There are two kinds of primary colors that are often confused. They can be called “physical primaries” and “visual primaries.” Sets of three “physical primaries” in the form of lights, paints, or inks can be mixed to produce a comprehensive range of other colors. These sets are “primary” because they have been found to deliver the most extensive and useful range of other colors. Colors are also recognized as “primary” by virtue of their appearance. These are the “visual primaries.”
Much of the confusion that surrounds the topic of color can be exposed in a discussion about primary colors. Primary colors, for some people, are just colors that are particularly vivid. For those who work with color, there are two kinds of primary colors that are often confused. They can be called “physical primaries” and “visual primaries.”
Sets of three “physical primaries” in the form of lights, paints, or inks can be mixed to produce a comprehensive range of other colors. These sets are “primary” because they have been found to deliver the most extensive and useful range of other colors. The particular lights or paints that work best can be recognized by their appearance: for lights, they appear orange-red, yellow-green, and violet-blue, but they are usually identified simply as red, green, and blue. The mixing process for lights is additive. For paints and inks, the mixing process is subtractive. Best results are obtained with paints or inks that appear purplish red, slightly greenish blue, and yellow. These are best identified as magenta, cyan, and yellow, but they are often loosely identified as red, blue, and yellow. The reds and blues for additive mixing appear different from those for subtractive mixing, but confusion can arise because of the range of different “reds” and “blues,” where an orangish red and a purplish red are both “red.”
Colors are also recognized as “primary” by virtue of their appearance. These are the “visual primaries.” In the continuous circle of hues, there are four that are recognized as “unique”: a yellow that appears neither greenish nor reddish, a red that appears neither yellowish nor bluish, a blue that appears neither reddish nor greenish, and a green that appears neither bluish nor yellowish. These visual primaries serve as conceptual reference points and, together with white and black, anchor the Natural Color System (NCS), which orders colors according to how they appear. Confusion can arise when visual primaries are thought of as the same thing as physical primaries. Confusion can be overcome when the distinction is maintained between visual appearances on the one hand and, on the other hand, the lights, paints, and inks which give rise to those appearances.
Before going deeper into the two main questions – what are the primary colors, and what makes them “primary” – it will be helpful to consider the nature of color itself.
The Nature of Color
Color means different things to different people. There are the colors that physicists measure, the colors that chemists analyze, the colors that psychologists study, the colors that artists brush onto paper or canvas, and the colors that people talk about in general conversation. All these may be related, but they are not the same kind of thing.
It is possible to identify seven different “kinds of color” , but for present purposes a broad distinction can be made between the physical aspects of color and color as it is experienced in perception. In general conversation a rose or the setting sun might be described as red, and for most people that red would be a physical property of the rose or the sunlight. They might also recognize as red the subjective visual experience they have when they look at the rose or the setting sun. The one word “red” is used for two different kinds of thing. The lack of distinction between the physical and visual aspects of color is responsible for much of the confusion that exists in connection with color in general and with primary colors in particular.
Confusion Between the Physical and Visual Aspects of Color
The confusion is evident in what might be assumed to be an authoritative definition. In the Concise Oxford Dictionary  a primary color is defined as “any of the colours red, green, and blue, or (for pigments) red, blue, and yellow, from which all other colours can be obtained by mixing.” In this definition the meaning of the word “colour” changes halfway through. Pigments are physical substances. A particular pigment might appear red, but that redness is not the same thing as the pigment itself. It can help to remove the confusion if (physical) and (visual) are added after the word “color” to clarify the distinction between color meaning something physical on the one hand and color meaning visual experience on the other. Then that definition could be reworded: “any of the colors (physical) … from which all other colors (visual) can be obtained by mixing.”
The same change of meaning can be recognized in another definition, this from The Color Compendium : “Simply stated, primary colors are ones that cannot be mixed or formed by any combination of other colors.” This could also be reworded: “… primary colors (visual) are ones that cannot be mixed or formed from any combination of other colors (physical).”
Evolution of Ideas about Primary Colors
Although the definitions above confuse the physical with the visual aspects of color they do illustrate the two main ideas about what should be understood by primary colors. One is that primary colors are physical and are used in mixing processes; the other is that primary colors are visual and serve as perceptual and cognitive reference points. The evolution of these ideas can be traced through history.
In the fifth century BC Empedocles recognized four principal colors as corresponding with the four elements: black, white, red, and green with earth, air, fire, and water [4, p. 15]. For Aristotle white and black, combined in different proportions, were the source of other colors. In a line between white and black he identified five other principal colors – yellow, red, violet, green, and blue. “… from these all others are derived by mixture.” [5, p. 698].
Leonardo da Vinci
white is the first among the simple colors, and yellow the second, green the third of them, blue is the fourth, and red is the fifth, and black the sixth. And white is given by light, without which no color may be seen, yellow by earth, green by water, blue by air and red by fire, and black by darkness … [6, p. 70]
I call simple colours those which are not compounded and cannot be compounded by means of a mixture of other colours … After black and white come blue and yellow, then green and tan, that is to say, tawny, or if you wish to say ochre, and then deep purple and red. These are the eight colours and there are no more in nature, and with these I begin the process of mixing, first with black and white, and then black with yellow … [6, p. 72]
It seems likely that these eight colors represent the pigments which Leonardo used in his paintings. His eight simple colors would correspond with physical primaries, but their number would have had more to do with the limitations of available pigments and his own approach to the practicalities of painting than any of the theories of primary colors that are more familiar today.
… there are but few Simple and Primary Colours (if I may so call them) from whose Various Compositions all the rest do as it were Result. For though Painters can imitate the Hues (though not always the Splendor) of those almost Numberless differing Colours that are to be met with in the Works of Nature, and of Arte, I have not yet found, that to exhibit this strange Variety they need imploy any more than White, and Black, and Red, and Blew, and Yellow; these five, Variously Compounded, and (if I may so speak) Decompounded, being sufficient to exhibit a Variety and Number of Colours, such, as those that are altogether Strangers to the Painters Pallets, can hardly imagine. [7, pp. 219–220]
There are two significant features of this passage. First, it shows that the possibility of producing a complete sequence of hues with only three paints (physical colors) was well established. Second, it introduces, by implication, the concept of color gamut. A gamut is the range of colors (visual) that can be produced with a given set of colors (physical). Although with Boyle’s Red, Blew, and Yellow it was possible to produce all the hues, it was not possible to produce all the colors – colors of greater “splendor” were out of reach. With Red and Blew it was possible to produce a range of purples but not the vivid purples of some flowers. The word “hue” is often used as a synonym for “color,” but in a discussion of primary colors it is important to recognize that hue is just one of the dimensions of color. Boyle’s “splendor” is another dimension which refers to how vivid a color may be. Words used today for this dimension include intensity and saturation as well as the more specialist terms “chroma” and “chromaticness.”
Jakob Christoph Le Blon
While Boyle was a scientist, Jakob Christoph Le Blon was an artist. Le Blon developed a system of printing with transparent inks from three or four plates on white paper that is essentially the same as the process used by printers today. Three plates, one inked with yellow, one with red, and one with blue, were printed one over the other. “Painting can represent all visible Objects, with three Colours, Yellow, Red, and Blue; for all other Colours can be compos’d of these Three, which I shall call Primitive” . Sometimes Le Blon used a fourth plate, inked with black, much as printers today use black to deepen the colors and sharpen the image.
Isaac Newton experimented with prisms and demonstrated that different wavelengths of light are refracted by a prism to a greater or lesser extent and that the refracted light appears in different colors. When projected onto a white surface the refracted light is spread out to reveal the colors of the spectrum. In Newton’s words, “The Light whose Rays are all alike Refrangible, I call Simple, Homogeneal and Similar … The Colours of Homogeneal Lights, I call Primary, Homogeneal and Simple …” [9, p. 4]. For Newton, therefore, a primary color was light of a single wavelength.
The spectrum can be divided and subdivided many times over, so the number of primary colors in this physical sense is very large indeed. But Newton chose to name just seven: “The original or primary colours are red, yellow, green, blue and a violet-purple together with orange, indigo and an indefinite variety of intermediate gradations” (Newton, quoted in [10, p. 225]). Newton developed a circular diagram, divided into seven segments, which could be used to determine the likely appearance of a given mixture of his primary lights.
Now, as it is almost impossible to conceive each sensitive point of the retina to contain an infinite number of particles, each capable of vibrating in perfect unison with every possible undulation, it becomes necessary to suppose the number to be limited, for instance, to the three principal colors, red, yellow, and blue … and that each of the particles is capable of being put in motion less or more forcibly by undulations differing less or more from perfect unison … each sensitive filament of the nerve may consist of three portions, one for each principal color. [11, p. 147]
James Clerk Maxwell
Where Le Blon had demonstrated that an image could be reproduced in a print by the subtractive mixing of three transparent inks, James Clerk Maxwell demonstrated how an image could be reproduced by the additive mixing of colored light. The first color photograph, taken by Thomas Sutton, was used as an illustration for a lecture by Maxwell in 1861 [12, p. 321]. A ribbon had been photographed successively through a red, a green, and a blue filter. The three photographs were then projected through corresponding filters so that the three images were superimposed on the screen.
Maxwell also developed a means of measuring color. When a disk with segments in different colors is spun at high speed the colors blend and a new color is seen. Three paper disks, one in each of the additive primary colors (physical), are interleaved in such a way that the visible proportions of each can be adjusted. A disk in the color that is to be measured is fixed over the three interleaved disks. The superimposed disks are then spun in a series of trials, with the proportions of the three primary colored disks being adjusted after each trial, until the combination of three primary colors matches the color to be measured. The measurement is then expressed in terms of the relative amounts of each of the primary colors needed to achieve the match. In this process colors (physical) and colors (visual) come together. The colored papers are physical, but it requires a visual judgment to establish a match.
… there are four outstanding loci in the series of hues that make up the closed circle: first the locus of the yellow that shows no remaining trace of redness, and yet reveals no trace of green; second the locus of the blue for which the same is true. These two hues may be called primary yellow and primary blue (Ürgelb and Ürblau). Likewise we can name, third, the red, and, fourth, the green that are neither bluish nor yellowish primary red and primary green (Ürrot and Ürgrün). [14, p. 42]
The CIE System
In 1931 the International Commission on Illumination (CIE) established the system of color measurement that is today’s world standard [15, pp. 59–63]. All possible colors (physical) can be plotted on the CIE chromaticity diagram. The position of a color on the diagram is determined by the relative output of three primary lights that, when mixed together, would match the color. The measurement can be expressed in terms of tristimulus values. As with Maxwell’s system of spinning disks the matches that underpin the CIE system were initially the result of visual judgments made by human observers. But there is no set of three colored lights (physical) that can match all possible colors (visual). The most vivid colors (visual) are those of Newton’s homogeneal lights – the pure colors of the spectrum. In order to measure such colors it was necessary to reduce their “splendor” by introducing negative values. This led to complexities that were overcome by clever mathematics and the introduction of theoretical “super light sources” that could not be physically realized but that were able to deliver matches for the spectral colors.
Primary Colors in the Twenty-First Century
The different ways in which the term “primary colors” is used today can be related to these ideas that evolved in the past.
Further Problems with Standard Definitions
The definitions of primary colors, quoted in the section “Confusion Between the Physical and Visual Aspects of Color” above, not only confuse the physical and visual aspects of color, they are also misleading in other ways. As Boyle observed, and the CIE confirmed, there is no set of primary colors (physical) from which all other colors can be obtained by mixing. It is possible to obtain all the hues but not all the colors – only those within a limited gamut. The CIE had to resort to its theoretical super lights. A full gamut of all other colors, right out to Newton’s homogeneal lights, can only be achieved in theory, with the CIE super lights, and not in practice.
The second definition, that primary colors cannot be mixed from a combination of other colors, is also misleading. It would be true of the CIE theoretical super lights but not in the case of normal lights, paints, or inks. It would make no sense to claim that primary colors (physical) cannot be mixed from a combination of other colors (physical). The primary colors that do make sense in the context of this definition are visual colors. Leaving aside black and white the visual primaries should be the unique hues – Hering’s Ürfarben – and there is nothing unmixable about these. A paint that appears yellow can be mixed with paint that appears magenta to produce a red that is neither bluish nor yellowish and with a paint that appears blue to produce a green that is neither bluish nor yellowish.
Primaries for Mixing other Colors in Print and on Screen
Primary colors (physical) for mixing a large range of other colors (visual) have particular application for the reproduction of colored images in print and on the television or computer screen. There is scope for further confusion here because the mixing processes are not the same. When layers of transparent ink are superimposed in a print the result is a subtractive mixture. Additive mixture occurs when pools of colored light are projected to overlap or when the small points of colored light blend on a screen.
It was not until the nineteenth century that the difference between subtractive and additive processes was recognized. Then it became clear that the primaries (physical) for each process were not only different in that pigments and dyes are physically different from colored lights but also that the most useful sets of three colors (physical) appear different. The two sets are listed in the first definition set out in the section “Confusion Between the Physical and Visual Aspects of Color.” Additive primaries are listed first, then the subtractive primaries. They are described as red, green, and blue or (for pigments) red, blue, and yellow. This suggests that the only difference between the two sets is that one set has green and the other set has yellow. However, the lights that work best for additive mixture appear orange-red, yellow-green, and violet-blue, while the inks for subtractive mixture appear purplish red (magenta), slightly greenish blue (cyan), and yellow. The differences between the orange-red and magenta, and between the violet-blue and cyan, are masked by the way the names “red” and “blue” stretch to cover a range of different hues. What makes these two sets primary is that they have been found to deliver the most useful gamuts of other colors (visual). The gamuts could be increased for each process by adding further inks or lights, but any slight improvement would not justify the very large extra costs that this would involve.
Mixing with a Larger Number of Colors (Physical)
When a color is required in a print that is beyond the reach of the cyan, magenta, and yellow process inks it may be achievable with a mixture of inks formulated according to the Pantone Matching System®. This system has 14 separate inks that can be premixed before being used on the press.
A similar number of different colors (physical) are used to mix the very large range of different colors (visual) that are available for painting the interior and exterior of buildings. The tinting machines for Dulux Australia have 4 main bases and 13 different tinters that add pigments to a base according to given formulas.
Today’s artists, like Leonardo before them, need not restrict themselves to a set of three primary colors (physical). They use as many different paints as they need to achieve the results they want.
The confusion surrounding the different notions of primary colors, especially that between primary colors (physical) and primary colors (visual), was the subject of a paper that Green-Armytage gave at a conference on color education . This confusion is especially evident in what he calls traditional art school color theory.
… let us develop the 12-hue color circle from the primaries – yellow, red and blue. As we know, a person with normal vision can identify a red that is neither bluish, nor yellowish; a yellow that is neither greenish nor reddish; and a blue that is neither greenish, nor reddish.
These definitions are reminiscent of Hering’s definitions except that green is not included as a primary. However, green is included in the definitions of primary yellow and primary blue. Itten did not (could not?) describe primary blue as neither reddish nor yellowish or primary yellow as neither reddish nor bluish.
Itten goes on to describe secondary colors as “… three mixed colors, each composed of two primaries.” Later in the books [17, p. 136, 18, p. 88] he describes green as “… the intermediate between yellow and blue. According as green contains more yellow or more blue, the character of its expression changes.” Did he mean that there could be more or less yellow or blue paint in the mixture which resulted in the perception of green? Or did he mean that in the green percept there might be more or less resemblance to yellow and blue as primary colors (visual)?
It is not clear from the books whether Itten associated his primary yellow, red, and blue with any particular pigments. There is certainly no claim that with three such paints all other colors can be obtained by mixing. But what is clear from experiment is that three paints which might fit Itten’s definitions would not work so well for producing a comprehensive range of other colors (visual) as would paints that look the same or similar to the cyan, magenta, and yellow process inks used by printers. Itten’s primaries would not be a good choice as primary colors (physical). And without green they are not satisfactory as primary colors (visual).
Plotting the Gamut for a Given set of Colors (Physical)
On the left, cadmium yellow pale, cadmium red, and ultramarine are typical of the paints regarded as “primary” in traditional art school color theory. While these paints can deliver all the hues, the greens and purples are very dull. If more vivid greens and purples are required an artist would need to include more paints on the palette. A more extensive gamut can be achieved with the addition of a turquoise or cerulean blue and a vivid magenta such as Bengal rose. The paints used on the right were lemon yellow, Bengal rose, and turquoise blue. These paints appear similar to the yellow, magenta, and cyan process inks as used in printing. Their status as preferred primaries for subtractive mixing is clear from the more extensive color gamut that they can deliver. Within the mixture lines are the colors (visual) that could be achieved with that set of paints. If a desired color (visual) would be located outside the mixture lines there may be a paint that could be added that would bring that color (visual) within reach .
The Role of Color Names in the Confusion of Different Kinds of Primaries
There has been a long history of singling out particular colors as being somehow special. The realization that a comprehensive range of colors (visual) can be mixed from a limited set of paints, inks, dyes, or lights led to the development of ways to reproduce fully colored images in prints and projected photographs, techniques pioneered by Le Blon and Maxwell. This realization also inspired ideas about the physiology of the eye and how it is that humans can experience such a great variety of colors.
Concepts of primary colors can help people to work successfully with colors. These concepts can also be confused. The key is to recognize that there are two kinds of primary colors: physical primaries, which take the form of lights, paints, inks, and dyes to be used in mixing a comprehensive range of other colors (visual), and visual primaries as very specific visual experiences – pure white, black, and the four unique hues. By using the visual primaries for reference the physical processes involved in the production and reproduction of colors, which may involve the physical primaries, can be monitored and controlled.
- 1.Green-Armytage, P.: Seven kinds of colour. In: Porter, T., Mikellides, B. (eds.) Colour for Architecture Today. Taylor and Francis, Abingdon (2009)Google Scholar
- 2.Thompson, D.E.: The Concise Oxford Dictionary. Oxford University Press, Oxford (1995)Google Scholar
- 3.Hope, A., Walch, M.: The Color Compendium. Van Nostrand Reinhold, New York (1990)Google Scholar
- 4.Beare, J.: Greek Theories of Elementary Cognition. The Clarendon Press, Oxford (1906)Google Scholar
- 5.Aristotle: Sense and Sensibilia. In: Barnes, J. (ed.) The Complete Works of Aristotle. Princeton University Press, Princeton (1984 [4th c. BC])Google Scholar
- 6.da Vinci, L.: Leonardo on Painting. Yale University Press, New Haven (1989 )Google Scholar
- 7.Boyle, R.: Experiments and Considerations Touching Colours. Johnson Reprint Corporation, New York (1964 )Google Scholar
- 8.Le Blon, J.C.: Coloritto. Van Nostrand Reinhold, New York (1980 )Google Scholar
- 9.Newton, I.: Opticks, 3rd edn. Bell & Sons, London (1931 )Google Scholar
- 11.Young, T.: On the theory of light and colours. In: Peacock, G. (ed.) Miscellaneous Works of the Late Thomas Young. John Murray, London (1972 )Google Scholar
- 12.Kemp, M.: The Science of Art. Yale University Press, New Haven (1990)Google Scholar
- 14.Hering, E.: Outlines of a Theory of the Light Sense, transl. L. M. Hurvich, D. Jameson. Harvard University Press, Cambridge, MA (1964 )Google Scholar
- 15.Berns, R.: Billmeyer and Saltzman’s Principles of Color Technology, 3rd edn. Wiley, New York (2000)Google Scholar
- 16.Green-Armytage, P.: Primary confusion. In: Arnkil, H., Hämäläinen, E. (eds.) Aspects of Colour. The University of Art and Design Helsinki UIAH, Helsinki (1995)Google Scholar
- 17.Itten, J.: The Art of Color. Van Nostrand Reinhold, New York (1961)Google Scholar
- 18.Itten, J.: The Elements of Color. Van Nostrand Reinhold, New York (1970)Google Scholar
- 19.Green-Armytage, P.: Paints as ‘magnets’ to guide the mixing process. In: Hansuebsai, A. (ed.) Proceedings AIC 2003 Bangkok Color Communication and Management, pp. 309–311. Chulalongkorn University, Bangkok (2003)Google Scholar