TOWARD A THEORY OF MEANTONE (AND 31-ET) HARMONY

Siemen Terpstra

ABSTRACT

The author develops a syntax for meantone harmony based on the structural characteristics of the temperament. These features include a description of the Regions with their boundary functions, and also the three genera. An abstract topology of triad progression is introduced, as well as an introduction to the organization of functional tables for 31-ET.

INTRODUCTION

The great one-quarter-comma meantone tuning system has had overriding importance in the history of European music culture, where it was put to use for longer than any other temperament. Moreover, it was the tuning system associated with the greatest flowering of that culture, being used from the high Renaissance through the 16th and 17th centuries and on well into the 18th century. The only evident reason for its demise was a problem of technology. No adequate keyboard was developed which enabled the performer to implement all 31 notes to the octave. Since the keyboard became more and more popular over this period of time, musicians moved to an inferior system which better suited the medieval keyboard. However, the long-term influence of this tuning can still be felt in the continued use of a musical notation which employs both flats and sharps, a legacy of the meantone era.

Modern electronic technology makes possible the fabrication of efficient music keyboards for 31-ET. Remember that most other instruments, for instance wind instruments and strings, have variable intonation and so can readily adapt to whatever temperament is at hand. The keyboard was the only impediment to the use of extended meantone music in ensemble settings. Enharmonic keyboards would be happily accepted by violin players who have no qualms about the use of the enharmonic, thus liberating the potential of the string ensemble. Hence we have the real possibility of a new flowering of meantone music. At the same time, many composers now feel that the 12-ET system has been exhausted of its new harmonic possibilities. As a result there is a new willingness to explore alternative systems of harmony. 31-ET offers immense harmonic resources which are as yet largely untapped. With the aim of promoting the exploration of meantone harmony I offer a systematic layout of the structural features of the system. Such a layout will suggest a natural syntax for meantone music, without dictating such matters as musical styles.

My aim in this endeavour is to develop a theory which is descriptive, and not proscriptive (as was the case with 19th century music theory). Musical styles change as the progressive working out of the implications of the structure develop over time. I have already used this observation with regard to the evolution of 12-ET harmony in the 19th century, an evolution which developed often in spite of the theoretical structure which was overlaid on the materials. We could say, for example, that the structural characteristics of 12-ET included an inherent tendency toward atonality. No matter how badly the theoretical structures of the late century fit the musical patterns, composers felt compelled, nevertheless, to explore the implications of the actual musical structure with its functional ambiguity and potential for rapid modulation. The theoretical language in use, meanwhile, was largely inherited from the meantone era, with its distinction of sharps and flats. Eventually this theoretical language became overly complex and stilted, since it did not suit the actual structure of 12-ET. To a large extent, this situation came about because of the lack of an adequate model with which to compare the divergent structure of 12-ET and 31-ET.

In the same way as 12-ET, 31-ET has its own structural character which endows it with inherent tendencies. The history of the evolution of harmony during the 16th and 17th centuries attests to the working out of various implications of the extended meantone fabric. One prominant example which comes to mind was the increasing use of chord relations which depend on the mediant during the course of the 16th century, so that by the end of the century we see the 'avant-garde' progressions of Marenzio and Gesualdo. Another example was the Baroque distinction between the usual type of Dominant Seventh tetrad and another type which is typified by the function 6-1-3-4: called a "German Sixth Chord". This chord displays the Harmonic Series Seventh between 6 and 4, instead of the usual Seventh which would have the function 6-5, thus giving it a different structure and sound. (This distinction is meaningless in 12-ET). Hence we see a progressive working out of the implications of the true, underlying structure of the system; moreover, we see the conservative maintenance of this theoretical language long after musicians were using 12-ET in fact, and not 31-ET. The 31-ET distinction between sharps and flats remains 'in theory' alone, and not in the actual structure of 12-ET.

The appropriate theoretical language for a system of harmony should arise from its true underlying structure, and not from some other system. Such a practice results in confusion if there is little understanding of the divergencies between the systems. The music cultures of the 19th and 20th centuries have displayed a deplorable ignorance in their comprehension of meantone harmony. Although the modern situation has improved, there is still the need of an integrated overwiew of the system anatomy. The composer who may wish to work in the milieu of extended meantone tuning needs a theoretical structure which reflects the inherent tendencies of the tuning. With such tools he/she can better utilize the materials at hand. The proper aim of a theoretical structure is to clarify the nature of the materials so that the composer or improviser can command his medium to produce the most satisfying results.

I have already written at length concerning the extended meantone fabric. In the article The Meantone Series of Cyclical Temperaments - An Introduction and Comparative Appraisal the most important structural features are presented within the context of the other great tuning cycles, so that an understanding of the structure can emerge from its contrast to Just Intonation. In the article The Extended Meantone Style of Gesualdo I have focused on historical issues so as to put the evolution of 16th century style within the context of meantone harmony. Both of these articles complement this one. In the interest of brevity, this article will develop issues that are only just touched upon in those two articles. In like manner, themes which are well worked in those two articles will be presented in an abbreviated form here, so that we may look upon these three articles as mutually complementary. Of course, a certain amount of overlap cannot be avoided. We are observing a jewel from many angles in order to appreciate its luminous structure.

THE EIGHT REGIONS OF THE FUNCTIONAL FIELD

31-ET is fundamentally a "microtonal" scale of 31 pitches to the octave. This "ultra-chromatic" scale uses the diesis or enharmonic (the difference between a sharp and a flat) as the basic step interval of the system. All other intervals are made up of aggregates of this basic step. Consequently, a good workable preliminary notation for the system is a numeration of the pitches from 0 to 30, pitch 31 being the octave equivalent to pitch 0. This numerated scale arbitrarily commences at 'C'=0. The Baroque notation for the system used naturals, sharps, flats, double-sharps, and double-flats. This notation is preserved in the modern Fokker notation, but it has been augmented in a clear and rational fashion so that double-sharps become semi-flats, and double-flats become semi-sharps. Additionally, flats may be renamed as sesqui-sharps, and sharps may be renamed as sesqui-flats, for purposes of clear staff notation. The complete rationality of this notation can be seen on the Circle of Fifths which this system exhibits. See Figure 1. chromatic wheels

31-ET is a cyclical temperament, meaning that the fifths are tempered (mis-tuned) in such a way that they form a closed circle of fifths, thus allowing unlimited modulation. The regional structure of the system emerges from the 'background' nature of Just Intonation, which has its own structural features. The anatomy of Just Intonation is carefully covered in the Meantone Series article. Briefly, we may say that the framework of Just Intonation stems from the restriction to the use of the pure 3rd and 5th harmonics in setting up the tuning. Consequently, a field or matrix of relations results which arises directly from the lattice of just musical triads. This lattice is open-ended, it does not form a closed cycle of fifths, due to the nature of number relations. However, several musical intervals crop up which suggest the possibility of alternative closed cycles. Such closed cycles can be achieved by slightly mis-tuning the Just norms in controlled ways resulting in an system of Equal Temperament. The schisma interval lies at the very edge of our pitch discrimination perception. Consequently, we could resolve the field into a scale of schismas, by tempering to the closed cycle of 612-ET. Such a scale is maximally accurate in its 'resolution' of the Just lattice, but its complexity makes it musically impractical. The presence of the Syntonic Comma interval (11 schismas in size), and the Ditonic or Pythagorean Comma (12 schismas in size), leads to a natural resolution of the field of Just Intonation into a scale of Commas, if we temper into the closed cycle of 53-ET. This great system of temperament can rightly be named Quasi-Just Intonation since it preserves so many of the essential properties of pure Just Intonation within a reasonably manageable complexity.

Figure 2: The field of Just Intonation showing the comma based notation: field of just intonation

The regional structure of 53-ET discloses three primary regions: a central region in which intervals are not comma-altered, and two flanking regions in which intervals are raised by a comma (on the dominant side), and lowered by a comma (on the sub-dominant side). See Figure 2. The important boundary between these regions of Just Intonation I have called the Meantone Corridor. 31-ET tempers away the comma, but preserves the Diesis, which is about two commas in size, as its basic step interval. This is done by tempering the fifths by one-quarter of a comma, so that the two pitches D and \D of Just Intonation form a compromise pitch of about midway in size - the meantone. Such a compromise strongly orients the lattice of triads along the Major Thirds Axis of Just Intonation, since the major thirds are tuned pure while the fifths and minor thirds are compromised. The notion of a triad lattice is preserved, but it is given a decidedly 'vertical' cast. Figure 3 shows the alteration of the Just Field in parts of a comma. Figure 4 indicates the resultant field of pitches with the modern 31-ET notation. deviation meantone-JI

31-tone lattice

We see that certain aspects of the Just Field are preserved, while other aspects are eliminated. The Schisma boundary (which is elsewhere also called the 'East-West Boundary' of the Field) is eliminated along with the comma-alterations, leaving the Meantone Corridor as the true 'east-west' boundary of the field. At the same time the Enharmonic Gateway (the North-South boundary) of 53-ET and 19-ET is eliminated, while preserving the Harmonic Antipodes with its neutral intervals. The orientation of the lattice of triads is quite 'vertical' along the Major Thirds Axis of Just Intonation, where the enharmonic displays itself.

This lattice of triads is the key to understanding the field of potential chord relations for the system, since it may also be viewed as the field of functional relations to a given Tonic. Figure 5 shows this Functional Field for 31-ET using the numerical notation (which may thus also be employed to describe functional relations) as well as the more standard functional notation in which "do-re-me" becomes "1-2-3". Thus the tonic major triad has the function 1-3-5, and the tonic minor triad has the function 1-3-5. It is important to remember that this chart is independent of any particular pitch, since any pitch in the system may act as the Tonic. It shows all of the ways in which various tones can relate to the Tonic. These tones combine to form triads, tetrads, etc. leading to the complex relations of functional harmony.

31-ET functional field

The 'vertical' nature of the functional field leads to a natural stratification of the lattice into eight regions with the Enharmonic Region occupying both the 'south' and the 'north' of the triad web. The Meantone Corridor forms the appropriate boundary between the regions. Triads may be classified according to where they 'sit' in the lattice. For instance, the 6 major triad (6-1-3) lies squarely in the middle of the Minor Region; whereas, the 6 minor triad (6-1-3) lies within the Major Region. Note that the Enharmonic Region lies at the furthest removed position from the Major Region. The top and the bottom of the chart meet in a 'strange loop' since it is a closed cycle. Consequently, the regions may also be indicated on the Circle of Fifths, as shown on Figure 6.

Certain functional triads seemingly lie in two regions at once. For example, the 2 major triad (2-4-6) appears in the Major Region as well as the Supra-Major Region, as does its Relative Minor - the 7 minor triad (7-2-4). Such triad functions which have two componants within the Meantone Corridor are called boundary functions. These functions have peculiar significance to meantone tuning. They do not have the same character in Just Intonation or in 12-ET where this whole regional structure collapses to essentially three: Major, Minor, and 'Enharmonic'. Thus in 31-ET the 5 minor triad (5-7-2) lies on the boundary between the Major and the Minor Regions, as does its relative major triad, 7 major triad (7-2-4). Suspended triads and tetrads, for example the function usually called the 'Tonic Suspended Fourth Triad' (1-4-5) straddle the Major and Minor Regions. They are 'ambivalent' about their major or minor character and do not 'disclose' their region. Certain other harmonic structures, on the other hand, may cover two regions for even more) at once. For example, the Diminished Triad called 6 Dim. Triad (6-1-3) covers the Major and Minor Regions at once. The Tonic Wolf Major Triad (1-3-5) spans three regions: the Minor, Flat-Minor, and Semi-Major Regions. The two pitch functions which stand at the furthest removed position from the Tonic, in the very heart of the Enharmonic Region are the neutral third and sixth, 3 and 6. Note their position on the field lattice of Figure 5.

Figure 6:
The eight regions shown on the Circle of Fifths-Fourths. The functional notation indicates major triads (outer part of the circle), and minor triads (inner part of the circle). The majors and minors are paired by the traditional relation known as "Relative Major-Relative Minor" (Type 3 pairing).
The eight regions are highlighted. Major-minor triad pairs associated with the boundary between regions (that is, triads with two tones within the meantone corridor) are marked by asterisks:
major and minor regions
Major and minor triads naturally pair along these lines so that the circle of fifths becomes the circle of major and minor triads in functional array around the Tonic. Note the symmetrical nature of the regions. The Enharmonic Region lies 'opposite' the Major Region in which the Tonic resides. Modulations to related major and minor key centres can be mapped on the circle of fifths according to movement in a clockwise or a counter-clockwise direction. An awareness of the regional structure of 31-ET leads to increased understanding of the specifics of meantone progression.

THE THREE GENERA

The ancient Greek classification of musical 'types' according to the notion of Genus is perfectly suited to the meantone system. Intervals, chords, and scales can all be classified according to whether they are Diatonic, Chromatic, or Enharmonic. Such a classification is meaningless to 12-ET where the enharmonic is eliminated. This feature of 31-ET can already be seen in the functional field of Figure 5. The Ptolemaic structures which we associate with the diatonic major and minor scales of meantone and of Just Intonation form a 'double strand' of fifths on the graph. Within this graup of intervals, for example, is the diatonic semitone 3-4 (eg. E-F). If we now 'skip' a level we get the semitone associated with the Chromatic Genus - the chromatic semitone 5-5 (eg. G-G). Within the chromatic grouping are the important septimal intervals, for instance 2 and 6. Thus we can also associate the diatonic with the '5-Limit' intervals, and the chromatic with the '7-Limit' intervals. If we now jump one more 'vertícal' level we integrate the enharmonic intervals; for instance, we now have the enharmonic 'semitone' between 7 and 1 (eg. C-C). We have jumped these levels in a 'northerly' direction, but we can also do so symmetrically in a 'southerly' direction. The notion of the three genera is another result of the 'verticality' of the system.

The field of Just Intonation also embodies a similar structure, but it has in addition a 'horizontal' component in the comma-altered regions. In 31-ET this horizontal component is eliminated so that the 'vertical' component takes on much greater structural importance. Hence we have the classification of patterns in the system according to the three genera alone and not the additional comma-alterations. We will use the notion of Genus in investigating the intervals and also the more complex harmonic structures of the tuning system.

THE CLASSIFICATION OF THE INTERVALS

Musical intervals are the primary building blocks of a harmonic system. Hence an understanding of the interval types is essential in mastering the materials for composition. 31-ET offers a rich assortment of intervals, which may be grouped according to the notion of Genus.

It is helpful to group an interval with its inversion, so as to form a class of interval type in aggreement with the principle of octave equivalence. Such a grouping of the intervals of 12-ET leads to seven interval classes, which may be ordered according to degrees of consonance. This order stems from the progressive appearance of the interval types within the Harmonic Series:
1. Unison - Octave } Perfect Consonances
2. Perfect Fifth - Perfect Fourth
3. Major Third - Minor Sixth } Medial (Imperfect) Consonances
4. Minor Third - Major Sixth
5. Wholetone - Minor Seventh } Dissonances
6. Halftone - Major Seventh
7. Tritones
In order to classify the intervals of 31-ET we may keep this overall pattern, but superimpose over it the notion of Genus. Hence the class 6 intervals (the wholetones-minor sevenths) have a diatonic version, a chromatic version, and an enharmonic version. All of this information may be conveniently laid out in the form of a table which is Figure 7. Note that the numerical notation is useful here since it indicates the size of the interval (the number of steps involved in making up the interval). Only the first and second class intervals do not have representatives in all three genera.

Figure 7: Overview of interval classification:

Class Diatonic Genus Chromatic (Septimal) Genus Enharmonic Genus

1 0 - unison
31 - octave

2 18 - fifth19 - "wolf" fifth
double-flat sixth
semisharp fifth
13 - fourth12 - "wolf" fourth
sharp third
semiflat fourth

3 10 - major third11 - septimal major third
semisharp third9 - neutral third
flat fourthsemiflat third
23 - major sixth24 - septimal major sixthdoublesharp second
semisharp sixth
doubleflat seventh

4 8 - minor third7 - septimal minor third
sesquisharp secondsharp second22 - neutral sixth
sesquiflat thirdsemiflat sixth
21 - minor sixth20 - septimal minor sixthdoublesharp fifth
sesquisharp fifthsharp fifth
sesquiflat sixth

5 5 - wholetone6 - septimal wholetone4 - semiflat wholetone
semisharp seconddoublesharp prime
doubleflat third
26 - minor seventh25 - septimal minor seventh27 - neutral seventh
flat seventhsharp sixthsemiflat seventh
sesquisharp sixthsesquiflat seventhdoublesharp sixth

6 3 - diatonic semitone2 - chromatic semitone1 - diesis, doubleflat second
flat secondsharp primeenharmonic
sesquisharp primesesquiflat secondsemisharp prime
28 - major seventh29 - flat octave30 - semiflat octave
semisharp seventhsharp seventh

7 16 - flat tritone15 - sharp tritone14 - semisharp fourth
flat fifthsharp fourthdoubleflat fifth
sesquisharp fourthsesquiflat fifth
17 - semiflat fifth
doublesharp fourth

The class 3 and class 4 intervals have been somewhat 'shuffled' to reflect the fact that the major third and major sixth embody more consonance than the minor third and the minor sixth. The chromatic and enharmonic forms of the thirds and sixths obviously fall within the status of dissonances rather than consonances, with the neutral (enharmonic) forms exhibiting strong dissonance.

In spite of the notion of octave equivalence, the particular voicing of the interval is a big factor in judging the strength of its consonance. The minor third is the most significant example. In closed position the minor third embodies more consonance than its chromatic cousin, the septimal (or sub-minor) third; however, in open or compound form (as a minor tenth) the sub-minor form gains in consonance over the minor form. This Sub-Minor Tenth is on the verge of being considered as a consonance, just as the diatonic minor sixth is on the verge of being considered as a dissonance. Some intervals gain in their compound expression (eg. fifth, major third, wholetone) and others lose (eg. fourth, major sixth, minor sixth, minor third).

Of course, the intervals can also be arranged according to their size, leading to Figure 8. This table is the appropriate place to include other technical information relating to the 31-ET scale. The fractional ratios are useful for tabulating positions in the layout of fretboards. The Frequency numbers are useful for synthesizer tuning programs which have the option of direct input of frequency data. The two columns on the right indicate the size of the interval in cents and also the size of closely related just intervals in cents, so that we can compare the deviation of 31-ET from Just.

Figure 8: Technical data on 31-ET intervals:

Numerical
notation
Ratio
in cents
Interval
notation
Fractional
ratio
Relation to
just ratio

0 0.000 C 1.000000 1/1
1 38.710 C D 1.022611 64/63
2 77.419 C D 1.045734 21/20
3 116.129 C D 1.069379 16/15
4 154.839 C D 1.093559 35/32
5 193.548 D 1.118286 9/8
6 232.258 D E 1.143572 8/7
7 270.968 D E 1.169430 7/6
8 309.677 D E 1.195873 6/5
9 348.387 D E 1.222913 128/105
10 387.097 E F 1.250565 5/4
11 425.806 E F 1.278842 9/7
12 464.516 E F 1.307759 21/16
13 503.226 F 1.337329 4/3
14 541.935 F G 1.367568 48/35
15 580.645 F G 1.398491 7/5
16 619.355 F G 1.430112 10/7
17 658.065 F G 1.462449 35/24
18 696.774 G 1.495517 3/2
19 735.484 G A 1.529333 32/21
20 774.194 G A 1.563914 14/9
21 812.903 G A 1.599276 8/5
22 851.613 G A 1.635438 105/64
23 890.323 A 1.672418 5/3
24 929.032 A B 1.710233 12/7
25 967.742 A B 1.748904 7/4
26 1006.452 A B 1.788449 16/9
27 1045.161 A B 1.828889 64/35
28 1083.871 B C 1.870243 15/8
29 1122.581 B C 1.912532 40/21
30 1161.290 B C 1.955777 63/32
31 1200.000 C 2.000000 2/1

DIFFERENCES BETWEEN TRUE MEANTONE AND 31-ET

It is useful to conceptualize the extended meantone fabric and 31-ET as equivalent because they are sonically almost identical. However, we should keep in mind that, strictly speaking, subtle distinctions exist between the two tunings. The 'true' extended meantone fabric is essentially a line of musical fifths each of which is flattened by exactly one quarter of a syntonic comma. Such a line of fifths will never meet to form a circle. In the 31-ET tuning the size of the musical fifth is slightly altered from 'pure' meantone so that a circle of fifths results. The 31-ET fifth is widened by about 0.19 cents (an amount which is sonically imperceptible) from the meantone fifth. All of the intervals of the two systems are hence slightly skewed to each other.

We can best compare the two tunings by projecting the line of fifths in the two directions from the Tonic, culminating in the Enharmonic Region. The interval sizes, expressed in cents, can then be contrasted and their differences tabulated. See Figure 9. Note that the greatest deviation of the 31-ET temperament from the 'true' me antone temperament amounts to only 3.13 cents, and occurs at the most remote functional relations of 3 and 6. We can consider these amounts to be negligable, especially for the more common functions. Consequently, the two systems may be considered as identical for practical purposes, sonic effect and in theoretical structure.

Figure 9: The difference in interval size between 31-ET and the extended quarter-comma meantone tuning (all values are in cents):

Number
notation
Interval
notation
31-ET 1/4-comma
meantone
Difference MT-31

22 G A 851.613 854.746 3.133
9 D E 348.387 351.324 2.936
27 A B 1045.161 1047.902 2.741
14 F G 541.935 544.480 2.545
1 C D 38.710 41.059 2.349
19 G A 735.484 737.637 2.153
6 D E 232.258 234.216 1.958
24 A B 929.032 930.794 1.762
11 E F 425.806 427.373 1.566Subdominant
29 B C 1122.581 1123.951 1.370direction
16 F G 619.355 620.529 1.175
3 C D 116.129 117.108 0.979
21 G A 812.903 813.686 0.783
8 D E 309.677 310.265 0.587
26 A B 1006.452 1006.843 0.392
13 F 503.226 503.422 0.196

0 C 0.000 0.000 0.000Tonic

18 G 696.774 696.578 -0.196
5 D 193.548 193.157 -0.392
23 A 890.323 889.735 -0.587
10 E F 387.097 386.314 -0.783
28 B C 1083.871 1082.892 -0.979
15 F G 580.645 579.471 -1.175Dominant
2 C D 77.419 76.049 -1.370direction
20 G A 774.194 772.627 -1.566
7 D E 270.968 269.206 -1.762
25 A B 967.742 965.784 -1.958
12 E F 464.516 462.363 -2.153
30 B C 1161.290 1158.941 -2.349
17 F G 658.065 655.520 -2.545
4 C D 154.839 152.098 -2.741
22 G A 851.613 848.676 -2.936
9 D E 348.387 345.254 -3.133

Meantone fifth = ratio 45 = 1.4954 = 696.578 cents. The 'pure' meantone tuning can be expressed as a line of fifths in which the ratios are all multiples of two and the fourth root of five.

It is interesting to note that he 'true' meantone fabric is made up of intervals which can all be expressed as powers and roots of the numbers two and five. For example, the meantone itself is the ratio 5 / 2, and the musical fifth is the ratio 45 (the fourth root of five, which means the square root of the square root of five). The number five and the fifth harmonic are, of course, associated with the Major third axis. The entire extended meantone field can be expressed in ratios which stem from the numbers two and five. (See Figure 39 in the Meantone Series article). Although 31-ET deviates from these ratios in slight amounts, the essential quality is the same. Since these ratios belong to the family o£ ratios associated with the Golden Section, we can associate 31-ET with the Golden Section just as well as the 'true' extended meantone tuning. Thus in practically every aspect, we can equate 31-ET to the extended meantone tuning.

THE LAYOUT OF THE FUNCTIONAL TABLES

The cyclical nature of the system results in the ability to 'map' out all of the possible functions of the tuning in a manner similar to that used for 12-ET in The Dictionary of Harmonic Functions for 12-ET The organization is essentially that of non-ordered sets or pitch class. The 'patterns' are first grouped according to internal structure into a class of functions within a modulating framework. For example, we have the functions associated with the class of major triads, and in addition the functions associated with the class of wolf major triads, and so on. Each class of pattern occupies one page. Then the various and sundry classes of patterns are organized according to structural complexity, into duads, triads, tetrads, pentads, and so on theoretically to the one and only 31-tone pattern (the ultra-chromatic scale).

It is obvious that any given pattern must embody 31 possible Functions. For instance, any given tone can relate to the Tonic in exactly 31 different ways. These ways may be described by the usual functional notation of 3, 4, and so forth; or alternatively, the numerical notation is also possible. For the sake of convenience and brevity, I have used the latter approach in the tables. The tables are also accompanied (as in the 12-ET Dictionary) by a representative of the class on the Chromatic Wheel and the Wheel of Fifths circular graphs. These graphs make clear and visual the interval connections between complex patterns, as well as such structural characteristics as symmetry, complementarity, and sub-set groupings.

Note that all of these mappings of structure: the triad lattice, the Wheels, and the functional table, assume the notion of octave equivalence - a natural companion of the concept of pitch class. Hence a vast amount of data is presentable in an elegant form. For example, the totality of relationships possible between pitches taken 'two notes at a time' (Duads) amounts to only fifteen classes (shown at the bottom of Figure 8). These are all of the possible intervals paired with their inversions. As a first example of a representative page from the functional tables, I present the duad associated with the neutral third and its inversion, the neutral sixth. See Figure 10. Functions which contain the Tonic are called primary functions, the others are called secondary functions. Of course, here there are only two Primary Functions: 0-9 and 0-22. The table can not only be read to refer to functional relations, but also pitch relations. It shows all of the ways that two pitches can relate to each other by the internal structure of a neutral third and its inversion. From the two Wheels we see that the pattern is symmetrical.

Figure 10: A sample from the book of functional tables for 31-ET.

As a practical application of this data, suppose we want to find the function that is a neutral third from the Dominant (function 18). Simply find 18 on the table and read 27 to its right. Function 27 is the neutral seventh. The table is simply a modulating framework for the internal relation which defines the class of pattern.

As in the 12-ET Dictionary of Harmonic Functions, the various classes of patterns are organized according to whether they are triads, tetrads, etc. As in that system, there is only one pattern representing the empty set (silence). This pattern may be linked to the one pattern which is the 'full set' - the ultra-chromatic scale. Then there is only one class of patterns called the Monad, associated with the interval of the Unison and its inversion, the octave. The Monad and the fifteen Duads are lumped together to constitute all of the possible interval patterns.

The total number of possible patterns in the entire 31-ET system amounts to 231 (two to the 31st power), a huge number! Hence a Dictionary of all of the possible patterns in the system is impractical. It is, however, possible to tabulate all of the patterns and organize them according to class, consonance, symmetry, and so on. Such a comprehensive mapping is practical for 12-ET which has only 212 (4096) patterns, which may be grouped into only 352 classes. 31-ET, on the other hand, has 230 classes. Consequently, in my tabulation of the patterns, I have only mapped the most prominant harmonies which are of greatest interest to me. However, the methodology for a complete mapping exists.

In order to show some further examples from the functional tables, I present the data for two forms of the 'Dominant Seventh' tetrad (alluded to earlier in the article). They are presented as Figure 11 and Figure 12. Note that a typical resolution format is provided in brackets. These two patterns can be grouped according to genus as diatonic and chromatic respectively.

Figure 11: A sample tetrad from the book of functional tables for 31-ET.

In a similar manner, all possible functions for any given structure or 'pattern in the system may be conveniently mapped so that its full functional possibilities may be guaged with a minimum of time imput. Patterns of great complexity, such as Hexads, Heptads, and Ogdoads tend to be regarded as 'scales' rather than 'chords'. The table can be viewed in both a vertical and a horizontal manner. In complex patterns the primary functions become the 'modes' of the pattern. Unfortunately, the short format of this article does not allow me to discuss at length the specifics of all the types of patterns in the system. Such information can best be gleaned by examining the functional tables themselves, which speak for themselves. Two more examples of pattern classes appear as Figure 13 and Figure 14. The former figure shows a dissonant triad class of the enharmonic genus. The latter figure maps the Functions for an eleven note segment of the circle of fifths. Of course, very many other classes of patterns exist in the system. Every class of patterns has its own functional table, defining and delineating all of the possible ways in which that class of patterns can be used (independent of specific octave voicing). These tables must be considered as the practical heart of the meantone system.

THE USEFULNESS OF THE MODEL OF THE TWO WHEELS

In addition to the mapping of the triad lattice, a given harmonic pattern is also charted on the Chromatic Wheel and the Wheel of Fifths. These two maps each give useful and distinct information relating to the structure of the pattern. Hence the Wheel mappings are compiled together with the lattice graph and the functional table.

The Chromatic Wheel lays out all of the potential voicings of the pattern, and reveals the interval (subset) structure of the pattern. Remember that, even though octave equivalence is assumed in designing our model, it is still possible to look at the Chromatic Wheel as an 'octave-specific' model. Movement on the Wheel in a clock-wise direction indicates falling pitch, and counter-clock-wise movement rising pitch. A complete 'round' is an octave. Thus the interval vectors can be viewed in an octave specific manner. In addition, the vector sizes are colour-coded, so that the interval size is easy to 'read'. (Unfortunately, the colour code must be eliminated for this article). Hence the Chromatic Wheel gives a direct, visual 'analogue' of the interval relationship and potential layout of the pattern.

The Wheel of Fifths, on the other hand, is never associated with a 'Staff' voicing layout. The relations are truly 'abstract' and strictly octave equivalent. This model gives information on the harmonic spread' of the pattern, its consonance and its region of the circle. Strong harmonies tend to be tightly grouped clumps of vectors; whereas, dissonant harmonies tend to be 'spread out' forms of vectors. These wide 'splays', for example the tetrad in Figure 12, define a domain of the circle of fifths in which the dissonance has its most natural resolution. (Of course, some would argue, that this particular pattern is not a dissonance at all, and that it has no natural tendency to resolve like its diatonic cousin. Even if we grant this value to the Harmonic Series Seventh tetrad, it still most naturally 'progresses in the way outlined).

The Wheel of Fifths gives information which is purely 'harmonic or abstract, much like the triad lattice mapping of the pattern. The lattice and the fifths wheel are closely related. The study of the pattern on both models together indicates a clear mapping of the harmonic 'tendency' of the pattern. This inherent tendency to movement is then 'expressed' through a specific function and a specific voicing.

AN ABSTRACT TOPOLOGY OF 31-ET TRIAD MODULATION

We have emphasized elsewhere that meantone tuning is more conducive to an extended tonality (or modality) than to atonality. Since it is a prime system, it embodies no 'reductionist' functions which play such a pivotal role in the structure of 12-ET. Every type of chord displays 31 possible functions, and there can be no ambiguity of functional intent. This situation does not mean, however, that tonality must necessarily follow traditional channels. The gravitational pull of tonality may be greatly diffused by using dissonant patterns and distant modulations. Nevertheless, some form of tonality has a tenuous hold on meantone progression.

The tendency of 12-ET patterns toward atonality is also aided by the 'restless' nature of the major and minor triads, with their significantly tempered thirds. 31-ET triads, on the other hand, are much more euphonious and 'restful'. Hence the status of the triad patterns is characteristically enhanced in meantone harmony. This situation also contributes to the tendency for meantone music to explore 'extended tonality' rather than atonality.

The early history of the use of meantone harmony was closely associated with the choral traditions of Renaissance and Baroque music. It could even be argued that this choral tradition was made possible through the implementation of this tuning system over the medieval 'Pythagorean' tuning system. Anyone who has worked with meantone is aware of the eminantly 'singable' quality of the triads. The 31-ET triads are truly a treasure having a 'sweetness' unmatched by other systems of temperament.

The quality of these triads means that progressions connecting such chords are natural to the system. Indeed, the music of the 16th and 17th centuries was largely triadic for probably this very reason. In the article on Gesualdo I examined the types of chord progressions between major and minor triads which prevailed throughout the 16th century. These progressions may be conveniently and systematically organized according to their relations on the triad lattice. Such a structuring of progression types is a topology of relations between triad chords.

The topology presented in that article was not comprehensive, since my interest was specifically the 16th century. Only such experimental composers as Gesualdo at the end of the century ventured into the more 'distant' progressions. As usual, this abstract topology of triad progression can be organized according to Genus. Figure 15 is a comprehensive reworking of this topology to include all of the possible 'progression types' between major and minor triads. Note that the majority of the 63 progression types fall in the enharmonic genus. Most of these progressions were largely untried before meantone went out of fashion in the 18th century, although Gesualdo used some of the enharmonic progressions in the 16th century. Thus we can see the immense scope for original harmonic progressions in the meantone system.

As an example of an unusual triad progression from this topology, I refer you to the progression marked as number 56 in the series. Here we have the unique situation in which a major triad moves to a minor triad (or the reverse in which a minor triad moves to a major triad) in such a way that all of the intervals shift by an enharmonic alone. Thus, for example, we have the move from C minor triad (C-E-G) to C major triad (C-E-G). Such a progression is peculiar to meantone tuning.

Of course, I do not want to leave the impression that only triad progression is possible in the meantone system. The vast array of possible structures in this system is truly dazzling. A new and largely unexplored universe of harmony is opened up - a universe which has many inhabitants, both beautiful and bizarre. It is hoped that the use of the triad lattice and the Fifths Wheel introduces a sense of order to this complexity. The movements of meantone harmony can be guaged according to these criteria. Rather than dictate as to matters of style, the aim is simply to create an awareness of the structure of the movement, so that the composer can make intelligent choices.

CONCLUSIONS

This article has served as an introduction to the structure of the meantone system. Due to the vast assortment of materials in the tuning, only the conceptual framework has been given, along with a few examples to illustrate the concept. The real 'meat' of the system lies in the functional tables, with their delineation of 'mode' and pattern. These tables rightly accompany this article, and this article properly serves as an introduction to the intelligent use of these tables. Such tables are by their very nature voluminous, even if they are easily organized into books of triads, tetrads, heptads, etc. The composer will find in these patterns not only the material for new composition, but also the basis for an intelligent analysis of past composition, in whatever style the composer had used.

It is hoped that this conceptual framework complements the Fokker notation for the system, a notation which has done so much to improve the intelligibility of the system, and make it applicable to the musical Staff. The combination of a clear notation, a clear comprehension of structure, and a suitable keyboard technology for the composer's exploration, opens the pathway to a glorious future in the evolution of musical harmony.