§45.3. Color and Partial-Tone Coordinates
If at this point, at the only point in this book, I stick my hand into the wasp's nest that is audition colorée,
it is certainly not with the intent of solving this problem
harmonically. A full treatment of this problem in itself, with its
factual and non-factual scientific, aesthetic, and symbolic
considerations, would fill not just a book, but many books, and
whatever harmonics specifically would have to offer would require the
establishment of a whole series of approaches before a survey could be
undertaken. In my Hörende Mensch,
Ch. 5, section 2, I discussed the phenomenon of “tone-spectra” at first
from a few of these harmonic approaches, but encountered a fundamental
difficulty of a purely technical nature: being unable to provide any color
illustrations or diagrams. There is not much sense in writing or
talking about colors if one cannot demonstrate at least the most
important things that one wishes to impart ad oculos
in color. But of course any color printing significantly increases the
price of producing a book, and thus we must again rely on the reader's
collaboration. Figure 442a is constructed so that the black and white
values and their gray intermediate steps, i.e. all c-values,
are correctly printed, whereas the color tones are indicated only with
words. Refer to the instructions accompanying the table, according to
which the reader can fill in the individual colors himself, with
colored pencils.

Figure 442a
Amendment to the text (§45.3): For this color table, the reader will
find the 6 primary and secondary colors pasted in, besides the printed
gray-values-a verbal designation of the colors of the fields would
therefore be superfluous. Admittedly, the reader must complete the
table by pasting in, or drawing with colored pencils, (1) the identical
color-tones, and (2) their upper (brighter) and lower (darker) octaves.
The model for how to proceed here is given by the c-values already printed. For example, the identical 1/1 c-values of the generator-tone line are all the same medium gray, the identical 2/1 4/2 8/3 8/4 ... c-values
are a lighter gray than the next octave up, and so forth. The 6 basic
colors should be developed upwards and downwards in the same manner. If
the tone 2/3 f, is a medium bright green, then the identical tones 4/6 f, 6/9 f, 8/12 f, and 10/15 f, will be the same shade of green, the next octave up 4/3 f will be a somewhat lighter green, and the next octave down 1/3 f,,
a somewhat darker green-and so on. It will be very enjoyable for the
reader, then, to locate all the colors by means of the rays of the
“equal-tone lines” on the monochord string starting from 0/0. This analysis represents only one
attempt to coordinate color and tone. I have not as yet been able to
persuade myself of an absolute connection between “this” tone and
“that” color. Just as every person is tuned to a certain tone, every
color can, in my opinion, be identified with a “generator-tone,”
according to the various individualities of the being-values. If 1/1 is defined, the arrangement of the tone-colors among themselves is something different. If, for example, I identify 1/1 c
with red, then all the following tone-color or color-tone arrangements
will proceed within the “P” system; admittedly, a few variables must be
fixed: which is “blue,” which complementary, etc.
Here we will discuss only a single problem for a future harmonic study
of color, namely the color analysis of the partial-tone coordinates.
The reason this is undertaken only here, and not in earlier sections
(such as in §20 or §21) regarding the “P” itself, is connected with the
type of analysis. Specifically, we will use only one element of the “P” for our color analysis, namely the “interval constants” of this chapter-title.
For a correct understanding of our analysis, a brief account of the Goethe-Schopenhauer-Newton situation is necessary.
Schopenhauer (Über das Sehen und die Farbe,
Introduction) writes: “Goethe provides, in his admirable work, all that
the title promises: information about the study of color. The
information is correct, complete, and significant; rich material for a
future color theory. He has not undertaken to provide this theory
itself, since, as he admits in §39 of the Introduction, he has advanced
no actual explanation for the existence of color, but has postulated it
as an appearance and teaches only how it appears, not how it is.”
Since Newton based everything upon an exact physical foundation of
color, whereas Goethe based everything upon a physiological and
psychological characterization that was as all-encompassing as
possible, it is not initially understandable why Goethe argued so
acrimoniously against Newton. Both viewpoints, right up to modern
times, are well founded alongside each other, and one might think that
neither stood in the other's way. It was not possible 100 years ago to
bring Goethe's extremely precise and conscientious analyses and
observations physically to a norm, and Goethe did not succeed in
unifying his views with Newton's equally meticulous experiments. Thus
Goethe became an embittered opponent of the Newtonian “heresy.” Our
high esteem for both geniuses today forces us to ask the question:
shouldn't a reconciliation between both views be possible?
The main quarrel, as is known, centers around the fact that Goethe and
his adherents say that color is an appearance “on” light, i.e. it is
actually only generated in our eyes, in the retina; whereas the
Newtonians, and modern physicists with them, claim that color is a
function of the wave-lengths of individual light rays, and thus is
already present a priori
in the nature of light. Goethe did not completely repudiate the
spectrum, although in accordance with the current state of research he
ascribed no great value to it. I am sure that if Goethe could have seen
with his own eyes the marvel of polarization phenomena as perfected
today, and learned of the improvements in spectral analysis, he would
certainly have changed his opinions and tried to bring his ideas into
agreement. So Goethe was undoubtedly wrong there. Holding this single
point against all of Goethe's studies of color would reduce him to
nothing, especially since he did not touch upon the essentials of this
study at all. But in one place, Goethe's intuition hit the right spot: The world of colors emerges between light and darkness, and the nature of color itself is polar.
One is easily tempted to say that this intuition transgresses the
bounds of science. But if one perceives both these theses in all their
power and depth, one will sense how far they penetrate the innermost
areas of nature and art. Modern science, however, has no need of that.
The spectrum itself appears in colors out of the darkness, and returns
to darkness; and the fact of the “subjective” complementary colors is
at least so incontestable psychologically and so well established
physiologically (as Herr J. Itten of Zürich told me, complementary
shadows in color can be photographed objectively!) that there can be no
talk of a repudiation of this polarity on the part of science. Besides,
Goethe relinquishes the main problem that has occupied all of
scientific optics since Newton-namely, a mathematical foundation of the
phenomena of color and light-although he does not in any way consider
it impossible (see Farbenlehre, “Didaktischer Teil,” §722 ff.).
These two main ideas of Goethe's are completely harmonic. Envision the
tone-development as it groups itself in the partial-tone coordinates
around a center, the generator-tone, upwards (to the light) on one side
and downwards (to the darkness) on the other side; consider, further,
the dual construction of the tone-system, in which each tone has its
complement in its reciprocal (3/5 es ↔ 5/3 a
etc.) and all the ratios together emerge from two opposing impulses.
This is the exact counterpart, in tonal studies, of Goethe's basic
view. In my “Tonspektren” (Abhandlungen), this polarity is brought to the same basis as that of the optic spectrum.
Admittedly, the exact link between the two opinions is missing, because
the arrangement of spectral colors differs from Goethe's and
Schopenhauer's ordering of their color-evaluation. But this deficit is
not the fault of Goethe's followers, but of modern science, which
despite many energetic attempts has not yet been able to provide a
natural mathematical rationale for color studies. Here, also, harmonics
can be a competent mediator, simply because it has a number-form at its
disposal in the tone-number, which is firmly rooted in a sense
coordinative to the eye (the ear), and is simultaneously anchored in
nature (partial tones). In Hörende Mensch,
and in the abovementioned “Tonspektren,” I attempted to show how we can
use this tone-number to undertake a harmonic analysis of light as the
generator of color. Since every spectral line marks a specific color,
this may lead us to a standardization of color that is not abstract and
artificial, but in accordance with color, and thus visual and natural.
I will now disclose an attempt made, arising from the desire to find a
direct harmonization between Goethe and Newton, which might at the same
time be important for a tonal analysis of color. Goethe and
Schopenhauer consider the complementary colors red and green to be the
midpoint between white and black, followed by the complementary orange
and blue, then yellow and purple. Schopenhauer (Über das Sehen und die Farbe,
Ch. 2, 5) holds these six colors, red, green, orange, blue, yellow, and
violet, “although in nature these very seldom appear in pure form,” to
be “certainly a priori,” i.e. they must be perceived a priori
“like the regular figures, which in reality cannot be perfectly drawn
and yet which we perceive perfectly and understand with all their
attributes.” Schopenhauer now (ibid.) gives the foundation for his
view, and the following scheme:
0 |
1/4 |
1/3 |
1/2 |
1/2 |
2/3 |
3/4 |
1 |
black |
violet |
blue |
green |
red |
orange |
yellow |
white |
└──────┘
└──────────────────┘
└───────────────────────────────┘
└────────────────────────────────────────────┘ |
The ratio numbers have no meaning to us, since they are used by
Schopenhauer only as psychological “quantities.” The brackets, on the
other hand, are of great importance, because they embrace the pairs of
colors (complements) and likewise the series-progression of colors,
which strictly follows the psychological evaluation. The reader will
already have noticed that this series-progression violet blue green red orange yellow is not in concordance with the series of the spectrum:
violet blue green | yellow orange red (physical)
→ ←
However, a connection is evident, in that the spectrum exhibits the
reversed series-progression of the Schopenhauer series in its second
half. Both series are well-founded, the first psychologically, the
second physically.
I will now place Schopenhauer's pairs of colors in the partial-tone
coordinate scheme according to their evaluation: (1) green-red, (2)
blue-orange, and (3) purple-yellow, assuming that the overtone basal
series 1/1, 2/1, 3/1 ... approaches the light (white), while the undertone basal series 1/1, 1/2, 1/3 ... approaches the dark (black). The diagonal 1/1, 2/2, 3/3
..., i.e. the generator-tone, is a medium gray, in which all
complementary colors dissolve before our eyes as they are mixed. The
identical tone-values (equal-tone lines) always contain the same color,
the octaves upwards are always the next step towards white, and the
octaves downwards are always the next step towards black. All
generator-octaves thus have gray-values, upwards approaching white,
downwards approaching black. G and F appear
in the tone-development as the first tone-pair differing from the
generator-tone. In accordance with Schopenhauer's view, they should
have the colors red and green (or vice versa; but I chose the colors
according to their brightness-value, placing green, blue, and purple in
the < 1 domain, and red, orange, and yellow in the > 1 domain),
and in their upper and lower octaves, the corresponding brighter and
darker values of red and green. E-As and A-Es appear as further reciprocal tone pairs. From the above viewpoints, and because E and As appear first and closest to the generator-tone, these tones receive the colors blue-orange; and the tones A-Es receive
the last color pair purple-yellow. The brightness value of the color is
always defined so that the “purest” colors, i.e. those with the highest
saturation, lie next to the central diagonal 1/1, 2/2, 3/3.
Naturally, any number of other colors-even an endless number-can be
inserted between the above six colors, and will appear and require
places in further tone-development. For the sake of clarity, only the
color-values of these 6 colors are inscribed, along with the gray
values with their upper and lower octaves. I recommend that the reader
draw Table 442a himself in color according to the specifications given.
It would be better to leave the table in this book as it is, to draw a
grid of lines on a large sheet of cardboard, and to fill out the
corresponding fields with gray and color values.
If we now observe the parallel series lying next to the diagonals 2/1 3/2 4/3 5/4 6/5 ... and 1/2 2/3 3/4 4/5 5/6 ..., we see that the colors green, blue, and violet appear in 4/3 5/4 6/5, and red, orange, and yellow appear in 3/4 4/5 5/6, with a pronounced color-harmonic direction of the two series toward the middle line n/n - 1 = gray. It is interesting to note the switching of red and green in the two previous steps 2/2 and 3/3,
which cannot at present be interpreted purely physically. The substance
of the experiment, however, is shown by the fact that the overtone and
undertone parallels of the middle diagonal from the ratios 3/4 and 4/3
upwards show a definite spectral structure, whereby the division of the
spectrum between red and green remains, for now, yet another physical
puzzle. But perhaps the spectrum is not a homogenous band at all, but
is superimposed from two impulses, beginning with yellow and green
(emerging from a gray stage) or touching each other? In any case, we
see that a harmonic analysis of the Goethe-Schopenhauer color
arrangement leads to a spectral arrangements, albeit of a rudimentary
type. Otherwise, our table shows that every color-complement
corresponds to an exact tone-complement, and thus, regardless of its
physical significance, the table is useful as a color-tone designation
table, which even in this small version clearly illustrates a great
number of color-chords, color-intervals, and color-tone steps. It is
mentioned here only as a side note that a color-body can be developed on this basis analogously to the partial-tone cube,
and Runge's color-sphere and similar spatial color constructions are
well worth contemplating for their inner logic. Perhaps this
“tone-prism” will shed light upon the entire polarization problem, and
lead to a connection with spectral analysis.
§45.4. Bibliography
For 3: As the most pertinent published work, predominantly from the
standpoint of the artist and pedagogue, I recommend Johannes Itten's
great work on color, published privately at the beginning of the 1920s,
which will hopefully soon be made generally accessible.
Furthermore: Carry von Bienna, Farben und Formen als lebendige Kräfte
(Jena, Diederichs), a student of Adolf Hötzel who offers Hötzel's
lessons as a whole and expounds upon them; Wilhelm Steinfels: Farbe und Dasein (Jena, Diederichs, 1926); the writings of the Swiss painter Aeppli: Die Symbolik von Licht und Dunkel. Die Farben und ihre Offenbarung, self-published, Uerikon am Zürichsee 1936, and Lebensordnungen, Farbe, Ton, Form,
Emil Oesch Verlag, Thalwil-Zürich 1944. Aeppli's works discuss color
problems from the religious-symbolic viewpoint. In the latter work
Aeppli analyzes the partial-tone coordinates in his own way, using red
and yellow as the generator-tone 1/1.
Victor Goldschmidt, on the basis of his crystallographic-harmonic
complication law, has published a tabular-work on color that I have not
yet been able to obtain. A rich bibliography on audition colorée can be found in Georg Anschütz, Farbe-Tonforschung I (Leipzig 1927), from whose school the dissertation of Johannes Hantzsch: Farbe-Formbeziehung bei Kindern und Jugendlichen (Hamburg 1935) also originates. Besides the well-known books by Ostwald (Farbenlehre und Farbenfibel), a good popular-science discussion of the color problem is in William Bragg: Die Welt des Lichtes (Braunschweig, Vieweg, 1935).
Hans Kayser, Hörende Mensch, 17 and Ch. V, section 2; Klang, 33; Abhandlungen, “Tonspektren”; Grundriß, 40, 169, 200
The so-called “color-keyboard” was explained as an absurdity by Herder
in his prize essay “Vom Ursprung der Sprache” (1770); see further his
“Fragment über Licht und Farben und Schall.”
I cannot further discuss Aeppli's beautiful and religiously founded main work, Lebensordnungen,
since I obtained a copy of the book from the friendly author only after
the completion of this manuscript. But the reader interested in
color-tone problems is referred all the more urgently to this book.