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

More thoughts on this:

I've been using traditional Roman numeral analysis on the Clementi sonatinas, including slash notation for secondary dominants, and multiple slash notation if needed. This has sufficed for just about almost everything in the sonatinas; perhaps for everything. I think that's useful to notice, and it may be harder to notice this if you (generic you) only use letter names for chords (although if you're using letter names but also keeping the knowledge in your head of what key you're in and where you are in that key and how it relates to the original key, then you have the same knowledge that writing Roman numerals shows me).

For example I know really clearly and explicitly that most of Clementi's harmonies lie within the realm of ii V7 I (or i) and vii° and vii°7 (or their half-diminished relatives) additions, with occasional forays to vi ii V7 I and even rarer forays to iii vi ii V7 I. That's from writing out and seeing pages and pages and pages of ii and V7 and I and vii°7 and viiø, and rare vi and even rarer iii. (Even IV is, perhaps surprisingly, pretty rare.) If I'd written everything out just in letters, I would have to be making extra implicit mental calculations to realize that same phenomenon. Even when a chord is not a diatonic ii, V or I, it's almost always a ii, V or I in a closely related key, and the progressions proceed in very predicatable ways, that is, the chords aren't just popped in, but are used in very typical relations.

Incidentally, this is related to why I don't like just saying of the chromatics in Clementi that they're used for colour. They may be being used for colour, but they're almost always, if not always (I'm just covering myself in case I've forgotten any exceptions) used in a small set of very specific ways, and because I have done explicit harmonic analysis rather than just skating by them, I know what those ways are, and on top of that I'm able to notice that Schubert and Mendelssohn use chromatics, perhaps also for colour, but in different ways.

So I think that's a useful tool to have used for the Clementi sonatinas, even if it's not a tool that is useful for all eras of music, or perhaps even universally applicable to music of some of Clementi's more adventuresome near contemporaries.

I'm using the same Roman numeral analysis toolkit for my Schubert and Mendelssohn analyses, and starting to find some chords -- not a lot, but more than "almost none" -- that don't fit into that scheme. So I've needed to expand to an improved notation that can handle the demands of this music. (I still like the (expanded) Roman numerals that I'm now incorporating because they show me explicitly when a chord has jumped out of the diatonic realm, in a way that letter names don't show me as explicitly.) But I wouldn't have noticed this new way of composing with chromatics as readily if I had just been writing letter names from the start. With my expanded notation, I'm looking for patterns that may not be explicit in the notation and so far are just things I keep in my head: for example, these chromatic chords are not chosen randomly but seem to appear in contexts where the chromatic notes can move by half-step. Eventually I am wondering if I will come up with a pattern for which chords are often used and in which contexts, e.g. I #iv°7 I but not I bIIIaug I (just to make up an example), the same way that ii V I is a very familiar pattern already, and getting a bit more subtle, tossing in a particular inversion of the tonic chord such as ii I64 V7 I is also now familiar.

I describe all that to try to illustrate the benefits for me of using an old-fashioned not-universally-applicable tool like Roman numeral analysis with secondary dominants slash notation, on music where it is appropriate to use it, and where using the latest conventions of all letter names, or expanded Roman numerals, on that same music would have obscured certain facts for me, and also made it harder for me to see the development in music from the Classical to the Romantic era.

One other thought, this time on the pro side for letter names for chords: especially if you're learning to identify chords, the letter names make clearer what you're doing. Using Roman numerals involves having to do two things: identify the letter name, and then translate it into a Roman numeral (at least, for me it takes those two steps). So for clarity in a learning situation, starting simple with letter names can be useful. (Thinking back to my harmonic analysis course where we did all RNA after the first week, I wonder if the professor in working with students and grading their work had to always be thinking about those two levels: was the problem identifying the chord, or was the problem picking the roman numeral? Or at an even more foundational level, was the problem in reading the notes?)

[ETA: and another pro for letter names, is that if you're going to be working from lead sheets, being able to quickly match letter names with chords is good, and to recognize common sequences of progressions of letters, and so on.]

An interesting read. You speak of Clementi and Schubert, et al, using specific chords in specific but different ways. Did it occur to you that they used specific notes and the chords were simply a result of those choices rather than choosing what chord to use for the harmony? And if you did would that make a difference?

For example, do you think Schubert came up with a sequence of chords in M65-70 (of this Moment Musical) that coincidentally formed a symmetrical pattern or do you think he moved the bass in contrary motion to the melody and kept the inner pedal and cared not a jot what "chord" resulted?

I'm not sure he wrote this many, Jeff!

M55 is Bb minor 6
M62 Yes, you can have a minor chord with major 7. It occurs on I in the harmonic minor scale, Db harmonic minor = Db-Eb-Fb-Gb-Abb-Bbb-C.
M70-72 What notes are in each chord and if E=1 what other numbers are present? And why might M70 be a rootless E chord if there's an E present?
M75 looks unusual!

I'm comfortable with the rest.

As I've been working through these pieces more, I'm coming to think largely the latter: that he's experimenting with moving his various voices in various ways, producing as a side-effect unconventional chords (usually sandwiched between conventional chords). So in that sense the names of these non-diatonic chords are not very important.

But I can't rule out that he also liked the sounds that his note-moving produced, and maybe even once he'd found that sound effect, looked for ways to get more of it.

I wonder how his audiences responded to his music. Did they find these harmonies strange, or pleasant, or unpleasant, or not very noticeable at all in the flow of the music, or what?

Schubert Op. 94 No. 96


yes, of course. Didn't I say that? OK, I missed the flat again and will try to be more careful. But, I do not believe it is minor either. Just Db6, same as M2 which I also had written down wrong.



Cool. No idea what I would have called it otherwise, but quite sure you wouldn't have liked it.



So much for my rootless flat nine. No idea what I was thinking here. It seemed pretty good this morning. It is just an E7/D, and no rootless involved. Darn ...



I'm looking more at 74-77.

Abm/Eb, Eb, Ab, Ab

Pouring over so much now, mixed up what I had, but know it wasn't right.

[b]Haydn Sonata in C Major - Score Download[/b]

Haydn! Hooray! Begone, ye mystifying Romantic chromatics! Welcome back, strict Roman numerals!

You all please feel free to toss any exceptions to strict sterotyped Baroque practice in this Sonata, especially ones that stretch my Roman numerals, back in my face. .

[ETA: oh, wait, as usual I'm confusing Handel (Baroque) and Haydn (Classical). And I don't expect Classical practice to so closely conform to a strict stereotype. OK, I retract my offer to toss things in my face. You'll have to just point them out to me gently .]

Haydn Sonata Hob. XVI:50 in C major

On a first listen: cheerful and playful. I like this a lot. I could hear the divisions between the exposition, development, and recapitulation. As usual, I didn't hear the change from tonic to dominant, or the modification in the recapitulation to stay in tonic (perhaps passing through subdominant). I did hear key change in the development, but couldn't tell if it was multiple key changes or just one.

I heard only one movement, which surprised me. Although perhaps it shouldn't. I don't know where Haydn stands in the history of the development of The Sonata.

[ETA: looking at the score now. There are three movements. That's more like what I expected. It seems that the posted Youtube video only has the first movment. Unless my ears have completely deceived me (haven't yet listened to it tracking with the score to be sure). OK, now I have 15 pages of Haydn Sonata analysis to do, which should keep me from posting for, oh, at least an hour .]

Haydn Sonata Hob. XVI:50 in C major

This is a three movement work.
1. Allegro
2. Adagio
3. Allegro molto

Still on number ninety-six, Jeff?



The notes are Db-F-G-Bb:-
Bb-Db-F-G = 1-b3-5-6 = Bb minor 6
Db-F-G-Bb = 1-3-b5-7 = Db 7b5
F-G-Bb-Db = 1-2-4-#5/b6 = F yuk!
G-Bb-Db-F = 1-b3-b5-7 = G min7b5

Personally, I would go with Bb or G as root.

oops ...

Schubert Op. 94 No. 96



Yep, and I have a problem.

M69 According to the score we have notes (reading up from bottom)

Eb bass, Eb, A, Cb; We called this a Cb7 which is fine. But playing it and listening to it, it isn't fine. The A should be flat and this should be a Abm/Eb.

Am I missing something? The A natural (Cb7) is very dissonant here and does not sound this way in the recording. I think it is a typo and should be Ab or G# (I don't care which as long as it is not A natural) and Abm.

I accidentally deleted a previous post and replaced with this one. Not sure if you saw it. I am absolutely 100% cool with Bbm6 for M55 ...

Can I just say that I completely love this thread, think it's a great idea, and more more more. This kind of analysis is extremely interesting and challenging, and it's completely impossible to find people who even know it even exists, much less are capable of talking about it. I just wanted to add this note of appreciation and gratitude. As you were.

Thanks, Mark. It is nice to hear from a devoted follower. We knew there were some -- with the 50,000 + views, but thanks for saying so, nonetheless. Please do not hesitate to join in. So long as you can tpye .

This is all brand new to me as of August of this year, and believe my overall musicianship is now advancing considerably more rapidly as a direct result. There are some real knowledgeable folks participating in this thread (I'm not one of them unfortunately) and everything is freely shared.

What a delight it is to see posts like this!

Thank you, Mark, and do feel free to join in. We do spend a little time naming chords and arguing about who's got them right but a lot of the time we also discuss the music, as much in lay terms as not, and what's happening in ways that doesn't need a lot of theoretical knowledge. There is also the odd practical tip regarding learning the pieces at the piano.

Your opinions would be welcome, they don't need to be qualified.

We're still here on the ninety-sixth Moment Musical!



What you're missing is a flat sign before the A! My score has one (Peters Edition). I see from the score you linked to (a month ago - wow!) it is absent. As was my attention when I OK'd you're Cb7!

Welcome, Mark VC! Glad to hear you're enjoying the thread. Do join in, the more the merrier.

Haydn Sonata Hob. XVI:50 in C major

With a piece of this scale, which I'm not familiar with, I'm finding I need new ways of approaching analysis of the piece. I've done the quick things -- listened to the piece, gotten a general sense of its flavour, heard the sonata-allegro form of the first movment, spotted the exposition/development/recapitulation and the starting and ending keys for each section in the score.

But my usual next step, of sort of obsessively starting to look at harmonies chord by chord, and during that investigation spotting phrases and themes, seems like starting out to examine an elephant with a microscope. I need to find an intermediate scale on which to start to understand the parts of the piece.

I think several more listens are in order, to find out aurally what starts to emerge for me. Another possibility is playing through the piece, or at least a simplified skeleton of it, and finding out what I discover that way.

Yes indeed ... I knew it ... and you were all probably thinking I was just another pretty face.

I wonder if the folks over at IMSLP would be interested in hiring me for Quality Assurance.

I'm very happy now ... and can move along to Haydn

Haydn Sonata Hob. XVI:50 in C major

If I might suggest a way forward, begin by isolating sections that LOOK different on the page and SOUND different in the recording.

The first section is in crotchet/quaver beats. At the end of M7 we hear a new figure that recurs up to the end of M14. The next section lasts to M19. From M20 we start to hear the first theme again but in the LH. At this point we can assume that Haydn will expound on these first three ideas so before we go any further look at each of these in isolation and get to know them well enough that you'll be able to hear Haydn work his magic on them in the next 32 measures. And that's just the exposition!