Classical Music Forum banner

Exposition repeats

1 reading
16K views 48 replies 27 participants last post by  Fredrikalansson  
#1 ·
I've always presumed that the repeat of the exposition had a specific purpose: to familiarize the listener with the thematic material, so that they will be able to appreciate its transformations during the development section as well as their reappearence during the reprise.

However, this would only really apply to audiences that were not familiar with the work. That is, to audience before the age of radio and recordings, who could only make themselves familiar with works either by repeatedly listening to them in concerts or by buying the scores or piano transcriptions and playing them at home.

Nowadays, since audiences are well familiar with Beethoven, Mozart, Schubert, etc., observing the repeats seems no longer necessary in the strict sense. So has it become a matter of taste? Or dogma?

Omitting repeats certainly affects the overall architecture of a piece. Take Dvorak's Ninth, for instance. If the conductor does not observe the repeat in the first movement, the weight shifts significantly toward the slow movement.

Or Beethoven's Ninth. If one observes the repeat in the scherzo and then plays the adagio in a somewhat swift tempo, they're almost equal in duration; I've noticed this with Zinman and Herreweghe, for example. However, by omitting the repeat and by broadening the tempo of the adagio, the latter can become almost twice as long as the scherzo (Furtwängler, Barenboim).

A curious case is a recording of Schubert's Great C Major by Solti. He omits the repeat in the first movement, observes it, however, in the finale, making the entire symphony somewhat bottom-heavy, so to speak.

Some composers have played with the idea of repeats. Brahms's Fourth does not call for a repeat of the exposition in the first movement, but the development section begins exactly like the exposition, creating the false impression of a repeat, before surprising the listener (though not anyone familiar with the symphony) with a sudden developmental turn.

Generally, what's your take on exposition repeats? I tend to prefer them being observed.
 
G
#16 · (Edited)
Agree; the repeats were put there for a reason and to tamper with a score is heresy IMO!! I always regarded Schubert's 9th ('Der Grosse') as being 'burdened' by repeats (responsible for that 'heavenly length') but the more I get to know the work the more I love every note, repeated or not. I suspect it was structural reasons which motivated composers to use the repeat/da capo. You find them in music much earlier than the 'classical period'. And, of course, the Da Capo aria in opera is a significant structural feature.

I'd say repeats have shaped even popular music today. Take the strophic song - it has a 'refrain' and a 'chorus' - and that 'chorus' is a repeat. The simple strophic song is built on the premise of the repeat. The modified strophic plays with that a little more loosely and then when something is "through-composed" it no longer relies upon the repeat, but moves into new territory. Think of Cole Porter's song, "I've got you under my skin". This is not relying on the repeat. Linda Ronstadt once said this song was like a cotton reel which had fallen onto the ground and unwound and it was hard to pick it up again.

The song goes: 'I've got you under my skin..' - let's call this A - repeated after another line, "so deep in my heart, you're really a part" - repeat A "I've got you under my skin" , THEN
"I would sacrifice anything come what may" (let's call this B) 'for the sake of having you near" - repeated, "in spite of a warning bell that comes in the night and repeats again in my ear", THEN
"don't you know little fool you never can win" - let's call this C
"use your mentality; wake up to reality" - let's call this D;
"but each time ...just the thought of you makes me stop before I begin" - let's call this E
then back to A, "I've got you under my skin". The orchestral 'riff' plays A-B-C-A. Then the pattern is repeated: C, D, E, C, A.

Here's the song:


So, repeats are structural and necessary and the extent to which they are ignored/extended/changed is a measure of a composer's innovation as well. (But these are not necessarily "Exposition Repeats", which are the subject of this thread!!)
 
#3 ·
Repeats played, please. This is my one beef with Celibidache; I listened to his Beethoven 5, and it was the only performance I've ever heard without the 1st movement repeat. You can get away without it in Eroica, not so the 5th..
Most pieces have better proportions with it played, and the transition to the development - or whatever - after the second playthrough has a totally different feel than if you never heard the first half twice, if you like.
cheers,
GG
 
#5 ·
Repeats played, please. This is my one beef with Celibidache; I listened to his Beethoven 5, and it was the only performance I've ever heard without the 1st movement repeat. You can get away without it in Eroica, not so the 5th..
Wow! I thought that repeat was something like a holy cow. Well, who better to slaughter it than Celibidache.
 
  • Like
Reactions: haziz
#6 ·
I usually prefer that repeats be observed. Usually the formal balance of a movement depends on it. That said, I can think of one repeat that I've never known anyone to take: the third movement of Alban Berg's Chamber Concerto has a repeat of nearly 200 bars (the entire movement minus intro and coda), and I just don't see the point, honestly.

But in Beethoven, Schubert, Mozart, and so forth, I find I enjoy a work more when repeats are observed.
 
#13 · (Edited)
My theory. Exposition repeats were used in the old days primarily to fix the themes and modulations and so forth in the mind, so that the development could be better appreciated and the recap recognized and welcomed. This was necessary because most people could only expect to hear a piece once, or at best every few years.

These days, when we've virtually memorized the music through repetition, the repeats are often unnecessary.
 
#14 · (Edited)
Omitting the repeats in a Mozart symphony....is like repeating the exposition of Sibelius's 2nd!!! Shouldn't be done!

Just thought I might give my own explanation of the importance of repeats...

Repeats grew out older music, especially dance music, to give musicians a chance to embellish the melody after it had been played once. Dance music was often short, dances were often long, so repeats were necessary and ornamentation and slight variations were improvised so the music wouldn't become boring. Most of the time repeat signs would be written in binary form music of the Baroque (just look at any dance suite by any Baroque composer) which later evolved in the 18th century into sonata form. A typical piece in binary form would have a basic tonal structure of...

A section
Subject in the tonic key, modulates to the dominant and stays there for a few bars until a candence. A section is repeated and ornamented or varied slightly in some way.
B section
Melody would be built on the same or similar rhythmic motifs, perhaps even fragments of the melody, from the A section starting on the dominant key where it left off. The music might pass through a couple of closely related keys before landing back on the tonic. This section is again, repeated and varied in some way.

Or to put it simply
A section
I...modulates...V (repeat with variation)
B section
V...longer modulations, exploration of more keys....I (repeat with variation)

It is easy to tell the beginnings of sonata form and the importance of repeats upon analysing its predecessor in the Baroque period. The modulation from tonic to dominant in the A section can be seen as a primitive example of the exposition, exploration if the same or similar thematic material in the B section while exploring new but closely related tonal centres can be seen as a primitive example of the development section in sonata form, landing on the tonic just like in a recapitulation.

As the binary form of the Baroque grew into sonata form in the classical period (look at sonatas by Scarlatti, symphonies by Sammartini, and notice how binary form is evolving) the repeats stayed with it and did help put emphasis on tonality and modulation, a very important part of how sonata form works.
 
#15 ·
There's an additional factor, although not so frequent, that the development will carry on such a lengthy and tortuous movement through many remote keys, that it becomes necessary to reinforce the tonic key by following the repeat. Secondary themes are typically in the dominant in major mode, and major mediant in the minor mode. By asserting the tonic and dominant twice, the listener has an "oriented confirmation" that will give them the ability to understand the instability of the development section.

Sometimes, and especially in Mozart, this need is doubly reinforced by tonic cadences at the end of regular length phrases [typically four measures] in the primary themes, and similar modules cadencing on the dominant in the secondary theme. This type of tonal over-determination gives the listener the message that the movement is an ambitious one and that one may expect an especially stormy development.

In my own playing, I always observe the repeats. Everyone has already mentioned the importance of balance, and that really is the key here. A well-proportioned exposition, twice played, happily and evenly counterpoises the development, recapitulation, and coda [if there is one].
 
#17 · (Edited)
Viva la repeats

A few years ago our community orchestra performed the Schubert Ninth. Our conductor orginally wanted to cut some of the repeats, like the exposition in the first movement because of time. The orchestra revolted and we played all of the repeats. It was one of the few times we won an argument.

Since then I have attended three performances of the work. In all cases conductor honored the repeats.
 
G
#19 ·
A few years ago our community orchestra performed the Schubert Ninth. Our conductor orginally wanted to cut some of the repeats, like the exposition in the first movement because of time. The orchestra revolted and we played all of the repeats. It was one of the few times we won an argument.

Since then I have attended three performances of the work. In all cases conductor honored the repeats.
A revolting orchestra, aye? Now, that's a thing of beauty - surely!!
 
#18 ·
Perhaps repeats were also omitted to give those sonata allegro movements a greater sense of a continuous narrative, of a continuous psychological development. To make them a little more tone poem-like so to speak. Because you woundn't have repeats in a tone poem, I guess.

This would also make pieces of the classical period somewhat more modern in the sense that later composers like Tchaikovsky, Mahler, Bruckner, Sibelius, Shostakovich rarely used repeats in their symphonies or even not at all.
 
G
#20 ·
Good points. And remember that first movement sonata form came from the operatic overture and these are laden with repeats. The treatment of melodic material changed from the classical period from smaller-scale "motivic development" through to larger-scaled 'thematic transformation' - this would have affected the need for repeats.
 
#23 ·
I think it's one of the blessings of the HIP movement that they've helped making it acceptable or even fashionable again to observe the repeats. Even Thielemann, staunch disciple of Furtwängler and Karajan, observed the repeats in his recent Beethoven cycle with the Vienna Philharmonic, and even in the Eroica and the Ninth.
 
#28 ·
Apart for familiarization with the new listener, and the fact you could not listen to 'that symphony' 800 times via CD, all part of the reason those repeats are there, but....

Also part of the deal and very much part of the plan, the loss of proportion to the overall movement / piece is, imo, very damaging to the whole.

Prefer the repeats, without which the whole movement / work is wonkily out of proportion.
 
#29 ·
Maybe the habit of not observing exposition repeats came from making recordings on physical records or CDs that would only hold so many minutes of music?
similar to Glenn Gould leaving out many repeats in the Goldberg Variations

I always play repeats except for the traditional avoidance of them when playing a da capo section (so the audience hears material 3 times instead of 4)
 
#31 ·
Could someone kindly tell me who put an end to the tradition of repeats?

I always assumed the end of repeats began with Beethoven's 9th.

Or was it Tchaikovsky and Bruckner who stopped Repeats for good?

PS: I really don't believe in the repeats frankly. Though I make exceptions for great recordings that do it. But I've never understood the necessity.
 
#33 ·
Could someone kindly tell me who put an end to the tradition of repeats?

I always assumed the end of repeats began with Beethoven's 9th.

Or was it Tchaikovsky and Bruckner who stopped Repeats for good?
Impossible to tell. There are a few classical era movements without repeats (despite some philological uncertainty, the first mvmt. of Mozart's symphony #35 is nowadays played without the repeat).

Beethoven has several without repeat (e.g. Appassionata, i, op.59/1i) before the 9th symphony but keeps it in other late works (such as op.111 or 130). Brahms kept it for the first 3 symphonies, Mahler "reinstored" it in his 6th symphony.
Very roughly, it was mostly compulsory until Beethoven, became optional with Beethoven and faded in the last 3rd of the 19th century.
 
#32 ·
Nowadays, since audiences are well familiar with Beethoven, Mozart, Schubert, etc., observing the repeats seems no longer necessary in the strict sense. So has it become a matter of taste? Or dogma?
By this logic, why play any famous works at all? The audience already knows them, surely there's no need to play them...
 
#35 ·
I prefer repeats on the whole, but I'm not dogmatic about it . I agree about how wrong it is to omit the repeat in the first movement of the Beethoven 5th, but Celibidache is by no means the only conductor to have done this on record or live .
Others who have done this include Szell in his Cleveland recording, but not on his Philips version with the Concertgebouw , and Bruno Walter . Years ago, I heard a radio broadcast with Maurice Abravanel and the Utah symphony, and he omitted it too .
Artur Nickisch, who made one of the earliest recordings of this iconic symphony with the Berlin Phil - an acoustic recording, omits it too . But I suppose the primitive recording techniques before the 1920-s , when electrical recordings began , may have been a factor in this .
Nickisch lived from 1855 to 1922 , dying just before the invention of electrical recordings . What a pity !
We would have been able to get much better picture of the things which made him the first superstar conductor . Regarding the repeats in the expositions of the Dvorak symphonies, which are found in number 1,2,4,5, 6 and 9 , Dvorak did not actually want them to be observed and never did so when conducting them himself . According to one story, he became angry at a conductor who observed the repeat in the 6th !
 
#36 · (Edited)
I believe Bruckner 7 (Karajan & VPO version) has a 1st mvt exposition that lasts 7:03 mins. (until a clarinet solo starts the Development). Someone please correct me if I'm wrong! But considering the length of that it would seem a little crazy to repeat it.


BTW, while on the subject of timings, I think the Recap happens at 11:15 (after the climax). But, again, please correct me if I'm wrong! I'm just an amateur listener. Anyway, the whole thing is just MAMMOTH. 19:30 for the 1st mvt alone!

I don't know any other composer using Sonata form who worked on such a huge scale. I know Mahler was pretty expansive, but wasn't he less formal with form? I haven't heard Mahler in a while.

Cheers! :)
 
#37 ·
I think you got the end of the expo correct but I'd put the recap at ca. 12:24 in the video with the minute before a transition towards it that uses the beginning of the first theme.

Bruckner's tempi are quite slow. If you count bars, the first movements of the Eroica and Beethoven's 9th are probably longer than several Bruckner movements, despite lasting only around 15 min.

The longest sonata movements in playing time with expo repeat (that can last up to or above 20 minutes, depending on tempo) I can think of right now are the first movements of Schubert's quartet D 887, his string quintet, piano sonatas D 894 and 960, Brahms' 2nd symphony and Mahler's 6th. Mahler's 2nd has a ca. 20 min. first movement w/o repeat, Mahler's 7th and 9th are even longer but probably not considered strict/typical sonata movements.
There are probably a few more candidates.
 
#40 ·
I think you got the end of the expo correct but I'd put the recap at ca. 12:24 in the video with the minute before a transition towards it that uses the beginning of the first theme.

Mahler's 2nd has a ca. 20 min. first movement w/o repeat, Mahler's 7th and 9th are even longer but probably not considered strict/typical sonata movements.
Ah, those pesky transitions! I heard the 1st theme so assumed it was the recap. Nope! Thanks, Kreisler. :tiphat: You must be a musician. Quick question: there are two woodwinds that follow one after the other, at 7:03. The first one is the clarinet? But what's the other one?.

I assume Exposition repeats (and Recap) were the Classical Era's version of ritornellos (or "return" or refrain). So it could have validity in terms of structure. But as they kept getting longer and more complex it seemed to move further away from some short, catchy tune (as in the Baroque Era) so it made Repeats superfluous. I think that was already happening with Beethoven. In fact, I think all his repeats (perhaps even including the scherzos!) could be done away with it and still work fine. I know that may seem sacrilegious, but honestly just listen to the many great performances that have no repeats! :D

That said, I can also love a great performance that has every repeat. Depends on the conductor. :) For example: Boulez takes the Scherzo repeat in the 5th, which interestingly even "purists" say you shouldn't do (irony!). But he pulls it off.