Piano World Home Page Forums Our Most Popular Forums Adult Beginners Forum Classical Sonata Analysis

Aw, shoot, and I was going to have fun with adding hypothetical roots all over the place .


See, I don't know what VIIdim to I sounds like, or V7 to I. What I mean is, they both probably sound familiar to me, but not in a way where I can identify "this sounded like V7 to I, but that sounded like V to I, and that other thing sounded like VIIdim to I" and so on. I doubt I can even hear "this sounded different from that other one I heard 8 measures ago" or "these sound the same, just different voicings; while these other ones sound different."

The aural analysis I can do is depressingly limited compared to the paper analysis I can do.

Since I believe that playing and hearing are dynamically connected, I still believe that there are many people who have potentially strong aural skills that remain undeveloped because of a combination of:

1) Fear, anxiety, self-consciousness.
2) Very very VERY strong reading skills that are not expanded to writing music.

The reason: any kind of composition requires a different focus. There is not longer right and wrong, just what works, or what does not. Once you brain kicks into creative mode, it is free to find things, and new pathways are formed. So if you come up with a tune, or some nice changes, or both, the process of discovering them and/or writing them down just changes the way you hear.

1) check.
2) check.

Guilty as charged .

Composition: I like that idea. A lot. Thank you.

I would also be interested to understand more about this, PS88. To be honest, there has been so much new information for me to absorb that I have not given much consideration to RN analysis. Know what it is/does, but that is about it.

There has been a lot to say about this movement. Great stuff. Also, great to see LadyChen joining the discussion, and hope still with us.

Moving right along: Hope not too basic for some here, but this is more on what I have to say about the recap. before we move along to next movement;

M48-M51 = M1-M4
M52-M55 = M13-M16 (2/3 of M16)

M56-M57 - I believe this is coming from development, but can not say for sure where.

M58-M60 = M17-M19
M61-M62 = M25-M26
M63 = M22
M64 = M20
M65-M67 = M20-M21
M69 = M28 inverted
M70-M71 = M29-M30

By this accounting, I still have the following measures unaccounted for from exposition:

M5-M12 (M9-M12=M1-4) so just leaves M5-M8
M23-M24 (same as M20-M21)

I'm still here! I work during the day, so I won't be doing any score analysis during the week, but on the weekends I'll be able to participate more with a score in front of me. For now, I'm enjoying the conversation!

OK, here's how I normally think about diminished chords. I'm going to present this in a few separate posts, leading up to dim7 chords at the end. This is essentially how I learned it in my music theory class. I've learned a lot of other stuff since then, but these form my basics.

This assumes that you know major and minor scales (or how to figure them out) and that you're comfortable with naming intervals and chords. If you have questions about those, please ask. I'm just pointing out the assumption because I'm going to start this series of posts assuming scales, intervals, and chord names, and then backtrack if needed.

Chords in a major scale

Take a major scale, and form the triad and seventh chords on each note in the scale, just using notes in the scale. For example, working in the key of F major:

Triads: F, Gm, Am, Bb, C, Dm, Edim
Sevenths: Fmaj7, Gm7, Am7, Bbmaj7, C7, Dm7, Em7b5

This pattern of types of chords is exactly the same in every major scale. For example, in A major:

Triads: A, Bm, C#m, D, E, F#m, G#dim
Sevenths: Amaj7, Bm7, C#m7, Dmaj7, E7, F#m7, G#m7b5

If we replace the letter names with a roman numeral for where the note comes in the scale, we get a general pattern that can summarize this information for all major scales:

Triads: I, IIm, IIIm, IV, V, VIm, VIIdim
Sevenths: Imaj7, IIm7, IIIm7, IVmaj7, V7, VIm7, VIIm7b5

Notice some patterns:

The only major triads are I, IV, and V.

The only dominant seventh chord is V7. (A "dominant seventh" chord is a major triad with a minor seventh added.)

The only diminished triad is VIIdim.

The only half-diminished chord is VIIm7b5. (A "half-diminished" chord is a diminished triad (root, minor third, diminished fifth) with a minor seventh added. We'll meet its cousin the "fully-diminished" chord when we get to minor scales.)

Exercises that may help:

A. Try this out in various keys. Build the chords at the piano, listen to them, name them. Do it in at least two ways: (1) pick a key. Build (play/listen/name) all the triads and seventh chords. (2) pick a roman numeral and build the corresponding triad and seventh chord in all major keys (for example, pick VII, and build VIIdim and VIIm7b5 in all major keys. Or at least several keys.)

B. Take some music in a major key, and identify a section without accidentals. Label all the chords, and see how often they fit into the pattern of one of the triads or types of seventh chords described here.

Good to know. I actually work during the day too. But lately, as little as possible

Well, isn't that clever.

So, these are the only chords that belong in F Major? We will still allow, 9, 11, 13 and + - iterations?

I should say that my music theory course was based on tonal music of the Baroque and Classical eras, and the examples we used were generally cherry-picked to show certain kinds of examples of the use of harmony. So what I'm laying out is not a complete story for all kinds of music, and not even the complete story for Baroque and Classical music, and not even the complete story of everything I learned in the course. Primarily, this is the material we focussed on as fundamentals in the first three weeks of the course, to prepare us for the rest of the semester.

OK, on to the next part.

Resolution in a major key

After analysing a lot of music, we started to notice very frequently a V-I or V7-I progression, or a IIm-V7-I progression, or even longer chains, for example IIIm-VIm-IIm-V7-I. All of these have the chord roots moving down by a fifth ("down" logically speaking: as actually used, the roots might move down a fifth, or up a fourth, or down a twelfth, etc.) For example, IIIm-VIm-IIm-V7-I in the key of F major is Am Dm Gm C7 F.

I want to look particularly at C7 to F. C7 (C E G Bb) contains a tritone: E-Bb. This is an unstable and dissonant interval (at least, I'm told that it sounds unstable: sometimes it sounds dissonant to me, and sometimes it just sounds interesting), and when moving between chords from C7 to F, it resolves "inwards": from E-Bb to F-A. What I understand is that this is part of why the V7-I resolution sounds so satisfying: because of the dissonant tritone moving by half-steps to the nice consonant major third in the tonic chord.

Slight change of pace: Let's look at Edim. Edim (E G Bb) contains that same tritone: E-Bb. So moving from Edim to F should provide a similar sense of relaxation or arrival of the dissonant tritone to the major third in the tonic chord. So VIIdim to I might be something we would also expect to see.

Similarly with the half-diminished chord, Em7b5 (E G Bb D). That also contains the tritone E-Bb, and again we might expect that Em7b5 to F should give a sense of relaxation or arrival. This is VIIm7b5 to I.

I don't normally do the following, but it has come up, and others do use it, so I'll mention it here: If we happen to decide to look at the chord C9 (C E G Bb D), we will see all the notes of C7, Edim, and Em7b5 in there. So we can call Edim a rootless C7 chord, and Em7b5 a rootless C9 chord. Then we can see VIIdim to I as a type of V7 to I, just with a rootless V7. And we can see VIIm7b5 to I as a type of V7, just with a rootless V7 that's been jazzed up to a rootless V9.

Without using accidentals, there is only one tritone in a major key: between the fourth and seventh notes of the scale. For example, in F major, between Bb and E. A tritone inverts to a tritone, so B-Eb and Eb-B are both tritones. Because there's only one tritone, this kind of "resolve the dissonant tritone" game can only be played around this one place in the scale. At least, without using extra accidentals.

There are other directions tritones can be resolved to, and other chords they can resolve to. But for now I want to just emphasize this basic pattern of V7-I, VIIdim-I, and VIIm7b5-I.

Exercises which might help:

A. Pick a major key. Try out the V7-I, VIIdim-I, and VII7b5-I progressions. Experiment with different voicings. Listen. Try these in other major keys.

B. Verifying that there's only one tritone in a key: pick a key. Go up the scale playing fourths, using only notes from the scale. E.g. in the key of F, play F-Bb, G-C, A-D, etc.Notice that exactly one of the fourths is not a perfect fourth, but is instead an augmented fourth (a.k.a. tritone). Now go up the scale playing fifths, using only notes from the scale (e.g. in the key of F, play F-C, G-D, etc.) Notice that exactly one of the fifths is not a perfect fifth, but is instead a diminished fifth. Can you hear the difference between the fourths and the tritone? Between the fifths and the tritone?

I'm glad you like it! Well, there are oodles of other chords that can be built. For example Fsus2, F6, C9, Bbmaj7(sus2), etc.

But if you restrict yourself to triads and types of seventh chords with note names skipping a letter, yes, these are the only chords of those types using just notes of a major scale. In other words, these are the only types of three- or four-note chords that can be built stacking major and minor thirds using notes from a major scale.

As far as allowing other types of chords: anything's allowed (although some things are more common than others). And once you allow accidentals, the possibilities skyrocket. I'm laying out here what I learned at the start of learning music theory, that starts out with a confined palette: three- or four-note chords, confined to notes of a major scale. And it is cool that there are only a small number of possibilities.

Good job, Jeff. You're looking more closely now and picking out things you wouldn't have when we started.

The exposition consists of two subjects. The first is in tonic and the second in the dominant. You've got the keys already but didn't highlight the two subjects. The second subject begins at M18 and is the core material for the development.

The development uses four measures of M18 (three obvious and one decorated) then it uses the four-note figure from M15-16 to effect a sequence down from F to D, then repeats those last two measures with a slight expansion then begins the transition from D minor to the dominant preparation passage before beginning the recap'n.

What alerts me to a problem in your analysis is the way the measures on the right of the equation are not in numerical order but jump backwards and forwards.

M48-51 = M1-4, yes.
M52-53 = M13-14 but in Bb (subdominant, not unexpected)
M54-55 = M15 or M16 repeated as a sequence to return to tonic for
M56-57 = functionally equvalent to M16 but repeated, I suspect, to balance the bars
M58-71 = M17-30, bar for bar but with minor modifications.

The minor modifications such as starting with a rising figure instead of a falling one might be more noteworthy in the development (if it leads anywhere) but is just variation/creativity in the recap'n.

Clementi has not just repeated the first subjects in his sonatinas but has continued to practise his craft. He has been more restrained with the second subjects but hasn't packed up early for lunch as he so easily could have.

My next sections leading to "why PS88 loves dim7 chords" will be on chords in a minor key, but it will probably have to wait until tomorrow. Just didn't want you to think major keys are all there is, and wondering what happened to the dim7.

Yes, this was among the top of my list of questions. But, no worries, I can hold out to tomorrow .



So no reversing allowed? I thought that might pose an issue for you. But, just wanted to keep you on your toes.



Not sure what this means. In exposition we were in C Major and in recap'n, at these measures in G minor, I thought. So, where is Bb coming from.



Should have listened to this more, instead of relying on the score.



Yes, indeed. Not sure how/why I missed this.

Not necessarily. But in the recapitulation it's less likely. The recap should be fairly straightforward.


Yes, you did say G minor and I looked briefly at the score and saw Eb and F# but in M52 the bass is F and A so it's not minor at this point, it's only M55 that touches on G minor. It's only transitional anyway. The important thing is that in the exposition it was in F and now it's not so an extra measure, M55, is thrown in to take us back to F - that's the point, and then a repeat of M56 to balance the measures.

The better you get at reading the less you feel inclined to actually go to the piano but when the accidentals start getting into the mix it really is worth sitting down at the keyboard and playing through, not just the printed score, but put the bass in root intervals and simple triads to really get a feel for what the composer is doing. But we all do it, Jeff, so don't feel bad!

When you kick yourself for making a beginner's mistake, you've moved on from being a beginner! You really are making progress.

Movement 2: Moderate (walking) pace; Bb Major

A - M1-M16
B - M17-M26
A - M27-M46

The second A has a closing tag where M41-M42 are repeated. These measures (4 now) are not present in first A, and are coming from B.

The B section has lots of accidentals but nothing with permanence. So, just adding colour I think with no real lasting key change. Also, this B section is coming from development of first movement.

In the A's (first A for example) I would call M1-M4 Antecedent, M5-M8 Consequent, M9-12 Antecedent, M13-M16 Consequent.

On further thought, we could call it A,A,B,A,A with a little bit of B in the last A. This is since, each A is mainly two phrases repeated.

Perhaps we are moving through D Minor in second phrase of B (M21-M26,) but back to Bb major by M27.

About 1st movement...

I have not read this whole thread, so if I am repeating what you have already said, my apologies to everyone.

More than anything else I sense V to I movements, in any key. One “trick” that countless composers use is to take a I chord in a key – in this case F (the chord) at the a tempo – then make it into a dom7 chord. Immediately we hear F7 in M52 just begging to go to Bb, which it does, and we can call it our “new key”, modulation, or secondary dominance, whatever big terms we wish. I call it “just visiting a key”. It’s like a very quick detour, interesting, then the composer goes to a new V and ends up on a new I. So in M55 he dances around a D7, to Gm (another V to I movement), then in M56 with a C7 he takes us right back to F, the key of the movement.

The reason I think this is so important is that in more difficult or more adventuresome music, these continual “jumps” to a new V7 in some other key can go on forever almost. And the only thing that links it all together if we do not use a million rules is that it sounds cool.

Just looking at a score and hearing it in your head is “audiation”, and it’s one of those “fancy words” that I never heard until coming to this forum. You can’t ever predict how good people will get at it, but I know from experience that it never stops developing. I can hear something like this as clearly in my head as if I were playing every note. But sooner or later I run into something so unusual or so complex that I stop hearing it all, and then I have to go to the piano to get the rest of it. When I was young, I didn’t hear anything in my head. It develops with experience and time.

I’ve made several over the last few years while working through music with advanced students. When I scan music, hearing it, and I have never looked at it before, I make the same stupid mistakes I made decades ago. The only thing that has changed is my ratio of correct answers to goofs.

I hear this, without the terms: M1-8 states something, but it ends in a question. M8 ends on F, a V chord.

Then M9-16 are more like an answer, with M16 completing a section.

Important to me would be that M1-4 keeps the Bb tonal center, but M9-12 not only varies 1-4 but also finds a way to end on Cm/Eb. I don’t much care for Roman numerals, but there is a sort of ambiguity between a II and a IV chord, in this manner.

Key of Bb:

I = Bb
IV = Eb, II (or ii or IIm) = Cm
V = F

If we consider II and IV as interchangeable, we tend to see the important chords in any key that a composer uses to weave music around.

When I do this kind of analysis with students – form – I tend to use shapes and diagrams rather than names. I’m interested in how people who use antecedent and consequence would apply them hear.

My gut says 1-8 for antecedent, 9-16 for consequence. But I’m not sure. These are still new terms for me.

I hear it this way:

M21, Edim7b5 to F
M22, “C#dim7” to Dm
M23, Gm, to C7

When I put a chord in quotes, it means for me that a note is missing. There is only C# G Bb, but we can sense that the E could be there – C# E G Bb.

In M24 we expect Bb, but he delays, going to Dm. This would be a deceptive cadence in F major, a V7 chord (C7) going to Dm (VI), then after that he settles down to F, in 26, but then he makes us hear that F chord as a V because he is taking us back to Bb in the next measure.

You are not repeating anything we've said, Gary, but the more you repeat what you've said here the sooner we'll start to see it ourselves.


Did you use the piano to determine this? If not, do so now (or after re-reading the discussion of the development sections of the opening movements of this and the previous sonatina). Again, when in doubt about the key look first at the bass and then at the chords.

Looking back at keystring's use of antecedence/consequence I think that antecedence should end on a dominant chord and the consequence on a tonic. That makes the antecedence M1-8 and the consequence M8-16, and A is consequentially ( ) M1-16. So back to ABA.

Note the use of the subdominant key to tone down the mood in this movement as he did in 1 and 2. In 3 he went to the dominant. Haydn stayed in the one key throughout as Bach did in the suites. (Haydn's sonata was written much earlier.)

The first 8 measures, and more so the filled out version M35-42, have such a strong correlation to the first 8 measures of the Con spirito and the introductory notes of the B phrase (M17) so resemble M7 while the M18 figure is lifted from the second subject at M13. The end of B, M21-26, is reminiscent of the transition phase of the development, M38-47.

In the reprise of A he adds decoration and splits the thirds, etc. into separate notes. The final bars rely on the M17 figure leading back to the M15/16 cadence.

I started this before I went off for breakfast. I see, Gary, you have since posted (it was a long breakfast ). There is some overlap.

Hadn't ... will.



Although I struggle reading notes this high, I believe I have them right now.

M21, Edim7b5?

I'd call it Bbm/E. Why would you need to flat the five in a dim7? Notes I am playing here are:

Bass: E
RH: Bb (root), Db (minor 3rd)

Although, I agree that it sounds diminished. Perhaps ... Eb7b9/E? Actually, I like Edim7 best now ...

I'm cool with the F

also fine with C#dim7 to Dm

Groovy with the Gm but over Bb ... and then to C7