002gage LocusClassicusColour 1981

A Locus Classicus of Colour Theory: The Fortunes of Apelles
Author(s): John Gage

Source: Journal of the Warburg and Courtauld Institutes , 1981, Vol. 44 (1981), pp. 1-26
Published by: The Warburg Institute
Stable URL: https://www.jstor.org/stable/751049
REFERENCES
Linked references are available on JSTOR for this article:
https://www.jstor.org/stable/751049?seq=1&cid=pdf-
reference#references_tab_contents
You may need to log in to JSTOR to access the linked references.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
https://about.jstor.org/terms
The Warburg Institute is collaborating with JSTOR to digitize, preserve and extend access to
Journal of the Warburg and Courtauld Institutes
This content downloaded from

122.225.227.202 on Fri, 01 Dec 2023 09:47:35 +00:00
All use subject to https://about.jstor.org/terms
A LOCUS CLASSICUS OF COLOUR THEORY:
THE FORTUNES OF APELLES*
John Gage
WRITING IN 1637 to Franciscus Junius, Rubens praised the diligence with which
the author of De Pictura Veterum had gathered together 'all the examples, opinions
and precepts relating to the dignity and honour of the art of painting, which have
been preserved, widely scattered among ancient writings, to this day and to our own great
advantage'. 'But', continued Rubens,
since those examples of the ancient painters can now be followed only in our individual
imaginations, I could wish that a similar treatise on the paintings of the Italian masters might be
extracted and compiled with like care. For examples of their work are still publicly exhibited today;
we may point to them and say, 'there they are'. Those things which are perceived by the senses
produce a sharper and more durable impression, require a closer examination, and afford a richer
material for study than those which are offered to our imagination alone, like dreams; being
sketched out in words, they are 'thrice grasped in vain' (like Eurydice's shade by Orpheus), and
often escape to disappoint us in our hope. This we say from experience; for how few among us, in
attempting an adequate reproduction of some famous work of Apelles or Timanthes that is
graphically described by Pliny or by other authors, will not produce something insipid or
inconsistent with the grandeur of the ancients?'
The painter's appeal to practical experience over and against literary imagination must
also strike a sympathetic chord in those many who have addressed the ungrateful task of
reconstructing the history of Greek painting on the basis of little more than documents;
but in singling out Apelles and Timanthes, Rubens chose exemplars whose posthumous
reputations had taken, and were to continue to take, a far from purely literary form. This
was especially true in the case of Apelles, whose extraordinary financial and social
successes, and his remarkable gift for defending himself against ignorant criticism from
both the humble and the great,2 made him a natural hero to later painters, who celebrated
the more favourable episodes of his life in many pictures between the sixteenth and the
nineteenth centuries.3 His most elaborate attempt at riposte, the Calumny, described by
Lucian, and the only known example of a painted satire in Antiquity, was also re-
interpreted repeatedly, both as a visual image and in literature, from the Renaissance
onwards.4 But Apelles had more to offer his admirers than an enviable example and a
* My greatest debt is to David Cast and Jean-Michel 2 Most of the literary sources for Apelles have been
Massing for many helpful suggestions. I also owe crucial gathered and translated by A.-J. Reinach, Textes grecs et
information to B. Cook, Alex Potts, Oliver Logan, Jon latins relatifs ai l'histoire de la peinture ancienne (Recueil
Whiteley, Philip Conisbee and the late Professor Edgar Milliet), 1921, nos. 40o-86. The fullest discussion of his
Wind. career is W. Lepik-Kopaczyfiska, Apelles, der beriihmteste
Maler der Antike, 1963, which uses these sources rather
1 M. Rooses and C. Reulens, Correspondence de Rubens,
uncritically.
2nd edn, 1975, VI, no. DCCCXXXI. My translation is based
on P. P. Rubens, Letters, trans. R. Magurn, 2nd edn, 3 A listing of many examples in A. Pigler, Barock-
1971, p. 407, and E. McGrath, 'The Painted Decorationthemen, 2nd edn, I974, PP. 366 ff. See also E. Kris & 0.
of Rubens's House', this Journal, XLI, 1978, pp. 245 f.
Kurz, Die Legende vom Kiinstler, 1934, PP. 49 f.
Junius approved sufficiently of the letter, which Rubens 4 The most recent study, with the earlier literature, is
had prudently composed partly in Latin as well as D.inCast, The Calumny of Apelles: A Study in the Humanist
Flemish, to print it as an introduction to the Dutch Tradition, I98 i.
edition of his book (Middelburg 1641).
Journal of the Warburg and Courtauld Institutes, Volume 44, 1981

122.225.227.202 on Fri, 01 Dec 2023 09:47:35 +00:00
2 JOHN GAGE
subject flattering to their vanity
survived even until Pliny's times
style, more than for those of any
to describe several aspects of Ape
to lifelikeness, Pliny recorded se
have given practical encouragem
such anecdote was the story of th
made the miraculously fine line,
many and, it must be said, very
Another concerned the equally m
was said to cover his completed p
his colours at the same time as h
good deal of attention in recent
colour: that he was a four-colour
among classical archaeologists,
posthumous reputation, where it
others. If Franciscus Junius, a
Apelles' four colours, preferrin
painter's admirable simplicity,7
century or in Junius' own time,
colour and its organization which
scholars and the practical experi
that experience is the subject of
THE ORIGINS OF THE FOUR-COLOUR THEORY
Four colours only - white from Milos, Attic yellow, red from Sinope
black called atramentum - were used by Apelles, Aetion, Melanthiu
immortal works; illustrious artists, a single one of whose pictures the we
suffice to buy, while now that even purple clothes our walls, and India
rivers and the blood of dragons and of elephants, no famous picture is p
that when the painters' equipment was less complete, the results were i
... we are alive only to the worth of the material and not the genius of th
Pliny's circumstantial account, although it is not the only one, has

subsequent discussions of the four-colour palette, the most rec
Bruno's Form and Colour in Greek Painting (1977). Bruno's essay is
of painted tombs at Lefkadia in Greece and at Kazanlak in Bulgari
s Many are discussed by H. van 'Controversial

der Waal,Methods'The and Methods of Controversy',
"Linea Summae Tenuitatis" of Apelles:
ibid. Pliny's Phrase
cv, 1963, pp. 90 ff.; O. Kurz, 'Time the Painter',
ibid. esp.
and its Interpreters', Zeitschriftfir Aesthetik undp. 94.
allgemeine
Kunstwissenschaft, xII, I967. See also E. H. Gombrich,
7 Pliny's story had passed without comment in the
1637 edition of De pictura veterum (p. 68), and in the
The Heritage ofApelles, 1976, esp. pp. 15-16.
6 Much of the documentation hasEnglish beenedition
brought
of the following year (p. I I8),Junius did
notby
together in a series of articles provoked even name
the the colours. Cf. also the further examples
National
Gallery cleaning controversy in 1961:ofE. H. Gombrich,
Apelles's restraint as a colourist given in The Painting
'Dark Varnishes: Variations on a theme from 1638,
of the Ancients, Pliny',
pp. 119, 285.
Burlington Magazine, cIv, 1962, pp. 51-55; J.Naturalis
8 Pliny, Plesters,
Historia, xxxv, 50, in the translation
'Dark Varnishes - Some further Comments', ibid. by K. Jex-Blake and E. Sellers, The Elder Pliny's Chapters
PP. 453 ff.; D. Mahon, 'Miscellanea for the Cleaning
on the History ofArt, reprint 1968, p. 97.
Controversy', ibid. esp. pp. 463 f.; E. H. Gombrich,

122.225.227.202 on Fri, 01 Dec 2023 09:47:35 +00:00
COLOUR THEORY AND APELLES 3
late fourth or early third centuries B.c., and are thus slig
c. 370 B.C. and flourishing in the 320s); but it also seeks t
Pliny's story by relating it to pre-Socratic views of the fo
elements; and by explaining the rather surprising absenc
palette on the grounds that some black pigments have bluish
that there was an overlap in the Greek terminology for b
arguments is new, although they have never before been bro
of the discussion; but neither are they very compelling. Plin
philosophical argument relating colours to elements, and he m
the basis of this analogy, but there is little reason to think t
painters in the fourth century B.c. Empedocles, the earliest (f
to whom the four-colour theory of the elements is attributed
surviving fragments of his doctrine, and his use of the term
refer to the painter's palette in Fragment 23 would suggest th
restriction in workshop practice.10 Aetius, writing in the firs
Stobaeus, in the fifth, attribute to Empedocles a theory relat
elements, but they are quite unspecific in detail." The earlies
basic (hapla) is in De sensu by Aristotle's follower, Theophras
the painters adduced by Pliny. Theophrastus states that De
fifth century B.c.) had regarded these simple colours as whit
(erythron) and green (chloron).12 Theophrastus did not su
related these colours to the elements, and he himself contested
other investigators propose white and black as the only s
particularly worried by the inclusion of chloron (the only one
tally with Pliny's list) which, he claimed, should have the op
to act as an opposite, as black did to white (?82). The
contemporary of Theophrastus, and perhaps even by Theo
colour equivalents for the elements, but they were yellow for f
three: earth, air and water (79Ia). It was not until the firs
A*tius (De placitis reliquiae, I, I 5, 8), in Galen (De elementis ex
9 Both of these solutions had been 12 G.suggested by

M. Stratton, Theophrastus and the Gree
earlier scholars: for the relationship of
cal the four colours
Psychology toAristotle, 1917, pp. 132
before
the elements, W. Seibt, Helldunkel, I. the
Voncrucial
den Griechen
term in bisthis context: the earli
Correggio (Studien zur Kunst- und Kulturgeschichte,
not seem to have given it a specifically ch
ImI), 1885, p. 31; for this view, andtent, but rather
for vine-black asthe
a connotation of 'p
'blue' pigment, E. Berger, Die Maltechnik des'moist'
'fresh', Altertums,
(E. Handschur, Die Farb- un
2nd edn, 1975, P. 54. This function of beiblack
Homer und
as a blueHesiod,
had I97o, pp. 150 ff.; E. I
already been noticed in the I8th century
Termsby in V. Requeno,
Greek Poetry, 1974, PP. 31 f.). H
translated
Saggi sul ristabilimento dell'antica arte de'Greci eitRomani
in this passage as yellow, a
pittori, 2nd edn, 1787, 1, p. 25, terms and by for H. Meyer,
yellow attributed to Empedocles
Hypothetische Geschichte des Kolorits, (Aitius)
in J. W. or
vonpyrrochrous
Goethe, (Galen), and the
Zur Farbenlehre: Historischer Teil, ed.biguous
Kuhn,use of chl6ron
I957, p. 59. as 'yellow' seems t
For the equivalence of black and blue in Greek termin-
Hippocrates and Galen (H. Diirbeck, Zur C
ology, W. Schulz, Das Farbempfindungsystem der Hellenen,
der griechischen Farbenbezeichnungen, 1977,
1904, p. 36; K. Schefold, 'The Choice of 113).
io8 ff., Colour
It isin
at least clear that chl6r
Ancient Art', Palette, xIII, 2, 1963, p. 5.
understood to mean a saturated 'primary'
10 This was pointed out by E. Keuls, 'Skiagraphia
13 H. B. Gottschalk, 'The De Coloribus and
Hermes, LXXIX,
once again', Amnerican Journal ofArchaeology, xcII, 1964,
1975,p. 85.
p. I5
11 The evidence is reviewed by V. J. Bruno, Form and
Colour in Greek Painting, 1977, p. 57.

122.225.227.202 on Fri, 01 Dec 2023 09:47:35 +00:00
4 JOHN GAGE
in the pseudo-Aristotelian De m
consensus of opinion that there
that those colours were black, w
for the four-colour theory is of
which were held to express them
enjoyed an especially high reput
Imagines 7), and he inherited an
differences in sex by different c
been expected to amplify this re
palette as described by Pliny did
The school of Hippocrates, wh
argued that man is composed of
white; yellow bile; and black bile
for a perfectly healthy organism.
doctrine applied to the colour of
which were interpreted in diagno
seems to have been applied to th
available to painters of the figur
the pseudo-Aristotelian Physiogn
concerned with the colour of com
there is very little evidence of th
surviving body of ancient portr
of Roman Egypt, which emplo
painting, and which were for t
interpretation of Hippocrates wa
If we turn to other examples
described by Pliny continues to
that atramentum may well be a b
positively blue, is convincing in it
as we know it. There were man
times, and they are not of this b
the most part from Egyptian blu
see used in the sixth-century tom
Tuffatore near Paestum and th
third-century paintings at Lefka
'Physiognomische
14 The fullest discussion is in Lepik-KopaA
Physiognomik
cit. n. 2 above, pp. 36 im Mitte
ff. This distinction o
especially striking inxxIx, 19I 1, p. 685).
the Herculaneum Th
Thes
recently discussed Hunain ibn
as a close Ishaq
copy oflists
a fo
Greek originalbyunhealthy
I. Scheibler,skin,'Zum Ko
includi
melancholy
griechischen Malerei', Pantheon, (E. Grant,
xxxvI, 197A
15 Hippocrates, The1974,
Nature p. ofMan,
707.). iv ff.; f
ness of phlegm, ibid.18 For the history of t
viI.
in Ancient
16 Galeni in Hippocrates librumTechnology,
de humoribu2
tres, Venice 1562, pp.19 For Kizilbel and K
8 ff.
American Scriptores
17 For example R. Foerster, Journal ofphy
Ar
1893, 1, pp. 74-75. A 14th-century
25I; French
for the Tomba del T
cal MS, which adopts the Hippocratic
Tuffatore, s
1970, pp. 1
four temperaments and four
Lefkadia andcolours,
Kazanlake
choly not with black
9 but
and with yellow
pls 5b, (lute
6, 1o, I i,

122.225.227.202 on Fri, 01 Dec 2023 09:47:35 +00:00
see in the fourth-century Kertsch-style ceramics which have
the style of Apelles himself.20 Hellenistic works which h
paintings by Apelles or other four-colour painters, such as th
Pompeii or the Alexander mosaic now in Naples, also mak
greens.21 The first hint of a four-colour palette in a work of
appeared in a recently excavated fragment of panel painting
in the British Museum, which is especially interesting in re
stated that he painted almost exclusively on panel. The frag
and is painted in a four-colour palette of white, black, re
mixtures, grey and pink. There is no blue.22 This small p
reflect the most advanced procedures of its period, but it d
substantial to outweigh the improbability that a group of ar
abandoned the use of a well-known pigment on aesthetic gr
restraint should have found no echo in the practices of thei
The key both to Pliny's account and to the modern r
practice is indeed aesthetic choice: Pliny was above all c
simplicity of the ancients was preferable to the modern
expensive materials. In this he was taking up a recurren
complaint against modern and exotic taste also voiced b
Seneca (Epistolae, LXXXVI, 6 f.) and by Petronius (Satyric
indeed was Pliny to establish the essential sobriety of his clas
rather serious inconsistency in his various accounts of Apell
he lists are all in the category of colores austeri, as opposed t
(cf. N.H. xxxv, 30). In a passage on early monochrome pai
accounts for the abandonment of the florid colours cinnaba
austere ochres, rubrica and sinopis, on the grounds that the
too vivid (nimis acre). But in his celebrated account of Apell
Pliny claimed that one effect of this material was to tone do
coloribus austeritatem occulte daret): precisely that class of
suggested Apelles had been studious to avoid.24 It may
distinguished between phases in Apelles's career, in which
under a dark-toning varnish, and then a restricted palet
22 B.
20 For example the pelike of c. 330 B.c. in Cook, 'Painted Panel from Saqq
the National
Museum (no. 17 18), which has a palette MuseumofYearbook,
red, black,II, 977, pp. 197-98. In th
white, blue and gold leaf. For the link with
Cook Apelles,
refers J. J.
to the chiton as 'pale blue', bu
Pollitt, Art and Experience in Classicalreport
Greece,(of which
1972, he kindly allowed me
p. 159.
21 P. Mingazzini, 'Una Copia dell'Alexandros Ker-
script) he modified this to 'bluish-grey', w
aunophonos di Apelle', Jahrbuch derappears
BerlinertoMuseen,
be. In,
1961, esp. p. 15. For the Alexander23 Mosaic,
Bruno,which has
op. cit. n. II above, pp. 68 ff
been regarded as reflecting the four-colour palette
out that Cicero and
includes his rather diverg
the style of Apelles since its discovery in 1832,
four-colour see esp. in a similar context
painting
H. Fuhrmann, Philoxenos von Eretria,although
1931, pp. his
203complaint
ff., A. is not as explicit
Rumpf, 'Zum Alexander-Mosaik', Athenische Mitteilun-
See also H. Jiicker, Vom Verhiiltnis der Rbme
gen, LXXVII, 1962, pp. 240 f. Fuhrmann
Kunst dernotes a green 1950, cited by T. Le
Griechen,
stone in Alexander's costume, and there
lehre', are many
in Reallexikon zur deutschen Kunstgesc
others in the foreground plants and
164. rocks, and in the
draperies to the bottom left. The presence
24 For the fullestof green
discussion has et austeri,
of Coloresfloridi
been denied by Bruno (op. cit. n.J.J.IiPollitt,
above,The Ancientp.View75), but
of Greek Art, 1975, s.v.
the dispute may be no more than austerus.
a semantic one,
Although he discusses thesince
four-colour theory in
this context, original
he allows (pp. 76-77) that the painted Pollitt does not seem
mustto see the inconsis-
have used mixtures including blue.tency.

122.225.227.202 on Fri, 01 Dec 2023 09:47:35 +00:00
6 JOHN GAGE
development as could also be th
Rembrandt - but it seems clear t
in the interests of propagating a
the palette used by the fifth-cent
theory seems to have been drawn
number of colours which was current in his own time.
THE PROBLEM OF MIXTURE
One of the most persistent arguments among modern apologists

theory is that the small number of basic pigments could be sub
intermixture; one eighteenth-century scholar calculated that the f
yield eight hundred and nineteen variations.25 It is an argument w
to commend it if it could be shown that mixing was the norm amo
and if it were not for a body of ancient opinion which condemn
Empedocles compared the mixture of the elements in the mate
painter's blending (meixante) of pigments in their preparation of t
in a discussion of the mixing-bowl mentioned by Herodotus (I,
'pigments ground together, losing their own colour in the process'
mixtures of red and yellow ochre and of black and white (De defectu
47 (436)). But elsewhere Plutarch voiced the painter's strongest obje
Mixing produces conflict, conflict produces change, and putrefaction is a
why painters call a blending of colours a 'deflowering' (phthorax, Aristot
away') and Homer [Iliad, Iv, 141] calls dyeing 'tainting'; and common usage
and pure as virgin and undefiled'.27
The context where we would most expect to find mixtures is in the

especially in portraiture, where the specific problems of matching
colouring materials must have become apparent at an early date. Pl
of several colours specifically to flesh-painting (polla synkerasantes
both here and in the encyclopaedia ofJulius Pollux, written in the
flesh-tint is written with a special term: (andreikelon: Onomasticon
evidenct that even flesh-tints were sought after in an unmixed sta
red ochre (miltos), Theophrastus states that it was found natura
tones, 'hence painters use it for flesh tints'.29
For, as Plutarch's remarks suggest, there was in Antiquity a l
the nature of mixture: was it, as the Platonists and the Peripatetic
25 Cited by E. Bertrand, Etudes sur la ing, and that

peinture et lathe phrase, 'harmonei meixant
critique dans l'antiquiti, 1893, p. I39. Diels
Forand
another editors have translated as 'harmo-
extended
discussion of the four-colour theory inniously mixed',
the light simply has the connotation of 'mixe
of the
account of primary colours published by Ogden
closely Rood
together'.
(Modern Chromatics, 1879, German edition 27 Quaestiones
I88o), Conviviales,
and 725c; cf. also De E apud Del-
based on the practice of mixture, E. Veckenstedt,
phos, 393c, and Plato, Philebus, 51-53.
28 Here,
Geschichte der griechischen Farbenlehre, I888, too,
pp. 29 ff.the
Theanalogy is with letters side-by-side in
possibility of substantial mixing is also a word, so that
implicit in the
theidea of a more intimate blending to a
theses of Lepik-Kopaczyiiska and Bruno tone (nn.
may not be in
2 and II Plato's mind. Forbes, op. cit. n. 18
above). tibove, p. 222, notes some pinks made from red an
white,Empidocle,
26 Diels-Kranz, I934, I, p. 32I. J. Bollack, but without going into details.
n, i, i969, pp. 122 ff. points out that the 29
context of this
De Lapidibus, 5I; cf. also Pliny, xxxv, 31.
passage shows that mixture is prior to the act of paint-

122.225.227.202 on Fri, 01 Dec 2023 09:47:35 +00:00
qualities of substances were truly mixed, while the subst
some looser form of association, or was it, as the Stoics mai
were themselves fused, and hence destroyed? The cruci
process was reversible, whether the compound could be r
Composition or juxtaposition (synthesis, parathesis: Aristotle,
327b; or mixis: Philo, Conf., 37, I84-87) was reversible; fu
apud Stobaeus, I, xi, 5a) was not, for it led to the destruction
Aristotle (ibid., 328a) distinguished the two types of mix
homogeneous physical mixtures and purely perceptual 'optic
view was attributed to the atomist Democritus by Alexander
century A.D.31 Plutarch's record of studio language suggests
and it is certainly true that the less drastic optical mixt
repertory of ancient painters: the building up of tones by s
fuse in the eye at a distance, rather than by tones establish
be seen both in the wall-paintings of Pompeii and in th
painting on panel, like the mummy-portrait in the British M
century A.D.32 Another form of 'optical' mixture practised i
was the glazing of a transparent colour over an opaque one,
technique which was noticed both by Aristotle (De sensu, 44
xxxv, 26).33 Here, too, the mixture was easily resolvable int
Apelles was known in Antiquity as a painter in encaus
Lucian, Imagines, 23), and this was the ancient method
modern 'painterly' techniques of mixing, especially in t
mummy-portraits of Hawara and Fayum and in the Earl
Sinai, we often have a sense of a more spontaneous blend
tones on the panel itself, which suggests the oil-painters of
sometimes even Rembrandt.34 Pollux's account of paintin
particular emphasis on mixing, and seems to refer chiefly to
is applied to wax, and seven terms are listed, relating to the
kerasasthai, mixai, symmixai, symmixasthai, syncheai, s
ode to a painter, possibly of the first century B.c. (no. 16), sp
the artist's mistress which should have an ivory brow but a n
with milk. And yet even the most painterly of the mu
second-century Priest of the Sun from Hawara, in the Briti
30 H. A. Wolfson, The Philosophy of

op. the
cit. n.Church
9 above, p. Fathers,
69). For the portrait from Fayum in
3rd edn, 1970, I, pp. 374 ff; R. B. theTodd,
British Museum, A. F. Shore,
Alexander ofPortrait Painting from
Aphrodisias on Stoic Physics, 1976, Roman Egypt,
esp. pp. 1972,For
59 ff. pl. 4.the
The literary evidence for
later medieval discussion in the West,
'opticalA. Maier,
mixture' An discussed
has been der by E. Keuls, op. cit.
n. Io above, pp.
Grenze von Scholastik und Naturwissenschaft, 2nd Io ff.
edn, 1952,
pp. 4 ff. 33 For glazing in Egypt, Forbes, op. cit. n. I8 above,
31 De Mixtione, 214 (Todd, op. cit. n. 30 above, pp. 229, 247, and in Greek encaustic, H. Schmid,
pp. I io-i I). The term for mixture here is krasis. Todd, Enkaustik und Fresko aufantiker Grundlage, 1926, pp. 86 f.
p. 184, argues that the attribution of the idea to Demo- For the literary references, L. Borelli, 'Qualche Scheda
critus is Alexander's own invention. sulla tecnica della pittura greca', Bollettino dell'Istituto
32 Hatchings in the shadows at Herculaneum were
Centrale del Restauro, 11, i950, pp. 55 if.
already noticed by Cochin and Bellicard in their Obser- 34 See especially the late Ist-century A.D. portrait of a
youth
vations sur les antiquites de la ville de Herculaneum, 1754, and from Hawara in the Sainsbury Collection at the
at Pompeii by the Nazarene painter Peter von Corne- University of East Anglia (No. 326) and, in the British
lius, who described them as like 'worn carpets' (Berger, Museum, NG. 1265 (Shore, op. cit. n. 32 above, pl. 5).

122.225.227.202 on Fri, 01 Dec 2023 09:47:35 +00:00
8 JOHN GAGE
use of hatching to model the fle
Pliny is a far from restricted on
in demand.36 Encaustic was one
the Middle Ages: it is documente
of sixth- and seventh-century e
Sinai is among the most impress
fourth century, Gregory of N
technique, compared the soul to
already mixed into its componen
resurrectione, Patrologia Graec
four-colour painters: black, whit
Perhaps the most telling indi
ancient painting methods is, on
namely the palette, from early p
ignorance of the principles and e
educated men. It has been well ar
(67c ff.) can be made to yield a p
understood.40 But in the same p
mixtures:
The law of proportion, however, according to which the several colours are formed, even if a m
knew, he would be foolish in telling, for he could not give any necessary reason, nor indeed an
tolerable or probable explanation of them.
And none of the mixtures listed in this passage is, strictly, inter-chromatic: all are ma
through the use of 'lighteners' or 'darkeners', elements with which Greek scientists fe
themselves to be rather more at home. In the second century A.D. Aulus Gellius report
an interesting discussion between the philosopher Favorinus and the ex-consul Fronto,
the subject of Greek and Latin colour-terms, which reveals the prevailing vaguen
about basic colours and mixtures. The simple colours were, following Democritus,
(rufus) and green. Fulvus (which was classified as a type of red) was a mixture of red a
green, andflavus (also regarded by Fronto as a red) of red, green and white.41 Bothfulv
Terminologie
35 Shore, op. cit. n. 32 above, pl. 8 (NG. 2912); cf. also der Gewerbe und Kiinste bei den Griechen und
NG. 3139 (Shore, pl. I6). Rimer, Iv, 1887, pp. 459 ff., and contested by Berger, op.
36 Natural History, xxI, 85; xxxv, 49. I. Scheibler cit. n. 9 above, pp. 173 ff., who showed convincingly that
also
there
suggests that the encaustic painters' palette was farwas no substantial evidence for such use. For the
paintboxes of the Egyptians, where colours were
brighter than that of the 'four-colour' artists ('Die "Vier
Farben" der griechischen Malerei', Antike Kunst,ready-mixed
xvii, before use, with separate brushes for each
I974, PP. 92 ff.). For the use of glazes in encaustic,
colour, Forbes, op. cit. n. I8 above, pp. 244 f.
Berger, op. cit. n. 9 above, pp. 206 ff.; Schmid, op.40
cit.
Bruno, op. cit. n. I I above, pp. 89 ff. The first Latin
translation
n. 33 above, pp. 86 f. I have not been able to consult E. of this chapter of the Timaeus (by Marsilio
Berger, Die Wachsmalerei des Apelles und seiner Zeit,Ficino),
1917, interpreted the terms in a similar way, making a
but from Schmid's critique (pp. 77-78, 85) it appears distinction between niger and nigredo, and characterizing
that it added little to his earlier study. xanthon as yellow (flavus) (Plato, Opera Omnia, Venice
7 G. Loumyer, Traditions techniques de la peinture medie'-
I581, p. 4I5)-
vale, 1914, pp. 147 ff. 41 Aulus Gellius, Noctes Atticae, II, xxvi. Diirbeck, op.
38 K. Weitzmann, The Monastery of St. Catherine cit. at
n. 12 above, pp. 38 ff., in the only substantial modern
commentary
Mount Sinai: The Icons, 1976, nos. BI, B2, B3, B5, B9, on this passage, has translated viridis as
BIo, BI6, BI7. 'yellow' to accommodate the sense.
39 The evidence for the use of the palette in ancient
painting was assembled by H. Bliimner, Technologie und

122.225.227.202 on Fri, 01 Dec 2023 09:47:35 +00:00
andflavus seem to us to be varieties of yellow, and only to be comp
by an additive, optical mixture of the kind hinted at by Aristotle
rainbow (Meteorology, 357a, 7-8).42 Gellius has no reference at all
does not suggest that a knowledge derived from the practical exp
current in his time. A century later, a discussion of the rain
Aphrodisias also suggests that mixture was not the usual pro
Aristotle, in his account of this phenomenon (Meteorology, 372a),
was impossible for the painter to represent, for, although he d
mixing could produce red, green or violet. Alexander, in his comm
considerably amplified the argument:
That the ... colours of the rainbow can neither be procured nor imitated
[phoinikoun, puniceus] is closer to white than green [prasinon, prasi
halurgus] is clear from the following. The natural red pigments are
dragon's blood [drakontion], which are made from the blood of animals;
mixture [mixis] of talc [koupholithos] and purple [porphyron, purpureum
to the natural colours. Natural green [prasinon; Moerbeke's Latin t
erroneously at this point] and violet are chrysocolla and ostrum, the one
other sea-purple. But the artificial colours cannot match them: green is
[kyanon] and yellow [6chron], but violet from blue and red, for the contr
yellow make green, but those of blue and red, violet. And in these cases t
inferior to the natural ... That red is closer to white than green and vio
origin. For red is made from talc, which is white, but green from ochre, w
a [gradually] darkened light appears first to be changed into this colour
green to white ... But again [it is clear] that green is closer than violet to
made from yellow, but violet from red . . and yellow is closer than red
From this account, which seems to be one of the earliest

colour-scale between white and black, it appears that although pai
cheaper substitutes for natural pigments, they did not, so far as
to match the colours of nature with mixtures on the palette.
The Noctes Atticae of Gellius, and Alexander's commentary on A
were both familiar texts throughout the later Middle Ages and du
when they were repeatedly reprinted. Their rather loose classifica
uncertainty about mixtures may well have helped to inhibit a
colours until well into the sixteenth century.
42 For flavus as xanthos and fulvus as45pyrros, Ficino's

For the Greek text: Commentaria in Aristotelem Graeca,
translation of Plato, loc. cit. n. 40 above; Keuls,
II, ii, ed. op. 1899,
Hayduck, cit. p. i6i; for the Latin trans-
n. 10 above, p. 15. Aristotle, however, lationrefers
by Williamto of the
Moerbeke (126o): A. J. Smet (ed.)
product of red and green as white. Alexandre d'Aphrodisias: Commentaire sur les Miteores d'Aris-
43 Cinnabar is the term usually used of
tote:red sulphate
Traduction of
de Guillaume de Moerbeke, Louvain 1968,
pp. 252-54.
mercury (HgS), but Alexander is probably Heinrich Bate
thinking of of Mecheln, who extracted
vermiculum or kermes, made from the this dried insect,
version coccus commentary in his early
of Alexander's
illicis (cf. Pliny, N.H. xxxIII, 7; xxxv, 14th-century
7). Dragon's bloodSpeculum divinarum et quorun-
encyclopaedia,
is a reddish resin produced by a variety
dam of palm, added
naturalium, but thein important qualifications that
antiquity it was regarded as the product of a duel
halurgus is '"violaceus", that cyanus is "fuscus" (dark),
between an elephant and a dragon (Pliny, N.H.isxxxiii,
and that ochre "vitellinus"' (the colour of egg-yolk)
7). On these pigments, F. Brunello, De(ed.
arte illuminandi,
E. van de Vyver, I96o, pp. 126-27, 129 f.).
1975, s.v.
44 Chrysocolla was in fact a basic carbonate of copper
(CuCO ).

122.225.227.202 on Fri, 01 Dec 2023 09:47:35 +00:00
i0 JOHN GAGE
NICOLETTO DA MODENA'S ENGRAVING
Although the reputation of Apelles had been kept alive during the Middle Ages in the
biographies of artists and in popular story, it was not until the fifteenth century that the
nature of his art itself became a widespread topic of discussion. The competition with
Protogenes attracted the attention of Alberti and Ghiberti in Florence, and the Calumny
became a subject of general interest in both literature and art. But the first portrait of
Apelles in his role as artist seems to be a print by the North Italian engraver Nicoletto
Rosex (or de Rubeis or Rosa), produced in the early years of the sixteenth century
(P1. Ia).46 The painter is presented in a romantic landscape setting as Poeta tacentes [sic],
an allusion to the celebrated dictum attributed to Simonides, that painting is mute poetry
(Plutarch, De gloria Atheniensium, III, 346f-47c). He silently contemplates a board bearing
geometrical figures, which is resting against the base of twin broken columns, the emblem
of strength, and perhaps an allusion to Apelles's famous power over princes.47
But it is the subject of Apelles's meditation that most excites our curiosity, for a board
decorated with simple shapes recalls that remarkable joint 'picture' by Apelles and
Protogenes, the three lines which Pliny recorded as having survived in Rome until th
time of the Caesars: 'and among the numerous works by excellent painters it was like a
blank (inani similem), and it was precisely this that lent it surpassing attraction and
renown' (N.H. xxxv, 83). The architect's square resting at the foot of the panel, and the
calipers depicted on the base of the ruined monument, suggest that the artist is indeed
primarily a geometer, as he had been presented in Ghiberti's first Commentary, where the
competition with Protogenes was held to concern conclusioni in prospettiva.48 But ifApelles is
simply a geometer, the presence of the lowest of the four figures, the octagon, on his panel
is puzzling, for the three 'primary' figures of the circle, the triangle and the square, were
generally considered in the Renaissance to be an adequate symbol of the whole of plane
geometry.49 If, on the other hand, we regard the figures of Apelles's diagram as symbolic
of solids: the sphere, the pyramid, the cube and the octohedron; we encounter another
numerical difficulty, for the doctrine of the primary bodies attributed in late Antiquity to
Pythagoras, and much discussed in north Italy around 1500, involved a conception offive
regular solids: the pyramid, the cube, the octahedron, the icosahedron and the dodecah-
edron.50 We are especially struck by the absence of the sphere, which crowns Nicoletto's
sequence, but Aitius had characterized the dodecahedron as 'the sphere of all', a notion
which was repeated by Luca Pacioli in his De divina proportione, published in Venice in
1509.51
46 A. M. Hind, Early Italian Engraving, 1938, v, no. 29. 48 L. Ghiberti, I Commentari, ed. O. Morisani, 1947,
Hind gives the fullest published account of this mysteri- p. 24.
ous figure (pp. o107 ff.). The figure of Apelles seems to 49 Cf. Geometria, in the Tarocchi series (Hind, loc. cit. n.
depend on that of the philosopher in Filippino Lippi's 46 above, pl. 343) and the tablet at the foot of the
Triumph of St Thomas Aquinas in S. Maria sopra Minerva astrologer in the 1524 Venice edition of Cecco d'Ascoli,
in Rome (A. Scharf, Filippino Lippi, 1950, fig. 76), which Acerba (repr. Gazette des Beaux-Arts, vIme ser, xxvII, 1945,
may allow us to date the print between 1507, when p. 209).
Nicoletto is documented in that city, and 1515, when
50 Aitius, Plac. II, 6, Dox. 334; Hermias, Irrisio Genti-
records of him cease. lium Philosophorum, 16, Dox. 655, both in M. C. Nahm,
47 For the broken column as representing fortezza in from Early Greek Philosophy, 4th edn, 1964, p. 59.
Selections
the Mantegnesque repertory, from which Nicoletto 5s Ed. Winterberg, 1889, pp. 84-85. Pacioli attributes
the notion to Plato, basing himself probably on a rather
drew so many of his graphic ideas, E. Wind, Giorgione's
vague reference in Timaeus 55c.
'Tempesta', 1969, pp. 2, I8-I9. See also Fabio Segni's
phrase in his early I6th-century epigram on Botticelli's
Calumny; 'Terrarum Reges parva tabula monet' (G.
Vasari, Le Vite ... , Milan 1963, I, p. 204).

122.225.227.202 on Fri, 01 Dec 2023 09:47:35 +00:00
But another series, of four geometrical solids,

discussion of the structure of the elements (Timae
Smyrna early in the second century A.D.52 Accord
based on a combination of triangles, progressing in
(pyramid), through air (octohedron) and water
which was a cube composed of forty-eight tria
pentagons. But if Nicoletto intended to refer to thi
including a circle, did he allude to the sphere, which
the Platonic doctrine, the circle in Apelles's diag
edron, for the pyramid, the cube and the octohedr
plane figures. The icosahedron, constructed from
solid that can be reduced to an easily legible plan
comprehensive solid, with the exception of the dode
and shows that it can circumscribe even the dod
Plato's discussion of water in the Timaeus by sug
surfaces in this figure led the philosopher to think
al moto dela cosa che spargendo scende: che de quell
was indeed intimately related to the sphere. If
schema, and expected his public to identify the
water, and the other three figures with the other el
this was appropriate to the subject of his print
Apelles as a four-colour painter, and of the associat
four elements, which, as we have seen, had been a
But Nicoletto's engraving is not a coloured one
forms as colours is not as straightforward as it m
ancient commentators we have noticed related spec
the exception of the author of the De coloribus, wh
white. On the other hand, the Democritan view t
coloured, that colour was indeed only a secondary q
the Middle Ages.54 The earliest writer to allocate d
to have been the second-century A.D. Athenian astr
Smyrna, drew up an elaborate table of corresponde
earth, red of air, white of water and yellow of fir
attributed by Pliny to Apelles's palette, but there-is
52 F. M. Cornford, Plato's Cosmology, 1937,

55 J. Seznec, The PP-
Survival 51,
of the 70.
Pagan Gods, 1953, p. 47.
53 Ch. LV (pp. 96 f.). Cf. Timaeus
Marius, op.49c,
cit. n. where
54 above, p.water
63, cites is
a contemporary
shown to perform a complete viewupwards and water
that earth is black, downwards
white, air yellow and fire
cycle. red; and a 15th-century French Lumen luminum de Colori-
54 The influential Ioth-century Arab writer Alfarabi
bus proposes the same four colours as 'principales', but
also sustained the view that the elements were them-apparently without reference to the elements (cit: D. V.
selves uncoloured, but manifested colour whenThompson,
mixed Isis, xxII, 1935, p. 468). Writing about the
1220O
(F. Dieterici, Alfarabis philosophische Abhandlungen, 1892, William of Auvergne gave the equivalents:
p. 139). See also the extended discussion by theair:blue,
12th- fire:red, water:purple (since it derives from a
century South Italian writer Marius, On the Elements, ed.
sea-creature), earth:grey (bissus) (De legibus, in Opera,
Dales, 1976, pp. 58 ff., and esp. p. 63: the ancients,
1674, I, p. 32) and in the most sustained medieval
discussing the elements, 'nullam de colore fecerunt
account of the relationship of colours to the elements,
unquam mentionem'. This view was restatedTheodoric about of Freiberg (I304/10) has - fire:red;
air:yellow;
12oo by Daniel of Morley, Liber de naturis inferiorum et water:green; earth:blue (De Iride, ed.
Wiirschmidt,
superiorum, ed. K. Sudhoff, Archiv fuir die Geschichte der Beitriige zur Geschichte der Philosophie des
Naturwissenschaften und die Technik, vIII, 1917, pp.Mittelalters,
I f. xII, 1914, pp. 82-83).

122.225.227.202 on Fri, 01 Dec 2023 09:47:35 +00:00
12 JOHN GAGE
known in the fifteenth century, an
his treatise on painting, his corr
regard black and white as basic col
air, green to water and ash-colo
Leonardo, rehabilitated black an
without them', but he identified
them: his equivalents correspond
excluded yellow and introduced gre
If we were to interpret the diagra
of Pacioli and Leonardo, reading
equivalents:
Circle = icosahedron = water = green
Triangle = pyramid = fire = red
Square = cube = earth = yellow
Octagon = octohedron = air = blue
But this would be to assume the universal acceptance of the Platonic equivalents, which
was not the case: Leonardo, for example, disputed the identification of the cube with
earth; for him, the pyramid was a more stable body, for it had fewer surfaces than the
cube, and was thus a more appropriate image of earth.58
The question is further complicated by the persistence into the Renaissance of another
four-colour scheme, which derived from the late-antique symbolism of the Imperial
horse-races. Here the association was chiefly with the four seasons, and Alberti intro-
duced it into his treatise on architecture in the form of green for spring, red for summer,
white for autumn and black (fuscus) for winter.59 But the late-antique authors who had
discussed the question, notably Tertullian and Cassiodorus, although they allotted red
and green to these same seasons, gave blue to the autumn and white to the winter.60
Theon of Smyrna had proposed the equivalents:
Spring = pyramid (i.e., fire: red)

Summer = octohedron (i.e., air: blue)
Autumn = icosahedron (i.e., water: green)
Winter = cube (i.e., earth: yellow)
56 L. B. Alberti, On Painting and on Sculpture, ed. Gray- directly to air) is interesting, but he already had a
son, 1972, p. 46. Although he refers elsewhere to the perfectly good earth-yellow available to him in ochria
four-colour painters (p. 86), Alberti does not attempt to (cf. Cennini, Ch. XLV). Nor does Alberti seem to have
relate their palette to the four elements. In his own been concerned to use colour to create rilievo, as is
Italian version (ed. Mallk, 1950, p. 63), he stresses that proposed by Gavel on p. og9. S. Y. Edgerton, 'Alberti's
ash-colour is a form of grey (bigio), and some I6th- Colour Theory: a Mediaeval bottle without Renaiss-
century MSS of Della Pittura expand on this still further: ance wine' (this Journal, xxxII, 1969, pp. 123 ff.) glosses
'E perche la terra e feccia di tutti li elementi, forsi non over some important differences between the outlook of
diremo male tuti i colori chiamarsi bixi come feccia de la Alberti and that of the ancients.
terra' (C. Grayson, 'The text of Alberti's De Pictura', 57 C. Pedretti, Leonardo da Vinci on Painting: a Lost Book,
Italian Studies xxiii, 1968, p. 76). Cf. also Marius (op. cit.I965, P. 56. Green and yellow had often been confused
n. 54 above, p. 58) that earth may be black, white, red orsince Antiquity: see above, n. 12.
yellow. The recent attempt of Jonas Gavel to interpret 58 A. Uccelli, I libri di meccanica di Lionardo da Vinci,
Alberti's bigio or cenericcie as a yellow, albeit an unsatu- 1940, p. I.
rated one, does not seems to be very satisfactory (Colour: 59 L. B. Alberti, De re aedificatoria, viii, 8, ed. G.
A Study of its Position in the Art Theory of the Quattro- andOrlandi and P. Portoghesi, 1966, p. 751.
Cinquecento, Stockholm, 1979, PP. 49-5I). Certainly the 60 For a useful tabulation of these equivalents, Ter-
suggestion that Alberti needed a colour-term to relatetullian, De Spectaculis, ed. Castorina, 1961, pp. lxxxiv f.
directly to earth (as his blue, celestis seu caesius related

122.225.227.202 on Fri, 01 Dec 2023 09:47:35 +00:00
and Antiochos of Athens had
Spring = red
Summer = yellow
Autumn = black
Winter = white.
Thus we are presented about 1500 with a number of apparently arbitrary and conflictin
attempts to establish analogies, to accommodate a fluctuating range of'basic' colours and
shapes to the more firmly established number of elements and seasons. No clear reasons
for preferring one colour to another had yet emerged, probably because there was still
remarkably little interest in that aspect of colour which we regard as the most importa
one, namely hue.61 It seems clear that, ifNicoletto's portrait ofApelles was understood t
refer primarily to the four-colour palette, his public would have been hard put to it to
identify which these colours were, even with the help of Pliny's story.
DURER, TITIAN AND THE CRITICS

Apelles's fame made his the most appropriate name to invoke in a learned compliment
to a modern artist, and it was so used in the Middle Ages, sometimes in the most unlikely
places.62 Luca Pacioli's references to Leonardo's art as surpassing that ofApelles, Myron
and Polyclitus fall into this purely adulatory category,63 but the growing interest in the
formal qualities and subject-matter of Apelles's work during the fifteenth century lead us
to expect both more critical and more concrete allusions to him in the High Renaissance,
especially in contexts where painters and scholars were becoming more interdependent in
their treatment of the Classical past. Two painters of the sixteenth century closely
associated with humanist scholarship were Diirer, whose Classical interests were for a
time remarkably close to those of Erasmus, and Titian, whose circle in the middle years o
the century included a number of scholars and polymaths working for the Venetian
publishing houses of Giolito and Marcolini. Both were compared to Apelles more
persistently than any artist before them, and in both cases it was their mastery of colour
that seems chiefly to have invited the comparison.
Diirer and Erasmus at first cultivated their interest in Apelles quite independently of
each other. The painter had been characterized as altero Apelle by Conrad Celtis as early as
I50O;64 and a few years later Diirer was using the loss of the ancient writings on art
including those of Apelles (Pliny, N.H. xxxv, 79) as a stimulus to the compilation of his
own treatise on painting, most of which was itself never to be published.6s Thus it was
Apelles the theorist who seems chiefly to have interested Diirer; for Erasmus, on the other
hand, it was Apelles the satirist: he had edited Lucian's Slander, the only early source of
61 I have discussed the problems of interpreting 63 De divina proportione, ed. cit. p. 33; P. Speziali,
medieval colour terms and symbolism in 'Colour in
'Leonardo da Vinci et la Divina Proportione de Luca
History: Relative and Absolute', Art History, I, 1978,
Pacioli', Bibliothique d'Humanisme et de Renaissance, xv,
pp. 105 ff. 1953, PP- 302 f.
62 A Ioth-century Byzantine example is discussed as 64 D. Wuttke, 'Unbekannte Celtis-Epigramme zum
'merely a display of learning' by C. Mango, 'Antique Lobe Diirers', Zeitschrift fur Kunstgeschichte, xxx, 1967,
Statuary and the Byzantine Beholder', Dumbarton Oaks p. 322. See also A. Diirer, Schriftlicher Nachlass, ed. Rup-
Papers, xvii, 1963, pp. 65-66. A German example of prich, I, 1956, p. 255; cf. p. 290.
around 1500 is recorded by H. Huth, Kiinstler und Werk- 65 Ibid. In, 1966, pp. 99 f.; also pp. 109, 13 (1512) and
statt der Spatgotik, 2nd edn, 1967, p. 69. Several other 135 (1523); I1I, 1969, p. 438 (1527/28).
medieval and Renaissance examples are adduced by
E. Panofsky, 'Erasmus and the Visual Arts', this Journal,
xxxII, 1969, P. 223.
2*

122.225.227.202 on Fri, 01 Dec 2023 09:47:35 +00:00
14 JOHN GAGE
knowledge about the Calumny of A
his edition of the New Testament t
version of this subject for the title-
the Netherlands in I520, when the p
he soon became nostrum Apellem, a
friend, Willibald Pirckheimer.67 Bu
made the most interesting comparis
into his Dialogus de recta latini graec
which he claimed that the painter w
had only been able to attain with se
that, whereas the Greek painter had
Diirer was supreme in his day witho
be reflecting his friend's own views
thon recalled in 1546 how the pai
complexity and the vivid colouring o
favour of greater simplicity, and D
drapery painting of about 1512/1
chapter in his projected treatise on
of natural relief as opposed to th
materials.69 But that Erasmus did n
refer to it either in his notes to Pliny'
the introduction to his new editio
damaged Venus, 70 suggests that th
and that it had no significance for D
Could Pliny's story have any direc
even the identity, of the four colou
seventeenth century that it found
and especially with Titian. In his lif
this painter was the first to employ
imitation of nature, especially in the
which was imitated by Giorgio with a f
express, which procedure was also follo
by the illustrious painters Apelles, Aet
four colours (colori) to constitute flesh-tin
66 R. F6rster, 'Die 69
Verliumdung
Diirer, op. cit. n. 64 above, I, p. 289; 11, pp. 393des
f.; Apell
Renaissance', Jahrbuch (cf. ibid.
der II, pp. 94Preussischen
ff.) Diirer's note is hardly the chapter Kunstsam
on colourDiirer's
vIII, 1887, pp. 93 f. For itself, as claimed by W. ownJ. Hofmann,Calumny
Uber f
Nuremberg Town Hall, Diirer,
Diirers Farbe, 1971, op. cit. n. 64
p. 17. D. B. Kuspit, '"Melanchthon
p. 267. The sketch isand Winckler
Diirer": the search for the Simple no.Style', Journal
922; of the pa
probably executed by Georg
Mediaeval and Renaissance Studies,Pencz.
III, 1973, pp. I88 f., Erasmu
1518 of the response shows thatto this andhisrelated remarks
earlier reported by Mel- edition;
mihi quidem adversus anchthon form part of the humanist's
istorum argument in
calumnias suff
mus bene sibi conscius favour of the...
simple'style(Erasmi
in language. Epistolae,
no. 809), Cf. also the70 Epistolae, Apologia cit. ni, 503 f.; vi, pp. 16
forf. and letterthe
1544. 1518/
p. 82. 71 C. Ridolfi, Le Meraviglie dell'Arte, 1648, ed. Hadeln,
67 Epistolae, cit., nos 1398 (1523) 1536, 1558 (1525)- 1914, I, p. 107. Giorgione's epitaph (ibid.) makes the
68 Diirer, op. cit. n. 64 above, I, p. 297. The fullest standard reference to Apelles and Zeuxis as his peers.
commentary on this text is in E. Panofsky '"Nebulae in
Pariete": Notes on Erasmus' Eulogy of Diirer', this
Journal, xIv, 1951.

122.225.227.202 on Fri, 01 Dec 2023 09:47:35 +00:00
Modern analysis of Giorgione's flesh-painting has shown an avo
ments,72 and it is particularly interesting that Ridolfi confines the r
instance of the Venetian method. It was as a flesh-painter, too, that
the closest emulator of Giorgione, although he was equally cert
interest in blue, which, in combination with red in drapery, '
(sconciare) the figures'.73 Another seventeenth-century Venetian
Boschini, referred to Titian's preference for a palette in his underpai
white and red, and he may also be referring to the preparation of f
interest in Apelles can hardly be doubted; his three-quarter-length
present on loan to the National Gallery of Scotland) was clearly insp
Apelles's Venus Anadyomene, whose lower portion had been dama
9i),75 but even in this picture Titian did not avoid the use of blue, w
present, both in sky and sea.
It became a commonplace to link the names of Titian and Apelles
Venetian art-criticism which grew up about the middle of the sixte
writings of Pietro Aretino, Anton Francesco Doni and Lodovico Dolc
direct contact with the painter, and all of them show a knowledge of
recounted in Pliny's Natural History.76 In several letters between 15
one to Titian himself, Aretino compared the painter to Apelles,
Titian's colour which chiefly aroused his admiration, he nowhere ad
story as an example of the Venetian painter's mastery." Doni, in his
of 1549, made a comparison between Apelles and Titian as physiogno
he showed a particular interest in pigments, and especially in th
72 See H. Ruhemann's technical analysis of the Berlin
Boschini based his account of Titian's tech
information
Portrait ofa Youth in G. M. Richter, Giorgio from the painter's pupil Palma
da Castelfranco,
1937, p. 126, which identifies only black, white,
and red
it has and accepted as authentic by most
been
brown. In the early Pala di Castelfranco, which
scholars hasKennedy, 'Apelles Redivivus', E
(R. W.
been the subject of a recent technical analysis,
Memory of Gior-
Karl Lehmann, 1964, pp. 167 f.; E. Pa
gione used a palette including four reds, three yellows,
Problems in Titian, mostly Iconographic, 1969, p
ultramarine, three greens and black and white, but
Grunewald no
(op. cit. n. 73 above, p. 202) links t
samples were taken from the flesh areas (L. the
with Lazzarini in
late painting of flesh, but both the wo
Lazzarini, Pedrocco, Pignatti, Spezzani
aboveand
(n. Valca-
71) are late, and yellow is not in
nover (eds), Giorgione; La Pala di Castelfranco Veneto,
alleged triad. The characteristic purplish glaze
1978, pp. 46-47). by her (p. 137) in the Borghese Sacred and Profan
73 Ridolfi, op. cit. n. 71 above, pp. 154,
also209. A i9th-
clearly present on the ankle of Lucretia in
century restorer, Palmoroli, thought that Titian used
and well-preserved Tarquin and Lucretia in Cam
ultramarine as an underpainting for flesh (0. Kurz,Museum:
(Fitzwilliam op. Catalogue of Paintings, in,
914).
cit. n. 6 above, p. 94): and it has been identified in the
shadows of Ariadne's flesh in the National Gallery
75 Kennedy, op. cit. n. 74 above, p. 162; H. Wethey,
Bacchus and Ariadne (J. Plesters in National
TheGallery
Paintings of Techni-
Titian, iii, 1975, no. 39. Cf. also the patent
cal Bulletin, n, 1978, p. 40). The most
of attentive early
knighthood granted by Charles V to Titian in 1533
modern analyst of Venetian flesh-painting
(Wethey, n, found nothe letter of 27 September
1971, p. 7) and
blue in Titian's usage (M. Grunewald, Das Kolorit
1559 from in
Titian to der
Philip II (E. Tietze-Conrat, 'Titian
venezianische Malerei, I, Die Karnation, 1912, pp. 133
as a letter-writer', Artff.)
Bulletin, xxvi, 1944, p. 120).
Grunewald stresses Titian's liking for yellow inmembership
76 For Titian's flesh, of the Accademia Pellegrina,
and there are several yellows in the back of the
which included Doni man
and Dolce, by 1552, P. F. Gren-
offering tribute in the Tribute Money dler,(NG. 224);
Critics of the and on i530-i560, 1969, p. 58.
Italian World,
the wrist of the Diana in the Death ofActaeon (NG.
77 P. Aretino, 6420),
Lettere sull'Arte, ed. Camesasca, I, 1956,
there is a bright yellow which seems to be 242
pp. 156, the same
f., ii, pp. 192,as 198, 200, 221. For Aretino's
that in the foliage of the nearby trees. frequent borrowings from Pliny, G. Beccatti, 'Plinio e
74 M. Boschini, Le ricche Minere della pittura veneziana,
l'Aretino', Arti Figurative, in, 1946, pp. 1-7. For his aid to
1674, p. 27. Crowe and Cavalcaselle convert this story
Titian as a letter-writer, Ridolfi, op. cit. n. 73 above,
into a dictum 'according to a traditionp.still
208. preserved'
(The Life and Times of Titian, 2nd edn, I88I, 11, p. 125).

122.225.227.202 on Fri, 01 Dec 2023 09:47:35 +00:00
16 JOHN GAGE
flesh, he, too, made no reference to
of the lack of interest in the four-c
was one of the most prolific poligra
vigorous supporters of Titian, w
L'Aretino, published by Giolito in
brownish softness in flesh-tints
particularly cultivated brown, but h
restricted palette.79 Dolce certainly
most modern form. In 1565 he pu
diversita" e proprietat de i colori, much
Coloribus of Anthonius Thylesius
Thylesius had linked Pliny's accou
palette, and Dolce reproduced this p
protest that he wrote as a philologi
'ci6 [the painter's view] apparte
attempt to suggest that Titian wa
four-colour theory had enjoyed an
1540s to the 156os, we may be sure
avid for copy, would have made the
Even in the latter half of the cent
more closely in the light of painter
Titian's methods did not seem t
remarkable illustrated catalogue com
in the I620S, which includes repro
Titian, has a Latin preface extracted
Montjosieu (Demontiosius). Montj
teenth century under the title Comm
early discussions of the four-colour
it must be noted here that, althoug
painting is reproduced by Vendram
is not.83 Clearly it was of no interest
been associated with Venetian six
78 A. F. Doni, 82
Disegno, There1549:
Venice is no on
discussion
colours,
on flesh-painting, Paolo
pp. 9v, Pino, Dialogo
I4v; on di
Apelles P
(an
pp. 37V ff. For a brief Pino, as a
account painter,
of was
Doni's career,
op. cit. n. 76 above, pp.nique, admired Titian, a
49 ff.
Apelles. and
79 M. Roskill, Dolce's 'Aretino' NorVenetian
is it especia
Art
the Cinquecento, 1968, nichi in the On
pp. 152-53. commentar
p. 299 Ro
this with Apelles's dark (Venice,
varnish,1561),
which althoug
is also r
as bruno in the lettererences
of G. B.to contemporary
Adriani to Vasar
For other referencesedn, 1573, pp.
to Apelles: io87,
pp. II
164-67
been174-75.
148-49, 150-51, 156-57, close to Doni in the
so Thylesius, op. cit.above,
in J. pp.
W. 52 ff.,
von 66 ff.
Goethe, op
83 T. Borenius,
above, p. I 18; Dolce, Dialogo The Pictur
... dei colori, 156
where he also suggested1923, pp.
that 12 ff.; Lodovicu
painters still use
de pictura,
white of Milo. He referred in Vitruvius,
to Thylesius's book
8s Thylesius, op. cit. p. I II;
Antwerp Dolce,
1649, op.
pt III, ci
pp.
above, p. 7r. Dolce also refers to Titian on
64v. For a brief account of Dolce, Grendler, op
above, pp. 65 ff.

122.225.227.202 on Fri, 01 Dec 2023 09:47:35 +00:00
four-colour palette to Giorgione in I648 is, like Pliny's own account, part of a
against the garishness of modern painting, and particularly against the flesh-tints o
orange and blue' which he had noticed among 'some moderns', by whom he pr
intended Barocci and Rubens.84 There is little reason to suppose that the sombre pa
was ever an issue with the earlier Venetian painters themselves, and they wou
found the exclusion of blue especially difficult to accept. It is an irony of history
through the work of philologists like Montjosieu in the sixteenth century, blue ha
now become firmly established as a component of the four-colour palette, and that
palette could now be at the centre of the development of the modern system of pr
colours.
THE IDEA OF THE PRIMARIES
Closely related to the question of colour mixture is the ques

number of colours necessary to re-constitute the whole range of v
which we generally take for granted, but which even nineteenth-ce
which we are heirs) did not settle, proposing as it did differe
primary colours according to whether colour was seen from t
physicist, the painter or the psychologist.85 The early history
colours is a complex one, and part of it has been introduced above
very late development of the modern subtractive triad of red,
context of painting, is itself witness to that reluctance to experime
I have attempted to document in this study. A further obstacle to
coherent doctrine of primaries was the linguistic difficulty in
concepts of colour classes from a range of specifically coloured su
account of the four-colour painters, did find it possible to disting
and abstract colour-terms: the white of Milo was 'ex albis', th
rubris', atramentum was 'ex nigris'; but here two of the designation
the places of origin of colouring materials, and his fourth colour, a
of origin, sil from Attica, was such a doubtful term that, as we sha
of an important controversy in the sixteenth century. Favori
Gellius, recognized the poverty of Greek and Latin colour-term
capacity of the eye to discriminate between nuances of colour, and
colour categories far more wide-ranging than we would care to
purple (ostrum) on the one hand, and yellow (crocum) and go
technical literature of the Middle Ages for the most part avoid
84 Ridolfi, op. cit. n. 71 above, p. 107.

WienerFor the
Jahrbuch fir later
Kunstgeschichte, xxx/xxxI, 1977/78,
interpretation of Rubens's flesh-painting in terms of a
pp. 102 ff.
few simple, bright and optically-mixed 86 colours,J.
But see Gage, op. Gage,
cit. n. 61 above, pp. Io8-o9, for
purple
Colour in Turner: Poetry and Truth, 1969, pp.as a62
red,f.and
andfor the
H. antique taste for reddish
von Sonnenburg, 'Rubens' Bildaufbaugold. und Technik;
The 5th-century B.C. II:
writer Ion of Chios also tells a
Farbe- und Auftragstechnik', Maltechnik-Restauro,
story about 'purple' which suggests that the Greeks
were perfectly aware of the discrepancy between colour
1979/3 n.p.
85 For modern views of the arbitrariness of the usually perception and colour terminology (see L. Gernet,
accepted 'primary' set, red, yellow and blue, E. E. 'Denomination et perception des couleurs chez les
Gloye, 'Why are there Primary Colours?', Journal of Grecs', in I. Meyerson, (ed.), Problimes de la couleur, 1957,
Aesthetics and Art Criticism, xvi, 1957/58, pp. 128 ff.; E. p. 319).
Frodl-Kraft, 'Die Farbsprache der gotischen Malerei',

122.225.227.202 on Fri, 01 Dec 2023 09:47:35 +00:00
18 JOHN GAGE
abstract colour terms, and simply l
late medieval texts show an aware
pass quickly to the discussion of p
illuminandi, for example, attribu
colours: black, white and red, and
proceeds to show that the illuminato
terms niger, albus, rubeus, glaucus,
classifies the pigments yielding t
artificial, and arrives at a figure of
Cennini, in the chapter on colours i
seven 'natural' colours, then opts fo
the last three of which are 'natural
ultramarine or azurite, and giallor
status of 'natural' colours, nor with
colours; he had little interest in theo
at the start of his account.
Several early sixteenth-century Venetian writers record and deplore the prevailing
confusion about the nature and number of basic colours,90 but with the increasing use of
mixtures, especially in oil-painting, we might expect that some more empirical approach
to the problem would emerge in the course of the century from a painterly milieu. It has
rightly been suggested that the most important feature of the first modern colour systems,
which make their appearance about 600oo, is the new prominence given to blue.91
Although blue had been an important colour throughout the Middle Ages, it does not
seem to have formed part of a set of 'basic' colours; we have seen that green was far more
usual in such sets. But in the sixteenth century when, as has often been noticed, Titian
87 See especially the list of thirteen colours in a I2th-clearly an error, some earlier editors have emended the
century Anglo-Norman MS of the Mappae Clavicula,reading Pliniam to physicam.
published by H. Roosen-Runge, Farbgebung und Technik 89 The Craftsman's Handbook, Ch. xxxvI. The division
friihmittelalterlicher Buchmalerei, 1967, I, pp. 185 ff. and theof pigments into 'natural' and 'artificial' is an ancient
glossary in vol. ii. A slightly amplified copy with fifteenone, but for Vitruvius (vii, vii) the 'natural colours' were
colours is in the 14th-century French (?) Liber de colori-yellow-ochre (sil), red ochre, minium, white, green and
bus, published by D. V. Thompson, Speculum, I, 1926, yellow (orpiment). His epitomiser, Faventinus (c. 300
p. 288. The 15th-century Portuguese Livro de como seA.D.) omits sil, but adds the blues chrysocolla, armenium
and indicum (De diversis fabricis architectonicae, ?27, trans.
fazen as CUres, lists ten cores principaes, most of them the
names of pigments (Portuguese text in Todd memorialW. H. Plommer in Vitruvius and Later Roman Building
Volumes, I, 1930, p. 8o; translation by D. S. Blondheim,Manuals, 1973, PP. 74 ff.) For the medieval interest in
Jewish Quarterly Review, xIx, 1928-29, P. 130.) M. F.Faventinus, in the context of pigments, K. W. Grans-
Edgerton points out that the word color in a 15th-centuryden, 'The Interpolated Text of the Vitruvian Epitome',
German Tractatus de Coloribus usually refers to a colour-this journal, xx, 1957, p. 370. For Michelangelo Biondo,
ing agent, rather than to a concept (Mediaeval Studies,Della nobilissima pittura, 1549, P. 21r, the 'natural' colours
xxv, 1963, p. 194). An exception to this general rule is in were blue, red, yellow and green, plus black and white.
the I3th-century additions to Eraclius de coloribus et artibus 9o For example, Mario Equicola, Libro di natura
romanorum, ?50: De diversis colorum principalium et interme-d'amore, 1526, in P. Barocchi (ed.) Scritti d'arte del Cin-
diorum speciebus ..., which lists black and white in quecento, uI, 1973, P. 2153; F. P. Morato, Del significato dei
several varieties, and then the intermediaries, rubeus, colori, 1535, in ibid. p. 2176. Cf. Also R. Borghini, II
viridis, croceus, purpureus, prasinus, azur and indicus, only Riposo, 1584, p. 230.
the last of which is clearly a pigment, although the list 91 C. Parkhurst, 'Camillo Leonardi and the Green-
includes two blues and two greens (M. Merrifield, Blue Shift in Sixteenth-century Painting', in Intuition und
Original Treatises ... on the Arts of Painting, I, I849, pp.
Kunstwissenschaft, Festschrift fuir H. Swarzenski, ed. P.
244-45). Bloch et al., 1973, esp. p. 425.
88 De arte illuminandi, ed. Brunello, 1975, PP. 36 ff.
Since the attribution of a three-colour theory to Pliny is

122.225.227.202 on Fri, 01 Dec 2023 09:47:35 +00:00
showed an especial liking for blue, the view that this colour was es
set began to gain ground and, by a curious linguistic error, to be i
context of Pliny's four-colour system. The problem was the pr
Pliny's Attic sil. In the Natural History (xxxIIi, 158) he had state
ochre) and caeruleum (probably azurite) occur in gold and silver mi
led to a confusion of the two pigments, a confusion all the easier i
when a term for yellow, cerulus (perhaps from cera: wax), seems to h
fifteenth-century Venetian commentator on Pliny, Ermolao Ba
first to make the confusion,93 and he was soon followed by a num
Cesare Cesariano, in his commentary to the i521 Como edition
categorically that sil was ultramarine (although he here confused la
and Thylesius in 1528 accepted simply that silaceus was 'inter coeru
until the middle of the sixteenth century that the question app
important topic in the painterly context of mixture. In a treat
probably after i563, when he became luogotenente of the Flor
Disegno, Vasari's friend Vincenzio Borghini adduced Pliny's at
expensive pigments, citing the four-colour story; but he was
colour-equivalent for attico, as he had done for all the other terms
whether the Greek painters could ever have managed without b
French encyclopaedia of the arts, the Syntaxeon Artis Mirabilis of Pi
92 The earliest unambiguous use of ceruleus or caeruleus

Library, cited by V. Rose, 'Aristoteles de Lapid
as a yellow I have traced is in Matthew Paris's
Arnoldus illus-
Saxo', Zeitschriftfiir deutsches Altertum
trated inventory of the jewels of St Albans - compiled
1875, P- 345. See also n. 97 below.
in 1257 - (British Library, MS Cotton93 Hermolai
Nero Barbari in C. Plinii Naturalis histor
D I, fol.
146): 'gemmam oblongam coloris cerulei, videlicet
Castigationes (I493), Basle 1534, P. 378. In
topazium', to refer to a stone painted yellow
(p. (Luard's
465), where he discussed the four-colour
edition: M. Paris, Chronica majora: Additamenta,
Barbaro was (Rolls
much less certain that sil could be a
Series) vi, 1882, p. 383, reads 'caerulei').
well The Summa
as yellow ochre. For Barbaro's otherwise
philosophiae which has been dated 1265/75
approach andtoattri-
editing Pliny, V. Branca, 'Ermo
buted to Robert Kilwardby (C. K. McKeon,
baro e A Study of
l'Umanesimo Veneziano' in Umanesimo
the Summa Philosophiae of the Pseudo-Grossteteste, 1948,
umanesimo veneziano, 1963, PP. 193 if.
pp. Io ff.) speaks of 'color caeruleus et maxime scintil-
94 Vitruvius, De Architectura, 1521, p. cxxv (w
lans, qualis est topasius chrysopassus itemque
erence to chrysoli-
Vitruvius, vII, xiv, where imitation At
tus' (i.e., yellow stones: Die Philosophischen Werke
described as andesinfusion of violets, although th
Robert Grosseteste, ed. L. Baur, 1912, p. ing631). Theodoric
colour, called sillacetus in the later Middle
of Freiberg, De Iride, ii, i, p. 60o (see yellow
above, n. 55) has
(Merrifield, op. cit. n. 87 above, I, pp.
'caeruleus, quem Xancton vocant', from the Greek
Thylesius, op. cit. n. 8o above, p. i 18. G. Phila
xanthos. The 8th-century Glossaria abstrusa and abolita
decem libros M. Vitruvii Pollionis de architectura ann
give caeruleus and ciruleus as viridis, glaucus vel niger
Rome 1544,(Gloss-
p. 232, notes Ermolao Barbaro's v
aria latina, ed. Thomson & Lindsay ni, prefers
1926,to pp.regard
20, sil as 'coloris purpurei violac
I io--I I and Mittellateinisches Wo6rterbuch,ianthinus dicitur)'. Veronese's patron Daniele
s.v. 'caeruleus';
and glaucus, although it is usually translated as 'grey'
took the more comprehensive view that sil was a
(see above, n. 17), could also mean 'yellow'.of ochre,See,'ma for
di colore alquanto diverso, o che
example, Roger Bacon, Opus majus, pt.all'azurro, o aled.
, vI, ch. xii, purpureo, & violino' (I dieci libr
Bridges, 1897, II, p. 197: 'dividendo glaucumchitettura in di
plures
M. Vitruvio (1556/67), Venice i629,
gradus, ut caeruleum et puniceum' and 95 especially Liber
P. Barocchi, op. cit. n. 90 above, I, 1971, pp.
de sensu et sensato, ed. R. Steele (OperaIthactenus
was partly inedita
at Borghini's request that G. B
Rogeri Baconi, xiv, i937, PP. 70 ff.)supplied
J. Maclean has
the resumiof the history of Greek paint
correctly translated Bacon's 'glaucitas' as 'yellow':
1568 'De
edition of Vasari's Lives, in which he incl
Kleurentheorie in West-Europa (I200-i500)',
stories ofScientia-
the competition with Protogenes, and
rum Historia, viIi, 1966, p. 40. Theodoric of Freiberg,
(bruno) varnish, De
but nothing on the four colours
Iride, xLIV, has 'citrinus sive glaucus', G.
as Milanesi,
does the I5th-
I, 1878, pp. 15 ff.).
century collection of lapidaries in Prague University

122.225.227.202 on Fri, 01 Dec 2023 09:47:35 +00:00
20 JOHN GAGE
following an earlier French comm
ianthinus, a shade of violet, but he a
colours all others could be mixed.96 But the most substantial discussion of the whole
question was made in 1585 by the French mathematician, philosopher and bear-le
Louis de Montjosieu, who had arrived in Rome with the Duc de Joyeuse in 1
Montjosieu was one of the first scholars to test the descriptions of ancient style
reference to contemporary practice; he rejected the view that the competition betwee
Apelles and Protogenes had been a demonstration of skill in drawing supremely thin l
for he had not found such an emphasis in the drawings of modern artists like Raphae
Michelangelo, Salviati, Polidoro da Caravaggio, Correggio and Titian.97 When he c
to discuss Pliny's account of the four colours, not surprisingly he found Philand
alternative of purple or yellow for sil confusing, and he stated clearly that one of the f
colours must have been blue (unum oporteat esse caeruleum),
For it is certain that these four colours, white, black, red and blue, are the fewest that are neede
painting, and from a mixture of which all the others are composed.98
However, Montjosieu went on to list a number of mixtures which may cast doubt upon
practical experience; we may accept his grey (cineraceus) composed of black and white,
his brown (fulvus) made from red and black, but his green is a mixture of red and blue,
his yellow (luteus) a mixture of green and red, which suggests that he was still thinking v
much in Classical terms, for there is no suggestion that he was concerned with op
mixture, which might conceivably produce these results.99 What is important for us i
that Montjosieu stressed the dependence of all colours upon these basic four, that Atti
(as opposed to other types ofsil, which might be purple or yellow) is always blue, and
his views were several times re-published, and widely read, during the sixteenth
seventeenth centuries.
For the several accounts of the primary colours which appeared about I6oo, and
which established the modern subtractive triad of red, yellow and blue, appealed to the
experience of mixture in painting, although for the most part they were written by
Ferri was inclined to return to the 'outline' theory (A.

96 Petrus Gregorius, Syntaxeon artis mirabilis (i574),
Minto, Le Vite dei pittori antichi di C. R. Dati e gli studi
Leiden 1576, pp. 563 ff. Cf. the remark on mixing in the
index: 'Coloribus quattuor omnes alios misceri'. Else- erudito-antiquari nel Seicento, I953, p. - I6).
where (p. 242) Gregoire describes black and white as98 'Certum enim est, in pictura minimum opus esse
his quattuor coloribus, albo, nigro, rubro & caeruleo.
'praecipui colores', and proposes a scale with five inter-
mediates: 'albus, glaucus, puniceus, ruber, purpureus, Quorum mixtura ceteri omnes componuntur', op. cit.
n. 83 above, p. 61.
viridis, niger'. J. C. Scaliger, De Subtilitate ad Hieronymum
Cardanum (1576), Frankfurt i6oi, p. 1047 included sil99 These mixtures were followed by G. J. Vossius, De
among the blues. quatuor artibus popularibus, I650, pp. 74 f., ?27, but were
97 Op. cit. n. 83 above, pp. 59-6o. Montjosieu pro-
already attacked by Johannes Schefferus, Graphice id est
posed instead that the 'lines' were tonal areas corre-
de arte pingendi, 1669, pp. 161 f., ?44 (see A. Ellenius, De
sponding to high-light, shadow and mid-tone (splendor)
arte pingendi: Latin Art Literature in Seventeenth-Century
in which the hue of each colour is clearest. This solution Sweden and its International Background, 1960, pp. 181 ff.).
has been adopted in a modified form by GombrichI. (loc.Schiffermiiller, Versuch eines Farbensystems, 1772, pp.
cit., n. 5 above), who has also solved the four-colour
38 f., cited these mixtures as an instance of how even
problem rather neatly by proposing a blue ground learned
for men could err if they neither experimented
themselves nor consulted artists who did.
the panel. It was probably Montjosieu's example which
led Carlo Dati (who cites him on the four colours in Vite
dei pittori antichi, 1667, p. 169) to consult the contempor-
ary painter Ciro Ferri on the nature ofApelles's line, but

122.225.227.202 on Fri, 01 Dec 2023 09:47:35 +00:00
physicians.100 Robert Boyle, writing in mid-cen

painters had guaranteed for natural philosophers
primary set:
... the mixing of Pigments being no inconsiderable p
Incroachment in me to meddle with it. But I think I ma
to the making of a Transient mention of some few of the
so far forth, as may warrant me to observe to you, tha
Colours (if I may so call them) from whose various com
For though Painters can imitate the Hues (though n
Numberless differing colours that are to be met with in
yet found, that to exhibit this strange Variety they need
Red, and Blew, and Yellow; thesefive, variously Compoun
being sufficient to exhibit a Variety and Number of Co
Strangers to the Painter's Pallets, can hardly imagine.10
But the growing conviction that the primary col

white stood in some sense outside this triad, even if
threw the validity of Pliny's account still further
Boulenger hedged his bets in the I62os by claiming
from either three or four colours, without naming
scheme, which Boyle clearly presents as a novel
generally accepted until the end of the century
discovered to be a blue, it could not also be a yellow
no closer to being assimilated to the new triad of pr
and identity of these basic colours continued t
writing at the end of the century, characterize
'capitales', but did not specify them, although in
100 V. A. Scarmilionii, De coloribus, Marburg

coloribus, as well as 16oi,
the work of T
printed
p. I 12, where, however, it is argued as appendixes
that to J. Actuar
painters cannot
mix certain colours from these I6th primaries.
century (cf.Anselm de the 1548
p. 259 of
Boodt, Gemmarum et lapidumrius).historia, Hanau 16O9, I, pp.
viii, 8. See C. Parkhurst, 'A Color-theory from Prague:
101 R. Boyle, Experiments & Considerations touching
Anselm de Boodt, 1609', Allen Colours, Memorial
1664, pp. 219-2 1. Art Museum
Bulletin, XXIX, I971, pp. 3 ff. 102 J. C. Bulengerus, De pictura, plasticawho
Scarmiglioni, & statuaria,was
professor of theoretical medicine
Lyons 1627, p.at Vienna,
io6. Cf. dedicated
pp. 1o, 14 on Pliny's four-colour
his treatise to Rudolph II, one
story,of whose
reproduced physicians
without comment. For a comparable was
de Boodt, and de Boodt's viewsvagueness, on
Pierre lethe primaries
Brun, Recueil des essais des merveilleshere
de
may depend on his. Another lacontemporary
peinture, 1635, in Merrifield, op.theorist
cit. n. 87 above, ii,has
also been discussed by Parkhurst:
pp. 771-73 (three 'Louis Savot's
or four colours), nova-
811 (four: white,
antiqua Color Theory, 16o9', Album Amicorum
black (obscure), red, green). J. G. van
Gelder, 1973, pp. 242 ff., but I 103
have
It is clearbeen unable
from a passage to trace
in the Schilderboek of Karel a
three-colour theory to the text he cites
van Mander (Nova,
that this painter seu
still regarded verius
Attic sil as
nova-antiqua de causis colorum yellow: he congratulated the
sententia, modernsSavot
I6o9). on having four
does
yellows where
indeed offer a four-colour theory ofthe ancients
the had but one (ed. Hoecker,
primaries (ibid.
Index and p. 6r ff.), which is1916,
clearly based
pp. 302 ff.). This is also theon Pliny
tone of a remark by(cf.
G. B. Birelli, sil
pp. I3v, 7v fif.), which interprets Opere,as
I, Florence,
blue i6oi,(bleu),
p. 345, on theand
superiority of the modern
which appeals to the 'daily experience' of palette
variousto the ancient;
crafts-'La
men, including painters. All Pittura of
in quei these
tempi non potevaauthors were
haver l'intera vaghezza
students of medicine, and it is
perworth
mancamento dinoting that
colori'. Schiffermiiller (op.colour
cit. n. 99
was of particular importanceabove,
to pp. 36, 38) also noted the for
physicians, need fordiagno-
a yellow
sis through the study of urine. The Peripatetic De
among the primaries.

122.225.227.202 on Fri, 01 Dec 2023 09:47:35 +00:00
22 JOHN GAGE
Venetian flesh-painting, he exclude
the anonymous French translatio
1725, glossed the four colours as 'S
during the eighteenth and nineteen
primaries became the orthodoxy
interpreted entirely in its terms
ville's Abrgge' de la vie des plus fam
yellow and blue, plus white and bla
Antiquity, and this verdict was
Hagedorn observed that Pliny had
of palette, an endeavour of which
nub of this story during the Neo
noticed that they had more than fo
white, and he concluded, somewh
Et quand ils y mettroient du mystire,
la palette 't la main, peint avec un si g
de ses couleurs.107
Yet Hagedorn was writing at a period when the setting of the palette was still a thoroughly
organized affair, and its range of pigments was far from restricted to four. How did the
artists of the eighteenth century imagine that Apelles has set his? The Renaissance
interpreters of the theme of Calumny had given no sign that they were anxious to follow
Apelles in respect of colour; they made, for example, an abundant use of blue.108 Several
artists of the early eighteenth century who showed Apelles at work were also happy to
represent him essentially as one of themselves. In an Allegory of Painting in the Mostyn-
Owen collection, Sebastiano Conca showed Apelles painting Campaspe as Venus, and
his palette seems to be set for flesh with only reds, yellow and white, but there is blue in the
picture on his canvas, as well as in other parts of Conca's scene.109 A similar subject by
Francesco Trevisani (Pasadena, Norton Simon Museum of Art) also presents Apelles
with a far from restricted palette, including vermilion, as does G. B. Tiepolo in two
versions of Alexander and Campaspe in the studio of Apelles. Apelles's palette in the later
version, in the Louvre (c. 1735-40), is set in the standard eighteenth-century sequence
from white, near the thumbhole, through yellows and reds, to black, a total of six colours,
including vermilion.110
104 R. de Piles, Abrege'de la vie des peintres, 1699, pp. 131, 109 Colour-plate in Apollo, LxxvI, 1962, p. 397.
257-58. In his Cours de peinture par principes, I7o8, p. 352, 110 The Trevisani is reproduced in colour in The Con-
De Piles concluded that the four colours could only benoisseur, cxcmi, 1976, p. 209, and the earlier, Montreal
an underpainting, which would then have been finishedversion of the Tiepolo subject in A. Morassi, G. B.
in the lighter colours he called 'aerial'. Tiepolo, 1955, pl. iiu. For the Louvre picture, A. Morassi,
105 Histoire de la peinture ancienne extraite de l'HistoireG. B. Tiepolo: A Complete Catalogue of the Paintings, 1962,
naturelle de Pline Liv. XXXV, 1725, p. 44. p. 38, fig. 284. The most useful study of the setting of
106 C. L. von Hagedorn, Reflixions sur la peinture, 1775, palettes from the sixteenth to the nineteenth centuries is
1, p. 201. The original German edition dates from 1762.F. Schmid, The Practice of Painting, 1948. The absence of
107 Ibid. pp. 202 f. Anxious to save the probability of blue was a standard feature in 18th-century palettes for
Pliny's story, Hagedorn also stressed that it could onlyflesh-painting, yet Anton Raphael Mengs's pupil,
apply to flesh-painting, since there was clear evidence inDaniel Webb, cast doubt on the authenticity of Pliny's
Pliny (xxxiii, I I) and in the paintings of Herculaneum, story precisely because he felt that the four colours cited
that blue was used by the ancients (pp. 201, 204). were incapable of forming 'a perfect carnation' (D.
108 For a colour-plate of Botticelli's Calumny, in theWebb, An Inquiry into the Beauties of Painting, 1760,
Uffizi, L. Venturi, Botticelli, 2nd edn, 1971, pl. 42. See p. 8o0 n.).
also the colour notes in R. F6rster, op. cit. n. 66 above,
PP. 35 if., 45-46, 48-49.

122.225.227.202 on Fri, 01 Dec 2023 09:47:35 +00:00
Possibly the first visual indication of the effect both of the
laneum, where a number of representations of artists at work had
the study of the four-colour theory, is presented by the otherwise
frontispiece by Friedrich Oeser to Winckelmann's Gedanken of
painter Timanthes, who had been characterized by Cicero (Brut
painter, at work on his most famous subject, The Sacrifice oflphigen
does not use a palette, but seems to be about to take his colours fro
closer examination, however, reveals a shadowy fifth pot behind th
that Oeser, like d'Argenville, intended to allude to the story of th
the most general way. Later painters were more precise: Jacq
unfinished canvas now at Lille, showed Apelles painting Camp
saucers of paint and without a palette;112 and in 1819J. M. Langlo
Medal at the Salon with a Generosity ofAlexander which also shows
pots of colour (although he does have another saucer in fr
palette).113 A French theorist of the 1820s, David's pupil J. N.
insisted that the three colours of the primary triad, plus black or w
four the ancients had used, and this view was re-stated by Ingres'
the 1850s.114 Ziegler was familiar with the Principles ofHarmony and Co
chemist and colour-theorist, M. E. Chevreul, which claimed summ
had used a palette of five primaries, including the red-yellow-blue
may have been produced 'spontaneously' (i.e. optically) by the effe
contrasts Chevreul had expounded.115 By the time of Chevreul, th
four colours in terms of some 'primary' combination had become
11 Cf: Pliny, xxxv, 73. The r1le of Apelles's

this imagecolour asas
hean
was in his line (Cf. Ingres raconte'par
embellishment to the Gedanken is obscure. Winckelmann
lui-meme et par ses amis, 1947,1, Ip. 57); in the Apotheosis of
stated in June 1755 that he had explained it to the King
Homer (Louvre, 1827), Apelles is shown, wearing the
of Saxony in a written commentary which has not standard blue cloak, and holding a palette (seen from
survived (Briefe, ed. Rehm, I, 1952, p. I72).Justi's viewthe back) and brushes (R. Rosenblum, Ingres, 1967,
that the subject represents Greek beauty yielding to pl. 35). In a preparatory watercolour in Lille, Apelles
sentiment (Winckelmann und seine Zeitgenossen, 1898, I, wears a pink cloak and his square palette faces the
p. 354) seems to be contradicted by Winckelmann him- spectator, but shows no colours. During the I82os the
self, who describes it as Nachahmung in a letter to Ber- English theorist Charles Hayter also proposed that the
endis (Briefe, I, p. 177). For Oeser's Tiepolesque proto- ancients had used the three primaries plus black (A New
type, see the Sacrifice oflphigenia (c. 1735-40) in Paris andPractical Treatise on the Three Primitive Colours, 1826, pp.
a replica (c. I740-50) in Hamburg (Morassi, Catalogue,14-15)-
n. I0o above, pp. 13, 42 and fig. 233)- 115s Trans. Martel, 1854, ?342. In the I82os, Benjamin
112 In the painting, Apelles is still at the stage of
Robert Haydon was showing Thomas Phillips that the
sketching his subject in white. Campaspe rests on a blue range of the Greek four-colour palette could be extended
drapery, and the painter himself is dressed in green and by contrasts, including the 'management' of black to
yellow. The colours in the pots are not clear. For look a blue (Autobiography and Memoirs, ed. Taylor, 2nd
reproduction, Paris, Grand Palais, De David a' Delacroix,edn, 1926, 1, p. 395). In his Academy lectures, however,
1974/5, no. 37, pl. 147. The catalogue gives it to c. 1814. Phillips remained doubtful that the whole range of
For a related drawing of 1813, D. and G. Wildenstein, colour could be represented with four colours available
Documents complimentaires au catalogue de l'oeuvre de Louis as pigments, rather than with the prismatic primaries
David, 1973, Pp. 218-19. (Lectures on the History and Principles of Painting (1829),
113 This canvas may be the same as the Alexandre cidant 1833, PP- 352-53). An earlier French commentator, the
Campaspe, shown at the Salon of 1817 as a commissionAbbe Arnaud, noting the discrepancy between the
from the Ministry of the Interior. It is now in theaccounts of Pliny and of Cicero, appealed to the 'simple'
Museum at Toulouse.
palette of Titian, and to his cunning handling of con-
114J. N. Paillot de Montabert, Traite complet de la to suggest that 'c'est des moyens les plus simples
trasts,
peinture, 11, 1829, pp. 245-46, viI, pp. 367-68. J. queC. sortent les plus grand effets' ('Memoire sur la vie et
Ziegler, Traite'de la couleur et de la lumiere, 1852, p. 15.les
It ouvrages
is d'Apelle', Mimoires de litterature... , XLIX,
not clear whether Ingres himself was as interested i8o8,in pp. 200 ff.).

122.225.227.202 on Fri, 01 Dec 2023 09:47:35 +00:00
24 JOHN GAGE
and it does not seem to have been o
Gladstone in the 185os raised the q
pointing again to the absence of bl
APELLES IN THE STUDIO
Painters in the Romantic period were still conscious that Pliny'

colours could be taken as practical advice. Reynolds had alread
current in the eighteenth century, that Apelles had laboured u
restricting his palette; he interpreted the story of the dark varnis
who had noticed the sobriety of Apelles's paintings, due, accordin
overall toning-down, but to
his judicious breaking of those [four] colours to the standard of nature.11
Later, in his notes to Du Fresnoy's De arte graphica, Reynolds has

account of the varnish as a description of glazing and scumblin
commend the four-colour palette:
I am convinced the fewer the colours, the cleaner will be the effect of tho
sufficient to make every combination required. Two colours mixed toge
brightness of them simple, nor will three be as bright as two .. .118
Too many mixtures would themselves be an infringement of t

Apelles's palette, and Reynolds's own practice towards the end
base his pictures on simple underpaintings of white, black, ind
'representatives', as a commentator put it, '(however negative)
colours'.119 In Reynolds, too, there is a hint of the lingering preju
more comprehensive sense. In a note to line 523 of Du Fresnoy
harmony which is produced by what the ancients called the corrup
of the colours, by mixing and breaking them till there is a gener
without any thing that shall bring to your remembrance the
original colours', which he found especially characteristic of
style.120 This method of achieving harmony he felt was inferior
best exemplified in Rubens, where
the brightest possible colours are admitted, with the two extremes of w
reconciled by being dispersed over the picture, till the whole appears like
It is especially ironic that Reynolds should have thus asserte

Antique virtues of the limited palette, for this must also have bee
enemy Blake, who attacked the President precisely because h
Rembrandtesque mixture, and who condemned the work of Titian
daubs'.122 Blake's own debt about 18oo to Rembrandt and the
116 W. E. Gladstone, Studies in Homerp.and the Homeric
328: 'breaking down these fine colours which would
Age, 1858, 1I, p. 488. Gladstone's hintappear
wastoo raw, tofirmly
more a deep-toned brightness'.
121 Ibid. p. pp.
developed by W. Schultz, op. cit. n. 9 above, 337, and p. 339 on the pre-eminence of this
I86-88.
system. Du Fresnoy,
117 J. Northcote, The Life ofSirJoshua Reynolds, 2nd lines
edn, 339-40, had characterized the
1818, I, p. 40o. corruptio colorum as a specifically Venetian practice. For
118 The Literary Works ofSirJoshua Reynolds, ed. Beechey, Reynolds's method as essentially a glazing technique,
1852, I, pp. 328 f. (Note xxxvii to Du Fresnoy, 1783.) H. Buttery in D. Hudson, Sir Joshua Reynolds, 1958,
119 C. L. Eastlake, Materials for a History of Oil Painting, pp. 248 ff.
1869, 11, pp. 255 ff. 122 Descriptive Catalogue: Poetry and Prose of William
120 Reynolds, op. cit. n. I 18 above, 11, p. 337. Cf. also Blake, ed. G. L. Keynes, 1956, p. 612.

122.225.227.202 on Fri, 01 Dec 2023 09:47:35 +00:00
been given its due,123 but it was probably in the context of h
colour study, described in a letter of November 1802, that
widely for appropriate models of colouring, and came across t
shortly after Blake's death, J. T. Smith, who had known the p
recorded:
As to Blake's system of colouring . . . it was in many instances most beautifully prismatic. In this
branch of the art he often acknowledged Apelles to have been his tutor, who was, he said, so much
pleased with his style, that once when he appeared before him, among many of his observations he
delivered the following: 'You certainly possess my system of colouring; and I now wish you to draw
my person, which has hitherto been untruly delineated'.124
It is not easy to date Blake's ostensible encounter with Apelles; he alludes to the
competition with Protogenes in the Descriptive Catalogue of 1809, and also in the Notebook,
which was in use at many periods from the late I780s.125 But the 'prismatic' palette,
which was apparently the product of this encounter, can be recognized in the water-
colours prepared for Thomas Butts in 1803, and in his letter to this patron of the previous
November, Blake showed that he had made an extensive study of Reynolds's writings,
where he will have found the discussion of the palette of the Ancients reviewed above. The
lesson which he claimed to draw from Reynolds' Discourses, and which he reinforced in his
later Marginalia to Reynolds's Literary Works, was that the 'broken' colour of the Venetians
was injurious to grandeur, which could only be the product of simplicity. 126 But it is not at
all clear how far Blake took Apelles's message to be a serious recomendation of the
restricted palette: the very florid tonality of many of the later illuminations, with their
frequent use of gold, suggests that austerity was certainly never a constant aesthetic with
him, and the technique of even the simplest of the late watercolours, the Dante series,
makes a good deal of play with mixed and 'broken' tints. Blake's early biographer,
Alexander Gilchrist, described his pigments as 'few and simple', but went on to list five of
them, including cobalt blue, which the painter would supplement occasionally with
ultramarine, gamboge and vermilion; and the palette of the figure of Painting in the Enoch
lithograph of 1821 shows a range of six.127 It must remain debatable how far Blake's
123 D. Bindman, Blake as an Artist, 1977, PP. 125 ff.fig. 166 and pp. 161-63, suggests a date-bracket of
124 G. E. Bentley, Jr, Blake Records, 1969, p. 468. c. I803-20, pointing out that the print is not strictly a
lithograph, but a relief-etching on stone, a 'hybrid'
Blake's portrait of Apelles may be identifiable with the
well known drawing of The Man who Taught Blake Painting technique first published in 182o. Although the design
does bear some relationship to an early watercolour of
in his Dreams (c. 1818; collection of Sir Geoffrey Keynes),
for which see M. Butlin, The Paintings and Drawings Enoch
of (Butlin, op. cit., n. 124 above, no. 146), the
William Blake, 1981, no. 753. subject seems to be that of chapter 92 of The Book of
Enoch, first translated in 1821, where the prophet
125 Poetry and Prose, cit. n. 22 above, p. 617, and p. 590,
where Apelles and Protogenes are characterized as 'fre-
addressed his children on the subject of his vision 'from
sco' painters. The Notebook of William Blake, ed. D. Erd-
a book' (The Book ofEnoch the Prophet, trans. R. Laurence,
man, 1973, p. 32: 'Ghiottos circle or Apelles line were1821, ch. xcii, vols 1-3 (p. 134); cf. also ch. civ, vols I, io
not the work of Sketchers drunk with wine'. (pp. 154-56)). Painting, poetry and music, 'the three
126 For a discussion of the letter to Butts, and of thePowers of Man in conversing with Paradise', were
colour of these watercolours, Bindman, op. cit. n. 123 personified by Enoch's grandson Noah, and his great-
above, pp. 136 ff. For the Marginalia, Poetry and Prose, grandsons Shem and Japhet in Blake's Vision of the last
cit. n. 22 above, esp. pp. 791 ff. Judgement of 18o0 (Poetry & Prose, cit., n. 23 above,
p. 643). For slight pencil sketches connected with the
127 A. Gilchrist, Life of W. Blake, ed. Todd, 1942, p. 6o.
The Enoch is reproduced in D. Bindman, The Complete Enoch print, Butlin, op. cit., no. 582 (given to c. 1807),
Graphic Works of W. Blake, 1978, no. 413, and dated and for the series of drawings illustrating The Book of
c. 1806-I2. R. N. Essick, William Blake Printmaker, 198o,Enoch (dated c. 1824-27), ibid. no. 827.

122.225.227.202 on Fri, 01 Dec 2023 09:47:35 +00:00
26 JOHN GAGE
interpretation of Apelles's colour-
Gerard, a reductive one.128
Thus, from its origins in Pliny un
was primarily a function of that
colouristic dimension, whatever id
current in each period. Only in the d
classical philology and studio pra
significantly to the elaboration of a
CAMBRIDGE UNIVERSITY
128 C. LUnormant, Girard, peintre d'histoire, 2nd edn, Romanticism circa I8oo', Burlington Magazine, cxviiin,
1847, p. 55, cit. J. H. Rubin, 'New Documents on the 1975, PP. 787-89.
Meditateurs: Baron Gerard, Mantegna and French

122.225.227.202 on Fri, 01 Dec 2023 09:47:35 +00:00
a-Nicoletto da Modena, Apelles, engraving, c. 1507/15 (P. Io) b-Friedrich Oeser, Timanthes painting
engraving, I755 (p. 23)

122.225.227.202 on Fri, 01 Dec 2023 09:47:35 +00:00

002gage LocusClassicusColour 1981

Uploaded by

Copyright:

Available Formats

002gage LocusClassicusColour 1981

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

002gage LocusClassicusColour 1981

Uploaded by

Copyright:

Available Formats

A Locus Classicus of Colour Theory: The Fortunes of Apelles

Author(s): John Gage

Stable URL: https://www.jstor.org/stable/751049

This content downloaded from

Journal of the Warburg and Courtauld Institutes, Volume 44, 1981

This content downloaded from

THE ORIGINS OF THE FOUR-COLOUR THEORY

Pliny's circumstantial account, although it is not the only one, has

s Many are discussed by H. van 'Controversial

This content downloaded from

9 Both of these solutions had been 12 G.suggested by

This content downloaded from

This content downloaded from

This content downloaded from

THE PROBLEM OF MIXTURE

One of the most persistent arguments among modern apologists

The context where we would most expect to find mixtures is in the

25 Cited by E. Bertrand, Etudes sur la ing, and that

This content downloaded from

30 H. A. Wolfson, The Philosophy of

This content downloaded from

This content downloaded from

From this account, which seems to be one of the earliest

42 For flavus as xanthos and fulvus as45pyrros, Ficino's

This content downloaded from

This content downloaded from

But another series, of four geometrical solids,

52 F. M. Cornford, Plato's Cosmology, 1937,

This content downloaded from

Spring = pyramid (i.e., fire: red)

This content downloaded from

DURER, TITIAN AND THE CRITICS

This content downloaded from

This content downloaded from

This content downloaded from

This content downloaded from

THE IDEA OF THE PRIMARIES

Closely related to the question of colour mixture is the ques

84 Ridolfi, op. cit. n. 71 above, p. 107.

This content downloaded from

This content downloaded from

92 The earliest unambiguous use of ceruleus or caeruleus

This content downloaded from

Ferri was inclined to return to the 'outline' theory (A.

This content downloaded from

physicians.100 Robert Boyle, writing in mid-cen

But the growing conviction that the primary col

100 V. A. Scarmilionii, De coloribus, Marburg

This content downloaded from

This content downloaded from

11 Cf: Pliny, xxxv, 73. The r1le of Apelles's

This content downloaded from

APELLES IN THE STUDIO

Painters in the Romantic period were still conscious that Pliny'

Later, in his notes to Du Fresnoy's De arte graphica, Reynolds has

Too many mixtures would themselves be an infringement of t

It is especially ironic that Reynolds should have thus asserte

This content downloaded from

This content downloaded from