Consonance and Dissonance

Author(s): Mieczyslaw Kolinski

Source: Ethnomusicology, Vol. 6, No. 2 (May, 1962), pp. 66-74
Published by: University of Illinois Press on behalf of Society for Ethnomusicology
Stable URL: http://www.jstor.org/stable/924667 .
Accessed: 01/10/2013 15:20

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp

.
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.

University of Illinois Press and Society for Ethnomusicology are collaborating with JSTOR to digitize, preserve
and extend access to Ethnomusicology.

http://www.jstor.org

This content downloaded from 129.64.99.141 on Tue, 1 Oct 2013 15:20:32 PM

All use subject to JSTOR Terms and Conditions
 CONSONANCE AND DISSONANCE
Mieczyslaw Kolinski

the characteristic features of the music of Laos, Humbert-

Discussing
Lavergne (1934:18) emphasizes the preference for dissonant inter-
vals and chords. By this the author obviously means tone combinations that
in conventional Western harmony are considered dissonant. Nothing would
be gained by merely criticizing such an inadequate application of Western
concepts on Oriental music. What really matters is the question whether it
is possible to approach the phenomena of consonance and dissonance in an
objective way giving these terms a universally valid meaning. The present
paper attempts to give a positive answer to this crucial question.
The claim that a Western student of non-Western music must complete-
ly lay aside his own musical concepts is only partly justified. As far as
consonance and dissonance are concerned, it is indispensable to investigate
the nature of these phenomena first within our own culture so as to estab-
lish a dependable basis for analogous research in the field of allocultural
music. The problems involved encompass two different but complementary
facets; the first one calls for a formulation of a satisfactory phenomenological
definition while the second one refers to the question of psycho-physiological
and psycho-physical interdependence. The two most widely accepted descrip-
tive definitions of consonance and dissonance could best be termed "aesthet-
ic" and "functional." The former conceives of consonance as of a "pleasant"
or "satisfying," and of dissonance as of a "disagreeable" or "shocking" si-
multaneous tone combination. Evidently, such an "aesthetic" appreciation
remains highly relative and subjective and largely varies according to style
and individual taste, but above all the aesthetic approach does not touch the
core of the problem; for example, theorists and musicians agree that a fifth
is more consonant than a third though many people might prefer the sound
of a third to that of a fifth. A similar ambiguity characterizes the "func-
tional" approach. It contends that an interval or chord is dissonant or con-
sonant according to whether or not it requires a resolution. The rules for
functional appreciation of consonance and dissonance have been established
within the framework of traditional harmony, but even there the same inter-
val or chord, depending on its harmonic function, may or may not require a
resolution and, therefore, be dissonant in one context and consonant in an-
other one. Even the perfect consonance of a fifth might become a pronounced
"functional" dissonance, as in Figure 1.

AK I

Fig. 1

Obviously, it would be senseless to evaluate consonance and dissonance in

allocultural music by means of Western standards of aesthetic or functional
appreciation which, even in regard to our own music, do not represent ob-
jective and comprehensive approaches.
It is impossible to arrive at a satisfactory phenomenological definition
of consonance and dissonance unless the fundamental tone qualities have been
clearly established and formulated. Strange as it may seem, the most
66

This content downloaded from 129.64.99.141 on Tue, 1 Oct 2013 15:20:32 PM

All use subject to JSTOR Terms and Conditions
essential musical tone quality had not been recognized as such until rather
recently. It is a well known fact that octave tones are identical in a certain
respect. For example, all C's have a quality in common that distinguishes
them from other tones. But none of the current qualifications of tone, such
as pitch, timbre or intensity, applies to this phenomenon. The identity of
octave tones is certainly not an identity of pitch; on the contrary, there is
always a more or less considerable opposition in pitch between tones that
are one or more octaves apart; on the other hand, octave identity persists
no matter whether timbre and intensity are similar or contrasting. Since no
appropriate name is available to denominate the quality that is identical in
octave tones but distinct in other tone relations, I suggested to term this
quality tint. For example, all G's have an identical tint; the same is the
case with all A's, all B's, etc. However, G, A and B represent different
tints. As a matter of fact, the Western tone system consists of a series of
twelve different tints.
Referring to concepts of Greek antiquity, Stumpf (1883 and 1890) linked
consonance to the phenomenon of fusion. He contended that a simultaneous
interval is all the more consonant the more it resembles a single tone.
This approach brings us somewhat closer to the core of the problem. Rather
independently of the aesthetic and functional concept of consonance, theorists
and musicians agree to give the octave the most prominent place in the gra-
dation of consonant intervals. This evaluation seems to be in perfect agree-
ment with Stumpf's approach since it evidently refers to the extremely high
degree of fusion inherent to the octave, as well as to its resemblance to a
single tone. Yet fusion as such does not constitute an appropriate criterion
for consonance because there are several factors that influence fusion with-
out altering the degree of consonance. For example, the fusion of a major
third is greater than that of a major tenth though Stumpf himself emphasized
that octave enlargements do not affect the degree of consonance; the fusion
of a chord is greater when it is played on one instrument than when its
tones are divided among instruments of contrasting timbre or when they
come from different directions though neither timbre nor localization charac-
terize consonance and dissonance; finally, Stumpf has proved that the phe-
nomenon of beats (that is, periodical oscillations of intensity) has nothing to
do with consonance and dissonance; however, the absence or presence of
beats might considerably alter the degree of fusion.
The factor that contributes in the first place to the high degree of fu-
sion of the octave and to its resemblance to a single tone is evidently its
tint identity; this phenomenon is no doubt viewed when the highest degree of
consonance is conferred to the octave. In other words, while fusion depends
on the interrelationship between various elements such as tint, pitch, timbre,
beats and direction, fusion is characteristic of consonance only in so far as
it is caused by tint identity or tint affinity. Hence, consonance is not fusion
or homogeneousness in general but homogeneity of tints; accordingly, disso-
nance is heterogeneity of tints. This concept of consonance and dissonance
might be termed basic, in contradistinction to the aesthetic and functional
ones.
Owing to the limitations of the aesthetic and functional approaches only
the basic concept is apt to serve as an objective principle for the grading
of consonance. It has been pointed out that the particularly high degree of
consonance of the octave is caused by the identity of tint. Among the inter-
vals that are composed of two different tints the fifth and its inversion, the
fourth, stand out as the most consonant ones. Evidently the high degree of
tint homogeneousness of the fifth and fourth is conditioned by the high degree
67

This content downloaded from 129.64.99.141 on Tue, 1 Oct 2013 15:20:32 PM

All use subject to JSTOR Terms and Conditions
of affinity between the two tints contained in these intervals. This means
that two different tints have the greatest affinity when they follow each other
in the cycle of fifths. For example, F is most closely related to C, C to G,
G to D and so forth. Consequently, there is a second degree affinity between
F and G as well as between C and D, a third degree affinity between C and
A, etc. In other words, the degree of affinity between two tints depends on
their distance in the cycle of fifths; therefore, the degree of consonance of
an interval depends on the distance in the cycle of fifths between the two
tints of which the interval is composed. Thus one arrives at the following
classification of intervals with regard to their degree of consonance:

I. Octave and multiple octaves (tint identity)

II. Fourth, fifth and their octave enlargements (1st degree tint affinity)
III. Major second, minor seventh and their octave enlargements (2nd de-
gree tint affinity)
IV. Minor third, major sixth and their octave enlargements (3rd degree
tint affinity)
V. Major third and minor sixth and their octave enlargements (4th de-
gree tint affinity)

There exists no direct tint affinity between tones separated by more than
four steps in the cycle of fifths; therefore, the minor second, the major
seventh, the tritone and their octave enlargements constitute basic disso-
nances. As a matter of fact, these intervals have never been considered as
consonant.
The above-mentioned classification of consonant intervals differs from
the generally accepted one in two respects: the major second and its inver-
sion, the minor seventh, are not only included within the group of consonant
intervals but range even before the thirds and sixths; in addition, the minor
third and major sixth precede the major third and minor sixth. However,
an analysis of the character of intervals shows that they follow exactly the
quintal principle of consonance grading and confirms, consequently, the valid-
ity of the basic concept. For example, when listening to the following set
of intervals

;,

1;]' I CLI l' !I !

.

Fig. 2

one first observes a gradual modification in their character going from the
hollowness of CG to the brightness of CE: CD is less hollow than CG but
not so colorful as CA; in turn, CA is more subdued than CE. This conti-
nuity is interrupted when proceeding from CE to CB and CF#, the latter two
intervals sounding relatively harsh because of the heterogeneousness of their
tints. A comparison of the inversions of these intervals, for example BE,
BA, BD, BG, BC and BF, yields a similar picture. The intervals are given
in octave enlargements in order to eliminate the effects of beats.
In accordance with these observations some medieval theorists con-
ferred to the major second and minor seventh a higher degree of consonance
than to the thirds and sixths, and to the minor third and major sixth a higher
degree of consonance than to the major third and minor sixth. For example,
68

This content downloaded from 129.64.99.141 on Tue, 1 Oct 2013 15:20:32 PM

All use subject to JSTOR Terms and Conditions
Guido of Arezzo considered the major second as more consonant than the
minor third, De Garlandia the minor seventh as more consonant than the
minor sixth, Tunstede the major sixth as more consonant than the minor one
and an anonymous theorist the minor third as more consonant than the major
one (Riemann 1920:77, 78, 116, 119, 121).
The quintal concept is further corroborated by an analysis of the char-
acter of chords; there the laws of basic consonance are even more manifest
than with regard to intervals. According to the quintal principle the most
consonant triads should be those formed by three tints adjacent in the cycle
of fifths, such as DEA, DGA or DGC. Compared with other triads the hol-
low character of these chords most closely resembles that of the highly con-
sonant fifth and fourth. The following phrase, which exclusively consists of
such chords clearly shows this specific quality:

T ?

r" r II r rr • r, ' t
'J •
, ii
L•
, i --2J
_

Fig. 3

Since in traditional harmony these chords have been treated as func-

tional dissonances, one might wonder whether past generations have failed to
recognize their high degree of basic consonance. This, however, is by no
means the case. One of the most prominent figures of 18th century music,
Carl Philipp Emanuel Bach, stated in his famous treatise (1753, Teil II, Kap.
10, par. 5): "Der Sekundquintenakkord fiir sich klingt stets leer, er mag drei-
oder vierstimmig sein. Die Aufl6sung erst macht ihn voll." (The five-two
chord in itself always sounds empty, whether it is in three or in four parts.
Only the resolution makes it full.) In other words, this chord has been
treated as functional dissonance because of its emptiness due to the high de-
gree of basic consonance. For the same reason traditional harmony avoided
the use of the fifth as closing sound.
A characteristic feature of 20th century harmony is the coordination of
functional and basic consonance. For example, the aforementioned five-two
chord, that is CDG, and its inversion GCD have been frequently used as
functional consonances, and in particular as closing chords, by Debussy,
Hindemith, Honegger, Stravinsky, Prokofieff, Bart6k and other composers.
In my study (Kolinski 1936), I gave 91 examples illustrating this trend to-
wards coordination of functional and basic consonance.
We know that any sensation of sound is caused by vibrations, and that
velocity, amplitude and shape of sound waves determine pitch, intensity and
timbre. It is safe to assume that similar sound waves produce basically
similar sensations regardless of the cultural or racial pertinence of the per-
ceiving person because nervous system and brain structure are basically
similar for all populations. It would be, indeed, absurd to suppose that, for
example, the Javanese does not perceive the contrast between the timbre
of a gong and that of a flute in the same way as we do, though, of course,
the musical function of these instruments might be quite different in the two
cultures. I noticed that people who are not familiar with the terms "high"
and "low pitch" sometimes apply them inversely; but this divergence has
obviously nothing to do with perception and simply results from using the
word "high" in the sense of "tall" or "conspicuous" and the word "low" in
69

This content downloaded from 129.64.99.141 on Tue, 1 Oct 2013 15:20:32 PM

All use subject to JSTOR Terms and Conditions
the sense of "small" or "thin." Possibly for the same reason the highest
note of the ancient Greek tone system was called "nete," low, and the lowest
one "hypate," high, unless the commonly accepted assumption is true that
this inversion stems from the inclined position of the kithara. Just as self-
evident as in the case of timbre and pitch is the universal validity of the
interdependence between amplitude of sound waves and the sensations of loud
and soft, and there is no doubt that, for example, the gradual change from
ff to pp characteristic of many American Indian song styles represents an
extreme diminuendo to the Western ear as well as to the Indian one.
It has been pointed out that the concept of basic consonance and disso-
nance is derived from the phenomena of tint identity and tint affinity. If the
latter prove to obey specific laws of psycho-physical parallelism, comparable
to those governing the sensations of pitch, timbre and intensity, we will have
to assume universal validity of the concept of basic consonance and disso-
nance.
East and west, past and present, a strict correlation between simple
ratios and fundamental musical intervals has been recognized. Though these
proportions have been discovered by dividing strings, they represent, as we
know, vibration ratios. Obviously, slight changes of physical stimuli will
hardly, if at all, alter corresponding sensations even if such deviations would
mean substituting very complicated ratios for simple ones. For example, an
ellipse with empirically equal axes will be perceived as a perfect circle in
spite of the fact that the relation between major and minor axis could be
expressed only through a quite complicated fraction; nevertheless, it is the
approximation to the simple ratio 1:1 which causes the sensation of a circle.
Similarly, slight deviations from theoretically exact sizes of consonant inter-
vals, such as octave or fifth, would result in a change from simple vibra-
tion ratios, such as 1:2 or 2:3, to irrational fractions without affecting the
specific character of these intervals. In such cases it is evidently the ap-
proximation to simple vibration ratios which produces the sensation of con-
sonance. Experiments with European musicians (Stumpf and Meyer 1898)
showed that consonant intervals have to be slightly sharpened in order to
appear perfect. Referring to these experiments, as well as to freedom of
intonation on the one hand and to equal temperament on the other, Kunst
speaks of "the fairy tale of the simple ratios." Nevertheless, the universally
accepted general concept of psychophysical interdependence between simple
vibration ratios and consonant intervals remains unchallenged in spite of a
certain latitude in the actual size of the intervals involved.
We saw that basic consonance results from tint identity and tint affin-
ity and that these phenomena evolve from the specific properties of the fun-
damental intervals octave, fifth and fourth due to the simplicity of their vi-
bration ratios 1:2:3:4. This means that, similarly as in the case of pitch,
intensity and timbre, there exists a close psychophysical parallelism between
tint relations and vibration ratios. Consequently, we have to assume univer-
sal validity of the concept of basic consonance and dissonance.
The question which ratios determine the qualities characteristic of the
main musical intervals has been answered differently in the course of mus-
ical history. The medieval principle was based on the assumption that the
so-called "Pythagorean" ratios constitute the physical equivalent of any mu-
sical tone relationship. This means that the fundamental intervals were the
octave and fifth from which all other intervals were derived. The gradual
decline of modal music in favor of the dualistic major-minor concept promp-
ted the theorists of the Renaissance to reexamine the medieval ideas on the
foundation of the tone system and to look for a theoretical justification of the
70

This content downloaded from 129.64.99.141 on Tue, 1 Oct 2013 15:20:32 PM

All use subject to JSTOR Terms and Conditions
new harmonic approach. This trend culminated in the work of Zarlino, the
great Italian theorist of the late 16th century. His achievements not only
revolutionized the millennial Pythagorean doctrine but had a decisive impact
on the development of musical theory up to the present day. Zarlino dis-
tinguished two series of ratios of string lengths, the "divisione armonica"
1 : 1/2 : 1/3 : 1/4 : 1/5 : 1/6 and the "divisione aritmetica" 1:2:3:4:5:6,
the former yielding an ascending major chord, the latter a descending minor
chord. Thus, the dualistic structure of post-medieval music with its char-
acteristic major and minor chords seemed to be supported by a system of
simple ratios. It is still taken for granted that the ratios contained in the
progression 1:2:3:4:5:6, etc., represent the physical equivalent of the phe-
nomena of consonance and tone affinity; however, both phenomenological and
theoretical evidence reveal that not the generally accepted so-called "natural"
approach of Zarlino, Rameau and their followers but the seemingly obsolete
and erroneous Pythagorean or "quintal" system reflects the true nature of
the psychophysical parallelism between musical intervals and simple ratios.
One of the main arguments against the Pythagorean concept has been the
seemingly high complexity of the Pythagorean ratios for the thirds (major
third 64:81, minor third 27:32) compared with the seemingly quite simple
corresponding "natural" ratios 4:5 and 5:6. The degree of simplicity or
complexity of a ratio is generally evaluated according to the size of the two
figures involved; however, observation on the nature of interdependence be-
tween ratios and intervals shows that the inner structure of the figures plays
a decisive part in determining the degree of simplicity or complexity. Let
us compare, for example, the ratios 1:17 and 1:64. According to the size of
the figures employed the former interval should be far more consonant than
the latter; but quite the contrary is true: Ratio 1:17 is a slightly narrow
minor ninth enlarged by three octaves and, therefore, a pronounced disso-
nance, while ratio 1:64 is a sixfold octave and, therefore, a perfect conso-
nance. Evidently, the latter ratio is perceived as 1:26 (2 to the sixth power)
rather than 1:64. When applying the same principle to the Pythagorean thirds,
one arrives at the ratios 33:25 for the minor, and 26:34 for the major third.
These ratios are simpler than the so-called "natural" ones in so far as the
basic figures do not exceed the value of 3, contrary to the "natural" thirds
which contain the value 5 as basic element. Helmholtz's overtone theory of
consonance has been refuted by Stumpf who proved that overtones could be
altered or even completely eliminated without affecting consonance or tone
affinity; therefore, the occurrence of the ratios 4:5 and 5:6 within the series
of partials is irrelevant in regard to the consonance of thirds. On the other
hand, phono-photographic measurements showed "that the violinist, when un-
accompanied, does not play consistently in either the tempered or the natural
scale, but tends on the whole to conform with the Pythagorean scale in the
intervals here studied' (Seashore 1938:224). These experiments are certain-
ly quite revealing, but the decisive argument in favor of the thesis of Pytha-
gorean ratios being the determinants of tint affinity and basic consonance is
the afore discussed and demonstrated fact that the grading of these two re-
lated phenomena follows the quintal order and not that of the ratios contained
in the overtone series. Most indicative in this regard is on the one hand
the greater tint affinity and higher basic consonance of the major second,
minor seventh and of triads such as CDG and CFG, in comparison to the
more brilliant thirds, sixths and major and minor triads, and on the other
hand the more subdued quality of the minor third in comparison to the
brighter major third. This situation exactly reflects the Pythagorean prin-
ciple since the Pythagorean vibration ratios for the major second and minor
71

This content downloaded from 129.64.99.141 on Tue, 1 Oct 2013 15:20:32 PM

All use subject to JSTOR Terms and Conditions
seventh involve the value 32, those for the minor third and major sixth 33
and those for the major third and minor second 34. Incidentally, this quintal
grading of basic consonance constitutes an additional strong argument against
Helmholtz's overtone theory of consonance.
The fact that major second and minor third are next to fourth and fifth
in their degree of tint affinity evidently accounts for the universal importance
of the former two intervals as melodic and structural elements, especially in
tribal and national folksong. The equally universal distribution of the so-
called anhemitonic-pentatonic scale can be easily understood if it is realized
that this structure embraces all existing direct tint relationships and that all
its parts are directly related to each other through tint affinity or tint ident-
ity. The frequent use as functional consonance of the so-called major chord
with added sixth and second in Western 20th century music represents an
exact harmonic equivalent of the melodic phenomenon of the anhemitonic-
pentatonic scale.
Because of the laws of tint affinity, all chords the tints of which are
placed within a section of 2-4 consecutive steps of the cycle of fifths are to
be considered basic consonances, and can be classified as tri-, tetra- and
pentaphonic chords according to the size of the cycle section involved. When
applying this rule to the music of Laos, we see that one of its character-
istic features is not the preference for dissonant chords, as Humbert-
Lavergne believed, but, on the contrary, a consistent use of basic conso-
nances. In Figure 4, an excerpt from a Laotian love song, played on a
mouth organ, the tetraphonic triad and tetrad G Bflat F and G Bflat D F
function as pedal points, while the pentaphonic pentad G C D F B flat serves
as closing chord (Humbert-Lavergne 1934:ex. 3). If we invert the latter by
making B flat the fundamental tone, we obtain the aforementioned favorite
modern Western major chord with added sixth and second. On the other
hand, seemingly dissonant chords occurring in early medieval polyphony, as,
for example, in the "Benedicamus," Figure 5 (Schneider 1935, ex. 112) prove
to be highly basic-consonant chords such as the triphonic triads DGA and
CFG and the tetraphonic tetrad CGDF. The same triad is used in parallel
motion, and manifestly as functional consonance, for example in Hindemith's
"Examen," (Figure 6). Listening to this passage strongly reminds us of
actual organal parallels in fourths, fifths and octaves in spite of the pres-
ence of continuous parallels in octave-enlarged major seconds.
This brings us to the last point of our discussion, namely, to the oc-
currence of parallel major seconds in musical styles as opposed to each
other as the Ambrosian chant of the Middle Ages, Figure 7 (Handschin 1943)
and the songs of the tribal inhabitants of the Caroline Islands, Figure 8
(Sachs 1943:49). Ferand's statement that "the explanation of the appearance
of consistent parallel seconds offers serious difficulties" reflects the general
view of present day research in regard to this phenomenon. But since the
basic consonance of the major second follows that of fourths and fifths, its
use in parallel motion is comparable to the parallels in fourths and fifths
practiced in various parts of the world, and corroborates the thesis of uni-
versal validity of the concept of basic consonance.

72

This content downloaded from 129.64.99.141 on Tue, 1 Oct 2013 15:20:32 PM

All use subject to JSTOR Terms and Conditions
 my- _7p pi

••"-".''i•
"" ..,' " ".......
........ ...... • ' i>•....".....
• ..........

Fig. 4

7W9•. .t ..4
Am I
- Id

Fig. 5

I Vol i _:7i
D4

Fig. 6

73

This content downloaded from 129.64.99.141 on Tue, 1 Oct 2013 15:20:32 PM

All use subject to JSTOR Terms and Conditions
 PADA- - - - ffA'n -

Fig. 7

•
ImaJ ,La_

Fig. 8

REFERENCES

Bach, Carl Philipp Emanuel

1753 Versuch fiber die wahre Art das Klavier zu spielen (2nd edition). Berlin,
C. F. Henning.
Ferand, Ernst T.
1939 "The howling in seconds of the Lombards," Musical Quarterly 25:313-324.
Handschin, Jacques
1934 "Zur ambrosianischen Mehrstimmigkeit," Acta Musicologica 15:2-15.
Humbert- Lavergne, M.
1934 "La musique a travers la vie laotienne," Zeitschrift ffir vergleichende Musik-
wissenschaft 2:14-20.
Kolinski, Mieczyslaw
1936 Konsonanz als Grundlage einer neuen Akkordlehre. Prague, Ver6ffentlichungen
des musikwissenschaftlichen Institutes der iDeutschen Universitit in Prag, No. 7.
Riemann, Hugo
1920 Geschichte der Musiktheorie (2nd edition). Berlin, M. Hesse.
Sachs, Curt
1943 The rise of music in the ancient world. New York, Norton.
Schneider, Marius
1936 Geschichte der Mehrstimmigkeit, vol. 2. Berlin, J. Bard.
Seashore, Carl
1938 Psychology of music. New York, McGraw-Hill.
Stumpf, Carl
1883-90
Tonpsychologie. Leipzig, S. Hirzel.
Stumpf, Carl, and
Meyer, M.
1898 "Massbestimmungen aiber die Reinheit konsonanter Intervalle," Beitriige zur
Akustik und Musikwissenschaft (Leipzig) 2:84-167.

New York City

74

This content downloaded from 129.64.99.141 on Tue, 1 Oct 2013 15:20:32 PM

All use subject to JSTOR Terms and Conditions