Piano World Home Page Forums Our Most Popular Forums Adult Beginners Forum Starting out with analysis, all invited

At the top of bar 11, the melody is on F, so that's clearly not the end.

When we get to the C major triad and C melody note at the top of bar 13, try something: Make that a whole note, and just skip the rest of it. Play it that way a few times, trying very hard to forget that you know the rest of the song.

I think it's mainly that we all *know* that there's more melody to come.

The ii7 - V7 - I thing is kinda like a wheel that goes around and around. But it's a special wheel that can only stop on I.

I'm very comfortable with the process of finding where phrases and pieces end. I'm much less comfortable explaining it to someone else what to look for. I seem to rely on intuition and experience.

M12-13 does look like a ii-V-I cadence but isn't conclusive.

I would venture that the melody falling from D to C without being precedeed by a higher E, let alone G and F, is insufficient to close the piece.

Not wanting to get into hot water with continued remonstrance of the importance of the 3-2-1 (preferably 5-4-3-2-1) descent to tonic or the 7-8 rise from the leading note in the melodic line can anyone else offer an explanation as to why this is not a final cadence and how it is we "know" there's more to come?

By that I mean only that the song has been world famous since 1943. It would be quite rare to find a musician who doesn't know it. Of course the whole tune is better than the truncated version. But if it were possible to forget it, there is a resolution of sorts there.

Because there are two more chords in measure 13 that move the music along. I'm unfamiliar with the 3-2-1 idea. I'm sure that music ends in all kinds of fashions, but very often on the tonic.

Since Richard first pointed out 3-2-1 endings last year, I have started to notice them in a surprising number of places.

I stumbled upon a series of very interesting videos on youtube that could in my opinion be really helpful for someone interested in analysis: http://www.youtube.com/user/AustinTPatty/videos?flow=grid&view=0&sort=da

This is the question I'm trying to answer.

Why does the C at the start of M13 not feel like a stopping point - apart from more music being written after it?

If I remove M12-13 the only change is the tonic being arrived at via the leading note instead of the supertonic. Is the change from 2-1 to 7-8 enough to finish the music?

If I made the second F in M12 an E could the song finish on lamp? Would it be musically complete with a 3-2-1 finish?

If either change makes the song musically complete would it need another two bars to balance it proportionally? If it needed more by way of proportion, and I genuinely mean no disrespect here, would it matter if the next two bars ended on a major seventh or a ninth?

Is there something else that that I'm missing?

Because with the overall rhythm of the piece, an ending for the third part which is only 2 measures long would be too short. If that same passage was in beat 4 of the last section, it would sound final. However, climbing up from the leading note to the tonic is also more effective and final than coming down from the supertonic to the tonic. That is why the leading note is the leading note. If you are a singer or violinist, and not restrained by the tuning of a piano, you can even heighten it by making B closer than a half step to C.

Having written this, I also know what niggled at me about the idea of a rule that says you should have 3-2-1, and why now I'm fine with the overall idea. This rule gives me no feeling to the music at all - it's just a rule. But if I actually look at how music feels, then it is clear. If Oh Christmas tree were in C major, then the ending would be "B D C" We feel B as leading note wanting to go to C, the B and D circle the C like moths circle a flame, and then land in the C. Here there is a musical feeling, rather than a rule (which isn't a rule --- you are not "wrong" if you write music that doesn't end in 3-2-1).

Similarly, D to C isn't that strong, because the supertonic is a whole step away from the Tonic. There is no strong movement pulling D to C. D is neutral and is just as happy climbing up to E, as it is to go down to C. So if you have a 3-2-1 or 5-4-3-2-1 movement, it's like watching something slide down a hill, and you expect it to keep going in that direction. There is a downward momentum and a landing. I can accept 3-2-1 in that sense. But otoh, I prefer to feel it as music and explore, rather than remembering a rule and then looking for that rule. Music is sound.

Brilliant, keystring! I never thought of switching the bars.

Yes, it does sound final if I swich M13 and M15. So proportion also plays an important role.

ETA: It also works by duplicating either M13 or M15.

I did not take 3-2-1 as a rule. I took it as an observation. It's a result of observing and experimenting with music, not a rule that states things must be this way, or are always this way, or should be this way, or are wrong if they're not this way.

I find it useful to notice patterns in music, and describe them. Otherwise music is just a bunch of random patterns and there's no unity and no understanding for me.

I think it's a mistake to interpret the results of observations and the naming of patterns, as "rules." They may become principles that it's useful to remember if you want to compose in a specific style, but they're not rules. My ear is not sufficiently developed to be able to discover these principles purely by noodling at the piano, or even by playing written music and only responding aurally without thinking analytically about it, and so I appreciate learning about patterns and their typical effects. Not as a way of straitjacketing music with rules, but as a way of understanding music.

And landing on C instead of B in M14 lets me finish on other, even non-diatonic, notes/chords in M15 albeit with different parting moods so I'm learning a lot about this proportion business.

Richard, you wrote something earlier in regards to phrases, of being able to perceive them due to experience and instinct, but not being able to teach how to do it. I think this brings us to some interesting things.

One is that when we play music, we absorb patterns, just like a child can speak grammatically before learning grammar in school. That is, the child can say "I saw a dog run across the street." instead of jumbling dog, street, run, across. Theory then tries to get at those patterns and if you got them first instinctively, you'll say "Ah yes - now I see why I always seem to know this! That's what's behind it." Reversing this, without some backup theory, you have to dig in an actually try to see what it is that makes you hear a phrase as a phrase. Apparently whatever we do instinctively is harder to define to others. I've been told that the easiest thing to teach are the things that you struggled with yourself, because you had to take it apart or have it explained.

This leads to a second thought, namely that the stuff of music exists in and of itself, and we can sense a lot of it. The explanations in theory are a way of codifying it. As such it's a crude approximation of the real thing. And maybe we need overlapping different code systems to get a better picture since nothing fits 100%. Knowing this might help with potential disputes as people with different "systems" (classical, jazz, ear) come in. In the same vein, any "absolutes" that go "This is how it is, totally, no other way" should probably be considered with caution.

OK, this discussion of knowing instinctively about resolving to the tonic and all that reminded me of something that may be helpful and even amusing. Here's another melody that virtually everybody will recognize (even though I did it hastily by ear and didn't get the durations right).

It's just a one note melody line, no chords, which shows that it's the melody that resolves, the chords just go along for the ride.

It also illustrates the business about familiarity with a melody, and knowing how it's supposed to go. (The last note here doesn't go where it's supposed to.... ;-) )


http://www.pianoworld.com/Uploads/files/ResolutionExample.pdf

http://www.pianoworld.com/Uploads/files/ResolutionExample.mp3


I'll be out of town on business all next week, so I'll see your responses perhaps on the 24th.

Hi, all. I've been away a while. But I find this discussion very illuminating.
Clearly the rhythm of LM requires the other 2 measures, so that does push it along.
(But then what about God Save the King, which seems to be missing a measure?).

I also found the 3-2-1 vs 2-1 makes sense when I listen to it.
Switching measures, to me it sounds OK with FDC (498) at the end,
but the FBC (47,1) in the middle seems harder to get past.
So I guess that reinforces your point.

And John reminded me of another issue:
why do we name the chords and do the harmony from the bass up?
Why not melody down--if indeed the melodic progression defines the tension-resolution?

I don't know whether I mentioned this before, but from singing in a choir,
I got the impression that the harmony follows the melody, not vice-versa.
Soprano of course has the melody,
the alto follows the soprano typically a third/sixth down,
the tenor line rides "above" the melody but an octave down,
and the bass pounds away on 1, 5 and 4 to keep the beat going.

My piano teacher of the last 2 years uses a "method"
that consists of learning some kind of "walking bass" line in LH (I may be mis-using terms),
rooted in the chord-name note,
with melody in RH and one or two notes of the "chord" (from lead sheet) in the RH below the melody note.
It's a bit different from the LH-chord RH-melody of typical "chord" methods,
and also from the two-part approach of "beginner classical" methods,
but I'm thinking it's probably more musically complete.

His approach to composition (at least for students) is to invent a chord sequence
(e.g. put I at the beginning, V7 I at the end, and "roll dice" for the middle),
make up a melody that fits the chords, then "harmonize" it as described above.

From where does the measure appear to be missing in G S the Q?

We tend to hear music from the bass up though this apparently isn't universal to all cultures. The bass is more fundamental to the music whereas the melody can adapt itself to lyrics, for example, that change from verse to verse. Early music forms such as chaconnes, passacaglias and grounds depend on the regularity of the bass.

The harmony follows the melody? can you be more specific as to whether you mean in the writing or the performing, etc.

The melody isn't always the highest note. Frequently in contemporary music the harmony is a third or sixth above the melody. And The Beatles used harmony a fifth apart in Love Me Do. Something of a novelty, I believe.

For choral harmony, it's logical for the alto to take the closest note down from the melody as long as it doesn't break the 'rules' of harmony as in parallel fifths, etc. The bass tends to keep to the root notes on the stronger beats and the tenor has to fit in with what's left, often having to double another voice when the harmony squeezes into close triads.

The composition varies with genre, style and effect.

To get something in the style of Springsteen, for example, where you want something anthemic and suitable for live performance, you want to keep the melody within a small compass using only four or five notes most of the time (Philadelphia, Born in the USA) then the chord first method is ideal. If you're writing for a woman's voice like Annie Lennox and want a wider vocal range (Love Is a Stranger, Right By Your Side) I prefer to write the melody on piano or a recorder and consider the harmony as an afterthought.

For beat music it's often a good idea to start with a rhythmic base such as a drum machine pattern and the poetic style of Dylan might best start with the lyrics. Elton John's music is always lyrics first, of course, and being composed on piano rather than guitar stands out as more poetical, melodic and often more adventurous harmonically.

If you're thinking of moving in that direction I would try different styles and forms regularly and often.

Has anybody done a harmonic analysis of Bach's Jesu Joy of Man's Desiring,
either the Hess piano version or the full chorale?
I saw couple posts in the Classical Sonata Analysis forum but they didn't really get into an analysis
--presumably it's not a Sonata. Does it even have a named form?

I'm not aware of a published analysis.

The intention of the analysis thread was not to analyse Classical Sonatas alone but to follow that style of analysis, i.e. structural, detailed and full harmonic as opposed to a more practical style geared towards listening, learning and playing the piece on the piano. It was touched on briefly in the thread because Jeff was going to learn it.

What sort of analysis were you looking for. Most folks here balk at the thought of a full harmonic analysis. I myself haven't done one since the thread was last active; not really my preferred style.

The Sonata Analysis thread was meant for understanding sonata form. At some point it deviated to other forms and that became confusing. I have not studied all forms, and at some points I was trying to figure out what was being said in the context of sonata form which had been carefully laid out, and could not make head or tails out of it until I figured out these were not in sonata forms. Other times one had to take a "crash course" in some other form being mentioned by doing research and studying it, in order to follow what was being said. I consider the Sonata Analysis thread to be about Sonata Analysis. Otherwise it gets too confusing.

We started out with the Moonlight but had been through preludes by Bach and Chopin and then we started on Clementi. We chose sonatinas not because they were (partly) in sonata form but because they were 'easy'.

We were analysing music and since sonata form is about as tough as it gets, fugues aside, that's where we were aiming.

There are several times, through the life of the thread, where I wished I'd used more foresight in the naming of it.

For me the term 'Classical Sonata Analysis' sums up the process of analysing music by its harmonic content, not the type of music being analysed - and how would you know before you analysed it? I use the same process for symphonies, concertos and all the forms used therein.

Sonatas and sonata form are not the same thing. I think that's where you may have confused yourself. There was never an explicit restriction to sonata form, or forms using only the sonata principle. It was about analysing tonal music.

I cannot, off the top of my head, think of a sonata where all the movements were in sonata form. Sonatas cover many forms.

I am definitely not confused about the difference between sonatas and sonata form. The idea of the thread came from the earlier thread, where it was decided that basic form first had to be established so that everyone would be up to speed. Perhaps the title was misleading. The confusion came when, after logical order starting with simple sonatinas and the earlier sonatas, suddenly music appeared that were not sonatas and also had no sonata form in them. If it was not meant to be about sonatas and sonatinas exclusively, perhaps it should have had a different title. Somewhere in the middle of that thread it was sort of, maybe, decided to make it more general.

The thing is that we started with a context, we defined what sonata form was, and what sonatas were. Everyone could follow because there was that context. If other forms of music are to be considered in such a thread, then those forms would have to be defined just as carefully or people who are not advanced in music will be lost. I had to scramble a couple of times and do some hasty reading on whatever form suddenly cropped up.

In any case, in regards to tinman's question, is there any problem with it being addressed here rather than in the "sonata" thread? I think that was the issue though at the moment I'm not sure.