MERTON SHATZKIN's HomePage

Music Theory, Medical Arts Symphony, NewEar, Orchestration, and Exercise machine.
 

My URL is http://s.faculty.umkc.edu/shatzkinm/

my email address is: shatzkinm@umkc.edu
 
 
 
 
 

CONTENTS

About Me
About the Medical Arts Symphony
About New Ear
ABOUT MY DATABASE OF COMPOSERS' WHEREABOUTS
SOME MUSIC THEORY TOPICS

About My Homemade Exercise Machine
 Orchestration Analyses I taught in the Conservatory of Music of the University of Missouri Kansas City from 1963 until 1997. Courses were in the area of theory. I published articles in BACH, PERSPECTIVES OF NEW MUSIC, and THE JOURNAL OF RESEARCH IN MUSIC EDUCATION. I am the author of WRITING FOR THE ORCHESTRA(Prentice-Hall, 1993).

 As a violinist, I have been a member of several orchestras--the Indianapolis, the Chattanooga(concertmaster),Rochester, Tulsa, and Kansas City. I was a member of resident string quartets at Pittsburg State University and the University of Missouri Kansas City. Also at the latter institution I was a member of the contemporary music ensemble and the baroque ensemble.

I am a former member of the contemporary ensemble, NewEar, and am presently the conductor of the Medical Arts Symphony.

 I have written compositions for orchestra, chorus, voice, piano, and ensembles.
 TOC

ABOUT THE MEDICAL ARTS SYMPHONY

The Medical Arts Symphony has been in existence for over 30 years. Its members are amateur musicians, many of whom are either active or retired members of the medical profession. Some are retired professional musicians. The orchestra gives two concerts yearly in Kansas City--usually in late November and late April. Professionals fill out the ranks for these concerts. Notable compositions performed were Berlioz's HAROLD IN ITALY, Rachmaninoff's RHAPSODY ON A THEME OF PAGANINI, and Brahms' Second and Third Symphonies. Several contemporary works were performed--several of them premiers of local composers'.

 The orchestra for many years sponsored the Leopold Shopmaker Violin Competition. This awarded three prizes and a performance with the orchestra for the first-prize winner. A recent winner went on to win the Queen of Belgium Competition.

 We welcome new, qualified members. There are openings in strings and bassoon. If you would like to audition, call me at 816-523-7376. Rehearsals are held at Battenfield Auditorium at the corner of Olathe Blvd. and Rainbow Blvd. each Monday evening at 7:15. Our next concert (at the same location) will be on November 16, 2002 at 8:00 PM. There is no admission charge. The program consists of  Beethoven's Symphony No. 7, and Liszt's Les Preludes, Straus' Tales of the Vienna Woods, and Oklahoma.  Our program carries advertisements which help support the orchestra. If you wish to advertise, please call me.
.
 

(TOC)

ABOUT NewEar

NewEar is a professional ensemble that performs contemporary music in Kansas City, Mo. It normally gives five concerts yearly at St, Mary's Episcopal Church at 1306 Holmes St. Its repertoire concentrates on works of the last two decades and has included several premieres, some by local composers.

 They encourage the submission of scores, particularly those calling for some combination of clarinet, saxophone, bassoon, violin, piano, and percussion or a subset thereof. Address scores with return postage and envelope to: Jan Faidley, 5530 Roeland Drive, Mission, Kansas. 66205 (816-671-7638) (JFAID74702@AOL.COM).

(TOC)
 

I am compiling a database that shows where composers were on each day of their life--as far as I can determine, using letters, biographies, etc., as sources. At this time (September 2002) there are over 400,000 entries. Composers covered are: Bellini,Donizetti, Mendelssohn, Chopin, Schumann, Rossini, Meyerbeer, Berlioz, Mussorgsky,Wagner,Liszt,Bizet, Tchaikovsky, Brahms, Bruckner,Gounod,Smetana,Borodin,Moussorgsky,Verdi,Saint-Saens, Wolf, Mahler, Dvorak,Debussy, Rimsky-Korsakov,Franck,Puccini, and Faure. The database is in Microsoft Excel 97. I hope some day, when it is much more complete, to make it available in some useful, searchable , form.
   TOC
SOME MUSIC THEORY TOPICS

A device I developed for analysis and teaching of SET THEORY and 12-TONE ANALYSIS, as well as composition is the MUSICAL SLIDE RULE. It consists of twelve slides mounted on a slab, held in place by two guides. There are twelve plywood slides that are 12 inches long and one centimeter wide. Each is covered with paper and marked off with squares of equal size throughout its length. The squares are filled with numbers as follows: from the top down, the numbers "0" through "11" twice in black ink, then the same numbers twice in reverse order in red ink. The slides rest on a slab of plywood that is 9 X 12 inches. They are held in place by two plywood guides, each 7 inches X 1 centimeter. These are screwed onto the slab 1 inch from each of the long sides of the slab, parallel with those sides, and centered. The slides should rest between them snugly enough to hold them in place when individually moved up or down, but also with freedom to move readily. A piece of clear plastic of about 7 X 5 inches is screwed into the guides so that it covers the central area over the slab(which is covered by the slides). With all slides in the same orientation, the display looks like this:

 0 0 0 0 0 0 0 0 0 0 0 0     (in black)

 1 1 1 1 1 1 1 1 1 1 1 1

 2 2 2 2 2 2 2 2 2 2 2 2

 3 3 3 3 3 3 3 3 3 3 3 3

 4 4 4 4 4 4 4 4 4 4 4 4     

 5 5 5 5 5 5 5 5 5 5 5 5       


 6 6 6 6 6 6 6 6 6 6 6 6

 7 7 7 7 7 7 7 7 7 7 7 7

 8 8 8 8 8 8 8 8 8 8 8 8

 9 9 9 9 9 9 9 9 9 9 9 9

101010101010101010101010

111111111111111111111111

 0 0 0 0 0 0 0 0 0 0 0 0

 1 1 1 1 1 1 1 1 1 1 1 1

        etc.





 0 0 0 0 0 0 0 0 0 0 0 0 

111111111111111111111111

101010101010101010101010

 9 9 9 9 9 9 9 9 9 9 9 9

        etc.                  (IN RED)
The Slide Rule has many applications for dealing with the set theory described by Allen Forte in his book THE STRUCTURE OF ATONAL THEORY(Yale University Press, 1973). It is assumed here that the reader is familiar with at least the basics of the theory.

 To work with a set, in most cases it is convenient to use the number of slides that corresponds to the cardinality of the set; e.g., the number of notes in the set. To illustrate, let us take 3-4 {0 1 5}.

 THE MATRIX

 If any three of the adjacent slides are arranged so the numbers of the prime form are aligned at the top of the display, all 24 of the equivalents of the set are shown:
 
 

 0 1 5   (IN BLACK)

 1 2 6       "     

 2 3 7

 3 4 8

 4 5 9

 5 610

 6 711

 7 8 0

 8 9 1

 910 2

1011 3

11 0 4

 0 1 5

  etc.



 011 7    (IN RED)

1110 6        "   

10 9 5

 9 8 4

 8 7 3

 7 6 2

 6 5 1

 5 4 0

 4 311

 3 210

 2 1 9  

 1 0 8

 011 7

   etc.
The matrix, arranged in this way--with the prime form at the top, shows all the equivalents in a manner that allows each equivalent to be quickly found. First of all, the transpositions of the prime form are in black and the inversions in red. Secondly, The left-most number of each equivalent is always the transposition number relative to the prime form: e.g., in black, the row (of the matrix) reading "8 9 1" is the prime form transposed up by 8 semitones, and in red the row reading "6 5 1" is the inversion of the prime form transposed up by 6 semitones.

 FINDING PRIME FORM
The prime form of any set can quickly be found by aligning the numbers of the unkown set at the top of the display. This still creates a matrix of the set in the display, so, naturally, the prime form will be shown in one or more of its rows. THE PRIME FORM IS THE ROW CONTAINING THE SMALLEST NUMBER, READING THE DIGITS IN THE ROW IN ASCENDING ORDER AND AS A SINGLE NUMBER.

 Let us suppose we have the set [3 1 6] and wish to identify its prime form. We align three slides with these numbers at the top:

 3 1 6  (in black) 

 4 2 7

 5 3 8

 6 4 9

 7 510

 8 611

 9 7 0

10 8 1

11 9 2

 010 3

 111 4

 2 0 5

 3 1 7   

  etc.



 0 2 9  (in red)

11 1 8

10 0 7

 911 6

 810 5

 7 9 4

 6 8 3

 5 7 2

 4 6 1

 3 5 0

 2 411

 1 310

 0 2 9

  etc.
By inspection, we find that the row containing "2 0 5" indicates the prime form, because these numbers, arranged in ascending order(025), form the smallest number of those that result when the numbers in each of the rows are so arranged. To use the matrix as it is now aligned to find transpositions, use the column(or slide)containing "0" in the prime form--in this case "0" appears in the second column. Thus, the numbers in the second column of the matrix can be used to identify transpositions of all equivalents.

I used the areas in the"margins" of the slide rule (to the left of the left guide and to the right of the right guide) to display all the prime forms of the set universe in ascending order--e.g., 012, 013, 014,--with the ordinal number of the short name in parentheses after the prime form; thus:

012(1)
013(2)
014(3>
etc.
Specifically, this shows, for example, that prime form {0 1 2} is also called "3-1."

 Transpositional relationships between equivalents can easily be determined by the index column--that is, the column containing "0" in the prime form. For example, in the matrix for {0 1 5} shown above, [6 4 9] has an index of 4 since that number is in the second column (the column containing ) in the prime form. Finding the transpositional relationship of this set to [1 11 4} is simply a matter of subtracting 4 from 11, 11 being the number in [1 11 4] that is in the second column. The difference, 7, is the number of semitones [1 11 4] is "above" [6 4 9].

 Common tones, or invariances, between equivalents are easily found by inspection. In the partial matrix of {0 1 5 6} shown here, the presence of pc 6 is easily seen in three of the equivalents and the dyad [5 6] in two of them:

 0 1 5 6
1 2 6 7
2 3 7 8
3 4 8 9
4 5 910
5 61011

It is easy to find common tones between members of different sets simply by setting up the matrices of those sets side-by-side on the Slide Rule. The matrices for {0 1 5 6} and {1 2 3 6}, shown in part below, reveal that the dyad [5 6] is common to the prime form of the first set and of t3 of the second:
 
 

0 1 5 6     0 2 3 6

 etc.       1 3 4 7

            2 4 5 8 

            3 5 6 9
Although the interval vectors of each set can be accessed in Forte's Appendix I, in the absence (OR presence) of that appendix they can be derived from a matrix in the following manner:
Beginning with the leftmost slide, find the row containing "0". Tabulate each of the numbers on the slides to the right of it in thatrow once. Do the same for the remaining slides(except for the rightmost one). e.g.:
For the set {0 2 3 6}, if the slides are aligned with the prime form on top--"0","2","3","6,"--the row containing "0" on the leftmost slide is the top one ( [0 2 3 6] ). Thus we tabulate the numbers to the left of "0" each as one count of their pc representation--one pc2, one pc3, and one pc6.
The second slide (the one to the immediate right of the leftmost slide)contains "0" in the row containing t10: [10 0 1 4 ]. We tabulate the numbers to the right of "0"--one pc1 and one pc4. The third slide has "0" in t9: [9 11 0 3], thus we tabulate one pc3. Compiling the tabulations we have the vector of 1121011 .
 

To find the inverse of a set using the matrix, take any of the numbers in that set and find its inverse on the same slide in the opposite "side" of the matrix (i.e., in the other color). The row in which that inverse occurs is the inverse of the whole set. e.g.:
In the matrix of {0 2 3 6} to find the inverse of the prime form we can look for "0" in the leftmost slide in red (assuming that the prime form is set up in black). The row containing "0" in red on the leftmost slide indicates the set [0 10 9 6], which is the inverse of [0 2 3 6].
  TOC

12-TONE ANALYSIS
 

Using all 12 slides, the Slide Rule can be used to display all 48 forms of a row (row-class). Each is easily identified by its initial pc number. To illustrate, let us take the row of Schoenberg's Fourth String Quartet: 0 11 7 8 3 1 2 10 6 5 4 9. The P transpositions are shown as the rows descend--P1,P2,---P11. In red the I forms appear in "reverse" order--I0, I11,---I1. The R and RI forms are read from right to left, their transposition number being identified (as by many theorists)by the leftmost pc number.
 

Hexachordal combinatoriality can be determined by inspection of hexachords. For example, note the first hexachord of P0 and inspect the hexachords of the remaining forms for the same pc content. In the matrix of the Schoenberg row we find the pc content of the first hexachord of P0 ( [0 11 7 8 3 1] ) in the second hexachord of I5 ( [3 7 11 0 1 8] ). This not only reveals that the row is hexachordally combinatorial, but also that it is so at the level of I5.
 

Unordered sets can be discovered within rows and row-forms by thinking of the entire matrix as the union of submatrices--that is, by using only some of the slides at one time.
Every unordered set is contained in every row-form since each form contains all 12 pcs. To begin, let us determine the prime forms for the tetrachords ( i.e., order numbers 0-3, 4-7, 8-11) in the Schoenberg row. To do this, we treat the leftmost four slides as a submatrix, and the same for f the four slides on which order numbers 4-7 and 8-11 are represented, respectively. Using the method for determining prime form shown above, we fing that the first tetrachord is {0 1 4 5} (this prime form is found in I0), the seond is {0 1 2 5} (see I3) and the third is also {0 1 2 5} (see P8].
If we do the same for trichords, we find the succession {0 1 5}, {0 2 7}, {0 4 8}, and {0 1 5}, but more interestingly, if we analyze EVERY trichord formed by adjacent row members, we find {0 1 5} in the order-number sets 0-1-2, 2-3-4, 7-8-9, and 9-10-11.
 

THE INTERVAL-CLASS SERIES(IC SERIES)
A twelve-tone row is most generalizably represented not by its P0 (the precompositional set), but by the ordered set of directed interval classes that relate the successive members of the row. This can be represented by ic numbers preceded by a + or - (except for pc 6, for which + and - entail yield the same result). Thus the ic series for the P forms of the Schoenberg row is:

 -1-4+1-5-2+1-4-4-1-1+5.

 Specifically, the first "-1" represents the directed ic relationship between order number 0 and order number 1 in any P form--in P0, these order numbers are pc 0 followed by pc 11 and 0-11 = -1. Similarly, from pc 11 to the following pc 7 in P0 the directed ic is -4, etc. This ic series holds for all P forms, thus is more general than P0 alone. For I forms, the series has the same order of numbers with signs reversed:

 +1+4-1+5+2-1+4+4+1+1-5.

 The series for R is I series in reverse:
 

-5+1+1+4+4-1+2+5-1+4+1

 because when pitches are reversed, so are the directions they move from one to the next.

 As one might expect, then, the series for RI is the series for P in reverse:

 +5-1-1-4-4+1-2-5+1-4-1.

The ic series can be used (in the absence of a matrix or magic square) to construct any form, given its first pc. It can also be used to analyze statements of rows in 12-tone works. The ic series can also be used to discover ordered pc sets that are common to more than one form. This is done by using the following formulas and procedure:

 FINDING ORDERED SETS

 1. Examine the ic series for subsets whose elements are
1) the same,
2) the same, but with opposite signs,
3) the same, but reversed, or
4) the same, with opposite signs AND reversed.

To illustrate, using the Schoenberg series:

-1-4+1-5-2+1-4-4-1-1+5

we find three two-number elements that have these similarities:

1. [-1-4] and its reverse [-4-1] 2. [-4+1] and its reverse [+1-4] 3. [+1-5] and with opposite signs [-1+5] If such similarities can be found within an ic series, the same pcs can be found in the same order in more than one row form.
Formulas for each of the four possibilities given above follow:

1. If the elements of a subset are the same, the pcs represented will appear in the order numbers of the second subset in the form Px-y, where x is the first pc of the first subset and y is the first pc of the second subset AND the pcs of the second subset will appear in the orcder numbers of the first subset in the form Py-x.

 To illustrate, suppose that order numbers 0-2 in P0 consist of pcs 0-5-4. The ic series for these pcs is therefore [+5-1]. Now let us suppose that in order numbers 3-5 there is the progression 9-2-1, which also has the series [+5-1]. Then according to the formula, x = pc 0, the first pc of the first subset, and y = 9, the first pc of the second subset, and the pcs of the first subset will appear in order numbers 5-7 in P3(0-9). AND the pcs of the second subset will appear in the order numbers of the first subset in P9(9-0). Let us show these three row forms as far as order number 5:
 

P0: 0 5 4 9 2 1- - -
P3: 3 8 7 0 5 4- - -
P9: 9 2 1 61110- - -

2.If the elements of the second subset are the same, but with opposite signs, the pcs of the first subset will appear in the order umbers of Ix+y and the pcs of the second subset will appear in the norder numbers of the first subset also in Ix+y, where x is the first pc of the first subset and y is the first pc of the second subset.
To illustrate:
P0:0 5 4 7 2 910
I2:2 9 107 0 5 4
Here the first subset contains [0 5 4] and the second [2 9 10]. The ic series is +5-1+3-5+1. The formula results in I2(0+2).

3.If the elements are the same, but reversed, the pcs of the first subset will appear in the order numbers of the second subset in the form RIx+y where x is the first pc of the first subset and y is the last pc of the second subset and the pcs of the second subset will appear in the order numbers of the first subset in the same form.
To illustrate:

 P0:0 4 1 6 3 8 5 9


RI9:- - - - - - - 0 4 1 6 3 8 5 9
(Note that in this case the subsets have four elements.)
 

4. If the elements are the same, but with opposite signs and reversed, the pc of the first subset will appear in the order numbers of the second subset in Rx-y where x is the first pc of the first subset and y the last pc of the second subset. The pcs of the second subset will appear in the order numbers of the first subset in Ry-x:

P0:0 4 1 6 9 5


R7:- - - - - - - 0 4 1 8 711


R5:             10 211 6 9 5
I have written a program in BASIC that accomplishes all of these and displays all the results. For a copy send me a formatted disc and a stamped, self-addressed envelope to 100 Morningside Drive/Kansas City/MO 64113

 EXTRACTED UNORDERED SETS

It is well known that Schoenberg manipulated row statements to create set-class unity by leading instrumental parts between nonadjacent row members, using that freedom to allow a specific set-class to be heard several times. A composer working with 12-tone procedures can prepare a chart that shows all the possibilities for deriving a given set-class among nonadjacent members of row that is used, using the matrix. The process is:
 

Make two columns. Label the first "row form" and the second "order nos." Then examine each row of the matrix (in the sense of "row" vs. "column") for the order position of the numbers in the prime form of the set-class. The row is then entered in column 1 as "P0" or "P1"---"I11" and the order numbers identified in the second column. To illustrate, let us begin a chart that show all of the possibilities for deriving set-class {0 1 5} in the Schoenberg row. For completeness, let us include possibilities where the pcs are adjacent:
Examining P0---0 11 7 8 3 1 2 10 6 5 4 9 for the numbers in the prime form of {0 1 5}, we find that pc0 is in order position 0, pc1 is in order position 5, and pc 5 in order position 9; therefore, we enter "0, 5, 9" in the second column . Moving to the second row of the matrix--P2, we find that the pc numbers 0, 1, and 5 are found in order positions 1, 0, and 10. Thus the first column entry is "P2" and the second column entry is "1, 0, 10". In this manner we make 24 entries in the two columns. This completes the chart and we are now in a position to associate certain order number groups in ANY of the 48 forms with the set-class {0 1 5}. For example, no matter what row form is chosen, if orders no. 0, then 1, then 10 are heard, the three pcs will be equivalent to {0 1 5}. The same is true for order nos. 1, 0, and 10. Thus the ordering of the row is satisfied while a desired unordered set-class is produced.
 

(TOC)

ABOUT MY HOMEMADE EXERCISE MACHINE
 I have constructed an exercise machine out of very simple, inexpensive materials that works very well for me in upper-body development/maintenance. It has two two-by-four pieces of wood about three feet long which are nailed to a flat board that is about 3 ft. square. The two pieces, which I will call the "arms," project from the board about two feet and run parallel, about one and one-half feet from each other. This assembly is placed on a pile of strong boxes which rest against a wall. The pile comes to about shoulder height and the arms point away from the wall. On top of the assembly is another pile of boxes, one of which contains weights. This holds the assembly in place. Two poles, resting on the floor, are wedged under the projecting arms to resist downward force on the arms. Two holes are drilled through the outward ends of each of the arms and through them are placed U-shaped rings that are about 3 inches wide so they extend downward under the arms. These have screws at their ends so that they can be fastened by bolts on the other side of the arm.

The weights to be moved during exercise are placed on a weight bar--I use five pounds on each side. Nylon cords are tied at one end near the weights on the bar and run up to, behind, and through the U-shaped rings. A single cord is used for each arm. The cord is run through holes drilled near the ends of a tubular piece of wood. The wood, which serves as a handle, is about 4 inches long and an inch wide. The cord, which goes "in" one hole and "out" the other, is led back through the ring and down to be fastened to the bar near the other end of the cord.
 Thus, the machine consists of a support that allows weights on a bar to be raised and lowered by means of cords attached at one end to the bar and that run through two handles. The amount of weight, of course, can be changed at will. Other factors, such as the height of the support and the lengths of the cords can be changed to suit the desires of the user.
 EXERCISING
The machine allows gentle and smooth resisted pulling and pushing in almost any direction while the user is standing (it ought to be possible also to sit while exercising--I haven't tried it). My routine consists of the following:
 DAY 1: Standing very near the support and facing it, I lift straight up with my palms toward the wall (8 repetitions), then, bending down to accomodate the length of cord, I push down (avoiding collisions between my hands and the rising bar) 8 times, then standing up again, lift 8 times with palms facing me and finally, bending down, push down with palms facing inward. I then repeat the routine. I vary the routine by standing with my back to the machine and lifting with palms facing away from me. This is like the overhead press using the bar directly.
 DAY 2: 1. Facing the machine and about three feet away, I pull the handles all the way into my body, keeping palms down--ten repetitions with my left foot forward and ten with my right foot forward.
Interspersed with these lifts is some chinning (not on the machine!), then some pulls to the side,using the machine: facing it, about two feet from it, I hold the handles with palms facing each other and twist my whole body to one side as far as I can 8 times, then to the other side 8 times. More chinning, then back to the twist for 5 repetitions on each side.

2.Turning around, my back to the machine, I push away from my body horizontally as before (10 with left foot forward, 10 with right foot). I vary this by pushing with one hand at a time.
3.I stand closer to the machine, facing it, and pull outward with my hands moving in opposite directions, then inward. The hands cross over/under--I change the hand that goes over each time. 15 repetitions out and in. I vary this by pulling with one hand at a time.
4.I stand a little further away, with palms down, and alternately pull down and toward my sides and lift up and over my shoulders. This is varied by one hand at a time and by pulling down with one hand while pulling up with the other.
Having gone through these routines, I repeat them, this time with palms facing in the direction opposite to the earlier one. In addition, for the "facing-away" portion of 1. I vary by holding my arms out straight to my side at various starting positions, varying from straight down to straight up, and pull forward, like a baseball pitcher using both hands (one hand at a time is also good).
After these lifts I do some chinning (not on the machine!), interspersed with some pulls to the side,using the machine: facing it, about two feet from it, I hold the handles with palms facing each other and twist my whole body to one side as far as I can 8 times, then to the other side 8 times. More chinning, then back to the twist for 5 repetitions on each side. (I finish this session with some lifting directly with another weight bar.)
  TOC

ORCHESTRATION ANALYSES
Analysis #1
RIMSKY-KORSAKOV, SCHEHERAZADE,  mm. 1-13
The opening 6 measures is a "unison" passage that descends. Its highest pitch is E4, which makes it dark, and it  uses all instruments that can play the soprano notes of the line except the horns and trumpets. The horns and trumpets are omitted after the first note possibly  because they would not in  be in their best playing registers.
The line of the first 6 measures is doubled in three octaves. The tuba leaves the lowest octave in m. 2, again, possibly because the composer thought that, if it continued, it would not sound its best that low, and that it would sound more effective in the middle octave from that point.
In m. 3 the trill is not given to to the trombones, tuba, or bass. It could have been given to the tuba and bass, but Rimsky might have thought their trills would be unclear. In the two scores I've seen, the marking for the clarinets shows a trill sign above the staff only, which could mean that the second clarinet does not trill. I assume that it does. The effect of some instruments trilling while others do not is, to my ear, a bit muddy or "messy."
The violins do not play the last note before the GP, of course, because they can't. This is not noticed because there enough instruments--especially the trombones--that do play it with strength.
After the GP, the trombones are omitted, with a considerably softening effect that is more effectively achieved than a simple "diminuendo" mark would be, and it enforces a change of color. The low clarinets emerge here.
The next passage is a woodwind group that sets a simple, chorus-like phrase in a very non-straightforward manner. This passage surely was inspired by the beginning of Mendelssohn's Midsummer Night's Dream overture, which it resembles in obvious ways.
M. 8:  The middle register clarinets enclose the low register flutes.
M. 9:  fl. 1 crosses over cl. 1 to continue the soprano while cl. 1 crosses over both flutes to take the tenor. The clarinets are in their weakest register and the vibrato of the flutes (as opposed to the non-vibrato of the clarinets) gives greater warmth to the passage. Otherwise, the change in color from clarinet to flute in the soprano is hardly noticeable because of their respective registers.
M. 10:  As the texture expands in contrary motion of the outer parts, bn. 1 enters on a relatively weak G4, taking over the tenor line from cl. 1 and avoiding parallel fifths in the clarinets. Entering in the middle of the texture, its new sound (double reed) is less noticeable than it would be as the bass. (In the Mendelssohn overture, instruments are added with each chord, and always BELOW the previous notes.)
M. 11:   Here is a considerable change in scoring, one that matches the change in tonal implication (the F# major triad, which implies V in B minor, is surprising after the modal E minor beginning): fl. 1 drops out (if it  continued, it might be superflous?) and is replaced by ob. 2 while ob. 1 doubles fl.1 (the first unison doubling). It is in this pitch area that the flutes and oboes match most in quality and strength; thus the change in color is the least it could be for these instruments. But the passage becomes suddenly more intense with the introduction of three more double reeds--bn. 2 joins in, like bn. 1, as an inner voice to minimize intrusiveness, while cl. 2 continues as a warm and increasingly resonant bass.
M. 12:  In the final chord the flutes double the oboes an octave higher in the soprano and alto. Along with that, the cl. 1 suddenly jumps up over the oboes and flutes (of m. 11) to double the 2nd alto note A while cl. 2 likewise crosses the bassoons to double the bass an octave higher. The bassoons unite on the bass note that "should have been given to cl. 2," while hn. 1 enters to take the tenor that "should have been given to bn. 1," again, disguising its presence by being surrounded, pitch-wise. The register of the horn is borderline in terms of balance--very much higher, and it would become rather intense.
The final chord completes the gradual increase in the number of instruments while keeping the spacing between instruments vertically and reinforces the implied crescendo by the increased strength of the flutes and cl. 1, which are now in stronger registers, adding to the greater intensity of more double reed instruments. The manipulation of the lower instruments avoids parallel fifths in every chord change! The sudden octave doubling of the soprano prepares the surprising and effective entrance of the solo violin on E6, which, with the harp (playing the whole chord an octave lower), suggests an abrupt scene change  to the heroine, singing to the harp! The crossings of instrumental lines are less noticeable than they might be in other scorings because the instruments are all woodwinds (including horn?) and mostly in inner parts and because the overall voice-leading (irrespective of scoring choices) is mostly stepwise.
Copyright c by Merton Shatzkin 1999
 TOC
   ANALYSIS #2
          WAGNER, TRISTAN UND ISOLDE, mm. 1-25
   This excerpt illustrates two techniques very well--highlighting (using one or more instrument(s) to double only one or a few notes of an ongoing line in other instruments) and linking (the transfer of a line from one instrument to another).
mm. 1-2:  The vc. are in their strongest and most singing register. The tempo is probably too slow for the slur to be executed literally--the players probably need to change bow--I recommend that they change in different places, to retain the effect of the slur.
mm. 2-3:  The e.h. links from the vc. on D#4. Ob. 1 and the bassoons are highlighted by ob.2 and the clarinets, respectively, creating an emphasis on the downbeat. Each of the four pitches is played by two "instruments" (counting the vc as "one").  After the clarinets and vc drop out, only double reeds are left, which should make a good blend.
mm. 4-7: This is a sequence fo mm. 1-3 with similar format: the oboe and clarinet roles are reversed. This causes a mixture of single-reed (clarinet) and double reeds. The clarinet, now in a stronger register, is more effective in the soprano than it would have been in the first phrase.
mm. 8-11:  The vc. are now quite high and somewhat strained-sounding. The pitch relationship between the three opening phrases is such that the vc. leap up to the first note of the previous woodwind phrase: e.g., to G#4 in m. 5, which retakes that note in m.2 and to B4 in m. 8, which retakes that note in m. 6, thus creating a sort of delayed, or implied, linking.
This phrase is another quasi-sequence, but with different harmony. The instrumental assignments are the same as in mm. 1-3, with the addition of hn. 2, which (together with cl. 2, which is in a weak register) highlights C4.
We see that in m. 10 cl. 1 highlights the soprano ob. 1, whereas in m. 2 ob. 2 is used for the same purpose. This "change" is possibly because the clarinet is now in a stronger register.
mm. 12-13:  This is an echo, 8va, of mm. 10-11. The phrases, from the beginning, exhibit an alternation in the sopranos of double-reed and non-double-reed (oboe, clarinet,oboe, flute). Fl. 1 is in a brighter register than it would have been for any of sopranos of the previous phrases. Now, Wagner apparently did not want emphasis on the downbeat--there is no highlighting.
mm. 14:  The violins (reinforced by the lower octave) echo the soprano of m. 13   and m. 11. This resumes the alternation between woodwinds and strings.
m. 15:  The alternation and echoing continues. The upper octave is given to only two flutes while there are two clarinets and e.h. on the lower octave. For balance purposes, the stronger register of the flutes requires more instruments below since these instruments are in weaker registers. Also, including e. h. in the mix retains the instrumental sound of the previous woodwind phrase.
mm. 16-17:  The adding of instruments helps the notated crescendo from m. 14. The lower-string pizzicato highlights the notes marked "sf", and is resonantly full, covering many notes of the harmony, stretching from the bass to just below the soprano,  A secondary line, G#4-B4-A4, important because it recalls those notes (in rearranged order) in mm, 2 and 3, is brought out by doubling in ob. 1 and hn. 3, both in strong registers.
The surprising F major triad ("deceptive" resolution of V) in m. 17 is emphasized also by fuller texture, but evaporates when the vc. return melodically, taking the violins' last note an octave lower.  The bass clarinet, unobtrusively entering, takes the bass line of a sustained woodwind group.
mm. 18-21:  The vc. are accompanied by woodwinds continuing from mm. 16-17, now articulated by the addition of vla.-cb. pizzes (this resumes the pizz. of m. 16 in a softer manner). In m. 20 hn. 3 holds the G5 that the vc. have arrived at while the vc. move away and back to it while hn. 4 reinforces bn. 2 below.
The vc. are in the same register as they were in mm. 1-10 and again rise sequentially, again reaching Bb 5 as part of the climax (as they did in m. 9). In m. 21, on the surprising Neapolitan sixth chord (beat two) on a hushed "p" marking, the pizzes are replaced by sustained arco and the warmer-sounding strings replace the bassoons.
In m. 22, second beat, the vn. 2 cross over the vc. to take over the soprano, having entered as if they were continuing the viola part. They play on the G string with its darker, huskier sound--otherwise, they might have sounded thin after the vc. In m. 23 vn. 1 takes over the line, linking from vn. 2's D#4, while the vc. move to a unison C#4 with vn. 2 and then drop a seventh, anticipating the same notes an octave higher they will play when they resume the lead in m. 25. Vn. 1 move between the same notes in reverse order (from D#4 up to C#5) as they ascend, with increasing intensity and straining on the G string.  At the high point, after the horns and bassoons come in to fill out the lower part of the texture, the vc. take over the line again, linking with the vn. 1.
Copyright c by Merton Shatzkin  1999
 TOC