdoor conversations. These challenges are that which I'm mostly interested in and are quite enough of a challenge for me.

Light and pigmnet as we know, are two very different things-in pigment we, with opaque paint-- must imply light.

Virgil Elliott
10-24-2002, 10:36 PM
Indoors or outdoors, it's important to understand what affects the color of shadows if the objective is to create the illusion of reality. Place a white horse on a green pasture on a sunny day, and you'll see green in the shadow on his belly and on all planes on that horse facing the ground. If he's standing on red dirt, you'll see that color in the shadow on those same surfaces. Lay an orange blanket on the ground under him, and you'll see orange on his belly. The bluish shadows will be confined to the areas in shadow facing the sky. The sky is a secondary light source. Reflected light off the ground is another secondary light source. When light reflects, so does its color. Stand your white horse next to a red barn on the sunny side, and the red will register on the shadow side of the horse on all the planes that face the barn. This is simple enough to test, if you have any white object and a few sheets of colored paper of different hues. I've been showing this to my students for years, using white drawing casts.

Theories and axioms can be very misleading. It's best to test them to see for ourselves whether or not they are borne out in reality.

Virgil Elliott

10-25-2002, 12:40 AM
Another article addressing this issue in Dec 02 issue of American Artist magazine.... starting on pg33 about theories of Camille Przewodek. Cobalt Fingers' statement may be correct, thread may have brought alternative ideas into the open.

cobalt fingers
10-25-2002, 10:25 AM
I agree completely. I stopped pondering long ago and started simply looking or setting things up and looking. When I see the studios and piles of sketches of artists I long admired it's always reassuring to see tons of color sketches. John Clymers sketches were to me more interesting than most of his finished works.

10-26-2002, 02:39 PM
Originally posted by cobalt fingers
I stopped pondering long ago and started simply looking or setting things up and looking.

Tim, I've been through this thread a few times now (searching for understanding) and have pulled out a lot of wisdom. The thing that keeps bothering me--I'm not sure what I'm missing--is in the example you post. Do you mean for us to understand that this lovely girl naturally had such glowing violet shadows on her face? To me it reads as the (much mentioned) secondary light source, and an unusual one at that, almost a blacklight, or fluorescent light bouncing off a strongly purple side wall or drape. But an effect, and perhaps a "pondered" one, that for me draws too much attention to itself.

Do you offer this as an example of "warm light/cool shadow" or "simply looking" or "trust what your eyes tell you" (Virgil) or "if it's wrong and it looks right-it's right. If it's right and it looks wrong-it's wrong." I don't mean to pick on you so directly, but your comments/example have left me befuddled.

For my own work, I tend to agree with what Scott said in the second post of the thread: "Shadows done well aren't noticed. They make the highlight the highlights."

cobalt fingers
10-26-2002, 02:50 PM
That sounded so earnest let me try and explain. I paint what I see because all else is just unreliable guess recall. Light is too complex to figure out w/o looking. That example was posted to show that black in shadows is pretty rare-even the famous Carravaggio's dark passages had some color and fill. Many of my other paintings have displayed my methods but here are two more that argue for the painting what you see from life concept.

cobalt fingers
10-26-2002, 02:52 PM
maybe these help. I'd never have known what color any of these shadows would have been had i not worked from life.

10-27-2002, 07:01 PM
Hello Tim,

There is no doubt that direct observation is important for someone looking for such a high degree of realism. This kind of realism may not be obtained without direct observation.
However, there are advantages to knowing, even a little, about how we percieve forms in space? Dont you think? Im referring to how our minds order shapes and values on a two dimensioanl plane. Do you draw on any such knowlege at all when painting or do you ONLY trust your eyes? Do you ever change what you see to make it more readable for the viewer?


cobalt fingers
10-27-2002, 10:41 PM
Last question first, yes I do move stuff around and leave stuff out. If we don't our work will often be busy.

I think that all learning and knowledge is interesting and fun. I love to listen to anyone tell me about how they do something or how a thing works. The discussion about light was interesting too-even the part that I feel is not relevant to oil painters. As for the act of painting what we see, maybe trying to understand a thing can use mental power and distract from the work of seeing.

Art is not a science. Artist see and feel and create.

10-28-2002, 05:02 PM
Ahhh, more examples. I guess I was hoping you'd explain in more detail than "not black" why you chose the violet in that particular painting. I'm not going to argue about painting from what you see, but you cannot deny that one makes choices every second while working from life. We can see from the work you've posted that you're a very accomplished painter; what you have to say about how you make those particular choices could benefit those of us who are less skilled. But thanks for the response.

MikeN, along the lines of your questions: what I see is not a picture; the picture is what I make of what I see. Whatever Tim insists, a good painter (as he is) is making a picture and not just recording what he sees. Careful looking tells you a lot--but not how to compose, not what's important, not what's meaningful, not how to feel. A good painting does all that. While still looking like a pretty girl or a pot on a table. JMHO.

10-28-2002, 06:16 PM
Originally posted by mirza

Whatever Tim insists, a good painter (as he is) is making a picture and not just recording what he sees. Careful looking tells you a lot--but not how to compose, not what's important, not what's meaningful, not how to feel. JMHO.

Hello mirza,

Thanks for your reply but I never disagreed with the above statement. I am in absolute agreement about the importance of personal interpretation. This is the core of any piece of art. Machines copy we compose.

And yes I agree that Tim is a good painter.

My point is that knowing how forms repeatedly act in certain situations can help us to CHANGE what we see to better fit our personal vision. Also, sometimes what we see is a spatial absurdity and knowing why can help us to fix it. We have the power to make the subject more "readable" to the viewer.

I also believe that introducing these things to students can help them to "see" better. For example knowing to at lest look for reflected light whether or not its there. Or, that as objects move back in space they become smaller optically. These things help to open their eyes. These are the reoccurring things Im referring to.

I use this sort of information in my own work to help communicate my personal vision.

I feel like Im speaking another language.


10-28-2002, 06:23 PM
One other point,

This drawing is a three hour demonstration for some students and the gourd at the bottom right was only imagined to be in the still life *for my compositional needs.*

11-03-2002, 08:21 PM
I sort of stumbled onto this thread and have read it with interest.
Just out of curiousity, I took those red and cyan eggs and posterized them to only 8 values, to see what colors the shadows would be. Computers are too stupid, (or too smart) to be fooled by optical illusions.
Here's what I found.

11-03-2002, 08:21 PM
Now the red one.

cobalt fingers
11-04-2002, 09:28 AM
You write well and observe sharply, here's what I don't understand. You dislike my use of the rich reflected light on page 2 and then state on page 3 that we should all make personal decisions as artists. Which is it?

11-04-2002, 11:54 AM
You seem to be speaking to me, but I think you must have confused me for someone else. My first post is the one just before yours, right here, and I expressed no opinions whatever.

cobalt fingers
11-04-2002, 01:50 PM
I am sorry, I should have said Mirza.

Keith Russell
11-04-2002, 03:03 PM

Has anyone read Howard Pyle's wonderful insights into this very subject?

(I have an excerpt from a book, taken from one of Pyle's students. Good stuff!)


cobalt fingers
11-04-2002, 06:19 PM
Please share, he's a great teacher and had influence upon many great artists.

11-04-2002, 07:54 PM
I found this thread very ... illuminating :D

And it strikes me as a bit funny that there is disagreement. I think ikeda explained very well how the laws of physics work and the effects of light and shadow. At the same time anyone can see that it is not really possible to calculate the "color" of shadows unless you know the exact temperature of all light sources, the exact temperature and reflective properties of every object in the scene. And even if you knew all that, It will probably take you too long to do the math :D

So, you do have to observe what it really looks like, and accept what you see without any judgement if it is the right color or not. If you see it, it is the right color. The laws of physics made it that color.

11-04-2002, 07:57 PM
Oh, and here is another rule of thumb.

If your paint by number kit has n colors, you simply add n/2 to the color number to get the shadow color number, and if the number is greater than n subtract n. Works wonders :D

The math for W&N oil color numbers is more complicated and is left as an exercise.

cobalt fingers
11-04-2002, 09:25 PM
[QUOTE]So, you do have to observe what it really looks like, and accept what you see without any judgement if it is the right color or not. If you see it, it is the right color. The laws of physics made it that color.

dang, 2 months of discussion down the drain!

The trips half the fun!:evil:

11-12-2002, 03:07 AM
Originally posted by henrik

So, you do have to observe what it really looks like, and accept what you see without any judgement if it is the right color or not. If you see it, it is the right color. The laws of physics made it that color.

Its about tapping into a language with the viewer. If what is seen at one particular moment communicates sufficiently we might consider using it. If it doesnt we should change it to make it better. There are times when the laws of physics, at a particular moment, do not communicate enough to the viewer. We may have to draw on less specific experiences.

Keith Russell
11-13-2002, 07:20 PM

I'll try to remember to bring the Pyle excerpt to work tomorrow.

(I've been moving over the past month, and the studio is in far more disarray than usual...)


cobalt fingers
11-13-2002, 08:51 PM
I read about NC's student days of studying with Pyle. It sounded heavenly-read that someday-it's poetic.

There's a great story about Harvey Dunn too correcting a guy at a banquet in honor of Pyle-good story.