Introduction to Post-Tonal Theory Joseph N. Straus, Queens Colege nd The Graduate Comer Of the City Universiy af New York Prentice Hall Englewood Clifs, New Jersey 07632 || 0~ Finnie b EditoriaVproduction supervision and vey interior design: Arthur Masel ~ ~~ * Cover design: Dine Saxe Manufacturing buyer: Ray Keating A sights reserved Nov ptf tht hook may he reproduced, in say form or by any means, ‘withont permission in writing from the publisher Printed in the United States of America wo8765 4.99 4 ISBN 0-23-b8bb92-1 Prentice-Hall International (UK) Limited, London Prentice Hall of Austalia Ply. Limited, Sydney Prentice-Hall Canada Ine, Toronto Prentice-Hall Hispanoamericana, S.A., Mexico Prentice-Hall of India Private Limited, Neve Delhi Prentice-Hall of Japan, Ine., Tokyo Simon & Setwstor Asia Pre Ltd. Singapore Baitora Prentice-Hall do Brasil, Ltda, Mio de Janeiro doer Contents Preface ’ CHAPTER 1 Basic Concepts and Definitions ij Pitch class 2 Thieger notation 3 Pitch intervals 6 ‘Octave equivalence 1 Enharmonic equivalence 3 Mod 124 Intervals 5 Ordered piteh-clats intervals 6 ‘Unorderedpite-class intervals 7 Interval class 7 Inervalclas content 9 Exercises 12 Analysis 1 ‘Weber, “Wie bin ich froh” from Three Songs, Op. 25. 16 Schoenberg, “Nacht,” from Plerrot Lunaire, Op. 2123 CUAPTER 2 Piteh-Class Sets 26 Pitch-class sets 26 Normal form 27 ‘Transposition 30 Inversion 34 Index number 37 Set class 39 Prime form 47 Exercises 44 Analysis 2 Schoenberg, Book ofthe Hanging Gardens, Op. 15, No. 11 47 Bartok, Sting Quaret No, 4, fist movement 33w Contents curren Some Aatiinal Relationships 59 Preface Common ne mdr anosion (59 Gommon tres un neon 1) 62 Depetotemnety ean 67 Eipewceietor 2 Rimwentiqeacenion 5 apes conpecnemt 72 Bunce 76 Analysis 3 Webem, Movements for String Quartet, Op. 5, No, 4 89 Borg, “Sehlafend trigt man mich," from Four Songs, Op. 2, No. 2 84 cuarrer 4 Centricity and Some Important Referenial Collections 99 ‘Tonalty and centricity 89 The diatonic collection 93 ‘Compared to tonal theory, now in its fourth century of development, post-tonal theory is in ‘The datnic octal '96 "The ocatonie collection 97 felfany. Ara fs, oe ae sl itr arc of degree and eae ignorance Inversion as 101 Exercises 208 ‘the su ine abroad consensus has bop to cinerge ering hs ase asia elements fuchsia soe, tamony, an election. This boo reps a somes an aes anabeet It available to’an adince outske the professional theoretical community. It inrodees basic Sitavinaky, Oepus Rex, rears nos. 167-70. 110 theoeal concept forthe pst ont nse othe vent cen Bary Sons, fist movement 114 “Tao oaks tended for an alone of undergradte msc majors. Viwally all callegzs ad niversies now recone the mprtane ofthe study of tweet eaary music cuarren 5 Sd most requ atleast oe couse fo twenty Techies and analysis 18 for asic Twelvecone Operations Such cure tht hie book i designed mem ducecn ore ae ‘Basic theoretical concepts are presented in six chapters, illustrated with music drawn Parents St | geen 20 largely from the ‘“classical’’ pre-war repertoire, by Schoenberg, Stravinsky, Bartok, Berg, Subect structore 128 — Enercions 136 ‘and Webern. Three principal kinds of post-tonal music are discussed: free tonal music, twelve- daapare ine musty anf ccnne muse. Kestonahly- diet theories have grown wp aud ach, Sigh thre cn be conical Daring ofthe bandas, bth musical and heretical Schoenberg, Suite for Piano, Op. 25, Gave 136 ach af te Horta chap Ir fllowed by # pai of abt analacs, designed to Siravindys In Memoriam Dylon Thomas 143 ply the nce once in traning tas! context "The ass ar tended to ‘Plasterer tan comple "The works anaacd ave all ean Wily ascsse and cuarre 6 Snilogze: hese ae the werk tnt tacers re font Hey Uo kw end deat to find More Twelve-Tone Topics 7 tat ani he ns det, neo apron ews cen Invariants 147 Combinatoialty 152 Twelvetone areas 158, EBSA rh Gs ay a eee nace es peered eral 02s xacioes 108, “This book makes no pretense to comprehensiveness, either chronologically or theoretical Rather, like books on scales, ads, and simple harmonic progressions in tonal music, ‘consis base theoretical framework within which students ean undertake serious inquiry ig Oa Ops Bt into the great and representative works of this century. ‘Schoenberg, Piano Piece, Op. 33a 173 Inevitably, a book of this kind owes profound intellectual debts to many individuals, PEE | bibliozraphies that follow each chapter, but my debt to the pioneers of post-tonal theory— List of Set Classes 180 “Mitton Babbitt, Allen Forte, David Lewin, Robert Morris, George Perle, and John Rabn—is a re eee a cae ele ares a Se ee ee eee ae ene SIHDNGA ep ‘of the City University of New York for their gentle forbearance and helpful suggestions regarding ree secu ae eater ee er ae meen ea Indes Veo 2 ee i Oe a Ear cae peel ‘Alexander Brinkman. At Prentice Hall, Bud Therien nurtured the projet from is inception fn Athur Mail hsiped shape a from a ough manuscript ico its presen polished form. Ta Index 215 nalters oth tangible and olan, Sally Goldfarb has offered continhng guidance and sropert ‘beyond my abit to deseribe or repay Josegh N. Straus _— rea Preface CHAPTER | Basic Concepts and Definitions OCTAVE EQUIVALENCE ‘There is something special about the octave. Pitches separated by one or more ‘octaves ate usually perceived as in some sense equivalent, Our musical notation teflects that equivalence by giving the same name to actavesrclated pitches, ‘The name A, for example, is given not only to some particular pitch, like the A a minor third below middle C, but also to all the other pitches one or more octaves above or below it. Octave-related pitches are called by the same name because they sound so much alike and because Western music treats them as functionally equivalent. ‘Equivalence is not the same thing as identity. Example 1-1 shows a melody. from Schoenberg's String Quartet No. 4, first as it occurs at the beginning of the movement and then as it occurs a few measures from the end. Vita2 asic Concepts and Definitions The two versions are different in many ways, particularly range, ‘The range of the second version is so wide that the first violin cannot each all of the notes; the cello has to step in to help. AC the same time, however, itis easy to recognize that they are basically the same mclody—in other words, that they are equivalent, They are not identical, but they have something basic in common, That something is expressed precisely inthe concept of octave equiva- lence. ‘In Example 1-2, the opening of Schoenberg's Piano Piece, Op. 11, No. 1, compane the first three notes of the melody withthe sustained notes in measures 4s. 1 Mabige J Piano a xample 1-2 Two equivalent omic! eas (Schonberg, Piano Pies, Op. 11, nob. “There are many differences between the two collections of notes (register, artcula- tion, rhythm, ete.) bul a basic equivalence also. They are equivalent because they both contain a B, a Gl, and a G. When we assume octave equivalence and other kinds of equivalences we will discuss later, our object is not to smoot fut or dismiss the variety of the musical surface. Rather, we seek to discover the relationships that underlie the surface and lend unity and coherence to musical works. PITCH CLASS ‘We will need to distinguish clearly between a pitch (atone with a certain frequency) and a pitch class (a group of pitches with the same name). Pitch class A, for ‘example, contains all the pitches named A. To put it the other way around, any pitch named A is a member of pitch class A. Sometimes we will speak about Specific pitches; at other times we will talk, more abstractly, about pitch classes. ‘When we say that the lowest note on the cello is aC, we are referring t0 a specific pitch, We can notate that pitch on the second ledger line beneath the bass staff. When we say that the tonic of Beethoven's Fifth Syraphony is C, we are referring not to some particular pitch C, but to pitch class C. Pitch class C is an abstraction and cannot be adequately notated on musical staves. Sometimes, for convenience, we will represent a pitch class using musical notation. In reality, however, a piteh claas ip not a single thing; i is a class of things, pitches one. for more octaves apart Basie Concepts and Definitions 3 ENHARMONIC EQUIVALENCE In common-practice tonal music, a Bb is not he same as an Af. Even on an cequal-tempered instrument like the piano, the tonal system gives Bb and AF ity. The melodies in Example 1-3 are enharrmonically equivalent (although the first one is much easier to read). ‘Bxample 1-3 Eahannonic euialence, IVTEGER NOTATION cause of enharmonic equivalence, there are only twelve different pitch classes. All the Bis, Cis, and Dis are members of a single pitch class, as are all the fs and Dis, all the Ces, Ds, and Es, and so on. Composers in the twentieth fcenuiry have generally continved 1 use waditional staf notion, where Ab is jolated differently from GE. However, for our theoretical and analytical purposes, ‘we will also use integers from 0 through 11 to refer to the different pitch classes. Figure 1-1 shows the twelve different pitch classes and some of the contents of each. linteger name pitchclass content BI,.C, Dib ch by Cr. D, Bh Di, Dw E, Ph ELF, Gb F4, Gh FG, Abb Ghat Ge, A, BL Ab bb An B, Ch gue 1-1‘ sie Conepts and Defiions For now, we will use a “fixed do" notation: The pitch class containing the Cs jg arbitrarily assigned the integer 0 and the rest follows from there. ‘We didn't have to use integers—we could have assigned arbitrary names to each pitch class—but integers are simple to grasp and to manipulate. They are traditional in music (igured-bass numbers, for example) and useful for repre- senting certain musical relationships. We will never do things to the integers jon't have musical significance. We won't divide integers, because, while ing 7 into I] makes numerical sense, dividing G into B doesn’t make much ‘musical sense, Other arithmetical operations, however, will prove musically useful We will, for example, subuact mwnbcis, because, as we will see, subtract gives us a simple way of talking about intervals. Computing the distance between F and 11 by subtracting 7 from 11 makes numerical sense, and the idea of ‘computing the distance between G and B makes musical sense. We will use numbers and arithmetic to model interesting aspects of the music we study. The ‘music itself is not ‘'mathematical” any more than our lives are “mathematical” Just because we count our ages in integers. In this book, we will identify pitch lasses with either traditional letter notation or integers, whichever seems clearest and easiest in a particular context. Mob 12 Every pitch belongs to one of the twelve pitch classes. Going up an octave {adding twelve semitones) or going down an octave (subtracting twelve semitones) ‘will just produce another member of the same pitch class. For example, if we start on the Eb above middle C (a member of pitch class 3) and go up twelve semitones, we end up back on pitch class 3, In other words, in the world of pitch classes, 3 + 12 = 15 = 3. More generally, any number larger than 11 or Smaller than 0 is equivalent to some integer from 0 to II inclusive. To figure ‘ut which one, just add or subtract 12 (or any multiple of 12). Twelve is called the modulus, and our theoretical system frequently will rely upon arithmetic ‘modulo 12, for which mod 12 is an abbreviation, In a mod 12 system, ~12 = 0 12 = 24, and so on. Similarly, ~13, ~1, 23, and 35 are all equivalent to 11 (and to each other) because they are related to 11 (and to each other) by adding or subtracting 12. eis easiest to understand these (and other) mod 12 relationships by envision ing a cireularclockace, like the one in Figure 1-2. a Figuie 1-2 Desie Concepts and Definitions 5 Ina mod 12 system, moving 12 (or a multiple of 12) in either direction only brings you back to your starting point. As a result, we will generally be dealing, only with integers between 0 and 11 inclusive. When we are confronted with a rhumber larger than 11 or smaller than 0, we will usually waite it, by adding fr subtracting 12, as an integer botween 0 and 11. We will sometimes use negative numbers (for example, when we want to suggest the idea of descending), and ‘we will sometimes use numbers larger dan 1 (for example, when discussing the distance between two widely separated pitches), but in general we will discuss, such numbers in terms of their mod 12 equiva INTERVALS Because of enharmonic equivalence, we will no longer need different names for intervals with the same absolute size—for example, diminished fourths and major thieds. In tonal music, such distinctions are functionally crucial; intervals are defined and named according to their tonal function, A third, for example, is an interval that spans three steps of the diatonic scale, while a fourth spans four steps. A major third is consonant while a diminished fourth is dissonant. In music that doesn’t use diatonic scales and doesn’t systematically distinguish be tween consonance and dissonance, it seems cumbersome and even misleading to use traditional interval names. i¢ will be easier and more accurate musically just €0 name intervals according to the number of semitones they contain. The: intervals between C and E and between C and Ps both contain four semitones find are both instances of interval 4, as are BI-FL, C-D, and so on. Figure 1-3 gives some traditional interval names and the’ number of semitones they contain, tratonal name ‘no. of semitones minor 2nd smajr 2, diminished 3d ‘minor 3, augmented 2nd ajo Sy diminished th augmented 3rd, perfect 4h gente th, diminished Sth perfect Sth diminished 6s igmente th, minor 6h tajor th, diminished 2 fgmente 6h, minor 2h majo 7h tino majo rina 10th raj 1 Figue 136 asic Concepts and Defiions PITCH INTERVALS [A pitch interval is simply the distance between two pitches, measured by the num- ‘bef of semitones between them, Sometimes we will be concerned about the direc- tion of the interval, whether ascending or descending. In that case, the number ill be preceded by either a plus sign (to indicate an ascending interval) or a jinus sign (to indicate a descending interval). Intervals with a plus or minus sign are called directed or ordered intervals, AL other times, we will be concerned ‘nly with the absolute space between two pitches. For such unordered intervals, ‘We mill just provide the number of semitones between the pitches. Whether we consider the interval ordered or unordered depends on our Example 1-4 shows the opening melody ‘and identities both 8 ordered and particular analytical interests at the ts From Schoenberg's String Quartet No. ‘unordered pitch interval. ae 4 3 5s 4 +S: a s a ie Fo, ters 2.55.) > = —Sge Ss — ts +. @ te 5 a4 oe el phe ne, Sig et “The ordered pitch intervals focus attention on the contour of the line, its balance of rising and falling motion, The unordered pitch intervals ignore contour and ‘concentrate entirely on the spaces between the pitches. ORDERED PITCH-CLASS INTERVALS A pitch-class interval is the distance between two pitch classes. As with pitch intervals, we will sometimes be concemed with ordered intervals and sometimes with unordered intervals. To calculate pitch-class intervals, it is best to think ‘gain of acircular clockface. We will consider clockwise movement tobe equivalent to movement upward, and counterclockwise movement equivalent to movement downwacd. With this in mind, the ordered interval from C¥ to A, for example, is —4 or +8. In other words, from pitch class CH, one can go either up eight semitones or down four semitones to get to pitch Class A. This is because +8 and —4 are equivalent (mod 12). In general, we like to avoid negative numbers ‘where possible, so we will simply call this interval an 8. To state this as a formula, we can say that the ordered interval between pitch class x and piteh class y is y — x (mod 12). Notice that the ordered pitch class interval between ‘and Cl (1 — 9 = ~8 (mod 12) = 4) is different from that between CF and ‘8 @), since, when discussing ordered pitch-class intervals. order matters. Figure 1-4 calculates some ordered pitch-class intervals using the formula, aie Concepts and Definitions 1 The orderd ptch-clas interval trom CH to Eb is 3~1 from ES to CH is 1-3 fom B to F is S11 fiom D 1 Bh is 10-2= 8 from Bb wo Ch is 1-10 = 3 igure 14 ‘You will probably find it faster just to envision a musical staf, keyboard, oF a clockface. To find the ordered pitch-class interval between C¥ and A, just envision the CF and then count the number of half-steps you will need to go upward (if you are envisioning a staff or keyboard) or clockwise (if you are envisioning a locktace) wo the nearest A. LUNORDERED PITCH-CLASS INTERVALS For unordered pitch-class intervals itno longer matters whether you count upward cor downward. All we care about is the space between two pitch classes. Just count from one pitch class to the other by the shortest available route, either up fof dovin, The formula for unordered pitch-class interval is x ~ y (mod 12) or Y— x (tod 12), whichever is smaller. The unordered pitch-class interval between Ch and A is 4, because 4 (1 ~ 9 = —8 = 4) is smaller than 8 (9 ~ I = 8). Notice that the unordered pitch-class interval between CH and A is the same as that between A and Cl, Its 4 in both cases, since from A to the nearest CF is 4 and from Ci to the nearest A is also 4. Including the unison, 0, there ate only seven different unordered pitch-class intervals, because, to ‘get from one pitch class to any other, one never has to travel farther than six semitones Figure 1-5 calculates some unordered pitch-class intervals using the formula ‘The correct answer is underlined “The unordered pteh-clss interval between CH and EB is 3-1 = 2 or Band Chis 1-3 =10 or Band Fis $—11= 6 oF D and By is 10-2 = 8 oF Bhand Chis 1—10= 3. or Figure 1-5 ‘Again, you will probably find it faster just to envision a clockface, musical ‘if, or keyboard, To find the unordered pitch-class interval between Bb and Fy, for example, just envision a BS and count the number of semitones to the nedxest available PE (4), INTERVAL CLASS ‘An unordered pitch-class interval is aso called an interval class. Just as each piteh-class contains many individual pitches, so each interval class contalns many® ‘ase Concepts and Defiins individual pitch intervals. Because of octave equivalence, compound intervals— intervals larger than an octavo—are considered equivalent to their counterparts within the oetave. Furthermore, intervals larger than 6 (the tritone) are considered ‘cquivalent to their inversions with respect fo the octave. In other words, 0 = 121 = IL, 2 = 10, 3 = 9,4 = 8, 5 = 7, and 6 is its own inversion. (These equivalences are also observed in many aspects of tonal music and tonal theory.) Thus, for example, intervals 23, 13, 11, and 1 are all members of interval class 1. Bure, 1-6 contains a chart showing the seven dilflerent interval cleses and come of the contents of eat interval cl ° L ome jimeerrrg orn mn pitehimervals: 012,26 M113 210,14 3915 4816 5,717 6.8 gure 1-6 We thus have four different ways of talking about intervals: ordered pitch interval, unordered pitch interval, ordered pitch-class interval, and unordered iteh-class interval. If in some pisce we come across the musical figure shown In Example I-5, we can describe it in four different ways. = onde pte tere: 1 cede pital itera: worded pits ners 5S ‘Brample 1-5 Four ways of describing an interval IF we call it a +19, we have described it very specifically, conveying both the sizeof the infeval and is direction. If we call it 19, we express onl its size. It we call ita 7, we have reduced a compound interval to its within-octave equivalent. If'we call it'a 5, we ave expressed the interval in its simplest, tnost abstract way. None of these labels is better or more right than the others its just that some are more conerete and specific while others sre mote general and’ abstract. Which one Ne use will depend on what musical relationship we axe trying to describe. T's like desenbing any object in the workd—what you see depends upon where you stind. Ifyou stand a few inches away from a Painting, for example, ‘you may be aware of the subtest details, right dow to the individual brushstrokes, iF'you stand back a bit, you will be beste able to sce the larger shapes and the ‘overall design. There (s n0 single “ight” place to sland Te. appreciate the punting fully, you have to be willing to move from place to place, One of the Specially nice things. about rousic is that you can hear a single object like: an imerval in many different ways al once. Our different ways of talking about infervals ‘will give us the Mexibilty (0 describe many diferent kinds of musical relationships asic Concepts ond Definitions 9 INTERVAL-CLASS CONTENT ‘The quality of a sonority can be roughly summarized by listing all the intervals i containg. To keep things simple, we will generally take into account only Jnterval classes (unordered pitch-class intervals). The number of interval clases 4 sonority contains depends on the number of distinct pitch classes in the sonority. ‘The more piteh classes, the greater the number of interval classes. Figure 1-7 ‘Suanmatizes the number of interval classes in sonorities of al sizes. (We won't bother including the occurrences of interval class 0, which will always be equal te the number of pitch elascor in the sonority ) no. of pitch classes no. of interval clases 1 0 2 1 5 3 4 6 5 0 6 Is 7 a 5 2B 5 36 0 45 0 58 2 6 Figue 17 For any given sonority, we can summarize the interval content in scoreboard tashion by indicating, In the apprupiate column, the number of occurrences of teach of the six interval classes, again leaving out the oceurrences of interval lass 0. Such a scoreboard conveys the essential sound of a sonority. Example 1-6 refers to the same passage and the same three-note sonority discussed back in Example 1-2, 1) Mani Pian miewalaas ea faals [a tino mcrae OTE YE JO ample 1-6 Interac content of theeote moive (Schoenberg, Piano Pies, Op. 11, No. Dh0 Base Concepts and Definitions Like any three-note sonority,it contains thee intervals, in this ease one occurrence ceach of interval classes 1,'3, and 4 (no 2s, 5s, oF 6s). How different this is from the sonorities prefesred by Stravinsky in the passage from his opera The Rake's Progress shown in Example 1-7! These sonorties contain only 2s and 5s. [iment cines Tia Ts of sera [DTT YO] Example 1-7 Iuerval-class content of » contrasting three-ote mative (Stinky, The Rake’s Progrse, Ae. ‘The difference in sound is clearly suggested by listing the interval-class content of the sonorities. The interval-class content is usually prosented as a string of six numbers with no spaces intervening. This is called an interval vector. ‘The frst number in an interval vector gives the number of occurrences of interval class 1; the second gives the number of occurrences of interval class 2; and so ‘on. The interval vector for the sonority in Example 1-6 is 101100, and the interval vector for the sonority in Example 1~7 is 010020. ‘We can construct a vector like this for sonorities of any size or shape. A ‘ool like the interval vector would not be nearly so necessary for talking about traditional tonal music. There, only a few basic sonorities four kinds of triads and five kinds of seventh chords—are regularly in use. In_post-ional. music, jowever, we will confront a huge variety of musical ideas. ‘The interval vector will give us a convenient way of summarizing their basic sound, ‘Byen though the interval vector is not as necessary x tool for tonal music as for past-tonal music, itean offer an interesting perspective on traditional fora tions. Example 1-8 caiculates the interval vector for the major scale. Basie Concept and Definitions u inenat cass [T]2] [4 [5 eo feccutenes: [2] 4] fo [1 ‘Example 1-8 Interval vector fo the major sae Notice our methodical process of extracting each interval class. Fist, the intervals formed with the first note are extracted, then those formed with the second note, and so on. This ensures that we find all the intervals and don’t overlook ‘ny, ‘As with any seven-note collection, there are 21 intervals in all. Certain intervalic properties of the major scale are immediately apparent from the interval vector. It has only one tritone (fewer than any other interval) and six occurrences of interval-lass 5, which contains the perfect fourth fifth (more than any other interval). This probably only confitas what we already knew about this scale, but the interval vector makes the same kind of information available about less familiar collections. The interval vector of the major scale has another interesting property—it contains a different number of occurrences ‘of each of the interval classes. This is an extremely important and rare property (only three other collections have it) and itis one to which we will return. For tow, the important thing is the idea of describing a sonovity in tasty oF its interval-class content.2 asde Concepts and Deftions BIBLIOGRAPHY “The material presented in Chapter 1 (and in. Chapters 2 and 3 as well) is also ‘Tpcussed in three widely used books: Allen Forte, The Structure of Atonal Music {Now Haven: Yale University Press, 1973); John Rahn, Basic Atonal Theory (New GESK: Longman, 1980), and George Pesle, Serial Composition and Atonaiiy, 3th ca Gbetesley and Los Angcles: Univesity of California Press, 1981). Two important Seen tk offer profound new perspectives on this Basic material, and much ‘doc besides: David Lomin, Generalized Musical Intervals and Transformations {Now Haven: Yale University Press, 1987% and Robert Mortis, Composition with Pick Classes (New Haven: Yale University Press, 1987. EXERCISES 1 Integer Notation: Any pitch can be represented by an integer. In the commonly used Saxed do” notation, C ='0, Cl = 1, D2, and soon, 1. Represent the following melodies as stings of integers 2. Show at lost two ways each ofthe following stings of integers canbe notated ‘on musical staf a. 01392114107856 bi 24124676424212 €.01178312106549 4,1187954 HL. Pitch Class and Mod 12: Pitches that are one or more octaves apart are equivalent members of 2 single pitch class, Because an octave contains twelve semitones, pitch sacs can be discussed using arithmetic modulo 12 (mod 12), in which any integer larger than II or smaller than © can be reduced fo an integer from 0 to LT inclusive Basie Concepts and Defiaions B 1. Using mod 12 arithmetic, reduce each ofthe following integers to an integer fiom 0 t0 1 inclusive: 15 2 a9 13 “3 =10 os 2. Tint Teast tuce integers that are equivalent (mod 12) to each of the following integers: a5 6 10 9 5 4 the following subsrations (mod 12): 29-10 bao e2-10 43-8 IHL, tntervals: Intervals ae identified by the number of semitones they contain 1. For each of the following tational interval names, give the numberof semitones in the interval major third be perfect ith , hugmented sixth 4. diminished seventh f, minor ninth major tenth 2. For each ofthe folowing numbers of semitones, give at east one (ational imerval name: pitch interval is the interval between two pitches, counted in semitones, + indicates an ascending interval; ~ indicates a descending interval 1. Construct the following onlered pitch it (Cote yr leting pin vals on rwsial staff, using mide a 413 br end a3“ ase Concepts and Definitions 2, For te following melodies, idemly the ondered pitch interval formed by each pair of edjacent notes. Fate lp are 'V. Unordered Pitch Intervals: An unordered pitch interval is simply the space between {wo pitches, without regard to the onder (ascending or descending) of he pitches, 1. Construct the following unordered pitch intervals on x masical staff, using middle Cas the lowest note. a. 15 bea e7 au eB 2, For the melodies in Exercise 1V/2, identify the unordered pitch interval formed by each pair of adjacent noses. VL. OnderedPitch-Class Intervals: A pitch-class interval is the interval between two pitt classes. On the piteh-class clackface, always count clockwise frm the rst pitch Clas to the second, 1, Poreach ofthe melodies in Exercise 1V/2, identify the ordered ptch-class interval formed by each pair of adjacent nove 22. Which ordered pitch-cass intervals are formed by the following, ordered pitch VIF, Unondered Pitch-Clas Intervals: An unordered pitch-clase interval isthe shortest stance Between two pitch clases, regardless of the order in which they oceur, To Basle Concepts ond Definitions s calcul an uno pts interval, tke the shots ute 0m he fist i las tothe second, going ether hckvae ocomtcockwise on he picasa, 1, For each ofthe melo in BxreiseIV2, ie th unordered pitch-class interval formed by each pair of adjacent notes. sae 2. An unordered pth-,F member of se cle 5 (0126), ‘The chor! in measure 4 verticalizes that set; meanwhile the same pitch classes ate stated, slowly and melodically, inthe bass. In terms ofthe large-scale statement of 4-17 (0347) discussed above, the C and C} in measures 5 and 6 are notes that embellish a more important Ds. But those embellishing tones themselves have an important local role to play in composing out this form of 4-5 (0126). Like set claes 4-17, set class 4-5 is developed in this music through transposition and inversion, through melodic and harmonic presentation, and through statement ‘over shorter and longer musical spans. Analysis 2 ‘You have probably noticed that our explanations sometimes have a way of overlap ping one another. Consider, for example, the melody note C in measure 4, We Rave described it in at least throe different ways. Its part of a vertical statement fof 4-5 (0126), as shown in Example A2-9; itis also part of a statement of 4~ 17 (0347) in the right hand and a melodic statement of 3-3 (O14). But these interpretations do not contradict each other. ‘The C can function in lots of different ways, depending on how one looks at it, Even better, it can function in lots of different ways simultaneously. A single fone can be, at one and the same time, part of a chord, part of a rogisteal grouping, and part of a melodic line. The Fehness of associations in this music is one of its most attractive qualities, tier networks of motivie association operate in this song, interwoven with the ‘ones we have discussed. For example, the chord in measure 2 is a form of set lass 4-18 (0147). The same set class returns in measure 8 (Fa, A, C, CF) and ‘again at the end of the song. It is also developed elsewhere in the song. As nother example, the first four notes heard in the song—Bh, G, Fa, and P— make up set class 44 (0125). The same set, transposed at ;, that familiar interval of transposition, occurs as a chord in measure 3 and is also developed clsewhere in the song. ‘The fabric of the song is built up from the interweaving of its motive threads, Barték’s String Quartet No. 4 is superficially a very different kind of piece from ‘Schoenberg's Book of the Hanging Gardens. It has much more overt repetition, particularly of small melodic fragments, It has a sharper, more propulsive rhythmic profile. Yet underlying these differences, both works share a pitch organization Based on the manipulation and interaction of pitch-class sets, Listen to a recording of the quartet, concentrating particularly on the first movement. ‘The music for measures 1-13 is given in Example A2-10. Niegro,g-10 Viotint (SEE Viotintt : oe Viole = Wolonceto| ‘continued Example A2-10 Bar, String Quaret No.4, frst movement (im, 1-13) 3_ Analysis 2 <0 ample A2-10 (oned) Let's begin our analytical examination by looking closely atthe music in measures, 5-7 (see Example A2-11). st Analysis 2 Example A2-II. Expansion from 4-1 (012) to 4-21 246), Begin by playing this pasage on the plano, Something striking happens onthe second eighth-note of measure 6, where the chord {C,CI,D,DH] expands to [B1,C,D,E}. The first chord is a member of set class 4-1 (0123) and the second is a member of set class 4-21 (0246). These two set classes, and the idea of ‘expanding or contracting from one to the other, are basic to the way this music ‘e008. Play each of the parts in turn up through the moment of that expansion. The celio enters on Et and descends five semitones to its goal, BS, taking eleven cighth-notes to do so. Then the viola enters, one semitone higher. It descends a shorter distance, four semitones from E to C, and takes a shorter time, nine cighth-notes, to do so, The second violin descends even more rapidly from F to Di(ihren semitones), while the fest violin enters las and covers ite allotted span, two semitones from F¥ to E, the most rapidly of all. ‘The passage thus contains aregistral and shythmie aecelerando as both the intervallc and durational distances get shorter and shorter. Play that much of the passage and listen for the sense of propulsion toward the goal [B},C.D.E}. Now play the rest of the passage in Example A211. In measure 6, each note of the second chord, [Bb,C,D,E}, is embellished with neighbor notes. The lower instruments have upper neighbor notes, while the higher instruments have lower neighbors. This gives a sense of contracting from 4-21 (0246) toward 4-1 (0123), although the instruments never move at exactly the same time. On the downbeat cof measure 7, we find ourselves still on [B4,C,D,E]. At that moment, the cello sates the main melodic motive of the movement. That motive is another member of set class 4-1, The last time we heard that set class was back at the end of measure 5; there, 4-1 was presented as a chord, with C as the lowest note, If wwe designate that as Ty, then the melody in measure 7 is Tyo. ‘The idea of ‘moving downward by (wo semitones—and, more specifically, from C to Bk—is ‘one we will return to. This passage ends with a sforzando statement of [Bb C. Partly because it is stated loudly by all four instruments and then followed by a silence, that chord sounds like & cadentil arrival 35 aeAnalysis 2 “The musi from the pckap to measre 5 though the fist cighth-noe of measure 6 fils the chromate spece between C and Fa. Starting on the seeond eighth note of measure 6, where 4-1 (0123) expaods to 4-21 (0246, te chromatic Space between Bhan Bis smi sacred (Se Example A212) [eee nee es eee eee pe ee ate om BE sample A2-12 Shift fom CFE to BLE, ‘The music in measures 1-13 involves a shift from a focus on C-FI to a focus on BIE, a shift that takes place precisely on the second eighth-note of measure 6, Play the passage in measures 3-7 once again, and listen for this downward ‘The opening measures ofthe piece use set classes 4-1 (0123) and 4-21 (0246) to foes on C-FI. ers lok closely a the fist two measures (oe Example Example A2-13 Meloiic and harmonic statements of sot cass 4-21 (0246) and an cenbelisitg statement of 4-1 0123), 56 Analysis 2 Play just the first three beats of the first measure. The registral extremes there are the C in the cello and the FH in the first violin, At the moment when those extremes are heard (beat 3), the second violin his an E. This seems like an incomplete statement of set class 421 (0246). Let’s designate that form, IC,D.E,FA), as Tp. Now play the first violin part through the third eighth-note ‘of measure 2. That melody seems (0 “compose out” the same set, tracing descent from F to C by way of E and D. Embedded within that melodi¢ statement of 4-21 is an embellishing statement of 4-1: the first four notes in the first violin are [DI,E,F,FF]. Both sets lie within the tritone from C to FF If the beginning of the piece focuses on the tritone C-FF and on forms of set classes 4-1 (0123) and 4-21 (0246) that lie within that span, the ending of the first section of the piece shifts to the tritone BSE and to forms of 4-1 and 4-21 lying within that span. Look, for example, at a passage beginning in the ‘middle of measure 10 (see Example A2-14) Example A2-L4 The contraction of 4-21 (0246) into 4-1 (0123) and its eeexpansion ck into 42 ‘The passage consists of six chords, marked A-F on the example. At a glance, itis clear that chords A and C are identical, as are chords D and F. Now play the chords slowly and you will become aware of a deeper similarity. Because the voices cross, chords A, C, D, and F are all identical in pitch content ((B4,C,D,E) and all are thus members of set class 4-21 (0246). Similarly, chords B and E have the same pitch content, {C,CH,D,Es), and are both members of set class 41 (0123). Notice how the voice erossing, works. From measure 10 t0 measure 11, the first violin and cello exchange parts, as do the second violin and viola. The sets are tamed instrumentally upside-down without changing their identity. As in measures 5-7, the basic idea of this passage seems to be the contraction of 4-21 into 4-1 and its reexpansion back to 4-21. The expansion to 4-21 has the force of an arrival From the beginning of the movement, then, there is a constant interaction of 4-1 (0123) and 4-21 (0246), even as the musical orientation shifts from C-P# to B)-B, This shift is confirmed in the cadence that closes off the passage (see Example A2-15). srAnalysis 2 fe a ample A2-I5 The ral cadene—ihe merging of se clases 4-1 (0123) and 4-21 (O46) within the etane BOE, “There each voice enters in turn, outlining the Tyg form of 4-21 (Bb,C,D,E] ‘When the cello finally brings in its BS, the other Yoices crash back in. Now the entire space between Bb and E has been filed in. The chromatic tetrachord 4-1 (0123) and the whole-tone tetrachord 4-21 (0246) are merged inthis final sonority. ‘The two principal set classes of the passage thus are developed, progress from. cone to the other, define a large-scale shift in pitch location, and ultimately merge into a single cadential sononity. BIBLIOGRAPHY “The eleventh song from Schosnberp’s Book of the Hanging Gardens has been anaiyzed briey by Tom Demske ("Reglstal Cente of Balatwe in Atonal Works by Schoenberg and Webern,” In Theory Only 912-3 (1986), pp. 60-76), and in leat and compelling detail by David Lewin (“Towa the Analysis ofa Scioenbexg Song (Op. 15, No. 11)" Perspectives of New Music 12/1-2 (1973-74), pp. 43= £86). My own discussion is heavily indebted tothe later. Bartsk’s String Quartet No. 4 has been analyzed from many points of view. See Eliott Antokoletz, The Music of Béla Baridk A Study of Tonality and Progression in Twentieth Century Music (Berkeley and Los Angeles: University of California Press, 1984); Milton Babbitt, ‘The String Quartets of Bartok," Musical Quarterly '35(1949), pp. 377-85; George Pere, “Symmetrical Formations in the Sting Quartets ‘of Béla Bart,” Masic Review 16 (1955), pp. 300-312; Roy Travis, Tonal Cober- fee in the Fist Movement of Bastk's' Fourth String Quartet,” Music Forum 2 (1970), pp. 298-371; and Leo Treitler, “Harmonie Procedures inthe Fourth Quart of Béla Bart,” Journal of Muste Theory 3 (1959), pp. 292-98, se CHAPTER 3 Some Additional Relationships COMMON TONES UNDER TRANSPOSITION (7, When a piteh-lass set is transposed or inverted, its content will change entirely, partially, or not at all. Tones held in common’ between two different members Of the same set class can provide an important musical continuity. Conversely, an absence of common tones may emphasize the contrast between two different ‘members of the same set class. ‘When you transpose a pith-clas gt by interval, the number of common tones will be equal to the number of times the interval n occurs inthe set (with ‘one exception, (0 be discussed ates). IFa set contains three occurrences of interval class 2, for example, there will be thee common tones at T oF Ty (see Example 3a), The major scale contains six instances of interval class 5, so there will ie six common tones when the seale is transposed up or down by five semitones (For T). (See Example 3-1b,) Example 3-1 Common tons under tansposton6 Some Additional Relationships ‘Tyil produces four common tones: it maps the 3 and 8 onto each other and the 4'and 7 onto cach other. At T,;L, the set [3,4,7,8] maps onto itself. ‘A simpler way to figure out the number of common tones under Ty is to ‘construct an addition table. Write the set along the vertical and horizontal axes tnd add as indicated. Such an addition table for [3,4,7,8] is shown in Figure 37. elu o 3 Figae 3-7 4 ‘This table neatly peforms all ofthe addons required; it adds each element to teach other element twice and adds each element to itself once. As a result, each ‘ocurence of a number isi the able represents single common tone. The number Ti occurs four tines, so there wil be four common tones at Th the be 3 cay oe il be ran os so ‘any {0 rearrange this information in the form of an index vector, or simply to read it dietly from the table This addition table has another advantage—it shows not only how many tones wil be held n common under Ty bt also which ons. Each index number inthe tuble lies atthe intersection of tones, Those are the tones mapped fnto one another by that index number. Inthe table in Figure 3-7, for example, 10 oscars atthe intersection of 3 and 73 and 7 are thas held in common at Tio, Similarly. 8 occurs in the table tthe intrsetion of 4 with all, 50 4 wl be held in common at Tl. "Appendix 3 Hist index vectors forthe prime-form ofeach set clss the set related by Tal to the prime form, Unlike the interval vector, the index Nector isnot the sain for every member of set clas, Fortunately, once you iow the index vector forthe prime form and for its Tel the index vectors for allthe remaining members can be easly duced from th simple rues given in ‘Appendix 3.‘The interval vectors in Appendix 1 and the index vectors in Appendix 3 ahould enable you rapidly to find the number of common tones any pitch: “ss set wl retain under Ty of Ty for any vals of Common tones under" an! tan be an porn sou Of susie continuity. Example 3-3 contains the fs ten meastres ofthe tid of Weber's Movements for Suing Quartet, Op. 5, a composition that makes intensive we of set clas 3-3 (O14). 7 ae ne ll Tae & is c wo re f . gq a lt P| Gil HS [ ‘Le Gus SS ES |, = nm 0 EE [le aN) i | UST I, I er aay is Lya i 7 3 Bie A RAPE - . 6s wer a fe See Ouuses Op 58 Bangle 3-3 Conon oe lr osc an inveion Wehr, Moneeca Some Addlional Relationships Six pairs of 3-3s are marked on the score. Look fist at the T-telated pairs ‘The (ranspositio levels used, 8 and 11, produce one common tone each, as we know from the interval vector of 3-3: 101100. And notice the special treatment this common fone receives in each ease—it is always retined in exactly the Same repster. The common pitch class is expressed as a common pitch, ‘The same thing is (tue of the TyL-related pair in measure 3. The index vector forthe first set, [8,9,0} is 100012102200; the index vector forthe second Set, [0,34], is 10220121000. You could discover that ether by simply doing fan addition table forthe sets, as discussed earlier, or by looking up the veetors in Appendix 3 (and performing the proper rotations). Both index vectors show that each ofthese sels holds one common tone at Tol. Since they are related by ‘Tol, that rmeans they will share a sinale pitch class. That common tone is C, which i retained here not only in the same register but in the same instrument In measure 9, the two TsL-telated forms of 3-3 share two common tones, C and EL. Notice how Weber arranges these notes to sound together simultaneously He thas uses common tones udder both T, and 7, to create smooth, continuous Yoive leading asthe music progresses among, the members of set class 3-3 (014) DEGREES OF SYMMETRY Sots that are capable of mapping onto themselves (under T, and/or Tyl) have a special place among the pitch-class sets. Appendix 1 makes it easy t0 identify Sch sets. In the fourth column in Appendix 1, you will see for every set class 4 pair of numbers separated by a comma. The first of those numbers measures {tanspositional symmetry it tells the number of transpositional levels at which a set maps onto itself. This number is always at least 1, since every set maps Onto itself at Ty. If the number is larger than 1, that means that the set maps. Onto iself at other transpositional levels as well. The number afier the comma ‘measures inversional symmetry; it tells the number of inversional levels that imap set Ont0 Iself. Many sets cama snap unto themselves under inversion, ‘and thus have a degree of inversional symmetry that is 0. Some sets can map ‘onto themselves at one or more than one inversion levels. The set 3-6 (024), for example, has a degree of symmetry of (I, 1). It maps onto itself at one transpositional level (To) and one inversional level (in this case, Ty). The most sytnmetrial set of all isthe whole-tone scale; it maps onto itself at six transpositional and six inversional levels "The greater the number of operations that map a set onto itself the smaller the number of distinet sets in the set class. Most set classes have'a degree of symmetry of (1,0) and contain 24 distinct sets. For all set classes, dividing the number of self-mapping operations into 24 will give you the number of sets in the set class, Let’s use the prime form of set 4-9 (0167) as an example to see. ‘why this is so, 4-9 has a degree of symmetry of (2,2), The four operations that map it onto itself are Tp, T, Til, and Trl. (This can be figured out by looking fat the interval and index vectors.) Now consider another member of this set class, [1,2,1,8}. This is simultaneously 7, 7, Tal, and Tl of (0,1,6,7]. Each member of the set class can be created by four different operations: Some Addon Reeionships o (0.1.6.7) TooToo THT (wan 175, TH Ty aso Bete DB 8.9. 5s To, Ta, Ty snotty yi Ho (Gono), Ty, Teh TEL But there are only 24 possible operations in all—twelve values of n for Ty, and twelve values of n for Tyl. As a result, 24 divided by the number of operations that will produce each member of the set class equals the number of distinct ‘members ofthe set class. In this case, 24 divided by 4 equals 6. Z-RELATION Any two sets related by transposition or inversion must have the same interval- ‘lass content. The converse, however, is not true. There are several pairs of sels (one pair of tetrachords, three pairs of pentachords, and fifteen paits of hhexachords) that have the same interval-class content, but are not related to cach other by either transposition ot inversion and thus are not members of the same set class. Sets that have the same interval content but are not transpositions or ersions of each other are called Z-related sets, and the relationship between them is the Z-relarion. (The Z doesn’t stand for anything in particular.) Sets in the Z-relation will sound similar because they have the same interval- class content, but they won’t be as closely related to each other as sets that are ‘members of the same set class. If the members of a set class are like siblings within a tightly knit nuclear family, then Z-related sets are like first cousins. In Example 3-4, an excerpt from the thi of Stravinsky's Pieces for String Quartet, the occurrences of 4-215 (0146) in measures 24-26 are strongly linked to the similar occurrences of 4-229 (0137) in measures 27-28. 4-215 0146) Tutti sul tasto, a Violin Violin 1 cat (continued) ample 34 The Zrelation (Stravinsky, Meces for Sting Qua)o Some Additonal Relationship 4-229 013 "uth tioee ane archon tule se fonguenr —— 3 Tz Sat 8 ‘Puaveel xample 3-4 (continued) In Example 3-5, the familiar beginning of the frst of Schoenberg's Piano Pieces, Op. Il, there is a similar strong, connection between the opening six note melody and the left-hand accompanimental figure that follows. 6-210 ons Mabige 4 Pas 6-239 (23458) ‘Example 3-5 The Ztlation (Schoenberg, Pano Piece, Op. 11, No.1) Any set with a Z in its name has a Z-correspondent, another set with a different prime form but the same interval vector. On the setlist in Appendix 1, the Zrclated hexachords are listed across from one another, but you will have to look through the list for the Z-related sets of other sizes. (COMPLEMENT RELATION For any set, the pitch clases it excludes constitute its complement. The complement of the set [3,6,7], for example, is [8,9,10,11,0,1,2,4,5]. Any set and its comple ment, taken together, will contain all twelve pitch classes. For any set containing fr elements, its complement will contain 12 ~ n elements Some Additonal Relationships @ “There isan important intervallic similarity between a set and its complement. You might think that whatever intervals a set has lots of, its complement will have few of, and vice versa. It turns out, however, that a set and its complement always have a similar distribution of intervals. For complementary sets, the differ nce in the number of occurrences of each interval is equal to the’ difference between the size of the sets (except for the tritone, in which case the former will be half the latter). Ifa tetrachord has the interval vector 021030, its eight hote complement will have the vector 465472. The eight-note set has four more ‘of everything (except for the tritone, of which it has two more). ‘The larger set is like an expanded version of its smaller complement. “This inervallic relationship holds even ifthe wo sets are not literally comple- iments of each other, 80 long as they are members of complement-related set Classes. For example, [0,1,2} and [0,1,2,3;4,5,6,7,8] are not literal complements of each other. Infact, all the members of the first set are contained in the second. However, they are members of complement-elated set classes and thus have a similar distribution of intervals. Complement-related sets do not have as much in common as transpositionally or inversionally related sets, but they do have a similar sound because of the similarity of their interval content. “The complement relation is particularly important in any music in which the twelve pitch classes ate circulating relaively freely and in which the aggregate (@ collection containing all (welve pitch classes) is an important structural unit Consider the relatively common situation atthe beginning of Schoenberg's String Quartet No. 3, where a melody (here divided between frst violin and cello) is secompanid yan sina tht consns a the pth clases excluded by the melody. Moderato 4-100 ‘i Violin 1 Violin 1 Viola cello Pigs Toned) Example 36 Complementary ses i melody an accompaniment Schocnberg, Sting ‘Quartet No.3}.Example 3-6 (coubued) ‘The melody and the accompaniment have a similar sound because they contain 4 similar distribution of intervals. he final four-note chord of the second of Schoenberg's Litle Piano Pieces, Op. 19, is a form of 4-19 (0148), a set prominent throughout that piece and commoa in much of Schoenberg's musi. 4-19 148) 19 wrRIs6e) Example 3-7 The complement relation (Schoenberg, Litle Pano Pie, Op. 12, No.2). ‘The last eight notes of the piece (which, of course, include that final four-note chord) are a form of 8-19 (01245688), ‘he complementary sct class. Compare the interval vectors of these two sets: the vector for 4-19 is 101310 and the vector for 8-19 is 545752. Both sets are particularly rich in interval class 4. In fact, no four- or eight-note set contains more 4s than these do. And notice how prominently the 4s are featured in the music. Because of the complement relation, the final four-note chord sounds similar to the larger eight-note collection of whieh itis a part. ‘The listof setsin Appendix 1 is arranged to make it easy tose the complement relation. Complementary set classes are always ised right across from one another. If you look up 4-19 (0148) and $19 (01245689), you will see that this is $0. ‘As a futher aid, the names of complementary sets always have the same number following the dash. Thus, 4-19 and 8-19 are complements of each other, a8 are Some Additional Relationships n 3-6 and 9-6, 5-Z12 and 7-Z12, and so on, These features of the list make it Yery easy to look up large ses." Say you have a nine-note set that you want to find on the list. You could put it in prime form and look it up, but that would bo a time-consuming operation since the set is so big. It is far easier to take the three notes excluded by the ninc-note set and put them in the prime form, then look that trichord up on the list—the prime form of the original nine-note sot will be ditectly across from it. ‘You may notice that there are some sets, exclusively hexachords, that have nothing written across from them. Hexachords like that are self-complementary-- ‘hey and their complements are’ members of the same sot class. For a. simple example, consider the hexachord (2,3,4,5,6,7]. Its complement is (8,9,10,11,0-1] But both of these sets are members of set class 6-1 (0123456) In ther words, self-complementary hexachords are those that can map onto their complements under either T, oF Ty If a hexachord is not self-complementary, then it must be Z-related to its ‘complement. Remember that, with complementary sets, the difference in the ‘number of occurrences of any interval is equal to the difference in the size of the two sets. But a hexachord is exactly the same size as its complement. As a result, a hexachord always has exactly the same interval content as its complement. Its also related to its complement by T,, or T,f, then it is self-complementary. If not, then it is Zerelated to its complement. The hexachords on the list are thus either written with nothing across from them or they are written across from theit Z-comrespondents. This intervalic relationship between complementary hhexachords is extremely important for twelve-tone music, and we will discuss it further in subsequent chapters, ‘SUBSET AND SUPERSET RELATIONS If set X is included in set Y, then X is a subset of Y and Y is a superset of X A set of 1 elements will contain 2° (2 to the nth power) subsets. five-note Set, for example, will contain the following subsets: the null set (a set containing no elements), five one-note ses, ten two-note sets (these are also called intervals), fen three-note sels, five four-note sets, and one five-note set (the original set itself). That makes a total of 2° (2 to’the Sth power) of 32. subsets. ‘The null Set, the one-note sets, and the set itself will usually not be of particular interest as Subsets. Even so, that still Ieaves lots of subsets to be considered (2" — (n+ 2)), and naturally, the bigger the set the more numerous the subsets In onder not to be overwhelmed by the possibilities, there ate two things to bear in mind. First, some of the subsets may be members of the same set lass. Consider set class 4-25 (0268), for example, which is something of an extreme case. As Figure 3-8 shows, al ofits three-note subsets are members of the same set class, 3-8 (026). “The set Its subsets ‘Their setnames and prime forms (2.6.8) 38 02 16.8.0) 3-5 2s) 425 2605 8/02] 33.25) 10,256) 3-8 020) Figure 3-8 i eC aTn Some Addiiowal Relationships Most set classes are not as restricted in their subset content as this one, but there is often some redundancy. The final eight-note sonority of the second of ‘Schoenberg's Litle Piano Pieces, Op. 19, (discussed above) contains two promi- nent statements of 4-19 (0148). (See Example 3-8.) ws we “9 Example 3-8 The Sal eghtnoe collection, set clas $-19 (01245689), contains two ‘rominentsltements of 419 (0148), ss complement Actually, no fewer than cight ofthe four-note subsets of 8-19 (01245689) are members of sct class 4-19 (0148). Most of these, however, are not musically featured by Schoenberg inthis passage "That brings us 10 the second important limitation on the otherwise vast world of subsets and supersets: only a small number will be musically significant In any specific musical context. Like any eight-note set, the final sonority of ‘Schoenberg's Little iano Piece contains 70 four-note subsets. Only a small nomber ‘of those can be heard as meaningful musical groupings, identified by shared register of articulation. For example, it makes no musical sense to combine the Gin the middle register with the op three notesF¥, Bb, Deven though that Combination ereates smother for of set clas 419" (148). Those four notes Simply don't belong together musically With the same final cight-note sonority, Scwoenberg could have grouped G, F¥, Bb, and D together, but chose not to Similarly, he could have revoiced the Sonority to emphasize subsets that were members of st classes other than 419, but again he chose not to. The subsets fof a set are a Kind of absiact musical potential the composer chooses which to emphasize and which to repress “There is atleast a weak family resemblance between a set and its subsets ‘or supersets. These ‘inclusion relations are not as strong as many of the relation- Ships discussed earlier, like the Z-rlation or the complement relation, but can sill be musicaly interesting. Smallereollections can frequently be heard combining into larger ones and larger collections dividing into smaller ones. LARGE-SCALE COMPOSING-OUT “Most of the piteh-class sets we have discussed have been confined to short durations within given works. ‘They have usually consisted of notes that were directly adjacent in the mnsic—as a single melody. as a chord. or as a combination of the two. But pitch-class sets can also extend over large stretches of music. In fact, one of the most potent means of assuring large-scale coherence: in_ post tonal music involves projecting pitch-class sets over large musical spans. There are many different ways of projecting a pitch-class set over a phase or section Some Additonal Relationships n or entire piece. All that is necessary is that the notes that make up the set, although separated in time, be associated in some musical way. Imagine a piece in which 3-3 (014) isa principal surlace motive. If we were composing such a piece, there would be many ways of “composing out” ‘hat motive. We could isolate three notes as regstal high poins oF Tow points and make sure that these made up a form of 3-3. We could transpose a distinctive bit of musical material in such & way tha it occured fist at Tp, then at Ty and finally at‘T, (or following some similar transpositonal scheme). We eould have some paticular instrument play only three carefully chosen notes over the course of passage, thus creating a timbraly distinct form of 3-3. In short, any ways that Wwe can imagine of associating nonadjacent notes can be used To compose ‘ut @ motive ora piteh-lass set, [Let's look at two very different works that both project a form of set class 3-3 (O14) over a large, musical span. The first example isthe first section ofthe first of Schoenberg's Piano Pieces, Op. 11, a passage we have already looked at several times. We noted previously the extent to which st class 33 pervades the musical suriace. Example 3-9 shows two large-scale statements of the same set elass, one inthe upper voice and one in the ass == * Cr Piano ‘Example 3-9 Large-scale statements, in soprano at bas, of set cass 3-8 (O14) (Sehoenbrg, Phin Piece, Op 11, No.‘Some Adtonal Relationships, [As we observed in Chapter 2, the three melodic high points, B-G-GE, constitute a large-scale statement of set class 3-3. These three pitches, widely separated in time, are associated by their shared register and contour position. Furthermore, these afe the same three pitches with which the piece began. ‘A similar thing happens in the bass. The left-hand part begins with two ‘chords (measures 2-3); the bass notes are Gb and Bb. After contrasting material, two more chords aro heard at the end of the section (measures 10-11); the bass notes now are G+ and G. That final G completes a large-scale statement of fanother form of set class 3-3. Those three pitches, GI-BI-G, are associated by their shared register and articulation. Like the large-scale melodic statement, this large-scale bass statement draws together and unifies this section of music: ‘Stravinsky's Oedipus Rex composes out set class 3-3 (014) in a truly monu- menial Way. This work has three widely separated siateiwents uf wht lias besu called a “Fate motive.” This motive occurs at the beginning of the piece with BB in the bass, in the middle of the piece with B in the bass, and at the end of the piece, some forty minutes later, with G in the bass (sce Example 3-10). 8 eampe 10 Ainge of etches? (0) pean nie wrk Srv a Oedipus Rex). —_—_hn— Some Aditional Relationships 1s This large-scale statement of 3-3 (014) mito countess small-scale statements of the same set class. Twentieth-century composers have frequently used this kindof large-scale statement to assure the coherence of the deper structural levels and to integrate them with the musical surface. BIBLIOGRAPHY Common tones unde transposon inversion ae dcssedin Foe, The Sctre of front Mai, pp. 39°40, day ste Aone Theory. 94-113 eben Wa identi of aiden ko dc conn ca oa inversion in ht review of Ras book In Mate Theo Spent (982) [BE-35 and gun Hove Conpation mth Pach Clases Fors nore pe canon of sof hs KS Bo Algonce, he Ivarence atx RD. dissertation, Yale University, 1974). ee ‘The Zetclationship was rt described by David Lewin in “The Intervalic Content of a Collection of Notes,” Journal af Music Theory 4 (1960), pp. 98-101. The tse ofthe abel "72" to refer this elatonship fs coinage of Alen Forte's (ee ‘The Structure of Atonal Music, pp. 21-24). sit ‘The intervallic relationship of complementary sols was first discovered by Milton Babbitt with regard to hexachords, Generalizing this relationship to seis of other sacs was the work of Babbitt and Lewin (ee Lewin’s "The Interallie Content ‘of @ Collection of Notes"). Babbit discusses the development of his theorem about hhexachords and its subsequent generalization in Milion Babbitt: Wore About Music, 4. Stephen Dembski and Joseph N. Straus (Madison: University of Wisconsin Press, 1987), pp. 108-0.Analysis 3 ‘The idea of transposition by +5, of course, is associated with the relentless ascending 5s of the bass part. Furthermore, set class 3-3 (014), particularly with the pitch classes C, Eb, and Fs, occurs in important places ‘elsewhere in the song, The singer sings those very notes in measure 9, and the most heavily stressed otes in the vocal part—the Fb in measure 9, the Eb in measure 11, And the C in measure 15—compose out that set over a large span. The contrapuntal complex of the opening—a series of statements of 425 (0268) where the bass moves by *+5 and the upper parts by 1, with the singer weaving in and out in 4 motivically associated Way-—is typical of the song as a whole and, to some extent, of the other songs of Up. 2 a8 wel BIBLIOGRAPHY Webern's Movement for String Quartst, Op. 5, No. 4 has boon widely discussed dna analyzed. Sex Charles Burkhart, ibe Synmenicat Source of Webern's Opts SiN. 4 Music Forum 5 (1980), pp. 317-34, David Beach, “Pin Strulue Sha the Anale Process n Alona Bsc: An inrodtion to te Theory of Sets, ‘Music Theory Spectrom (1979), pp. 7-22; George Pere, Serial Composition dnd Arona, Sthed pp. 16-19, Dav Lewin, "An Example of Serial Techigue in Barly Webern.” Theory and Practice 7/1 (1982), pp. 40-43; Allen Fonte, “A ‘Thom of Set Compencs for Music,” Journal of Music Theory 72 (1964), pp TPT abn Howe Som: Caan Preis of Pek Sus Perspectives of New Muse 4/1 1968), pp. 57-59, Allen Fore, "Aspects of Rhytim si osers Anat Matis" Me Treory Spectra 20980), fy, 90-108, and Benjamin Borer, **Mta Variations, Part IV: Analytic Fallot (Dy Perspectives of New Music Nit (1972), pp. 217-23 6m Bers song, see George Perle, “Berg's Master Array of the Interval ete, Musical Quarterly 631, (197), pp. I-30; Crug. Ayrey, "Berg's “Scheidewes “Analycl sues in Op. 2, Mase Analysis 172 (1982), pp. 189-202; and Douglas Samm, "Alban Berg: The Origins of a Method,” Muste Analis 6/3 (1987), pp Za, a8 ——p>— soins CHAPTER 4 Centricity and Some Important Referential Collections TONALITY AnD cENTRICITY ‘music seems to invite the use of traditional tonal a by Stravinsky, Bart, Berg, and even Schoenberg has a ‘atleast in certain passages. Bt on closer inspection, we generally find that tonal theary hee litle to tll us shout most twentth-century music When twentieth-centiry composers create a tonal sound, they usually do so by using noatonal means. For a pioce to be tonal, it must have two things: functional harmony and traditional voice leading. Funetional harmony refers to things like dominans, subdominant, and tonics, and to the general idea that different harmonies have specific, consistent roles to play in relation to each other. Functional harmony is the bass of analysis using Koran numerals. When we appiy the Roman numeral V to some harmony, for example, we are saying that it functions as a dominant, defining and leading to & tonic, ‘Traditional voice leading is based on certain ‘well-known norms of dissonance treatment. The triad and its intervals (thirds, ftths, sists, and usualy fourths) are consonant. Other sonorities and other intervals ae dissonant: they tend to resolve to more consonant sonorites and interval. ‘There are other aspects of tonality, but these are probably the most fundamental If music does not make use of functional harmony oF traditional voice leading, ‘we will find that our basic tools of tonal analysis, Tike Roman numerals, simply ‘will not apply. To appreciate post-tonal music most fully, we must learn to approach it‘on its own terms, Father than drag it onto the Procrustean bed of (O0F (onal assumptions Because a piece is not tonal, however, does not mean it eat have pitch or pite-class centers, All tonal music is centic, focused on specific pitch classesat (d-51) 4, 4, 4 lit 4 4 ae tr HH is. aare alle [< LN, al : a le | me Hy ir = Pi aR i 3 alfh lh off i ti ey te Ee a) ik: 3 PSS [PERE || A i HEE Sehr bewe; ory 3 vi ‘Violin 11 Viola| Centricity and Some Important ReferenalColtectons ey or triads, but not all centric music is tonal. Even without the resources of tonality, ‘music can be organized around referential centers. A great deal of post-tonai music focuses on specific pitches, pitch classes, or piteh-class sets as a way of shaping and organizing the music. In the absence of functional harmony and traditional voice leading, composers use a variety of contextual means of reinforee- ‘meat. In the most general sense, notes that are stated frequently, sustained at Jength, placed in a registal extreme, played loudly, and rhythmically or metrically stressed tend to have priority over notes that don’t have those attributes. For a simple example, consider the opening of the thied of Webern's Move- ments for String Quartet, Op. 5 (see Example 4-1). A CF pedal runs through the passage. By brute repetition, the Ch is established as an important pitch center in the passage, We inevitably hear the other events in the passage in ‘elation it. The CH receives special treatment throughout the piece and fs the last note of the piece, played in octaves, wriple-forte, by all four instruments ‘Though the piece is by no stretch of the imagination in Ci major or CE minor, the CH certainly has a centric Function, It is also possible for a pitch-class set, or even a set class, to act as a referential center if itis sufficiently reinforced. Consider the concluding measure of the first movement of Bartok’s String Quartet No. 2, shown in Example 4_2 i” vn, P Yn Va = APR, cal] Pee ‘Bxample 42 Set class 4-19 (0148), paticulnty HEPA, has 2 conte fection (Gat, String Quart No 2). The arrows inicate occurrences of se class 4-19 (0148). That et class, particularly with the piteh-class content [CJ,E,F,A\ plays a crucial role throughout the move. ‘ment and forms is final sonority. (The notes inthe upper three voices are assumed to reverberate until the very end). ‘The movement is not in a key and is not tonal, but the focus on particular sets and pitch classes shapes its structure. Triads play an interesting role in post-onal music, They no longer act as tonies or dominants: rather, they are usually used to reinforce and support a pitch or pitch-class center by retaining their traditional sense of consonance and fepose. The firs! theme af Stravinsky's Symphony in C io not traditionally tonal but uses triads as its basic harmonies (see Example 4-5). a‘ a Centricity and Some Important Referential Collections 1.90 ge gia ah vot 4] vat 2 va LB rweceo) o ov 4 viata iat 2 Yiote va Example 43 A cenvc confit between C and B, and between C-E-G and E-G-B Garvie, Symphony in). Centricity and Sore Important Referential Collections 93 ‘The accompaniment consists only of repeated Es and Gs. (Static, nonprogressive barmony like this I of Stravinsky.) That accompaniment could support either B or C in the melody. Both B and C do occur frequently in the melody, ‘that it is hard to fell the relationship between them. Which ? If the B leads to C, then the prevailing harmony is the C- ‘major triad, supporting a notion of C-centricity. IF the C is an upper neighbor to B, then the prevailing harmony is the E-minor triad, supporting a notion of E-centrieity. The accompaniment is ambiguous—it supports either interpretation, ‘An ambiguity between C-centrcity and E-centricity is crucial to this passage, ‘and to the entire movement. “The basic tools of tonal analysis don't work well in the passage or elsewhere in this piece. We can't meaningfully apply Roman numerals, and we can’t clearly describe the voice leading in taditional terms. Though the triad is the basic, ‘consonant harmonic unit in this piece, it supports centric ambiguity, not traditional tonal harmony. Stravinsky reflects this awarness wher he calls the piece Symphony in C rather than Symphony in C Major, Tc is not, in the deepest sense, a tonal piece. Vestigial, surface references io functional harmony and traditions) voice feading are common in the music of Stravinsky and other twentith-century compos ers. But the deeper levels of structure are rarely tonal. If we wish to analyze and understand such music, our traditional tonal analysis simply won't be of ‘much use. ‘Discussing pitch centricity in post-tonal music is mote complicated than identifying the tonic of a tonal piece. In post-tonal music, we can talk about an ‘entire spectrum of centric effects. At one extreme, represented by much twelve- fone music, there is litle or no sense of centrcity, Even so, of course, the pitch classes are not treated identically, and it is important to be sensitive to ny kind of special treatment accor! 'ta pitch classes or pitch-elass sets, At the other extreme, many posttonal pieces are deeply preoccupied by questions of centricity.. THE DIATONIC COLLECTION ‘A sense of centrcity often emerges from the use of stable, referential collections. Composers often use certain large sets as sources of pitch material. By drawing, all or most of the smaller sets ftom a single large referential set, composers can nity entire sections of music, particularly if the referential set is associated ‘with a specific pitch or pitch-class center. By changing the large referential set andor the piteh of pitch-class center, the composer can ereate a sense of large~ Scale movement from one harmonic area t0 another. The diatonic, octatonic, ‘and whole-tone collections are probably the most important large referential collec tions in post-tonal music, but there are others as well. ‘The diatonic collection is any transposition or inversion of the seven ‘white notes” ofthe piano. It is set class 7-35 (0135681). This collection is, of course, the basie referential source for all of Western tonal music. A «ypival wunal piece begins within one diatonic collection, moves through other transposed diatonic collections, then ends where it began. All the major scales, (natural) minor scales, land church modes are diatonic collections, Diatonic collections are also common in carly twentieth-century music, Large stretches of music by Stravinsky andom” Centcity and Some Important Referenial Coltectons others can be referred to one oF more diatonic collections. In post-tonal music, hhowever, the diatonic collection is used without the functional harmony and {raditional voice leading of tonal music. . Example 4-4 illustrates nonfunctional, static diatonicism in Stravinsky's Petrashka. Altogrowfusto, J. Fomre~ Aligerogiysta dat Centricity and Some tmportantReferetial Collections 9s Although the centriity of the passage is clear (on G or the first eight measures, then shifting to A), itis not traditionally tonal—just try analyzing it with Roman numerals! It does, however, use only diatonic collections. These collections define Gistinet harmonic areas. In the first cight measures, only the "*white notes” are Used. In measure 9, F¥is replaced by FF, resulting in a different diatonic collection, 4 transposition of the first. With the change in collection, we have a sense of large-scale shift from one area to another. ‘The change coincides with a change in centricity, creating a clear musical articulation ‘The diatonic collection provides a strong link to earlier music, but it acts in a new way, as primarily a referential source collection from which surface ‘motives are drawn. Tn tonal musi, the diatonic collection is usually divided up (partitioned) vertically into triads. In post-tonal diatonic musie, triade are aloo used but other harmonies also occur. For example, 4-23 (0257) and 3-9 (027) are diatonic subsets that occur in tonal music only infrequently and as dissonant by-products of voice leading, In Stravinsky's diatonic music, however, they are Particularly common. Example 4-5 shows a diatonic passage from the beginning of Stravinsky's opera The Rake's Progress. [Example 44 Two diatonic collation (Stavnsky, Perak, Rasian Date) Example 4-5 Static, monic diatonic music (Stravinsky, The Rake’s Progress As we observed back in Chapter 1 (Example 1-7), on virtually every beat in this passage one finds either A-B-E or A-D-E, two forms of set class. 3-9 (021). Togethers, they form set class 4-23 (0257), a favorite of Stravinsky's The passage is clearly centered on A and on the perfect ith A-E. But Stavineky fills in that fifth not with the traditional thitd but with scconds and fourths, creating the sonorities most characteristic of his own music. The music is diatonic. but itis nether triadic nor tonal. Rather, the collection acts as a kind of harmonic field from which smaller musical shapes are drawn,% Centricity and Sone Important Referenal Collections We need a terminology for posttonal diatonic music, one that will permit 1s to talk about its uses without confusing the issue with al kinds of inappropriate ‘assumptions. To call the collections by their traditional tonal names, Tike C major or Fl minor, seems wrong. Those names imply the entire system of major/minor tonality, most of whieh simply doesn’t apply to posttonal music. Modal terminol~ ogy is @ possibility —we could call the collections in the passage from Petrushka in Example 4-4 Mixolydian (on G) and Dorian (oh A)—but also not entirely appropriate. Those familiar modal names have all kinds of traditional associations (with certain types of melodies and moods, for example) that are not appropriate for the murie We want to diseuss "The solution is a neutral terminology that simply identifies the principal tone (if one can be ascertained) and the kind of diatonic collection built upon it ‘There are seven different diatonic scales, one beginning on each note of the collection. Think of the seven white notes of the piano; you will get a different patter of intervals beginning on each of them. We will call these patterns “C- Seale,” ““D-scale,"” ““E-scale,”” ""F-scale,"* ““G-scale,"* “‘A-seale,"” and “B- Scale.”” Bach of these patterns can be transposed to ‘any of the twelve. piteh Classes. For example, one can speak of an E-scale on G. This is a scale that has the same pattern of intervals you get when you play the white notes from to E, but it begins on G. In the passage from Perrushka in Example 44, the ‘music begins with the G-seale on G then shifts to the D-scale on A. The change jn collection and the change of pitch center both signal a larger formal division in the music. THE DIATONIC OCTAD “Another familar referential collection in Stavinsky's music is the diatonic octad. ‘You can think of this collection as resulting from the union of any two diatonic collections that share six pitch classes. For example, the diatonic octad results from combining C-scales on C and G or those on C and F, or combining the C- seale on C with the A-scale on E. The name of the set class is 8-23 (0123578T). ‘fien this collection is associated not with a single pitch-class center but rather with an ambiguous clash of competing centers, usually centers that Tie, pitc Class interval 4 apart. We already glimpsed one instance of this in Stravit ‘Symphony in C (Example 4-3). There, the conflict between C and B as centers js often worked ont within a diatonic detad that combines the C-scale on C with the A-seale on E, “The beginning of Stravinsky's Serenade in A, a similar example, uses the liotonie octad that results from the union of the C-scale on F with the A-scale ‘on A (see Example 4-6). Within that collection, the F and A compete as centers, 1s do the F-major and A-minor triads. Look at the sonority formed on the downbeat ff each of the first five measures. I has A in the outer voices, but F is always. present in an inner voice. The (iad is F major, but the A. is the most prominent fone. On the last beat of measure 5, the right hand arrives on an A-minor triad, but beneath it the bass insists on F. That sonority combines triads on A and F just as the larger diatonic oetad combines diatonic collections on A and F. Conticty and Some Important Referenal Collections ” aL ample 46 A dntonie ocd that combines tbe Ascale on A withthe Cseale on F (Sirvinsy, Serenade in. THE OCTATONIC COLLECTION ‘The octatonic collection has been another post-tonal favorite, particularly in the imusi¢ of Bartok and Stravinsky. This collection, 8-28 (01346797), has many Gistinetive features, First, itis highly symmetrical, both transpositionally and inversionally. It maps onto itself at four levels of transposition and four levels of inversion, As a result, it has only three distinct forms (just like its complement, the diminished-seventh chord). Figure 4-1 shows the three octatonic collections. lection 1 (1,2.4,5;7.8,10,11] Collection? (23,5,6,89.11.0] Collction3—10,13,4,6,7,9,10) Figure #1 ‘When writen out ao a scale, the oetatonie collection consists of alternating Isand 2s (unlike the diatonic scale, where 2s predominate and the Is are asyrnmetti-oy Cemticiy and Some Important Referemil Collections cally placed). It can actually be written in two different ways, either begioning with a 1 and alternating 1 I-2-I or beginning with a'2 and alternating 2-1-2-1-2-1-2. Somewhat different sorts of musical possibilities flow from these two different orderings. The fist ondering tends (o feature tritone-related forms of set class 43 (0134) (with additional forms of the same set class beginning ‘on the third and seventh degrees of the scale). ‘The second ordering tends to feature tritone-telated forms of set class 4-10 (0235) (with additional forms of the same set class beginning on the third and seventh degrecs of the scale) Figure 4-2 shows these different possibilities with rogard to the two different orderings of Collection 1 any 4s. 4410 0235) 10 se —— ——, CHD EF G Ab BB Chm DEF GAB BS Gm © o as 10 #10) Figwed-2 Example 4-7 contains « complete short piece by Bart6k, one that features the second of the orderings. callstion2 pon ma, date 5 colletion Barth, Mitroloaos No. 101, “Diminished Conticty and Some Important Referetial Collections ” tion = = =a Raa: Collection 255 Cotetion 3 ctecton 2 leu 5 ray Pa fet For the most part, this piece has « 4-10 (0235) in one hand sounding against a tcitone-related form of 4-10 in the other. The larger collections are thus octatonic. ‘The piece as a whole uses all three forms of the collection. It begins and ends with Collection 2 and moves through all three collections in a symmettical layout (Gee Figure 43) SS ‘The symmetry of this plan is reinforced in many ways. For example, the statement of Collection 1, which lies at the center of the work, contains the melodic high points in both voices. The large-scale harmonic organization of the piece is deter ‘ined by the motion from collection to collection. In subset structure, the octatonic collection, because of its preponderance of trtones, is generally quite dissimilar from the diatonic collection. ‘The octatonie collection, however, is surprisingly rich in triads as subsets. [t contains no fewer than eight occurrences of set class 3-11 (037), the set class of the major and Minor trad. AS a result, many octatonic pieces have a triadic, tonal-sounding, surface. Furthermore, triads constitute a point of intersection between the diatonic and octatonie eallections in pieces that uso both. Many picees involve inteiae ti between intersecting octatonic and diatonic collections a( 100 Centricity and Some Important Reforetal Collections ‘The classical case is the beginning of Stravinsky's Symphony of Psalms ‘That piece makes extensive use of an octatonic collection that begins on E. and follows the first ordering. Sot class 4-3 (0134), featured in this ordering, was described by Stravinsky as the basic idea for the entire work. He referred t0 that set class as “two minor thirds joined by a major third."* The famous opening. chord, known as the “Psalms chord,” is immediately followed by music drawn from an oetatonie collection that contains it (see Example 4-8). af wiparper. af Exanple 4-8 The "Psalms chord” boa a a subse of an otatone collection on E Atrehearsal no. 2, the chord is stated again (for the fourth time). Now, however, the chord is followed by music drawn from a diatonic collection (B-scale on EB) that contains it (see Example 4-5). Example 4-9 The "Pals chord” beard asa sabet ofthe Esale on E. ‘The chord is an element common to the octatonic collection beginning on E and to the E-scale on E. It links the contrasting types of music in this movement, The large-scale harmonie organization of the movement involves contrasting octs- tonic and diatonic sections, with important interactions and Tinks between the wo. ‘Because of the extreme symmeity ofthe velatoic collection, itoften produce ‘acentrc conflict. Consider, for example, the position of triads within the collection. In the first of the orderings, major and minor triads can be constructed on the first, third, fifth, and seventh degrees of the scale (see Example 4-10) Centcty and Some Inporten Referenial Collections 10 Example 4-10 The trade escurces ofthe ecto eoletion. ‘Because the trad can be used to reinforce pith classes, this symmetrical disposition frequently results in a static polarity of competing centers. Sometimes tritone- related pitch classes are poised against one another, competing for priority. Some~ times, asin the first movement of the Symphony of Psalms, the competing centers tre piteh-class interval 3 apart, In that movement, E and G compete for centric priority. Their competition can be heard even in the first chord. That chord has Bin the bass, but G is the note that is most heavily doubled. A tension between and G continues throughout the movement, with G winning out in the end, ‘This kind of centric polarity is typical of octatonic music. Large collections other than the octatonic and diatonic may also be. found in posttonal music, The whole-tone collection, set class 6-32 (024687), is a favorite of Debussy and can also be found, more subtly, in some early works by Schoenberg. We also discussed its use in a song by Berg (see Analysis Three). ‘The whole tone collection is very distinctive. It has the highest possible degree of symmetry, both transpositional and inversional, and its set class contains only fowo distinct members. Its intervallic structure and its subset structure are predictably redundant. Different large collections such as these may interact and interpenetrate ver the course of a passage or a piece. In analyzing posttonal music, one ust be sensitive not only to the motivie interplay of the surface, but to the gor referential collections that lurk beneath the surface. INVERSTONAL AXIS In most of the examples we have discussed, centricity is established by various Kinds of direet emphasis and reinforcement: centric pitches are usually stated longer, louder, oftener, and higher (or lower) than noncentric pitches. Centricity in post-onal music can also be based on inversional symmetry. Inversionally symmetrical sets—those that can map onto themselves under inversion—are com ‘mon in post-tonal music. Such sets are symmetrical around one of more axes of Symmetry. An axis is that point in a set around which all the pitch classes balance symmetrically. An axis of symmetry frequently functions as a piteh or pitch- class center. Consider (3,5,6,8], a member of set class 4-10 (0235). The set list in “Appendix I tells us that it is inversionally symmetrical, and its symmetry is really apparent when we inspect it (see Figure 4-4). AA; Figure 4 A x oO is2 Conticity and Some Important ReferentialColletons {ts interval succession is the same reading from left to right as from right to left. Every inversionally symmetrical set can be ordered in this way, so that the interval succession js its own retograde. Usually, but not always, this ordering. will be the normal form, The axis of symmetry is’also apparent: the set balances around a point between the 5 and the 6. Actually, because this is a pitch-class system, it also balances around a point between {i and 0. "To appreciate this, it may be best to envision the pitch classes around a clockface (see Figure 4-5), + Qo- Figue ds The pitch classes in the set are balanced around an axis that is illustrated by @ line through the clocktace. Here is how to identify an axis of symmetry. First, discover the index ‘number that maps the set onto itself under inversion. This value of n is not hard to find. First, write the set so that its intervals are the same from right to left as from left to right (usually the normal form of the set). Figure 4-6 shows several sets written in that way, Bil ile aish «2h isis, ea’ WK ANAK AAAYA ae bw Saas YS Fewes ‘The fist two sets are writen in normal form The thid set is also eapable (of mapping onto itself uler inversion, but the sytometry is not apparent Unless you envision the st as wrapped clockface. The lst pitch clas is written Again atthe beginning to indicate the circular nature ofthe pich-class system, ‘The normal form for the fourth set is [5,7,9,0,2}. Tha set had to be rotated (0 make its inversonal symmeicy apparent. To find the value of n that maps these sets onto themselves under Trl, ust add first to last, second to second toast, and so on. For [3,5,6,8), 3 + 8 and 5 +6 both equal 11, so that set maps onto itself under Ty For (2,5,8], 2 + 8 and 3+ 5 both equal 10, so that set maps onto itself under Tj. ‘The third set, [1.3.8]. maps onto itself under Tel G8 1 3 ~ 4) and the fourth st, [5°7.9,0,2}, maps onto itself under TAQ +0=549=7+7= 2). Of coune, there may be more than one Cntrciy and Some Important Referential Collections 1, value of n that maps a set o ‘many there are.) Once we know the value(s) of n that map a set onto itself under Tl, we will nd dat the inversonal ans pases trough 3 and! +6, Por example, if 0 itself. (The set list in Appendix 1 tells just how & ems ono ie at Te favesonl ais wil be $$ + 6, or 4, If the set maps onto itself by an odd index number, the axis will pass between {wo pairs of notes. For example, ifa set maps onto itself at Ty, the axis, according {o our formula, will pass through 1% and 7%, Since it doesn't make ‘much sense to talk about one-half of a pitch class, we will say that axis passes between, 1 and 2 and between 7 and 8. We will write that axis: 1/278 ‘We can represent axes ‘of inversion by drawing a Line right through the middle of the set and through a point a trtone away. Figure 47 represents in that way the four sets discussed above: [3,5,6,8],[2,5,8], 11,3,8], and [5.7,9.0.2] Ne cg es Onc MLeny , Os 1 a © 1 C30 ©, %" fee age Figure 4-7 Sets that map onto themselves under T,L for more than one value of w will have ‘more than one axis of symmetry. ‘The dctatonic collection maps onto itself under Jal for four values of nT thus has four axes of symmetry, ay Figure 4-8 In pieces using inversionally symmetrical sets, the pitch classes that make uj the axis or axes of symmetry may play a centre role. : ‘The so-called “changing chord" from the third of Schoenberg's Orehestra Pieces, Op. 16, is symmetrical around the axis E—BF (soe Example 4-11).© 104 Centicity and Sore Important Referemil Collections Miiige View. Piano 1 aa Piano I Example 411 lavesional axis a ptch-lass center (Schoenberg, Orchestral Piece, Op. 16, No. . “The symmetry of the chord is not apparent from its normal form, but is when the pitch classes are reordered to begin on B (see Figure 4-9), ‘You can probably see immediately that the chord is symmetrical around E (and Bi), but you could also figure it out by noticing thatthe set maps onto itself at ‘Tyl and is thus symmetrical around the axis of 4—10. The chord is not arranged symmetrically in register; this is a symmetry of pitch class, not of pitch. Nonethe- less, the axial E plays a special centric role in the piece. Notice in the opening, measures that it i the first note of the changing chord that actually changes. Inversional symmetry can be easier to hear when it affects not only the pitch classes but the actual pitches. A centric pitch can be emphasized by having the pitches around it arranged symmetrically in register. The beginning of Bartdk's ‘Sonata for Two Pianos and Percussion is shown in Example 12, ‘The opening seven-note melodic figure (measure 2) is stated several times the passage at different levels of transposition. The figure is a symmetrical cluster of semitones balancing around an axis. The first statement, for example, comprises IDLE,ELFI,G,GI,A}. The pitch-class axis is C—F4 (the set maps onto itself at ‘TaD but the F? is particularly emphasized as the pitch axis; the other pitches are arranged symmetrically above and below it. The F# is also the first note in the figure. The second statement of the melodic figure (measure 5), (A,B>,B,C,Ch, D,Eb], is balanced on the same pitch-class axis (it also maps onto itself at ‘ToD, but now the C is emphasized. The following statements (measures 8 and 9) Centricity and Some Important Referenial Collections 0s fies *Hessicteoas 1 Assa leato, Doon ample 4-12 ‘The melodies balance fist on a CPE axis, then on « G-Db ais (Gard, Sonata for Tw Panes and Pteussion), balance on the axis G—Db, emphasizing first the G and then the Db. ‘The tues provide pitch centers foreach statement ofthe figure, and taken together— H,C.G, Dithey constitute another symmetsical set, 4°9 (0167), that figures prominently inthe music that follows. ‘Sometimes the idea of inversional balance around an axis can affect more than jasc # single pitebrelass sot or group of eets, Tt can expand to. encompass All twelve pitch classes. In that ease, each pitch class maps onto another (or tomo itself) around some axis. Barték's Bagatelle, Op. 6, No. 2, begins with repeated Ais and Bbs inthe right hand (see Example 4-13).106, Centcty and Some Important Referentiat Colectons {Negro glacase to oa Example 4-13 Taversona Blane around A atk, Bagatelle, Op. 6, No.2. ‘A melody begins in measure 3 on BA, a semitone above the repeated figure, then continues with G, a semitone below the repeated figure, Then come C and Gt (owo semitones above and below), Db and F (three semitones above ad below), D and Fs (four semitones above and below), and finally Es, a pitch lass that lies five semitones both above and below. ‘The only pitch class that ‘has not been heard is A, which lies right in the middie of the repeated figure, a kind of silent center around which everything balances (see Figure 4-10). Figure $10 Ais the piteh axis A—Bs is the piteh-class axis. The Eb does not play much of a centric role in this opening phrase, but later in the piece the opening music returns at Ty. At that point, the pitch-class axis is still AEs, but itis the ES that is particularly emphasized. ‘There ate no strict “‘ules"* for analyzing centric pieces because the nature of the centricity varies so much ftom piece to piece. ‘Three general guidelines will have to sullice. First, don’t be deceived by the presence of triads and other traditional formations into thinking that eentrie pioses are tonel. In most tweiticlr century music, including music that strongly evokes the sound of earlier music, Hea Comricty and Some Inport Referentat Cottections 7 traditional tonality plays little role. So don't expeet Roman mumerals and other kinds of tonal analysis to be of much help. Second, be aware of the stabilizing effect of large, referential collections. Composers frequently use diatonic, oct tonic, or other collections (or combinations of these) to create stable harnionic areas, Finally, be sensitive to the range of centric effects in post-tonal music ‘There usually won't be a single obvious pitch center. Ofteny there will be # clash, of polarity, of competing centers. You will need to be flexible in evaluating the different ways that pitches, pitch classes, and pitch-class sets pet emphasized and the ways that such emphasis shapes the musie BIBLIOGRAPHY ‘There have been many attempts to apply tonal theory to posc-ionat music, with bredictably uneven results, For examples of Roman-nueral analysis of twentieth ‘century Works, see Dika Newlin, "Secret Tonalityn Schoenbcry's Piano Concetto. Perspecives of New Music 13/1 (1974), pp. 137-39, and Will Ogdon, “How Toaality Functions in Schoenberp's Opus 11,'No. 1," Journal of the Arnold Schoenberg Insitute 5 (192), pp. 169-81. Tonal-style’analyses using. Scheaker’s theory Cd “prolongation” have been more numerous and we suecessful. See Telit Salzor, Structural Hearing: Tonal Coherence in Music (New York: Dovet, 1962), Roy) ‘Travis, “Toward a New Concept of Fonality?” Journal of Music Theory 3 (1959), Pp. 257-84; Roy Travis, "Directed Motion in Schoenbery and Webern,* Perspectives Of New Masic 4 (1966), pp. 84-88; Robert Morgan, “Dissonant Prolorgations, Theoretical and Compositional Precedents," Journal of Music Theory 20 (1976), pp. 49-91; Paut Wilson, “Concepts of Proiangation and BartOk's Opts 20," Mcre Theory Spectrum 6 (984), pp. 79-9; and Allen Forte, "'Tonality, Symbol, ant Structural Levels in Berp’s Wozzect,"” Musical Quarterly 71 (1983), pp. 474-99. ‘The prolongational approach is cridqued in James Daler, “Schenkeren Analyste and Post-Tonal Music," in Aspects of Schenkerian Theory, ed. David Beach (New Haven: Vale University Pred, 1983) and Joseph N. Stas, “The Problem of PLONE, ‘ation in Post-Tonal Music," Journal of Music Theory 34/1 (1987). pps 1-22 ‘Most discussions ofthe octatonic collection in Stravinsky's music, and its interaction With the diatonic collection, have taken as theie point of departure Arthur Berger's seminal article “Problems of Pitch Organization in Stavinsky,” Perspect of ‘New Music 2/1 (1963), pp. 11-42, The definitive treatment of this subjects Peter van den Toom, The Music of Stravinsky (New Haven: Yale University bres, 1983), For a discussion of Bartok’s octatoncism, see Eliot! Antokoletz, The Music of Béla Barc: A Study of Tonaity and Progression in Twenteth'Century Musto Berkeley and Los Angeles: University of California Press, 1984). The octatonic ollection is one of Olivier Messiaen’s ‘modes of limited ttansposition.” See The Technique of My Musical Language, trans. J. Saltcfiel (Pais Alphonse Lede, 1956), pp. 58-63, Paul Johnson discusses the diatonic octad in “Cross-Collectional ‘Techniques of ‘Structure in Stravinsky's Cenisic Music,” in Siravinaty Retrospectives, ed! than, Haimo and Paul Johnson (Lincoln: University of Nebraska Press, 1987), pp. 53 %. ‘The notion of inveriona balance has been diseussed by David Lewin in a sesiex vf icles: “Inversonal alance as an Organizing Foree in Schoonberg’s Music and Thought,” Perspectives of New Music 67 (1968), pp. I-21: "A Label Prosto Centicty and Some Important Referemtal Collections Dovelopment for 12-Ptch-Class Systems,” Journal of Music Theory 21/1 (1977), ‘pp. 29-4; and “Transformational Techniques in Atonal and Other Musi Theories, = Perspectives of New Music 21 (1982-83), pp. 312-11. The ceniticity induced by invefsional balances central theme of George Peres Twelve-Tone Tonality Berke ley and Los Angeles: University of California Press, 1977) and Biot’ Antokoet2's The Music of Bela Bartok EXERCISES 1. Some referential collections 1. For cach ofthe large collections discussed inthis chaper (the diatonic collection, the diatonic octad, the octatonic collection, and the whole-tone collection), da the following: a. Compare their interval vectors 1b. Compare them with regard to common tones under Ty nd Ty €. List the complete subset structure for each, identifying the set class to which each subset belongs, Then compare the cllections how are they similar fand how are they different? 2, Give neutral descriptive names (e.g, “B-scale on F") foreach ofthe following teaditiona scales. a. Fl-major scale Bb. Mixolydian mode transposed to D . Bbsminor sale 4. Lydian mode transposed to C 13. Kdentty each of the following collections: Cenicty and Some Important Referential Collections 105 I, Inversional axis: Inversionally symmettical sets map onto themselves under Tyl. The axis of symmetry for such aset is 32 + 6 1, For each ofthe following sets, determine if they are inversonally symmetrical. they ar, ind the axis (or axes) of symmety: 1145.8) 1. 10,0124) 1,2,3.4,8.9) I 19.10.11,3.51 (6,1) F 11.2;5,6,9,10) 2, Construct a eas two pitch-class sets chat ae symmetrical around the following axis (or axes). Give your answer in normal frm. 410Analysis 4 Analysis 4 Stravinsky, Oedipus Rex, rehearsal nos. 167-70 Ns sum quo nefastim 1 was bom af whom view fade, Cones nef et | mared whom divin law forbade, (Bartok; Sonata; Best movement, ‘kekidi quem nefastum est, [killed whom divine law forbade, Uefa Allis now made cea! ‘Example AI shows the central passage from Stravinsky's opera-ortorio Oedipus Ren he moment were Oedipis realizes who he is and whet he bas done J the music that precedes this passage, a shepherd and a messenger reveal 10 Oedipus his unhappy circumstances, Tey do'so in a kind of sage section The Shp and Mesegr dean that uses the D-sele on D and eeniers on the tad D-I“A and on pte lae D. The fist measure of Example, AG] summarizes their music with sinple miner ina played 8y the sng Inthe second care He wood Simpani answer with aB minor ad. Pay these two mesures and ten eae) hey raise a umber of interesting musical sus. Fist ete fs the Heo B versus D, an idea with symbolic resonance throughout the werk. D's generally Asscined with moments of revelation, as when the shepherd and messenger Feveal the tut about Oedpes. The By in coasts asset wth Oedipus Blindness, both the symbole bindncss of hs ignorance an the Tal blindness he later infits upon hime ‘The clash between B and D in these measures, and between the B-minor and Deminor triads, involves a further clash, between Ft and PF. Fi ig associated with D-F-A, while FF is associated with B-D-F¥, The Pt and Fi come into direct confict in the measure before rehearsal no. 168, where D-F-A and B- D-F1 are compressed into a single sonority: D-F-F4. That sonority is a member of set class 3-3 (014), a central musical motive in this passage and throughout Oedipus Rex Back in Chapter 3, we discussed a statement of 3-3 (014), articulated by the ‘Fate motive,” that spanned the entire work. As you will reall, the thiee. state: ‘ments of that motive were centered on Bb, B, and G (see Example A4-2). Example AA-2 A large-seae statement of 3-3 (014 that spas the entre work, ‘The initial BL—B in that large statement could also have led to D, since [Bb, B,D] is also a member of set class 3-3. That implied D does, in fact, arrive in the ‘work in precisely the passage we are discussing, the moment of Gevliws's self revelation, When that D arrives, completing a large-scale statement of 3-3, itis immediately associated with a surface statement of another member of the Same Example A-1 The moweat of Oedipus elighenment in Stravinsky's Oedipu Res set class: [D,F,F] (see Example Ad-3), no umAnalysis 4 Example A4-3 An akional middlepround and foreground statement of 3-3 (OL, ‘embeded within the larga work spanning statenet. “The PID clash is thus part of the large plan of the work: ‘The shepherd and the messenger had straightforward, diatonic music that reflected the simplicity of their characters. The intrusion of the B-minor triad as Oedipus prepares to speak immediately makes the music more complex. ‘The music leaves the diatonic realm of the D-scale on D and moves toward an ‘oclatonic collection: D, Eb, F, Fl, Gt, A, B, C. The passages does not contain fof the notes ofthis collection, but the central conflicts inthe passage—between B and D, between B-D-FE and D-F-A, and between F and Fall take place there (see Example Ad-4), —— = ce Example A‘-4 Cetriecoafcs within an ostaonie colleston Play the music up to rehearsal no. 168 and listen carefully for the juxtaposition of B and D and the musical conflicts that result When Oedipus sings, he arpeggiates a B-minor triad. His melody suggests a new diatonie collection: A-scale on B, Each of the competing centers, B and D, has both a small harmony (a minor triad) and a larger collection associated with it (see Example A¢-5). [Example ALS A confit between tells cents, nds, and diatonic collections m Analysis 4 Despite these centers, triads, and diatonic collections, however, the music is not traditionally tonal, There are no dominants and tonics, and no real sense of progression, Rather, the competing centers, triads, and larger collections are juxtaposed in a static manner. Oedipus’s melody is clearly B-centered. The harmonization of his melody, how- ever, is ambiguous, Moments of conflict between F and FF and competition between B and D occur throughout. ‘The first two times Oedipus concludes a vocal phrase-on,F¥, the note is harmonized by set class 3-3 (O14), including both F’and Sing Occipus's melody between rehearsal nos, 168 anid 10Y and notice how unambiguously Beoentered it seems to be. Then sing it again while Playing the accompaniment. The cenricity is suddenly much less clear. Listen Particularly for the clash between F and F¥ and for the associated centric clash between D and B. In the last five measures of the passage, Oedipus states his last vocal phrase, ‘again ending on Fi, Now, however, the harmony is clatified as the symbolic light breaks in on him. Play these measures and focus particularly on the final sonority, the dyad D-FF, The bass in these measures descends from B to D. ‘That would seem to suggest a symbolic journey from blindness to revelation Atthe same time, however, notice what happens in the top voice of the accompani: ‘ment, Ft (spelled as E#) wills with F4 before definitively resolving there. ‘The F (previously assoctated with D) thus moves to FF (previously associated with B) jst at the moment the bass maves definitively from B to D. The final sonority ‘of the passage, the dyad DF, thus represents a synthesis of the competing ‘elements. Itis not really a D-major triad because there is no confirming A. It is not really a B-minof triad because there is no B. It is just a dyad, poised between the realms of Band D. combines the F4 (from ihe B-centered music) wid the D (from the D-centered music). (See Example A4-6.) [Example A4-6 ‘Tho coats between Band D, a resolved within the Bal ‘The musical conflits and ambiguities aro reconciled at the moment of Oedipus's| illominat 13Analysis 4 The opening of the first movement of Bart6k’s Sonata is shown in Example 4-7, Allegro moderato, Jt 16 e 33101) 33a. 10. 4-9 (0160) SAE +3) ‘xample M7 The begining ofthe fist thee of Bats Sorta wth some important plelsclas sts nea). With a few isolated exceptions, the pitch material for this passage is drawn from a single octatonic collection: E, F, G, Gi, Ak, B, CF, D. Within that collection, the E has priority. tis the pitch-class center of the passage, established by several musical means. First, E oceurs in the bass on the strong beat of the frst thirteen measures, regstrally prominent and accented. Almost every time it ‘sccurs, it is harmonized by an E-major triad, which does not function as a tonic jn atonal conext, but as consonant support for the cenrl Ewin an otatonic ramewo oo Analysis 4 On the second beat of each of the 2/4 measures, the bass moves from an E- ‘major tiad to a form of set class 3-3 (014), consisting of D (lower neighbor to E), Ad (upper neighbor to GH), and B (common tone. In this way, 3-3 embellishes and reinforces the structurally supesior E-major ria (gee Example A-8) Example AG An E-najor rind embllished by a form of et clas 3-3 (014), Play the chonds in Example A4-8 and listen to the triad being embellished by the 3-3. Both the triad and the 3-3 are subsets of the referential oetatonic collection. Later in the movement, when the second theme begins, set class 3-3 will emerge from its secondary role. While the bass is altemating between the E-major triad and the embellishing 3 3 (014), the upper voice is siowiy arpegsiating from Gt (measure 1) to B (measure 2) up through E (measure 7) to G (measure 14). This arpegsiation helps 10 reinforce the central E, but E major and E minor seem to be joined within single gestae. The arpeggiation begins on GE, but its goal is a heavily stressed Gi. Wien the GY ip attained in measure 14, GH is heard in the ass, The clash between GH and Gt (the third and fourth degrees of the ectatonic seale on E) and the seeming blend of major and minor are typical of octatonic music. In fact, seming clashes or blends of major and minor inthe music of both Bartok and Stravinsky ofien suggest oetatonic thinking and the presence of an octatonie collection. We slready saw an example of that in the passage from Stravinsky's Oedipus Rex discussed ealer. There, the D-F-A-cemered music ofthe Shepherd and Messenger finally gave way to Oedipas's cadence on D-Fi. As we saw, {ha di nt mean a overnent from D minor to P major, Duc raker 8 synthesis of B-D-FF and D-F-A within an octaonie framework At the top of the arpeggiation in measure 14, Barték introduces two new octatonic subsets tha willbe prominent throughout the movement, The notes onthe downbeat ‘of measure 14, and on the downbeats of the next four measures, form set class 4-9 (0167), a perennial favorite of Barték’s. ‘The left-hand accompaniment in these measures, excluding the AJ, focuses on sot class 4-18 (0147). Example ‘A4-9 shows the referential octatonic collection and the subsets that Bartok draws from it 311 (087): moe rad 235 014) 8h.) 49 0167} 16,6.C4.D) 4418 147) 1E,D,F.GH hale AAA Ths veteenal aciatne eoleton FOr the AIS Meme a oR importa subs, us