PIGMENTARY PRIMARIES

Although Newton (1672) applied the term "primary" to colours with the meaning of "monochromatic", the expression as it is understood today has its origin in the historical concept that yellow, red and blue, initially alongside white and black, were the "simple", "primitive" or "primary" colours from which all others could be derived. The first visual representation of this concept is in a diagram in the Opticorum Libri Sex (1613) of Francois D'Aguilon (Figure 7.3) . D'Aguilon showed his "simple" colours on a linear scale comprising white, yellow, red, blue, and black respectively, and placed each of the three "composite" colours purple, green and gold  ("aureus") in relation to the relevant pair of simple colours, in an arrangement derived from music theory. From its mention in the commentary on the Timaeus of Plato by Chalcidius, the linear scale of simple colours is known to date back (alongside many other systems) to the fourth/fifth century CE (Kuehni, 2003). The system was repeated and elaborated in diagrams by other seventeenth century writers, and seems to have become increasingly widely held throughout the century.

The word "primary" was introduced by Boyle (1664), who reported:

"...I tell you, that the mixing of pigments being no inconsiderable part of the painters art, it may seem an encroachment in me to meddle with it. But I think I may easily be excused (though I do not altogether pass it by) if I restrain myself to the making of a transient mention of some few of their practices about this matter; and that only so far forth, as may warrant me to observe to you, that 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 art, I have not yet found, that to exhibit this strange variety they need employ any more than white, and black, and red, and blue, 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."

Boyle's statement shows an awareness of the concept of a gamut: while the primary colours suffice to mix colours of a full range of hues, some colours will, by their greater "splendor" (we would say chroma), lie outside this gamut. In reality any set of pigments produce a restricted gamut or range of colour mixtures, and are unable to mix colours outside this gamut. To accurately reflect this aspect of pigment mixing, the concept of primary colours needs to be formulated with this awareness, as a set of pigment colours whose mixtures enclose an optimal gamut of high chroma colours.

The step of removing white and black from the list of simple colours, though anticipated by Alberti in the fifteenth century, was not taken decisively until after Newton had demonstrated that white light was a compound of light of many colours. One of the first works to take this step, and also the first to show these colours on a "colour wheel", was the 1708 edition of the Traite de la Peinture en Mignature (Fig. 7.4). The author, believed to be Charles Boutet, defined the "Couleurs Primitives", yellow, red and blue, as colours "that cannot themselves be constructed from other colors, but from which all others can be constructed" (translation by Mollon, 2003). This naive definition, which neglects the inevitable restrictions of a gamut, has been sighted in classrooms as recently as 2007.

We have already seen the expression "primary" used in several other contexts. The conception of primary colours as colours with an optimal gamut of high chroma mixtures has been extended to the additive mixing of lights, where the optimal "additive primaries" are located in the red, green and blue-violet bands of the spectrum. (Note however that in the scientific literature on additive mixing, the term primary is used more broadly, for any coloured light used in mixing). The same conception applied to ideal subtractive mixing identifies the optimal "subtractive primaries" as yellow, magenta and cyan, the complements of the additive primaries. Finally, the term "psychological primaries" has been employed for the opponent colours red, green yellow and blue, because our colour experiences all result from combinations of pairs of these colours.

Some authors today treat the red, yellow and blue traditional primary colours as unquestioned fact, others insist that they are obsolete, pre-scientific, and so on, and that the "real" primary colours are yellow, magenta and cyan, and others again treat the whole concept of primary colours for paints as a "useless fiction". All of these extreme positions are somewhat wide of the mark. In retrospect it can be seen that the choice of traditional primaries reflects the paint colours found by experience to give an optimal gamut of colour mixtures, but subject to an irresistible unconscious assumption that these colours should all be psychological primaries. Thus red and blue owe their status as traditional primaries to the fact that they are the psychological primaries closest to the ideal subtractive primaries magenta and cyan respectively. Ideal yellow, magenta and cyan colourants would be the optimal primary colours for paint mixing - if such pigments existed. However, the actual situation for artists' pigments is not quite so straightforward, because:

(1) No artists' pigments are ideal colourants: ideal yellow, magenta and cyan pigments simply do not exist.

(2) Not only are there no ideal magenta and cyan pigments, but no pigments presently in use are even very close to ideal magenta or cyan hue. The closest oil painting pigment to the hue of ideal magenta is Quinacridone Magenta (PV 19), which is distinctly redder than ideal magenta. Similarly, the closest popular oil painting pigment to cyan is the green shade variant of Pthalocyanine Blue (PB 15.3), which is far bluer than ideal cyan.

Thus even today our best magenta and cyan pigments are both well towards the traditional artists primaries of red and blue compared to ideal magenta and cyan (Figure 6.4). Yellow pigments are available in an essentially continuous range of hues, but the clockwise displacement of our other two primaries from the ideal subtractive hues demands the choice of a relatively greenish or lemon yellow to maintain wide spacing and hence obtain the maximum gamut of colours.

Several contemporary writers maintain that primary colours do not really exist for paints, citing with apparent surprise that the naive definition of primary colours does not hold, and that colours do not mix along straight lines on the colour wheel. However, the definition of pigment primaries as pigment colours whose gamut encloses an optimal range of high chroma mixtures in fact implies that paint colours do not mix along straight lines on the colour wheel, and that mixing paths differ according to the hue of the components. This is easy to confirm experimentally. Colours close to the ideal subtractive primaries mix along lines that bend outwards, staying relatively close to full chroma, while colours far from the ideal subtractive primaries tend to mix along lines that are straight or curve inwards, passing through relatively low chroma colours (see Fig. 6.7, 6.8). Consequently, mixtures of the former enclose a much larger gamut of colours than the latter.

PIGMENTARY COMPLEMENTARIES

We saw in the last section that even in ideal subtractive mixing, the hue of the subtractive complementary, while being at least close to that of the additive complimentary, commonly differs noticeably. With artist's pigments there is a further complication that tends to shift the complimentary relationships in a systematic way, particularly involving yellow-blue mixtures. In spectral terms, the non-ideal nature of artists pigments shows up as bell-shaped and round-shouldered absorption curves, instead of the perfect plateau-shape of ideal colourants. The overlap that this introduces creates the situation that two artist's pigments can be complementary in terms of the light they reflect, but not complementary in terms of paint mixing. It is easy to show how this can occur with two simple imaginary pigment reflectance curves (Figure 6.5).

Figure 6.5. Spectral transmission curves for two hypothetical colourants. Two artist's pigments can be complementary in terms of the light they reflect, but far from complementary when mixed together. Illustration by D. Briggs, after diagrams by Pope and Luckiesh.

The same effect can be readily demonstrated with actual pigments: Lemon Yellow and Ultramarine Blue give off light that is complementary (as can also be demonstrated with spinning discs), but when physically mixed together they make a distinctly green mixture (Figure 6.6).

Figure 6.6. Ultramarine Blue and Lemon Yellow, physically mixed. These colours, though additive complementaries (white line), physically mix to make a series of dull greenish colours. Photographed colours shown on CbCr plane and in side view of YCbCr space.

This drift towards greenishness means that we need to look for mixing complements of Ultramarine Blue among pigments that are more orange than Lemon Yellow, and several yellow-orange pigments are found to be pigmentary complements. Similarly we should expect the mixing complement of Lemon Yellow among pigments more purplish than Ultramarine Blue (Ultramarine Violet turns out to be about right). These relationships specific to pigment mixing are expressed in a general way by the positioning of yellow opposite violet and orange opposite blue on the conventional artists' colour wheel.

<< 1 2 3 >>