Do we need another "Goldberg Variations"?

114 posts / 0 new
Last post
More "Goldberg Variations"? Apparently yes!

To my intial question, two more labels answer rather emphatically "yes" with two brand new recordings:

- Warner Classics is coming this month with a new release of this monumental work, performed on Piano by the young and talented Beatrice Rana. She claims that she feels she is now...ready and "allowed to play it in public"! What can we say? Good luck, perhaps (not only to her).

- Pan Classics decided that it is ready to present to us another recording of the work on the Harpsichord with the fine harpsichordist Davide Pozzi.

Mrs. Rana is prone to broader tempi. So she brings the work to a 78 minutes disc, while Mr. Pozzi following Leonhardt's tradition avoids repetitions and opts for swifter tempi, so his CD lasts only 61 minutes.

For those still interested in more "Goldberg" recordings, I trust these new CDs might be good news. 

Parla

parla wrote:

parla wrote:

 Mr. Pozzi following Leonhardt's tradition avoids repetitions and opts for swifter tempi, so his CD lasts only 61 minutes

No repeats... swift tempi... 61 minutes...? Something is not right in this sentence parla...

The various "ways" of Goldberg Variations on record.

The multitude of Goldberg Varioations' recordings have a huge variety of performances and, subsequently, of the total time of the respective CDs. From the (probably) sortest one, i.e. the old one by Leonhardt (back in the early Sixties), lasting only 48 minutes (with no repeats at all and fast tempi), we have witnessed some quite extensive broad tempi and repeats bringing the work into the staggering time of up to nearly 85 minutes (Koroliov on the Grand modern Piano and Vinikour on the Harpsichord).

So, I believe that Pozzi's recording, lasting 61 minutes, is closer to the Leonhardt's tradition, with less repeats and not broad tempi. It is interesting that lately, both pianists and harpsichordists tend to adopt the broader tempi and enough repeats. So, just picking some of the most recent prominent recordings, both Ignacio Prego (on Glossa) and the well-advertised Post-Baroque Mahan Esfahani (on DG) last nearly 79 min., while both Zhu Xiao-Mei (on Accentus) and Angela Hewitt (on Hyperion), on their second recordings of the work, move up to 76 min and to 82 min. respectively.

Parla

parla wrote:

parla wrote:

The multitude of Goldberg Varioations' recordings have a huge variety of performances and, subsequently, of the total time of the respective CDs. From the (probably) sortest one, i.e. the old one by Leonhardt (back in the early Sixties), lasting only 48 minutes (with no repeats at all and fast tempi), we have witnessed some quite extensive broad tempi and repeats bringing the work into the staggering time of up to nearly 85 minutes (Koroliov on the Grand modern Piano and Vinikour on the Harpsichord).

So, I believe that Pozzi's recording, lasting 61 minutes, is closer to the Leonhardt's tradition, with less repeats and not broad tempi. It is interesting that lately, both pianists and harpsichordists tend to adopt the broader tempi and enough repeats. So, just picking some of the most recent prominent recordings, both Ignacio Prego (on Glossa) and the well-advertised Post-Baroque Mahan Esfahani (on DG) last nearly 79 min., while both Zhu Xiao-Mei (on Accentus) and Angela Hewitt (on Hyperion), on their second recordings of the work, move up to 76 min and to 82 min. respectively.

Parla

 

I was just asking for a little clarification on your post and instead you’ve added hugely to the confusion. From your original post I gathered he did not take repeats, which didn’t seem possible if the timing was correct. Now you say “less repeats”. Now that’s interesting. It is quite unusual (unfortunately) of interpreters taking only same repeats. Would you mind to clarify which ones? I feel genuinely interested. This would be better expressed maybe as the “Gould tradition”, from his 1981 recording, where he repeat only the first part of the canons. Kenneth Gilbert chooses to omit the repeats from the first variations of groups of three, the toccata like movements.

 

But unfortunately most players play ALL repeats, which means they play the whole music TWICE. In this you seem confused parla. Leonhardt, who does not take any repeats, is VERY slow at 48 m. (or 94 aprox if he took repeats). This is from his 1978 recording. He has an even slower one, which I cant find just now. So Koroliov at 85 m is actually much faster than Leonhardt, although not particularly fast.

 

This sort of thing has confused people regularly. For instance, Gould’s recordings last 38 m and

51 m, from where it has often been said that the second version is much slower. Gould himself didn’t help when he justified his 1981 recording by saying he had discovered slowness. But overall the second recording (not taking into account the aria, which is much is slower) is not actually slower, and some variations are even faster. To put things into perspective, Gould’s first recording, with all repeats, would take about 76 minutes.

camaron wrote:

camaron wrote:

 

Kenneth Gilbert chooses to omit the repeats from the first variations of groups of three, the toccata like movements.

That should "read from the second variations of groups of three", or the ones that come before the canons....

More complex than confusing.

I think the issue with the recordings and performances of Goldberg Variations is more complex than confusing. Quite a few of the soloists opt for most of the repeats and the tempi they choose. So, one has to go one by one the variations in the various recordings and better with the score to see what exactly is happening with each one of them.

Just comparing the timings of each variations in the Leonhardt's first and classic recording of nearly 48 min. (47:43) and 85 min. of Koroliov's (84:52), there some interesting features to observe:

While Leonhardt plays the whole work without repeats, Koroliov performs for example the Aria repeating the whole thing, but still he is slower (2:24 to 4:58), while he does not repeat the Aria da capo and then we see Leonhardt at 2:35 and Koroliov at 2:56.

In the pivotal slow Variation 25 (Adagio), Leonhardt reaches 4:22, while Koroliov goes to...11:09. Likewise, in Variation 15 (Canone alla Quinta), Leonhardt is at 2:39, while Koroliov goes up to 5:43! Of course, in some faster Variations, he is swifter than Leonhardt.

As for Pozzi, he repeats the Aria and, at first glance, the Overture (Var. 16) and the Quodlibet (Var. 31). From what he says in the booklet, he has made some "modifications" in some fifths in the Variations in minor, so that they can sound more "clear and relaxed", while he attaches great importance on the instrument he used for the recording.

Parla

Parla

 

 

 

parla wrote:

parla wrote:

I think the issue with the recordings and performances of Goldberg Variations is more complex than confusing.  

 

The issue is complex. It is YOU who is confusing. Leonhardt at 48 minutes is miles away from being "(probably) the shortest one", and it is a slow performance by any standard, even if far from the slowest, which might or might not be Turek in the 50s. So total time of the recording says NOTHING about tempo unless we know whether there are repeats or not (or much better, we actually listen to the music).

Again, if Koroliov takes 11 minutes in variation 25 (which is often not repeated, even by those who play all repeats, to confuse things further...) it is very likely that he plays one or possibly both repeats, which would bring the tempo very close to Leonhard's. This I cannot ascertain since I don't know/have his recording.

 

Then again, I find it very unlikely that Pozzi only repeats the numbers you say since a) there doesn't seem to be any structural or otherwise reason for it and b) given the overal timing it would still be a remarkably slow performance.

 

But leaving these things aside, and going back to the original question for this thread, what I find clear is that one thing that new interpreters of the Goldberg could explore is exactly the repeats. Gould's solution from 1981 makes a lot of sense: to repeat only the first half of the canons, and in this way stress their structural importance in the overall layout of the piece. I find Gilbert's choice more arbitrary. Another very obvious solution: to repeat EVERY first half of the variations, but I don't know if anyone has done this. But really, as far as I'm concerned, Gould in the fifties and Lenhardt do the right thing: no repeats at all.

camaron wrote: The issue is

camaron wrote:

The issue is complex. It is YOU who is confusing.

I fear this is yet another case of Parla being caught out. Anyone who knows the Goldbergs well understands the need to take repeats into account when comparing timings...........Rather a schoolboy error, in his case.

As for whether the repeats should be observed or not, I have to say that I used to feel much as you did. Then I heard Angela Hewitt's first recording. Repeats can be boring or "repetitive" or spoil the flow, but the second time she plays a passage, she always alters it, bringing out different voices and different rhythms, so you feel as if the journey is being deepened, rather than simply being repeated. Not another lap of the track, therefore, but a mile further along the road, as it were. In her case, I find I always look forward to the repeat for just the very reason.

I should add that I am not always convinced by Hewitt (though I certainly respect her total mastery). In the final analysis, I think she probably lacks the daring of a great interpreter; she can sound a bit tame, I mean. But her Goldbergs are truly wonderful, probably the most satisfying account I have ever heard. 

From Goulds gay and

From Goulds gay and irreverent 38 minutes in 1955 (or 37 in the live performance at Salzburg the year after) to Turecks cosmic 94 minutes in 1957… obviously the first one with no repeats, Turecks with all the repeats. The difference is so big as to almost question whether it is the same music, the same work. It is not the same aesthetic experience, that is clear. This, of course, is not an argument for or against either of them, but it comes to show: repeats (or the lack of) are important!

 

I don’t know Hewitt’s version, but for what you say it makes full sense. It is was I was refering to: if you are going to play the repeats do it creatively, either by they way you play them or by choosing which ones to play

I expect you have already

I expect you have already heard it, but there is an interesting interview with Gould on the three disc "A State of Wonder" set. It has been a while since I listened to it, but the interviewer had worked out that when you took the repeats out of the later version, the timings were much the same as the earlier one. Yet, the earlier one sounded incredibly rapid, while the later sounded much slower. To his ears anyway. Gould, if I remember rightly, responded by talking about the "pulse" of his playing, which he evidently differentiated from the tempo. While the tempo may have been equivalent, the pulse was not. Or something like that. Anyway, worth listening to; probably on the internet, if you don't have that set. At some point, Gould also performs a rather cringeworthy impersonation of a "British" person........

Pages

Log in or register to post comments

Gramophone Subscriptions

From£67/year

Gramophone Print

Gramophone Print

no Digital Edition
no Digital Archive
no Reviews Database
no Events & Offers
From£67/year
Subscribe
From£67/year

Gramophone Reviews

Gramophone Reviews

no Print Edition
no Digital Edition
no Digital Archive
no Events & Offers
From£67/year
Subscribe
From£67/year

Gramophone Digital Edition

Gramophone Digital Edition

no Print Edition
no Reviews Database
no Events & Offers
From£67/year
Subscribe

If you are a library, university or other organisation that would be interested in an institutional subscription to Gramophone please click here for further information.

© MA Business and Leisure Ltd. 2018