This morning Wolfgang Knauss wrote something on bar 28 of Bach’s Andante in the Italian Concerto which recalled me an issue of mine. Probably it’s only an issue for me, still here it is.
Bars 48-49, the very last two bars of the Andante.
The g at the tenor, last note in bar 48, is a suspension 4-3 which resolves to f. So far so good. However, my ear loses the perception of the g and usually all I can hear is that the piece ends with a jump of a third minor from the bass d.
In short, this is how it’s written:
And this is what I hear:
Pianists often play the g with a little sf, so I can clearly hear the g-f melody. On the harpsichord, usually I can’t.
When I play this myself I try underlining the g by detaching it from the previous c# (so to make it stand out a bit more) and not making any rallentando, so to have the g not yet died when playing the final f.
Sometimes this does work, sometimes less so. A re-striking of the g would fix that but it doesn’t sound so fine. And by the way, a suspension is tied, no? Mozart does un-tie such suspensions here and there, but it’s 50 year later…
This effect is exacerbated by the 20th-21st tendency to add ritards to all endings. Others more knowledgeable than I may comment, but I find no support for this in the historical record. The closest example - and not indicated here - is the 2-3 bar ending adagio often found in Handel, which does not imply a gradual slowing either.
This additional slowing of course causes the “g” to disappear. When I play this, I do not slow down much at all, just a touch to add a sense of finality. The “g” is then heard for what it is: a suspension.
Start first with no ritard at all, then play with it to add the sense of finality without the drawing out. Alternatively, if one must, one could add a pause on beat 2 of the penultimate bar, then play through to the end as an extended Italianate ornament. Dangerous territory, as Bach indicates nothing of the kind. But he does similar things in other pieces.
Also, one must be careful not to add additional time to the final note. Bach (or his engraver) used perfectly good ink to indicate just a quarter and then two rests. At first this is awkward to contemporary ears, but I find that once I get my head around it, I can communicate it and it sounds perfectly natural.
It may be just an issue of yours. I just played it: I hear the g still sounding (in my head?) and it has never bothered me. Doesnt the c-sharp resolution of the much longer d worry you?
That said, I do myself have similar issues with other pieces – like the c-sharp in the second (complete) bar of Handel’s Harmonious Blacksmith, which I cannot avoid sounding like part of the melody.
Just tried this on my clavichord and agree with David. The g is
definitely not sounding physically, but I also definitely hear it. I
guess I’m expecting it to resolve so my brain fills it in.
I agree the audience may not “get it”, because we see the notes and mentally fill it in.
Try experiencing that effect by listening to a CD, especially where the performer adds a pronounced ritard. It seems like the “g” pops out of nowhere, almost like a mistake.
You are right, but until you mentioned it it had never bothered me, because I always heard it (in my mind?). That brings back to me a comment Christoph Wollf touched on, namely, that often the/a note is held even though it has become (very) weak, because the mind substitutes what it should sound like. But that struck me as an audio-psychology issue, which I di no want to touch.
Now I think of you every time I conclude the Andante.
This morning Wolfgang Knauss wrote something on bar 28 of Bach’s Andante in the Italian Concerto which recalled me an issue of mine. Probably it’s only an issue for me, still here it is.
Bars 48-49, the very last two bars of the Andante.
The g at the tenor, last note in bar 48, is a suspension 4-3 which resolves to f. So far so good. However, my ear loses the perception of the g and usually all I can hear is that the piece ends with a jump of a third minor from the bass d.
In short, this is how it’s written:
And this is what I hear:
Pianists often play the g with a little sf, so I can clearly hear the g-f melody. On the harpsichord, usually I can’t.
When I play this myself I try underlining the g by detaching it from the previous c# (so to make it stand out a bit more) and not making any rallentando, so to have the g not yet died when playing the final f.
Sometimes this does work, sometimes less so. A re-striking of the g would fix that but it doesn’t sound so fine. And by the way, a suspension is tied, no? Mozart does un-tie such suspensions here and there, but it’s 50 year later…
Also, one must be careful not to add additional time to the final note. Bach (or his engraver) used perfectly good ink to indicate just a quarter and then two rests. At first this is awkward to contemporary ears, but I find that once I get my head around it, I can communicate it and it sounds perfectly natural.
That is a very decent and simple suggestion, I think.
Wolfgang
This effect is exacerbated by the 20th-21st tendency to add ritards to all endings. Others more knowledgeable than I may comment, but I find no support for this in the historical record. The closest example - and not indicated here - is the 2-3 bar ending adagio often found in Handel, which does not imply a gradual slowing either.
This additional slowing of course causes the “g” to disappear. When I play this, I do not slow down much at all, just a touch to add a sense of finality. The “g” is then heard for what it is: a suspension.
Start first with no ritard at all, then play with it to add the sense of finality without the drawing out. Alternatively, if one must, one could add a pause on beat 2 of the penultimate bar, then play through to the end as an extended Italianate ornament. Dangerous territory, as Bach indicates nothing of the kind. But he does similar things in other pieces.
Also, one must be careful not to add additional time to the final note. Bach (or his engraver) used perfectly good ink to indicate just a quarter and then two rests. At first this is awkward to contemporary ears, but I find that once I get my head around it, I can communicate it and it sounds perfectly natural.
This morning Wolfgang Knauss wrote something on bar 28 of Bach’s Andante in the Italian Concerto which recalled me an issue of mine. Probably it’s only an issue for me, still here it is. Bars 48-49, the very last two bars of the Andante. The g at the tenor, last note in bar 48, is a suspension 4-3…
Several recent postings on this subject appear to deny the possibility of listening and hearing polyphonically unless the sound source gives each tone a constant volume for its full duration. I exaggerate, but only slightly. Schenker would be appalled.
Maybe I’ve been brainwashed into thinking that it’s possible to re-interpret a tone, by hindsight, as being the preparation of a dissonance, and that this characteristic is determined by the appearance of a dissonance-creating tone, and that this sequence of aural events requires the resolution of the dissonance. But it is really not necessary to have passed a course in harmony or counterpoint to be able to hear, I mean perceive, these phenomena, even if the listener doesn’t know their names, or doesn’t even know that they have names. As Stuart writes, “The g is definitely not sounding physically, but I also definitely hear it. I guess I’m expecting it to resolve so my brain fills it in.”
Many posters have stressed brain’s ability to fill or recreate the harmony when sounds are missing; or the knowledge of the players of that musical language. Of course they are right, it’s a well-known phenomenon and a lot of fine music rely on that effect.
I may have missed saying explicitly, but I do experience that phenomenon when I hear or play music, everyday. So, nothing new for me.
But in the particular case of the ending of the Andante, Ican’t help but hearing the bass line playing d-f, and I lose the suspension’s resolution and hear a melodically weird ending. I don’t experience this with other pieces, just that. So I wonder if there is something particular there that I hear that way.
It sure can be a personal issue, as David says, but others are testifying they hear something like that as well. @Pickett. David, I have no problems with the c#. My brain keeps it in memory until the final d. Plus, it doesn’t cause any weird melodic conduction.
As for remedy, I concur with all who say no or little rallentando, and the last note must be of exact length as written. I add I try to make the g more relevant by detaching it from the previous note, so the brain is aided in remembering it.
Maybe recording techniques have a role? For example, I’ve just listened to a Scott Ross recording, the g is almost inaudible from the start, and so the ending is for me a clear d-f in the bass.
My apologies, Domenico, as my response was (unintentionally) in blue color, so you did not recognize it as a reply.
It should have been
I am sorry! What a bad way to ruin a fine piece of music. I appreciate your “emoji” (if that is the right term)!! W
Wolfgang
“…But in the particular case of the ending of the Andante, I can’t help but hearing the bass line playing d-f, and I lose the suspension’s resolution and hear a melodically weird ending.”
One simple measure that hasn’t been explicitly mentioned yet is to keep on playing the c# along with the a (and under the g), and rising to the d. That makes it clearer that the final d is coming from below, and the tenor f is coming from above. Or, I suppose (though I hadn’t thought of it before), you could restrike the bottom A at the end of the penultimate bar.
…on the other hand, you could decide to ignore what I suggested in my previous mail and just enjoy the melodically “weird” ending as an echo of the spicy leaping melodic sequence found in the first movement of the concerto, bar 88. There I hear the sequence alto f -d-g#-bflat-a. Here are bars 87ff.
Quite possibly, in fact: probably, I am out of my depth here, among such advanced players, but one option I would investigate to remind my ear that the g is carried across the bar-line, is to resolve the suspension non-legato, i.e. release the g earlier, so that any remaining vibration is audibly damped and the plectrum can (audibly) reset - all while the d and d’ are still sounding. This might remind the ear that the note just released was not one of the d’s, but had to be the suspended g.
If it’s all nonsense, I’ll gladly go back into lurk mode.
This I have tried before but didn’t work for me. If I keep the c#, I can’t have the g detached from it, which I find necessary to properly stress the g on order to have it stay well in my memory.
I am appreciating your following suggestion, to get it as a reminder of another spicy melody in the first movement. In fact I have called the ending “weird”, but it’s not unpleasant to my ears, just “incorrect” to my brain, so I sure can stand it.
| 
| 
| 
