Movement I - Measures 71 to 74

Visually, measures 71 to 74 look similar to the preceding six measures.

There are, however, several differences:

(1) The harmonic motion speeds up. Previously, the harmony changed every two measures. Now it changes every measure.

(2) The previous passage, in B-flat minor, was harmonically static.This passage modulates, reaching D-flat major on the downbeat of measure 75.

(3) The melody is transformed. In the previous passage, we heard the same melody three times. In measures 71 to 72 that melody is (roughly) inverted. And in measure 74 it is interrupted and replaced with a new syncopated figure.

In the piano version, Beethoven retains the same accompanying figure in these measures that he used in the previous measures. The only clue the accompaniment offers that something different is going on is the octave drop in measure 74, which, along with the new syncopated melody, dramatizes the arrival in D-flat major.

In the string version, however, Beethoven helps us hear the newness of this material by changing the accompaniment. Had he followed the same pattern as in the previous section, we would have this:

Because the harmony now changes every measure instead of every two measures, this scheme doesn't work well. It sounds strange to arpeggiate the harmonies in the odd measures but not the even measures. Beethoven might have added arpeggios to the even measures as well. But he preferred to dial things back instead. He drops the arpeggios altogether and has the cello join the middle strings in their tremolo. (As we shall see next week, dialing things back now gives him the chance ratchet things up more effectively later on. This is a technique known to all good composers and horror film directors.)

Even though the cello joins in the tremolo, it retains its individuality by playing eighth notes to the middle strings' sixteenth notes. It also employs the device Beethoven has used previously to accentuate the downbeat: an octave dip on the first note of each tremolo.

Note Beethoven changes the point where the bass drops down an octave to the low A-flat. In the piano version, this happens at the beginning of measure 74. In the string version, it happens three beats earlier. Why? Presumably, it is because the very climax of the passage has changed. In the piano version, the climax is on the sforzando in measure 74. In the string version, Beethoven eliminates this sforzando and marks a forte at the beginning of measure 73, shifting the climax to this point. In addition, he lands emphatically on the final cadence, specifying a forte-piano on the downbeat of measure 75. In the piano version, this measure is marked piano, so as not to upstage the sforzando of the previous measure. (These quarter notes aren't in the real score, by the way. I added them to avoid ending the sound sample on an unresolved dominant.)

What are these changes all about? Beethoven is being sensitive to the realities of string writing. A pianist can land forcefully and effectively on that high G-flat. But this would not work in strings. The violin's sound is too thin in that register to produce a convincing sforzando with no assistance from the accompaniment. The downbeat of measure 72, assisted by the change of harmony and a low A-flat in the cello, is a more natural place for the climax.

This is the kind of change an arranger might be afraid to make. How can you move the climax of the phrase? Isn't that tampering too much with the original? But Beethoven, who doesn't have to worry about offending himself, has no qualms about making whatever changes the new medium requires. Fidelity to the original is desirable. But it's more important that the music sound good.

Movement I - Measures 65 to 70

As I mentioned in the last post, this passage is particularly pianistic:

A literal transcription is virtually impossible at this speed, so I changed the arpeggios to something more string-friendly:

Beethoven does something quite different in his arrangement. Instead of arpeggios, he uses sixteenth-note tremolos in the middle strings. To keep this from getting boring, he adds a new countermelody in the cello. This countermelody takes up the arpeggio idea from the piano version, but the arpeggios are now staccato eighth notes and span two octaves. The countermelody provides the motion when the melody has a long note but then drops out on each even measure when the melody is moving. This scheme results in a cleaner, less busy sound.

Beethoven also changes the dynamics. The piano score contains a crescendo in the last two measures. (A short-lived crescendo. The very next measure, measure 71, will be marked piano.) Beethoven drops the crescendo in the string score and adds a forte-piano to the first beat of each odd measure. The forte-pianos add to the excitement, and it is the tremolos that makes them possible. Forte-pianos would sound very strange in the piano version.

Beethoven begins this passage with a dramatic leap in the second violin, which helps to differentiate this new section. Beethoven went out of his way to make this leap a large one. We noted last week that he gratuitously widened the spread between the second violin and viola at the end of measure 64. This widening sets up the high D-flat in measure 65. A more natural voice-leading would have had the violin leaping, less dramatically, from a B-flat.

Why does Beethoven invert the voicing in the tremolo chord, giving the second violin the B-flat and the viola the D-flat? Perhaps one reason is to make the leap in the second violin that much larger. But I suspect the main reason is he simply didn't want the viola to play the same note on the downbeat as the in rest of the measure. He did something similar in the first measure of the piece, scoring it as

rather than

This movement, in each case, ensures that we hear the chord on the downbeat as an entity distinct from the ensuing tremolos.

Movement I - Measures 61 to 64

The development section begins with a modulation to B-flat minor:

The most curious note in this passage is the C on the downbeat of measure 63. Every other melody note is doubled at the octave (unless the left hand is in the way). Here, Beethoven unexpectedly tosses in a tritone. Clearly he wants to draw attention to the modulation. We feel a jolt at this point that we would not feel if he had simply doubled the G-flat at the octave.

The literal transcription below is unsatisfactory for a number reasons. For one thing, the texture is too thick in the lower register. For another, we don't feel the aforementioned jolt on the downbeat of measure 63. The lone C in the second violin doesn't do the job.

Beethoven, in his arrangement, fixes both these problems--and others as well:

As expected, Beethoven adds a chord on the downbeat of measure 61. His reasons for doing this in measure one still apply. But he now has the addition reason of voice-leading. He must resolve the dominant chord at the end of measure 60.

Beethoven also adds a chord to the downbeat of measure 63. This not only continues the pattern begun two measures earlier. But, more importantly, it supplies the jolt that is missing in the transcription. He draws attention to this chord by tying the G-flat in the melody. It is the change in harmony that is important here, not the melody.

After the downbeat, the lower strings leap up an octave. This leap continues the pattern from measure 61. But, conveniently, it serves two other purposes as well: (1) It brings the lower strings closer to the first violin, avoiding the muddy texture of the transcription. (2) It gives Beethoven the chance to have the bass descend over the next two measures as the melody ascends. The widening wedge to support a crescendo is a device we already encountered in measures 50-56.

In the piano, the device used to support the crescendo was a thickened chord texture (which was true in measures 50-56 as well). The wedge makes that unnecessary, and Beethoven sticks to four-note chords. In the first chord, he drops the seventh. In the remaining chords, he simply avoids doubling the melody. This means dropping the E-flat at the beginning of measure 64. Since the E-flat is in the bass, dropping the E-flat means choosing a new bass note, which fits right in with the widening-wedge plan. Look how nicely that all worked out.

The final chord could have been voiced differently. Beethoven might have retained the A in the second violin and had the viola drop down to a C. That, in fact, would be more typical voice leading. But Beethoven apparently wanted to extend the wedge device to the inner voices as well.

I made a suggestion a few weeks ago that any composers reading this blog should try to do their own arrangements before reading about what Beethoven did. That applies especially to next week's passage. It is highly pianistic, so a literal translation is virtually impossible. I'm curious how other composers might try to arrange this passage. So, if you feel so inclined, feel free to post your solution in a comment either on the blogspot.com page or on the associated Facebook page. I doubt you will come up with the same solution as Beethoven. But that's not necessarily wrong. There is a variety of possible solutions, and I'm sure that among us we can up with several ways to achieve the desired effect.

This is the passage:

Movement I - Measures 57 to 61

Measures 57 to 60 modulate back to the tonic for the repeat of the exposition:

In the transcription, I have, of necessity, eliminated the low Cs in the accompaniment:

I brought the first violin in on measure 59 to handle the extra chord tones. That's not an entirely satisfactory solution. The return of the main theme in measure 61 would be more dramatic if the first violin were tacit until that point. Perhaps it would be better to handle the extra chord tones with double stops in the viola? I'm curious to see how Beethoven solves this problem.

Before we get to his solution, however, let's take a look at how he arranges measure 57 and 58:

Beethoven eliminates the opening eighth note rest and has all four instruments play on the downbeat. Recall that he did something similar in the first measure of this movement.

became

In measure one, Beethoven voiced the first chord differently than in the rest of the measure, which helped to articulate the downbeat. In measure 57, however, Beethoven voices the chord the same way. So the effect is different here than it was in measure one. We don't feel the downbeat so strongly, and the passage feels more like a transition than a beginning, as it should.

If accentuating the downbeat is not Beethoven's purpose in making this change, then what is his purpose? I suspect he was simply giving the players a helping hand. A slight retard was implicit in the previous measure, so it will be difficult for the players to get back in tempo unless they begin this measure together. This is why Beethoven adds the apparently extraneous quarter-note C in the first violin part. The players will look to the first violin for their cue, so Beethoven must give him something to play.

So how does Beethoven solve the aforementioned problem? That is, how does he fill out the chords in the accompaniment without destroying the dramatic effect of the return of the main theme? He borrows a trick from Mozart.

In the retransition in the first movement of Mozart's Clarinet Quintet (a work we've already compared to this quartet in a different context), Mozart has the clarinet hold an E for three measures while the strings modulate back to A major. Only when we reach A major on the downbeat of the fourth measure do we realize that this E we've been hearing in the clarinet is actually the first note of the primary theme. The recapitulation has actually already begun; we just didn't know it yet. The clarinet began the theme by itself, then waited for everyone else to catch up. It's one of my favorite moments in all of Mozart:

(Score in concert pitch)

It wouldn't surprise me to learn that Beethoven had the Clarinet Quintet in mind when he scored this passage. Like Mozart, he has the soprano instrument begin the primary theme three measures early, then wait for everyone else to finish the modulation. But, instead of just holding the note for three measures, he elaborates it with a couple of turns.

These turns serve two purposes:

(1) They hark back to the secondary theme. The secondary theme began with the same opening gesture as the primary theme--a dominant rising to tonic--but with a turn thrown in:

This transition is an elongation of that same idea.

(2) They fill out the harmony in the accompaniment. The two turns are not identical. Beethoven manipulates them so that, on the first and last of the three eighth-note chords, the first violin plays the precise chord tones that are missing in the lower strings.

So Beethoven fits in the extra chord tones not by adding the first violin to the accompaniment as I did in the transcription nor by using double stops as I speculated he might do. Instead, he slyly incorporates the extra chord tones into a new melody that prepares us for the return the primary theme. All in all, it's quite an elegant solution.

Movement I - Measures 50 to 56

This passage consists of three phrases, rising from piano to fortissimo. In the piano version, Beethoven supports this crescendo by thickening the texture. He begins with four-part harmony and finishes with full fortissimo chords.

Since these chords aren't practical for string quartet, I've retained four-part writing throughout in my transcription:

In his string arrangement, Beethoven uses a different technique to support the crescendo: He gradually increases the spacing. The parts begin in the same octave and finish with the bass and soprano more than three octaves apart.

In phrase one, Beethoven raises both the bass and tenor one octave, placing the viola above the second violin. In the second measure, this close spacing crowds the viola out. There is no good note for the viola to play on the first beat of this measure. A D, as in the piano version, would anticipate the resolution of the melody (which will sound worse in strings than it does on the piano). Doubling the cello's F would be fine at the octave (as in my transcription). But a unison doubling on a note that screams for a strong vibrato is not a good idea. So Beethoven omits the viola part altogether and switches to a three-voice texture.

On the cadence, Beethoven has the cello drop down an octave for the second G. Repeating the high G would not work, since the final chord would not be in root position. This octave drop supplies Beethoven with a new motive, which he duly exploits. He gives the cello another octave drop at the end of phrase two and an octave rise at the end of phase three. This octave rise turns the last two notes of this passage into an echo. Very clever. Beethoven takes an idea born of necessity and turns into a unifying feature.

The final change Beethoven makes to phrase one is to add a sforzando. Perhaps he thought the six-voice chord gave the climax sufficient weight in the original. In the string version, the number of voices is actually reduced in the climax, so the sforzando becomes necessary.

Phrase two returns to four-voice writing, with the bass and tenor again raised an octave. This time, however, the second violin takes the raised tenor part (starting on beat three) and the viola takes the alto. Why didn't Beethoven do this is the first phrase as well? Apparently he prefers to give the second violin whichever part moves in similar motion to the first violin. In the first phrase, that's the alto part; in the second phrase, it's the tenor.

In phrase three, both the tenor and bass drop down to their normal register. (The new octave-drop motive now serves yet another function: preparing this register change.) The second violin doubles the melody an octave below, leaving the viola to cover both inner voices with double stops. The combination of the octave doubling of the melody and the lowered bass effects the grandeur supplied by the chords in the piano version.

At the cadence, Beethoven adds a retardation in the viola part so that it moves in thirds with the melody.This is one of those changes that are obvious in retrospect. Beethoven is simply paying more attention to good voice leading in the string version than he did in the piano version. Note the crisp effect of having staccato in the bass while the other three voices play legato.

The final measure puzzled me at first. Beethoven chose to fill out the final chord by adding a D. But why give the D to the second violin? Beethoven went out of his way to cross the parts, so there must be a reason. The reason, presumably, is that the B is the "melody." But so what? We are going to hear the higher note as the melody no matter which violin plays it. Then it occurred to me. Yes, we will hear the D as the melody. But Beethoven wasn't writing for us. He produced this arrangement for amateur string players who get together on Sunday afternoons to play through some popular pieces. The performers themselves will be aware that the B is being played by the first violin. So I suspect they will hear the B as the melody.

At least that's my guess. Another possibility is that the D was an afterthought.  Perhaps originally the second violin was silent on this chord. Then Beethoven decided to add a D. Rather then scratch out the first violin part and rewrite it, he just gave the D to the second violin. As someone who learned to compose before music notation software and actually worked with dip pens and onion skin, I can attest to the temptation to do such a thing. But, if you look at some of Beethoven's manuscripts, you can see he wasn't shy about crossing things out. So I'm not sure that explanation holds up.

Movement I - Measures 46 to 50

Measure 46 begins the cadential material:

 

Here is Beethoven's arrangement:

The most noticeable change on paper is probably the least notable change to the ear, namely replacing the cello's triplets with a sixteenth note followed by two thirty-second notes. At this speed, it's hard to see how it matters. It's more of a visual cue for the performer. Beethoven probably just wants to ensure that the cellist doesn't rush the first note so that we hear the added sforzando.

Adding a staccato to downbeats for both the cello and viola is to be expected by now. Beethoven has been making changes like this all along to ensure a crisp downbeat. What is perhaps not expected is his having the viola repeat the C--with a sforzando no less--on beat two of each measure. This is the change we hear most readily when listening to the two versions. The transcription, with its sforzando only on beat three, sounds almost march-like. In Beethoven's arrangment, the round robin of sforzandi on beats two, three, and four and half, keeps the passage from sounding square.

This is a shorter passage than usual for one of these posts. So I'll take this opportunity to make a few general comments about this blog.

Two of the earliest influences in my education as a composer were, strange as it may seem, Edgar Allen Poe and Benjamin Franklin. Poe will have to wait for another time. Franklin's influence comes from this excerpt from his autobiography where he discusses how he taught himself to write prose:

"About this time I met with an odd volume of the Spectator.... I bought it, read it over and over, and was much delighted with it. I thought the writing excellent, and wished, if possible, to imitate it. With this view I took some of the papers, and, making short hints of the sentiment in each sentence, laid them by a few days, and then, without looking at the book, try’d to compleat the papers again, by expressing each hinted sentiment at length, and as fully as it had been expressed before, in any suitable words that should come to hand. Then I compared my Spectator with the original, discovered some of my faults, and corrected them."

I ran across this passage when I was in college. It seemed Franklin's technique should work in the field of music composition as well, so I decided to give it a try. I began with the Bach chorales. I would take a chorale melody, harmonize it, then look at how Bach had harmonized it. Then I would slap my forehead and say, "Why didn't I think of that?" I improved as the project progressed. By the time I finished, I was "thinking of that" more often than not.

Much later, when DAWs and good orchestral samples became available, I tried the same technique with orchestration. I took a piano score of The Magic Flute and orchestrated sections of it. Then I listened to my realizations and compared them to Mozart's.

This method of learning was clearly better than simply analyzing a score. It's only when you try to do something yourself that you get a perspective on what the problems are. If you look at a score without first tackling the problems yourself, all the choices the composer made seem pretty obvious.

As you might have guessed, this project is another of that type. Each week, I try my own hand at a musically satisfying arrangement of the passage I analyze. So, while I show you only two SoundCloud files, I am actually producing three: a literal transcription, my arrangement, and Beethoven's.  At first, I thought I might include my arrangements in the blog. But I decided against it, since they are probably be of no interest to anyone but me.

I recommend that you give this technique a try yourself. Dig out your copy of the Beethoven sonatas and try your own arrangement of the next passage (after transposing it to F major). Next week, you can see whether Beethoven agrees with your choices or not. If for some reason you don't have a copy of the Beethoven sonatas, you can use this one. I predict that, as the weeks go on, you'll become progressively happier with your decisions, even if you don't make the same ones Beethoven did.

Movement I - Measures 38 to 46

I was curious to see how Beethoven handles this passage, since it is particularly pianistic:

Here is what a literal transcription would sound like:

And here is Beethoven's arrangement:

In the piano version, Beethoven alternates between staccato and legato eighth notes in the left hand. In strings, this doesn't work. The legato eighth notes don't sound as energetic as they do on the more percussive piano. So the Beethoven makes all the eighth notes in the accompaniment staccato. In addition, he adds a staccato to the melody on the last note of each phrase. (Although this latter change is advisable in any event, given the first violin's ensuing drop.) 

Beethoven also decides to score the two phrases differently. In the piano version, the phrases are identical except for the cadences. In the string version, he opts for a light, bouncy first phrase and a more dramatic second phrase. This is more typical of the Viennese style, so it is no surprise Beethoven chooses to do this. The real question is why he didn't do something similar in the piano version. Perhaps he thought such contrasts in texture are more important in string music than they are in keyboard music. Or perhaps this is simply another case of his changing his mind.

In the first phrase, Beethoven lightens the texture by raising the eighth notes in the bass an octave (giving them to the second violin) and by eliminating the alto line altogether. He also replaces some of the repeated notes in the cello with rests, specifically on beat three of measure 38 and on the weak beats of measure 41. And he replaces the cello's dotted half note at the end of the first phrase (on the downbeat of measure 42) with a quarter note. The viola pretty much stays out of the way. In the first two measures, it merely adds color by outlining the melody in staccato. In the next two measures, it is silent. All in all, these changes produces a lighter, cleaner sound than we hear in the transcription.

One effect of switching the bass and tenor lines is that it changes the way the phrase is harmonized, as the analysis above shows. We now have root position tonic chords in measures 39 and 40 rather than first inversion tonic chords. This is unusual. One generally shies away from root position tonic chords in the middle of a phrase, especially if that phrase is going to end with an authentic cadence, since they will weaken the cadence's effect. Beethoven solves that problem by tossing in a retardation at the end of the phrase. On the downbeat of measure 42, the second violin arpeggiates a dominant chord over the C in the bass, delaying the tonic until beat three. This is a nice attention to detail. The retardation isn't necessary in the piano version. But here the phrase would sound too static without it.

The descending arpeggio is a new motive, not present in the original. And Beethoven immediately exploits it. In the second half of measure 42, he repeats the motive in orchestral unison to transport the accompaniment down into the second octave.

In the second phrase, Beethoven eliminates the sustained notes in the accompaniment. Both the bass and tenor play the staccato eighth-note figure while the second violin doubles the first violin an octave below. 

To construct the eighth-note pattern for the bass, Beethoven alternates the original bass note with the root or fifth of the chord. In the second half of the phrase, however, Beethoven makes some additional changes. Had he followed the same pattern, the end of the phrase would be as follows:

The first problem with this version is the preponderance of fifths and octaves at the end of the first measure. Beethoven solves this problem by breaking the pattern, having the bass move in thirds with the tenor:

 A second problem is the poor voice leading in the tenor. Essentially, we hear alternating notes of the viola as outlining two separate parts. If we write out these two parts, we can see the voice-leading problems more clearly.

The B-flat at the end of the first measure wants to resolve to an A. Instead, it leaps upward to a D while the lower voice moves to an A. Similarly, the F at the end of the phrase wants to resolve to an E. So Beethoven raises the last beat of the first measure up an octave and has the line descend. Each dissonance now resolves naturally: