Artists inherit a landscape dense with “rules.” We are routinely advised to work from dark to light, to observe fat over lean in oil painting, that warm lights produce cool shadows, and that the subtractive mixing of any two colorants will result in a reduction of chroma.
Some of these guidelines are grounded in material science and long-term conservation concerns; others persist because they are useful enough under most conditions. What they share is not infallibility, but an attractive path to efficiency. These are heuristics. They are cognitive shortcuts that reduce complexity, lower error rates, and allow practitioners to function effectively without recalculating first principles at every decision point. Heuristics are indispensable in skill-based domains. Vision itself can be understood as “relying” on them. The human visual system can seem to be very accurate in many contexts, but the reality is that it operates on “what works” in the pursuit of successful behavior–not an accurate recording of the physical world. Artistic training mirrors this strategy. Rules emerge not because they are universally true, but because they are statistically reliable within a given range of conditions.
Because we rely so heavily on these cognitive shortcuts and “rules of thumb,” we are often surprised by (and at times resistant to) cases that violate them. A familiar example in painting is the assertion that any subtractive mixture of pigments must result in a loss of chroma. In general, this holds remarkably well. Mixing two chromatic oil colors yields a result that is almost always duller than either parent, an outcome readily explained by overlapping absorption curves and reduced spectral reflectance. The guideline is practical, predictive, and therefore widely taught.
But heuristics, by definition, are not optimal solutions. They are approximations. And approximations tend to fail at boundaries, particularly when material behavior departs from the assumptions embedded in the rule.
This article examines one such boundary case: pigment mixtures that initially increase perceived chroma rather than reduce it. Specifically, it explores how the optical scattering behavior of titanium white can disrupt the expected subtractive outcome when combined with certain transparent or semi-transparent chromatic pigments, producing a perceptual increase in chroma that appears to violate a foundational rule of color mixing. The aim here is not to diminish the indispensable role of heuristics, but to understand them with greater precision: where they function reliably, why they function as they do, and when their underlying assumptions begin to fail. Certain advancements in technical disciplines rarely come from discarding rules wholesale once their limitations are exposed; rather, they come from learning exactly where and why those rules may no longer apply.
Chroma: From Studio Experience to Color Science
Painters tend to develop an intuitive sense of chroma long before encountering a formal definition of the concept. We often colloquially recognize it as the “strength” or “punch” of a color, or how vivid it feels relative to a neutral gray or to neighboring colors. In everyday studio language, chroma is often lumped together with terms like saturation or intensity, and for practical purposes, this shorthand usually works. However, that intuition can sometimes get chroma problematically linked with another dimension, lightness. In reality, color can be very dark and still feel chromatically strong, or quite light and yet appear weak and washed out. This is why painters quickly adopt a useful heuristic that adding white tends to “kill” a color. As a general rule, tinting (adding white) raises lightness while reducing perceived intensity. The color gets lighter, but less vivid, or less chromatic.
Color science gives us a way to speak about this more precisely. In the CIELAB color space, chroma is defined as the radial distance from neutral gray. It is described as a radial distance because, in color models, colors are mapped geometrically. In geometry, a radius is a straight line extending outward from a central point. Chroma is therefore measured as the length of the line radiating from neutral gray to a color’s position in color space.
Within this framework, chroma tracks perceived color strength independent of lightness. It does overlap with what artists call “saturation,” but it isn’t identical: saturation is often described relative to brightness/value, whereas chroma is a geometric distance in a perceptually uniform color space. For many paint behavior scenarios, though, chroma is a practical stand-in for “color intensity.”
As nearly every painter discovers, experientially, under almost any condition, adding white (tinting) increases lightness and decreases chroma. This is one of the ideas that pushes many to prefer flake white for mixing over titanium white, as it seems to be less “impactful” on chroma.
This general tinting rule of thumb is reinforced by additivity of absorption (a.k.a. additivity of optical density), which describes what happens when two pigments share the same paint film: each keeps “soaking up” its preferred wavelengths. Together, their blocking powers stack, so the mixture covers a broader absorption range and reduces spectral contrast. With fewer sharp “highs and lows” left in the reflectance curve, the color usually looks duller (lower chroma).
But this relationship isn’t absolute. In certain boundary cases, chroma can briefly rise before it falls. It’s not a violation of the additivity of absorption so much as a reminder that absorption isn’t the whole story. To understand the bump, we have to consider scattering, film thickness, and how the eye encodes color, especially when a strong scatterer like titanium white enters the mix.
The Optics: Absorption, Reflectance, and Scattering
A common explanation for surface color is that it “looks red” because it reflects “red wavelengths” and absorbs the rest. This, like many low-resolution explanations and heuristics, can be incredibly useful but ultimately incomplete. We can get closer to what’s actually happening by imagining the paint film as a photon forest. When light enters, it doesn’t simply reflect or disappear; it undergoes a chain of events including interface reflection and refraction (i.e., tiny “bounce-backs” or “bends” where light hits a boundary like air to paint or paint to a pigment particle, so a little bit reflects or refracts before it goes inside), multiple scattering (i.e., light getting jostled and sent in new directions by tiny particles, like how fog makes headlights flare) by pigment and extender particles, as well as wavelength-selective absorption (i.e., the pigment “soaks up” some colors more than others often turning that light into heat while letting the rest continue). Some photons are back-scattered and re-emerge toward the observer; some are forward-scattered deeper into the layer; some are absorbed; and some transmit to the ground and are reflected back up. The aggregate of these interactions produces the paint film’s spectral reflectance (and transmittance), which, after the visual system’s own neural processing, yields a perceived color. In practice, models like the Kubelka–Munk model summarize this using an absorption coefficient and a scattering coefficient, whose balance determines how much light escapes the forest and at which wavelengths.
So to Recap and Reinforce:
Subtractive mixing stacks absorption profiles: each pigment removes the wavelengths it dislikes, so the mixture transmits/reflects what survives. That’s the ‘additivity of absorption’ idea painters hear: the mixture’s absorption is the (weighted) sum of the parts. Typically, this reduces spectral contrast (peaks shrink and valleys rise), pushing the mixture toward gray.
Absorption (K) is how pigments soak up certain colors more than others. Alizarin-type reds, for example, gobble green–cyan light but let more deep red sneak through. That “picky eating” is why they look red in the first place.
Scattering (S) is how pigment (and extender) particles jostle light into new directions. Every tiny boundary between a particle and the binder acts like a mini redirect, making light zigzag and spend more time inside the film. Titanium dioxide is an extraordinary player here: because it’s optically very different from the binder, it scatters light extremely well. You can think of it like a milkiness that brightens and diffuses without adding much color of its own.
Reflectance is the wavelength-by-wavelength fraction of incoming light that emerges back out after scattering/absorption.
So why could adding titanium white to alizarin crimson raise its perceived chroma?
Because adding titanium white doesn’t simply “add white,” it supercharges scattering far more than it adds absorption. That boost can lift and reveal a transparent red’s favored wavelengths (the ones it doesn’t absorb), reducing the “buried, near-black” look of the masstone. In this small window, the red can look cleaner and more vivid, with chroma slightly higher, before further whitening inevitably pastels it.
This isn’t just a lightness trick. The first touch of titanium produces a real, localized increase in spectral contrast: in alizarin-type reds, long wavelengths already have a survival advantage while green–cyan is strongly absorbed; a small increase in scattering lets more red escape while green remains suppressed. Consequently, reflectance in the red band increases more than in the green, the red edge steepens, and the spectrum changes shape (not a uniform lift as you’d expect from a simple rise in lightness). However, the rise in lightness does help you notice the effect, but again, it isn’t the real cause.
Compared to flake white, titanium white is a super-bright flashlight: a tiny bit makes everything lighter and, at first, can actually surprisingly boost chroma before quickly washing it toward pastel. Flake white is more like a dimmer switch: it raises lightness gently, scatters less, and often keeps the color’s character (and apparent chroma) intact over more mixing steps. That’s why many painters prefer flake white for subtle mixing, even though the titanium trick can give a brief chroma “pop” with certain transparent colors.
In closing, this isn’t subtractive color theory “failing.” It’s a reminder, like many heuristics that don’t allow us to have the whole picture, that we don’t see absorption alone. Scattering matters (especially in these boundary cases) and titanium white changes that balance in a big way. The critical point for painters is this: adding titanium white does not merely “add white.” It dramatically increases scattering far more than it increases absorption. This shift often reduces the dominance of deep absorption effects that suppress chromatic information in appropriate masstones. In certain cases, this enhanced scattering can unmask the pigment’s selective transmission and reflectance, revealing a purer undertone and, perceptually, a higher chroma.
Happy Painting!
