Hello all,

This two-movement sonata is one of the better known and definitely one of my favorite of the Haydn sonatas, especially the first movement. It is a bucolic theme in double variation form, abounding in Haydnesque rhythmic complexity yet conveying a mood of childlike simplicity and innocence (as we see in the marking Allegretto innocente, a very unusual tempo indication). I've never cared for the Presto as much though it is mordant and energetic.

Thanks for listening,

Joe

Haydn - Sonata in G Major, Hob. XVI No. 40, I: Allegretto Innocente

Haydn - Sonata in G Major, Hob. XVI No. 40, II: Presto

Very well and competently played, as usual. Most of my comments are probably down to editorial differences as previously, my edition being Peters/Martienssen 1937.

First movement:

A general criticism here is that you insert a break or hesitation of typically one quaver's duration at the end of nearly every repeated section, both when going back (except in the case of first time bars 8 and 44) and when going on. I really think the rhythmic continuity should not be broken at these points.

You play almost all the turns as 5 notes starting on the main note. My preference is for them all to be 4 notes starting on the upper. There is only one place (namely bar 5 and its equivalent bar 21) where you play it "my" way. I'm not sure why you don't do them all the same, even if it's not "my" way. One thing, though: the footnote recommendation in my edition is to play the first 3 of the 4 notes as a triplet taking up half the time, with the 4th note taking up the other half, but I don't like this at all, and prefer each of the 4 notes to take up an equal share of the time (as yours do, whether they're 4 or 5).

I feel that in bar 5 you are making a bit too much rubato.

More often than not, my feeling is that you hold pauses (fermatas) for too long. Bar 16 is a case in point. The 6-note flourish which leads into bar 17 acts as transition from the held chord, and makes less sense if the listener's memory of what it links from is all but lost. I feel that the moment at which the flourish notes start should be intuitively predictable from what came before, and I find that the rit which you make in the last half of bar 15 disorients me and makes me lose my place during the held chord in bar 16.

In bar 23 you have an extra decoration which isn't in my copy. I have just a turn on the last quaver, you add a mordent on the 4th. Personally I feel having both is too much, I'd omit the turn if putting in the mordent, but frankly I reckon the mordent is out of character with the rest of this movement.

I guess you must have a fermata on the chord in bar 34.

At the end of bar 77, which is the one with the fast scale in it, I have a turn on the last note, which you play instead as a mordent.

In bars 78 and 79 you play the grace notes exactly as I have them printed, i.e. as acciaccaturas (before the beat, accenting the on-beat main notes). Despite what is printed, I would play them as appoggiaturas (on the beat, accenting the grace notes themselves). There even exists an objective reason for this, namely that the grace notes are in (double) octaves with the notes played by the left hand, but that's just an excuse, I'd play them that way even if this weren't the case. I would almost be tempted to play this sequence as triplets (but would probably manage to resist that temptation).

Bar 88 is the one where the repeated high Ds change from semiquavers to quavers. I have calando printed here, but I feel you are making too much rit and also dim in bars 89 and 90. Noting that part of the slowing down is already written into the music, it doesn't need you to add much more.

Bar 92 is of course a tempo, but I get the impression that you might feel by the end of bar 93 that you haven't sped up enough, and accordingly you now race ahead and play bar 94 (the one with the arpeggio chords) too fast. Now you're in trouble because your speed is too fast for the fast scale in bar 95. You control your panic well by putting the brakes on slightly for the scale, but this does result in your arriving a little late on the top D. I think that had you not played the chords bar so fast, you would have handled the scale just as well as you handled the one in bar 77.

I have no fermatas on that top D in bar 95, nor in bars 96 and 97. But even if you do make these, I feel the last two bars (98, 99) should be back to tempo, you play them as though you had perdendosi printed here. You haven't, have you? The final chord is marked forte. Yours, especially the second time, sounds a bit apologetic. I think Haydn is being mischievous here and saying that this is really the end now.

Second movement:

Generally, I must say that the fact you don't care much for this movement doesn't show through at all in your delivery of it. It's nice and snappy, and I particularly like the absence of those inter-section hesitations I dislike in the first movement.

I guess you must have a rall marked in bar 14.

In bar 17 I have the last 5 RH notes all the same (Ds), just like in bar 16 (where they're all Es), but you play only 4 Ds with an E as the 5th last note.

Do you have a fermata in bar 18? And in bar 41?

For the following, bar 76 is the 7th bar from the end of the piece. There is a rest lasting a whole bar (which straddles bars 76 and 77). This is a measured rest which should, I think, be in exact tempo. Although you may well intend it to be so, unfortunately the slight rit you are making in the second half of bar 75 destabilises the listener's internal metronome and renders the long rest impossible to count. Then, when the notes start again, their position in time is not exactly where expected.

Hello rainer and thanks for your careful listening as always.



The slight pauses between sections are to some extent deliberate because I like to give a bit of a breath when the material is new. I will think, however, about whether it may be too much.



Yes, and I do have a rationale for this. The way I think about ornaments is that they should be played with the line in mind, according to the melodic direction. So, e.g. in bar 1, if the turn weren't there, the melodic note on which the turn is printed would be F-sharp. Since the accent for these ornaments naturally is going to be slated more to the first note of the ornament, I want to emphasize the F-sharp more as part of the melodic line. For the identical ornaments in bars 5 and 21, however, I start the ornament on the D, which I think is understood as part of the stepwise part of the line, but of course, not printed since the preceding third is of quarter-note duration.



You make a good point here. In listening back, I thought maybe it was a bit too long too. Fermatas I had always thought were supposed to be roughly equivalent to doubling the original note values, though I confess to often not them counting that carefully




Yes, my edition (Schott Universal editions) actually has a turn with a mordent above it in parentheses and I opt for the mordent. I do like the decoration, though it is rather difficult to get them all in (on the modern piano) and make it sound natural.



No, but I deliberately wait a little extra time to emphasize the climax and dynamic contrasts.



Yes, I deliberately changed that to a mordent in my score, basically because a mordent seemed a bit better to precede the tied note (turns seeming more fluid and continuous to me).



On this, I disagree. If you were to accent the grace notes, you would lose the D-C-C-B-B-A melodic line.



I can see your point here, but I think this is a place in Haydn (and there aren't all that many) for a little wistfulness,drama, and romanticism in the Lisztian sense (where sound itself becomes the issue). It also signifies the coda and breaking of the double variation form.



Good point, I may just need to have worked on the scale passage a little more. I'm not sure the preceding arpeggiated chords were too fast, but that note grouping of thirty-second and 16ths is hard to get in completely.



Right, this is just my interpretation, despite what Haydn marked (I know I can sometimes be branded as a philistine for not always following the score exactly ). I, in fact, view this more as a perdendosi, an after-thought as a coda can sometimes be. This one seemed so different and novel compared with the norm and I wanted to emphasize that, almost an a reverie-like ending



Thanks. I didn't like it so much when I was playing it but then liked it more when I listened back, which is good because it's usually the other way around .



Yes, my edition has 4 Ds and an E.



No. Bar 18 I think I do count correctly listening back (being an equivalent rest of four quarters and an eighth), but I think I wait a bit long in bar 41. It could be even the one in 18 (and 76, as you noted) are just a hair too long. It's a rather offbeat entrance, but I agree in this case that they should be exactly in time.

Thanks again for your comments. Much appreciated.

Joe

Thanks for your detailed feedback to my comments. I hope you don't mind if I take this just a little further.
I agree that it's a good idea to have a gap for breath, but would say that this gap should be created, wherever possible, without changing the duration of the containing bar. So instead of splicing in silence, you can make room by silencing part of an existing note. Taking the 2nd time bar 8 as an example, instead of playing this as a 7/8 bar as you more or less did, with three 8th notes, one quarter note, one added 8th rest, then the 8th upbeat note, consider instead playing it as four 8th notes, one 8th rest, then the 8th upbeat, thus preserving the 6/8 bar length.
I understand your rationale and can see the merit in it, but nevertheless I believe there are more important considerations at play in this particular instance. I'm not convinced that the F# really deserves any emphasis or accent at all here, for two reasons. One is related to its position in the bar; we think of 6/8 as generally being two beats of three sub-beats each. The first beat receives more weight than the second, and within each beat the first sub-beat receives more weight than the other two. The F# therefore falls on the least weighty sub-beat of the bar (or joint least with the E). The other reason is melodic; the line is D E F# G, with G the important target, the E and F# are just passing notes, and the F# is also a leading note.
You are thinking of an implied melodic line, in the absence of the turn, of C D E E D C? That's interesting. One might almost be tempted to shorten the quarter note E to an 8th and play the turn as four 16ths.
Hmm. I'm not sure D-C-C-B-B-A is the real melodic line here. If we do a thought exercise similar to where you said above "if the turns weren't there", then what would we see if the grace notes weren't there? What I see is that the sequence of 16ths D-C-C-B-B-A is really a line of 8ths C-B-A with appoggiaturas D-C-B added. Moreover, this line of 8ths continues right through the next two bars, i.e. it goes C-B-A-G-F#-A-G-E-C#D-D-D-D, and indeed for most of its length it parallels the left hand at an interval of a 6th plus an octave. I think of every one of these 8ths, except the very last D, as having an appoggiatura added. I even think of the syncopated part of the phrase as containing "disguised" appoggiaturas. If you were to break the syncops up into 16ths, bar 79 would go A-G-G-F#-F#-A-A-G-G-E-E-C#, with the first of each pair of repeated notes being a "main" note (part of the line of 8ths) and the second of each pair being the appoggiatura to the next main note, the novelty here being that the main notes are tied to the next notes' appoggiaturas.

I hope this makes some sort of sense. Tell me if I'm talking rubbish.

This in fact is the lesson to learn with the first two little pieces from Bartok's Microkosmos book 1.

Yes, I think I see what you're saying here. That would make the quarter into an eight plus eighth rest, thus preserving the time while giving the impression of a breath. Trying it out at the piano both ways, I'm not wholly convinced this is better, though you're right it's more correct for classical style to be as rhythmically precise as possible (and I'm sure my old teacher would have probably gotten after me about it too, being French and a stickler about rhythm ). It still does almost have the impression to me of rushing a bit into the next section, though, even though I know it's exactly in time.



I'm not quite sure I agree with your reasoning here. My point is not that the F# deserves any particular accent or emphasis with respect to the D and E preceding it. You're right that the greatest accent for the whole melodic phrase would fall on the first note of each of the two groupings, in this case the D, and that the F# would be the least emphasized of the three. However, the F# still would need to be slightly accented as part of the ornament (though a lighter accent in comparison with the one on the D), which accentuates the F#'s position as part the first triplet phrase, as opposed to the G, which is part of the next phrase. Likewise, if one played the four-note turn starting on the G, the G would have to be accented slightly, as is the custom for ornaments (i.e., to slightly accent the first note). So if one does that, the phrase that the ear hears more is D-E-G than like D-E-F#, which is a bit jarring to my ears.



A very interesting idea. I tried that too and liked it



Your reasoning is interesting here, except IMHO it may be reading a bit much into it. If you look for example at the parallel point in the prior "variation" at measure 42 in the lefthand where the secondary melodic line emphasized is D-C-B-G-F# and compare it with the passage in question of B-D (with C as a passing tone that precedes it)-C-B-A-G-F# (of course, each one being repeated before the F# ends the pattern), the pattern seems very similar. I guess it then depends to some extent on which hand is emphasized as carrying the melody (the left hand in the later variation of E-D-C matching the C-D-E in the earlier passage). In some ways, in the later passage, the left hand does that more (and maybe I should have brought that out better), and one could even look at the right hand accaciaturas (or appogiaturas ) as the accompaniment, which could mean that Haydn has inverted the variations with respect to each other.

Joe

This is up, Joe. Sounded nice!

Thanks. Monica.

Hi Joe,
I have listened twice the set of movements - instead, I confess, of reading the long and detailed comments you got. And it was a pleasurable moment ! Technically and stilistically perfect ! I did not know this sonata, so thank you for the discovery. Each time I listen more Haydn, I am astonished by the scope and the depth of his genius. More than sixty piano sonatas, 104 symphonies, 83 string quartets... Let's worship Haydn, which is not enough played by pianists, I think...

Really good. Thanks for sharing!