Recordings/Discussions
Background Information
Performer Bios

Poet/Composer Bios

Additional Information

Recordings & Discussions of Cantatas: Cantatas BWV 1-50 | Cantatas BWV 51-100 | Cantatas BWV 101-150 | Cantatas BWV 151-200 | Cantatas BWV 201-224 | Cantatas BWV Anh | Order of Discussion

Cantata BWV 116
Du Friedefürst, Herr Jesu Christ
Discussions - Part 2

Continue from Part 1

Discussions in the Week of April 27, 2003

Aryeh Oron wrote (April 29, 2003):
BWV 116 - Introduction

The subject of this week’s discussion (April 27, 2003) is the Chorale Cantata BWV 116 ‘Du Friedefürst, Herr Jesu Christ’ (You prince of peace, Lord Jesus Christ) for the 25th Sunday after Trinity.

The connection between the cantata and Jakob Ebert’s hymn on which it is based was well described by Francis Browne in a message he sent to the BCML couple of days ago. You can read it at the following page: Cantata BWV 116 - Discusssions
Francis’ translation of the Chorale is located at the page:
http://www.bach-cantatas.com/Texts/Chorale001-Eng3.htm

Recordings

The details of the recordings of this cantata can be found at the following page of the Bach Cantatas Website: Cantata BWV 116 - Recordings

This cantata has 5 complete recordings, all of which come from familiar forces: Richter (1977-1978) [1], Rilling (1980) [2], Harnoncourt (1981) [3], Leusink (1999) [4] and Koopman (2000) [5].

Additional Information

In the page of recordings mentioned above you can also find links to the original German text and various translations, three of which were contributed by members of the BCML: English (Francis Browne), French (Jean-Pierre Grivois), and Hebrew (Aryeh Oron). There are also links to the Score (Vocal & Piano version, located at the BCW) and to commentaries: in English by Simon Crouch and in Spanish by Julio Sánchez Reyes (CantatasDeBach).

Music Example

You can listen to Music Examples of the poignant aria for alto (Mvt. 2) from all five recordings of the cantata: through the following page: Cantata BWV 116 - Music Examples

I hope to see many of you participating in the discussion.

Thomas Braatz wrote (April 29, 2003):
BWV 116 - Provenance:

See: Cantata BWV 116 - Provenance

Dürr’s Commentary:

See: Cantata BWV 116 - Commentary

Neil Halliday wrote (April 30, 2003):
Aryeh Oron wrote:
"You can listen to Music Examples of the poignant aria for alto (Mvt. 2) from all five recordings of the cantata: through the following page: Cantata BWV 116 – Music Examples"
I decided to do a quick 'blind' test with these examples (by just pressing the left-hand links, and came up with: 2, 1, 4, and 5 as the order of attractivenss. (no. 3 had a very long buffering time, so I knew this was the Leonhardt/Harnoncourt example [3] - it's a problem I often have with this site; I hope to listen to it later.)

Notice this is also the order of tempi, from slowest to fastest. I feel the slower tempi best bring out the anguish expressed by the chromaticism in the music.

All four vocalists bring a good deal of expressiveness to the music; the big difference is in the orchestras. Koopman (no. 5) [5], in the continuo, displays an unfortunate tendency to 'primness', which is caused by excessive detachment of musical phrases, and overuse of staccato (and a short staccato at that).

While I'm on this point, my two major dislikes in HIP are this tendency to 'primness' (overuse of short staccato), and its opposite extreme - an extremely raucous style brought about by attacking the basses and cellos over-vigorously and at high speed. (We have all heard this in, for example, HIP versions of Vivaldi's "Four Seasons".)

Conductors such as Herreweghe, Suzuki and Gardiner, with their very polished performances which I often find attractive, also need to be on guard to avoid their performances sounding 'prim' in Bach's cantatas, which ought never to display this characteristic.

Thomas Braatz wrote (April 30, 2003):
BWV 116 - Commentaries:

See: Cantata BWV 116 - Commentary

Bradley Lehman wrote (April 30, 2003):
Eric Chafe's harmonic analysis (posted by Tom Braatz, thanks Tom!...see above) brings out some interesting points. This cantata is indeed in the extreme sharps, and Bach uses that to expressive advantage...especially in the G#-minor passages (as Chafe points out) and in the final recitative.

I don't think this is fully appreciated without a bit of explanation of temperaments. So, here goes:

- Regular meantone: as the word "regular" implies, all the fifths are the same size...except for the "wolf" that is not really a fifth, namely G# to E-flat, a diminished sixth. The 'circle of fifths' does not close. The best keys for music are: up to three flats, or up to four sharps, plus the exceptional F minor (four flats). [The same 15 keys used by Bach in the Inventions/Sinfonias....] There are various "flavors" of meantone, depending how tightly the fifths are all tempered: 1/3 comma, 1/4, 1/5, 1/6.... This is most noticeable in the resultant sizes of the major and minor thirds.

+++ In 1/3 comma, the minor thirds are pure and the major thirds are narrow (!)
+++ In 1/4 comma, the minor thirds are wide and the major thirds are pure
+++ In all the others of smaller fractions, major and minor thirds are all wide
+++ At about 1/6 to 1/8 comma, all the keys become musically usable
(although this is a matter of taste, of course)...but see below.

- Irregular (modified) meantone: start with a regular meantone on all the natural notes A-B-C-D-E-F-G, and then (to taste) raise the sharps F#-C#-G# slightly and lower the flats Bb-Eb slightly, so the circle almost closes; all keys become usable. This is a very "common practice" style of temperament, especially in 17th century France and Italy (documented), and I suspect also in England and Germany (not documented, but from evidence inside the music...I explored this in one of my doctoral projects). And I keep my harpsichord set up this way most of the time.

- Well temperament: the fifths are a variety of sizes, either narrow or pure; no fifth (or diminished sixth) is allowed to be wider than pure. Therefore the circle closes. All the keys sound different from one another, due mainly to the varying sizes of the major thirds: the keys with the fewest sharps or flats are the best in tune, and it gets progressively farther out as one adds more accidentals, but all keys are usable. These styles of temperament were known by the 1690s (or earlier), and were common into the 20th century.

- Equal temperament is technically a 1/12 comma regular meantone! All the fifths are the same size, and the amount of tempering is such that the circle closes (enharmonically, although it is still technically a diminished sixth in there). All the major and minor thirds are extremely sharp (like having a constant vibrato in them); and all keys sound the same. There's never any real repose anywhere, because all triads have those tense and rapidly beating thirds in them. Modulation doesn't really mean much, other than a change of general pitch level; there are no clear destinations. How do we know if we're home or in a foreign land, if the landscape is the same?

Back to the other meantones for a moment: regular and modified. Here is the reason why some of the keys are less usable than others:

The notes available are A-B-C-D-E-F-G, plus F#-C#-G# and Bb-Eb. That is, the other notes such as A#, D#, Gb, Db, Ab, E#, B#, etc. really DO NOT EXIST in the temperament, but they might come up in the music anyway. In effect, in practice, a wrong-note sharp (such as D# or E#) is considerably higher tit should be, and a wrong-note flat is lower than it should be. Depending how strong the meantone tempering is, and on any modification as I mentioned above, these are somewhat usable in some musical contexts...and they stand out as effects to the listener.

Here's a little summary, showing where the accidentals are in comparison with equal temperament:

C# - lower
(Db) - lower, sounds wild
(D#) - higher, sounds very sharp
Eb - higher
E# - higher
F# - lower
(Gb) - lower, sounds wild
G# - lower
(Ab) - lower, sounds wild
(A#) - higher, sounds very sharp
Bb - higher
(B#) - higher, sounds very sharp

The G# to Eb diminished sixth (as noted above) is also a special case since it's not really a fifth; therefore G#-minor is especially nasty (unless one gets off the Eb quickly after playing it...there are ways). And Ab-major is even worse, because the "fifth" and the major third (really a diminished fourth) are horrible.

And the double-sharps also make things very interesting, but I won't go into that here.

Any chord that is spelled incorrectly in the music (having the "wrong" enharmonics in it, for example "B major" of B-Eb-F#) will be more out of tune than a chord that is spelled correctly (for example, "A major" of A-C#-E). And any chord that is spelled correctly will be much better in tune than it is in equal temperament: that is, about 95%+ of practical tonal music!

Diminished triads and sevenths are all usable in all the meantone temperaments; and they make very strong dramatic effects because those minor thirds in them are nearly pure. (Really, the diminished chords are usable in ANY temperament because the ear can tolerate almost anything that is given to it as a minor third, whether it resembles a 6:5 or a 7:6; and diminished chords are built entirely of minor thirds.) Music that uses many diminished chords, such as this Cantata 116, makes an especially potent effect in these meantone-based temperaments. [An especially strong one: the opening stroke of the last recitative...Fx, A#, C#, E! That F-double-sharp is the wrong note G in the temperament, and the A# is the wrong note Bb.]

=====

All of the above is a matter of degree. The tighter the tempering of the fifths is, the more strongly tonal (focused and stable) the temperament will sound when playing in the good keys, and the more intensely expressive it will sound when playing the wrong notes. It's like choosing the appropriate amount of spice for food.

And the most important thing to realize is: the wind instruments (including the organ) were designed to play those particular notes: A-B-C-D-E-F-G, plus F#-C#-G#, plus Bb-Eb. Players can bend these somewhat, but those are the basic notes.

And the music is written with that practical awareness: notes other than those will cause more spicy intervals in the ensemble, and on the keyboard. Composers can use that knowledge to make musical effects...as Bach does here. Music in the extreme sharp keys, like this cantata, will make an especially bright overall effect due to all the "wrong" A#s and D#s (etc.) in it.

=====

I listened to the two recordings I have (Harnoncourt [3] and Leusink) [4], and noticed that Harnoncourt is using something close to 1/6 comma meantone temperament, and Leusink's sounds like a well temperament. Harnoncourt's temperament seems to be modified a little bit in the usual ways (noted above). Additionally, I played through the cantata here on my harpsichord, in regular 1/6 comma--no modifications--the temperament of some of the organs Bach knew (Silbermann's, most notably). It works beautifully in regular 1/6 comma! But it's understandable that they would want to soften it a little bit for a recording: it still makes the same effects, almost as strongly.

The effect in the music (due to the temperament), to my ears, is an overall brightness to this cantata, like a series of yellows and reds and oranges; and when Bach gets into the wilder keys, it is even brighter, like mixing a bit of garish green, pink, and salmon together into the overall yellowish. TOTALLY different from the world of F major and Bb major....

The whole cantata stays tensely strung like this, until the perfect spot to resolve it, which is exactly what Bach does: at the end of that last recitative, in the last three bars, he stops using all the "wrong" accidentals. And the words are "uns bestaendig Friede bringen." The temperament's tension FINALLY relaxes in those final three bars...the effect is like suddenly having peace and fresh air. No more crosses (or double-crosses) to bear! He's been depicting foes and strife and pain and harshness and guilt with those wickedly out-of-tune misspelled notes, and here (after 15 minutes of that) he finally lets it relax.

IMO, Leusink's well temperament is too tame for this piece, too moderate: there are some contrasts among the keys, but they're not vivid enough. (Better than nothing, though.) And anybody using equal temperament is missing a vast world of expression that Bach wrote here. All the features that Chafe points out here are for naught, if all the keys sound the same.

And I wonder if Chafe himself was aware of the practical differences between F#-minor and G#-minor, specifically. He wrote this: "The move from F sharp minor to G sharp minor in the aria represents a false move upward, prompted by fear, (...)" F# minor is a very screwy key to begin with, at the edge of usefulness; but G# minor is over that edge because it doesn't have a real fifth in the tonic triad. G# minor is wildly unstable. That's the point here: it's not just some rather generic move "upward," it's a step off the edge of the planet. There be dragons.

Bradley Lehman wrote (April 30, 2003):
And I forgot to sum up:

With an appropriate temperament, all this business with key contrasts and "extreme sharps" etc etc etc is not merely a collection of symbolically hidden Easter Eggs for the cognoscenti, clever jokes on the word "scharfen" etc etc etc. It is in-your-face audible drama that everybody is supposed to get, like being slugged over the head or biting into a chili pepper. When Bach cadences into G#-minor on the word "schreien" in the alto aria, that thing is supposed to hurt like hell.

Thomas Braatz wrote (May 1, 2003):
[To Bradley Lehman] Brad, your explanation of the temperament used and its probable, intended effects upon the listener is very helpful ‘in shedding light’ (and even color!) upon a deeper understanding of what Bach is doing here and elsewhere in his sacred music.

You stated: >>And the double-sharps also make things very interesting, but I won't go into that here.<<
How would you explain his use of double-sharps in Mvt. 2 (alto aria) ms. 45-52 on the words “wie du, o Jesu, selbst verlangst zu Gott in deinem Namen schreien;” Mvt. 4 (Terzetto) ms. 108 on the words “um Geduld;” and the opening long, held note in the bc and viola at the beginning of the alto recitative, Mvt. 5 which begins with “Ach, laß uns durch die scharfen Ruten?

Thomas Braatz wrote “(May 1, 2003):
Chafe on the compromise that the ‘circle of fifths’ involved:

Unlike earlier styles, the baroque – in a highly suggestive parallel with philosophy – developed a feeling for the spit between the external, material sides of art and the internal, spiritual, and expressive sides. Renaissance art predicated the unity of inner and outer structures; architecture, for example, numerically reconciled ‘ratio’ or abstract proportion, reflecting the divine harmony, with ‘sensus’ or perspective. The baroque intensified the concern with the relationship of ‘sensus’ and ‘ratio’ but saw it as the split between inner perception and outer or objective reality. This view went some way toward overturning the conviction that the intrinsic aspect of music represents, or at least aspires to, a higher, eternal, unchanging level. The recognition of a split soon led to the baroque emphasis on ‘sensus’ as internal, subjective, and ‘ratio’ as external. From that point it was a relatively short stepto the complete reversal: ‘sensus’ as the intrinsic aspect. Perfection was no longer, it seemed, an attainable goal of art, and baroque art readily acknowledged its imperfection in countless ways. Just intonation was abandoned even as a theoretical goal, and harmonic purity yielded to distorting expressive devices.

Moreover, there almost seems to have been resistance to the inevitable association of the full range of major and minor keys with the concept of the perfect, closed circle. At first the word ‘circle’ appears more as the key, or Ariadne’s thread, to a practical method of modulation then as a full-blown theoretical concept, and until well into the 17th century many of the tempered intervals seem to be viewed as necessary evils, providing us, in Werckmeister’s words, with a “mirror and representation of our mortality and imperfection in this life.” The first presentation of the circle in print recognizes “difficult” and “unusable” keys, as well as tonal “extremes” that still fall short of being enharmonically equivalent.

Despite rapid alterations in these perceptions – for example, Johann David Heinichen’s 1728 revision of his 1711 book – the notion of difficult and imperfect keys remained an important part of baroque theory, not as a sign of failure to rationalize the maze of keys but as an inevitable consequence of the sense of compromise the circle involved. Only with the generation of Mattheson and Heinichen is ‘sensus’ finally elevated above ‘ratio’ and the circle of keys becomes a fully articulated, visualizable presence, giving final definition to the tonal material of music and replacing a great deal of the sense of imperfection and loss of modal variety with what would soon become a ubiquitous musical symbol of resolution and completeness in its own right. The circle of keys would also ultimately become transparent, losing the materiality and artificiality that kept it from being taken for granted in the baroque.

Neil Halliday wrote (May 1, 2003):
BWV 116 - 2nd movement

After playing the (piano?) reduction score (David Zale site) of the 3rd movement on the organ, I have to conclude that all these examples give a poor realisation of this astoundingly chromatic, emotionally powerful piece of music.

What we basically get in the recorded examples are three musical lines - the alto voice, the oboe line, a soft bass (cello) line, with a barely audible chamber organ in the background. In effect, as far as being able to easily hear the pitch of the notes is concerned, we have a piece consisting of three musical lines only.

Is this what Bach intended us to hear? (Toss in an alto with heavy vibrato, and you have only two lines of definable pitch, especially in those places where the alto sings successive notes belonging to the chromatic scale.)

Those of you who have access to the above score (which is apparently a 19th century realisation of the figured bass by Reger or someone else whom Tom suggested) will understand my point. This music really takes shape in one's mind (or is it just me, I admit its a possibility) when played on the organ as written here. (Have a look at the two bars on "Angst", and the bars with the modulation out of the G sharp minor section - wow!).

I notice no one has commented on this movement (other than tangentially as with Brad's exposition on tuning) - is this because it sounds 'strange', 'boring' and 'unattractive' in these recordings? (which was my initial response to it).

Bradley Lehman wrote (May 1, 2003):
Double sharp in BWV 116

Thomas Braatz wrote:
< Brad, your explanation of the temperament used and its probable, intended effects upon the listener is very helpful â?~in shedding lightâ?T (and even color!) upon a deeper understanding of what Bach is doing here and elsewhere in his sacred music.

You stated: >>And the double-sharps also make things very interesting, but I won't go into that here.<<

How would you explain his use of double-sharps in Mvt. 2 (alto aria) ms. 45-52 on the words “wie du, o Jesu, selbst verlangst zu Gott in deinem Namen schreien; >
That passage is (harmonically) a pretty straightforward establishment of a key: cadencing into G# minor. And then in the five-bar instrumental interlude there he cadences into G# minor again...making sure we spend some time in this, one of the most painful keys available. Functionally, the double-sharped notes here are merely filling their normal harmonic roles: as raised 4 and raised 7 of the key that is being established. Those are part of the dominant, and the "dominant of the dominant," in G# minor.

And, of course, there's also that chromatic vocal line there: a striking rhetorical device, whatever key it happens to be in. [See below.]

< Mvt. 4 (Terzetto) ms. 108 on the words “oum Geduld”? >
He starts a rather long sequence of descending fifths there (harmonically). He jumps to a D#-major chord (dominant of G#), and then everything that follows is a sequence (look at the bass downbeats): G#, C#, F#, B. Then F#, B again...he wants us to land in C#-minor, which is where we were before this sequence started, but also to seem that we've gone somewhere in the interim.

< and the opening long, held note in the bc and viola at the beginning of the alto recitative, Mvt. 5 which begins with “Ach, laß uns durch die scharfen Ruten?” >
That chord Fx-A#-C#-E is the vii-o7 (fully diminished 7th chord on the 7th scale degree) of G# minor. Bach hits it without any preparation...certainly grabs our attention, doesn't it?! Then he surprises us again with the way he resolves it: not G# minor, but G# major (surprise both in the singer's part and in the figures), and the bass moves to an unexpected note also: making it into a seventh chord in 6-4-2 inversion. That would "inevitably" go to a C# minor chord in 6 inversion (E in the bass), but he shocks us again by sticking a vii-o7 (B#-D#-F#-A) before he gets to his C# minor. (Remember when I said some weeks ago that a 6-4-2 in the figures almost always goes to a 6 next? And when a composer breaks that expectation, it's a surprise.)

As I pointed out yesterday, due to the temperament, G# major is even more unstable than G# minor is, because the major third is misspelled (G#-C is really a diminished fourth)...plus, as I've pointed out, he's added the 7th to this and put it into the bass (the F#). This all makes the word "Ruthen" really hurt, like being whipped or caned: that excruciating sting immediately after being struck, before the skin goes numb. Text-painting? Bingo.

=====

Now, on the even wilder contribution to all this from the temperament: what do those double-sharps sound like? All three of these passages are cadences into G#. The dominant chord to G# is D# major: D#-Fx-A#. ALL THREE of those notes are not in the temperament. Instead we hear Eb-G-Bb, a chord that (internally) is spelled correctly, and is therefore very well in tune! So, the effect is: in a dominant-tonic cadence into G#, the dominant chord is nicely in tune, and stable...and then the resolution into the tonic drops off the edge of the planet, giving us the worst chord
available in the temperament!

That's what the double-sharps do here. They've crossed over, leaping the gap where the circle of fifths doesn't close, and sound innocuous because we hear them as belonging to the flat keys (instead of the sharps). All pleasant and sweet. E-flat major is such a lovely key. And then the natural harmonic and melodic motions, their resolutions, explode in our faces.

Reminds me of Monty Python's "Crunchy Frog" sketch. Take a nice chocolate from the box, pop it into your mouth expecting a treat, and BAM!--it shoots steel bolts out through your cheeks.

=====

One other useful thing to know about the meantone temperaments: there are two different sizes of semitones (half-steps). A "diatonic" semitone is one where the note name changes, as in F# to G, or A to Bb. A "chromatic" semitone is one where the note name stays the same, as in G to G#. (Chromatic, change of color, a differently colaspect of the *same* note functionally! This phenomenon can be heard very clearly on a keyboard instrument tuned in a meantone temperament: simply play a series of E minor and E major triads, alternating. It seems like the same chord, except for a subtle change of color...and that's what it is. Or C major and C minor; or G major and G minor...whatever, as long as all the notes in these triads are spelled correctly.)

In all the meantone temperaments, the diatonic semitones are larger than the chromatic semitones. (MUCH larger: for example, in 1/4 comma meantone, they are almost twice as big! 1/4 comma and 1/3 comma meantone are both almost like having 19 equally spaced intervals within an octave, but 7 of them missing. Yes, there have been split-key instruments built to handle this, offering more than 12 notes per octave....) That is, in meantone temperaments, "leading tones" from the 7th degree to the tonic (like F# to G in G major) are the big ones.

When a composer uses the wackier keys, like in this cantata 116, employing notes such as D# and A# that are not in the temperament, we're hearing chromatic leading tones there rather than diatonic ones. They stand out. When an Fx (really sounding as a G) resolves as a leading tone to a G#, it's one of those small chromatic semitones...the wrong one. That wrongness is an expressive effect, a useful device if the composer wants to make some point with it.

And, in any chromatic melodic line such as the one mentioned above (alto aria), it's those two different sizes of semitones alternating. That's why chromatic scales (melodically) are so weird.

They remain weird (but in a different way) in the well temperaments. The diatonic semitones remain larger than the chromatic semitones, but there are more than two sizes. There can be as many as 12 sizes. Fun, eh? Fun for other instrumentalists trying to match their intonation to the keyboard? You bet. That's why I pointed out (approx. a month ago) that a mild regular meantone (such as 1/6) is easier to play with than automatically slapping Vallotti or Young or Werckmeister or one of the other currently popular well temperaments onto the keyboard for ensemble music. Sure, those well temperaments are "all-purpose," but that flexibility doesn't necessarily make them a good solution. The regular meantones have *only* two sizes of semitones.

Thomas Braatz wrote (May 4, 2003):
BWV 116 - The Recordings:

This week I listened to the following recordings:

Richter (1977-8) [1]; Rilling (1980) [2]; Harnoncourt (1981) [3]; Leusink (1999) [4]; Koopman (2000) [5]

The Timings (from slowest to fastest):

TT: Rilling (17:57); Richter (17:22); (16:14); Leusink (15:52); Harnoncourt (14:34)

Mvt. 1: Rilling (5:41); Richter (5:07); Leusink (4:39); Harnoncourt (4:12); Koopman (4:08)

Mvt. 2: Rilling (3:53); Richter (3:22); Koopman (3:09); Leusink (3:06); Harnoncourt (2:58)

Mvt. 3: Rilling (0:51); Koopman (0:50); Richter (0:48); Leusink (0:48); Harnoncourt (0:44)

Mvt. 4: Koopman (6:12); Richter (5:46); Rilling (5:20); Leusink (5:17); Harnoncourt (4:36)

Mvt. 5: Richter (1:20); Rilling (1:12); Harnoncourt (1:06); Leusink (1:02); Koopman (0:58)

Mvt. 6: Rilling (1:00); Leusink (1:00); Richter (0:59); Harnoncourt (0:58); Koopman (0:54)

The total time picture reveals just what one would come to expect as a difference between non-HIP (Richter [1], Rilling [2]) and HIP (Harnoncourt [3], Leusink [4], Koopman [5]) versions of a Bach cantata. The over 3-minute difference between Richter and Harnoncourt is substantial and is due to a number of factors already mentioned here previously. This type of difference is bound to have an effect upon the presentation of the music and how the listener will perceive it.

An unexpected surprise occurs in Mvt. 4, where Koopman presents the slowest reading of this trio.

Mvt. 1 (Chorale Fantasia):

Koopman [5] provides an excellent example of what a truly lite version of an introductory chorale cantata mvt. sounds like. This is the goal that many current HIP practitioners (Gardiner {with a bit more energy and speed}, Herreweghe {lightly floating celestial, ethereal sounds} are striving for. This is excellent background music and some listeners who actually attend more closely might think “I did not know that Bach’s cantatas could sound so pleasant and entertaining. This is delightful dance music. Let’s hear more of this sort of thing.” Although Leusink’s version [4] is slower, it still retains the lightness that comes from simply tapping notes quickly and softly. There is no stamina or backbone in the orchestra except for the booming double bass in the bc. Compared to Koopman, Leusink’s choir gives a very lackluster performance. Leusink skips the horn part entirely. Bach certainly knew what he was doing here. The sopranos in Leusink’s choir are unable to sing in one breath “ein starker Nothelfer du bist.” The horn would have helped to sustain the chorale melody. Even with his forever shaky oboi d’amore, Harnoncourt [3] is able to obtain more energetic playing from his orchestra. Harnoncourt’s problem is that he uses too much staccato. The violins sound thin and flimsy. Harnoncourt dissects the beautiful fugal entries and provides crazy accents so that the entire flow is interrupted. Sometimes the choir creates hideous sounds (ms 47-49) and at other times he allows notes to die out prematurely in the middle of a word (ms. 41ff. “ein star-“ and ms. 56ff. “im Le-“) or the final notes to simply expire prematurely as if someone had pulled the plug on a vacuum cleaner ms.20, 27, 84 .) Of the two remaining non-HIP versions, Richter [1] is able to bring more energy into the presentation by combining both legato and staccato elements in a good balance. With Rilling [2] the effect is predominantly legato which deprives this mvt. of some of its forward momentum. The great drawback with the Richter performance is the organ squealing in the highest ranges a duplication of the choral parts (it is unfortunate that this mars an otherwise excellent recording.) Rilling’s slowest tempo allows for the expression of solid strength that is offered mankind in both life and death. This is best expressed through sustained sound devoid of shocking irregularities. For this reason I would place Rilling’s version above Richter’s which also has many good features that would allow it to come in as second best.

Mvt. 2. (Alto Aria):

There are three female altos singing this aria, all with more or less operatic-style voices: Schmidt (Richter) [1]; Watts (Rilling) [2]; and Markert (Koopman) [5]. Schmidt (Richter) provides the most musically satisfying version of the heartrending fear and anguish that Bach wished to project in this aria. Rilling wisely chooses the slowest tempo for this aria. This allows the singer, Watts, to develop the full gamut of feeling. A slow tempo is commensurate with the emotions that are to be projected here. [Watts is the only singer in this group of recordings to use a repeated note pattern in ms. 41 & 42 instead of a vibrato which the wavy line above the dotted half-notes is supposed to represent.] Unable to probe the depth of feeling in this aria is Markert’s demi voix version with a light accompaniment supplied by Koopman’s bc and oboe d’amwhich remain overly restrained and simply tap many of the notes lightly. There are two counter tenors singing this aria: Esswood (Harnoncourt) [3] and Buwalda (Leusink [4].) Harnoncourt hurries Esswood through this aria with the fastest recorded tempo. Combine this with a solo oboe d’amore that sounds as if it came straight from hell and Esswood’s gasping “Ach”’s and ‘shakes’ = coloraturas and you get a version that is dramatic in a very different way. This is drama-lite or a kind of artificial drama, but certainly not the type of musical presentation that moves the soul in its depths with a feeling of almost unspeakable misery and fear. Buwalda’s version sounds more like a caricature than Esswood’s does.

Mvt. 3 (Tenor Recitative):

I would place Schreier (Richter) [1], Equiluz (Harnoncourt) [3], and Prégardien (Koopman) [5] into the class of excellent singers. Harder (Rilling) [2] is also extremely good, but I am slightly uncomfortable with the nasality of his voice. Reasonably good, but not strikingly so, is Schoch (Leusink [4].)

Mvt. 4 (Terzetto)

The non-HIP performers are Mathis, Schreier, Fischer-Dieskau (Richter) [1] and Augér, Harder, Huttenlocher (Rilling) [2]. Both of these recordings make it quite evident that this terzetto is not easy to sing and even more difficult to sing together in such a way that it does not sound like the piece is falling apart or going out of tune. In the Richter group, it is Mathis who is unable to achieve any kind of togetherness with the other singers. See my comments at: Jan 20, 2003 http://www.bach-cantatas.com/Topics/Sing-Style.htm
Rilling has a double-bass and cello playing along much too loudly with the organ in the continuo and while the voices here are much more aware of each other, this piece still eludes these singers. Harder’s unique voice does not blend well with the other two singers.

The HIP performers are Huber, Equiluz, Huttenlocher (Harnoncourt) [3]; Holton, Schoch, Ramselaar (Leusink) [4]; and Rubens, Prégardien, Mertens (Koopman) [5]. Possibly Harnoncourt (who, with great delicacy, plays along on the bc with the organ) felt that he had to take this very fast tempo to help the poor Tölzer boy soprano, Huber, along. Huber simply is not able to sing such a difficult part as this meaningfully. There are also intonation problems throughout which make listening to this version very unsettling. The Leusink version also is faced with intonation problems, if not quite as apparent as those in the Harnoncourt version. This is definitely a demi voix version. You get to hear the notes, and the voices sort of blend with each other on a very low sotto voce level, but isn’t there more to this music than this? The Koopman version, with its much slower tempo has a lot more going for it: two excellent male singers and a sensitive bc accompaniment. Once again, the spoiler is in the soprano part. Rubens is simply not able to supply what is necessary: at times the bottom of her range ‘falls out’ entirely and at other times she simply howls the higher notes or loses control over her fluttering vibrato.

Mvt. 5 (Alto Recitative):

There are 3 versions of this recitative with female altos, all of which sound operatically dramatic in their treatment of this text: Schmidt (Richter) [1] and Watts (Rilling) [2] are fully trained voices with ample power to project their voices, while Markert (Koopman) [5] is a demi voix singing the same text primarily sotto voce as if in a chamber music performance. For this reason I would prefer either of the previous two voices even though their voices can be somewhat overbearing at times. Between the male counter tenors Esswood (Harnoncourt) [3] and Buwalda (Leusink) [4] there is no real contest. Buwalda has even less of a voice than Esswood has.

Mvt. 6 (Chorale):

Richter’s version [1] at least has the right amount of solid, sustained sound, the type of choral sound that offers strong confirmation of belief in the words being sung. The fermati over the quarter notes in the middle of the Stollen and the Abgesang (the dotted half-notes at the end of each line area without fermati) are held for 3 beats, thus making the ends of all the lines in the chorale equal. This method seems quite reasonable as it provides the type of balanced structure that Bach must have intended. The negative aspect of this version is again the miserable registration (squealing stops) of the organ which is heard over the choir and even enters prematurely before everyone else on one occasion. There is a noticeable ritardando at the very end as a final reminder to any listener that this is romantic performance with tempo flexibility and not a rigid HIP where fermati and ritardandi are usually (not always) frowned upon.

Rilling’s version [2] has somewhat less intensity and its effect is more one of a matter-of-factness than a strong confirmation of belief in the words being sung. There is a nice ritardando at the very end that is similar to Richter’s, but with Rilling there is also a lessening of tension rather than an effort directed toward sustaining the intensity until the very end as with Richter. Both Rilling and Richter have in common a sustained, legato effect throughout the chorale; however, Rilling adds 3 sharp caesuras after “Herz, Scherz, bist” on the very notes (the only ones that Bach marked this way with fermati!) which have fermati over them. I had searched extensively in the New Grove and MGG when I researched the history of the fermata and have been unable to document the ‘tradition’ that would have Rilling (and most of the HIP conductors) shorten or overlook completely the actual note value of the note under the fermata. What usually happens with fermati of this type is that the conductor (in the manner that Rilling uses here, although Rilling usually avoids abbreviating the value of the note as much as the others do) decides that note 1) loses its fully stated value; 2) becomes unaccented, hence is sung with less volume; and that the word under the fermata is 1) is pronounced hurriedly – the final consonant is reached faster than with any other words in the text; 2) becomes garbled or swallowed up in an indecipherable schwa-like grunt. Yesterday I found the type of ‘proof’ that I was looking for as I investigated how unreliable are the ‘main Bach editions’ [not the BGA or NBA – unless referring to the so-called ‘performing’ editions (scores and parts derived from them) – Brad Lehman has pointed this out and I later found confirmation in a comment by Alfred Dürr that these performance editions are not entirely reliable.] I was looking for confirmation why Günther Ramin would remove phrase markings, etc. from a performance score such as those provided by Edition Peters and Breitkopf & Härtel.

Now I find this evidence: BWV 88 mvt. 7 (final chorale) [NBA 17.2, p. 58] vs. Piano Reduction of the Score published by Breitkopf & Härtel (Leipzig) in 1907 as part of the “Veröffentlichungen der Neuen Bachgesellschaft, Jahrgang VII, Heft 2. The latter appears, for all practical purposes, like any other piano reduction from the turn of the century (both Breitkopf & Härtel and Edition Peters were involved in producing scores and parts for the Bach cantatas.) The editor of the latter piano reductiois listed as Max Seiffert and the piano reduction was prepared by Otto Taubmann [sic.] Both, probably more the former than the latter, took unbelievable liberties with the original score that was available to them in the BGA. In Mvt. 1, the NBA shows that the phrasing for the violins and oboi d’amore in 6/8 time joins all 3 notes of each grouping of 3 while the Seiffert version has a phrasing of only the 1st two notes of each grouping of 3, a very different ‘feel’ to the manner in which these notes will be played and heard. Non existent in the Bach Urtext are the frequent ‘ < >’ markings to indicated crescendi and diminuendi, ‘cresc. poco a poco’ ‘mp =mezzo piano’ and ‘mf=mezzo forte’ poco rit. and rit. but where Bach (in the NBA) marks expressly ‘pianissimo’ there is nothing at all in one instance or ‘delictamente’ in another.

Getting to the point: the final chorale of BWV 88 in the NBA shows a fermata on the last note of each line of the chorale, sometimes this occurs over a quarter note, and, on one occasion, the fermata is over a dotted half note. As heard in performance correctly sung according to the Bach score with the Stollen repeated, there should be 6 ‘holds’ or fermati evident; however, the Seiffert performance score (I have to assume here that Taubmann simply copied from the performance score which Seiffert had prepared from the BGA) has *only ONE* fermata at the very end of the chorale. The other fermati have been replaced by ‘commas’ = breath marks which imply the type of shortening alluded to above. In one place Seiffert/Taubmann even added an additional breath mark after ‘denn’ where the original text shows no comma. In this instance the editors probably added the comma after ‘denn’ and then added the breath mark.

Conclusion on this matter:

It is my distinct impression (excluding the editing of the BGA) that the last thing in the minds of most performers at the turn of the century (1900) was to remain true to the original Urtext. If these performers (including those trained as Thomaner) could change Bach’s original cantata texts [Voigt gives numerous suggestions on how to improve Bach’s ‘horrible’ texts and even music (by cutting out ‘all the boring and uninteresting’parts)], then they would ‘not bat an eyelash’ when it comes to adding and removing fermati from the chorales. Perhaps it is not amazing then to see and hear performers such as Leusink [4], who admittedly could not or did not check the NBA for the correct Urtext version, but relied upon editions that were edited/created at the turn of the last century, present romanticized versions based upon ‘loose’ attitudes that prevailed when these performance editions were created over a century ago.

Back to BWV 116, the present cantata:

Leave it to Harnoncourt [3] to come up with something different (which does not make sense anyhow): Harnoncourt completely rewrites the notation of Bach’s score; for example, where Bach has a fermata over a quarter note, Harnoncourt perversely adds an extra quarter-note rest, so that it almost appears that he has placed the fermata over the rest (which does not exist in the Urtext) and not the actual quarter note. This extra rest sounds strange and this is perhaps the only reason Harnoncourt has chosen to perform it in this fashion. There is less of the usual ‘chop, chop’ effect that Harnoncourt usually applies in a simple Bach chorale. The choir has amazingly poor diction and Harnoncourt simply allows the final consonants on important words such as “Gnad, schad, bist” to slip into oblivion. As an instrumentalist, Harnoncourt does not feel a need for words to be communicated clearly. He is more interested in “Effekthascherei” [“straining artificially for effect” “mainly interested in showiness”] and this is demonstrated in his performance of a simple Bach chorale where such a thing has no place.

Leusink [4], using performance materials that originated over a century ago, and not bothering with checking with the NBA and NBA KBs because he did not have time to do this during his marathon recording schedule, performs the fermati in a manner similar to Rilling, but with the notable and characteristic distinction (very typical for Leusink’s chorale singing style) that the note under the fermata is both abbreviated and deemphasized/unaccented, leaving the listener under the impression that the final words under the fermati are, for some reason, less important than the preceding ones. Nothing could be further from the truth! A careful listener who is concerned about the words as well as the music will be forced to conclude that something is definitely wrong here. Take, for instance, the line “O Jesu Christ, allein du bist,” with the fermata over the deemphasized (the volume is much lower on the unaccented word, ‘bist’) last note of the line. Understanding this line the way Leusink has it sung, it means something like: “O poor Jesus Christ, you are so alone that I feel sorry for you.” Singing the line correctly the way Richter does with the word “bist” equal in power and strength to all the other words and even receiving more emphasis as it is held out in full voice (no decrescendo or ‘dying away’) as it is directed without pause to lead into the final line, the chorale line(s) now read as follows: “O Jesu Christ, allein du bist, der solchs wohl kann ausrichten” [“O Jesus Christ, you *are* the *only one* who can do this for us.”] All of this occurs because there is a lack of continuity between the lines of a chorale when the final words/syllables are dropped to much lower volume levels and when the final notes under the fermati are abbreviated/truncated in note value. As usual a balanced sound among the voices is lacking and German diction leaves much to be desired.

Koopman [5] is a vast improvement over Leusink in many ways: there is a well-balanced choral ensemble sound also in good balance with the instruments. The German diction is much improved. The quarter notes under the fermati call too much attention to themselves because there is no fermata at all (like Rilling in this respect.) This is a less satisfactory method for performing these holds because the holds are missing. Why did Bach mark them this way in the first place? Before abandoning these markings and relying on the now proven to be very unreliable performance practices of more than a century ago, each conductor ought ponder long and hard what Bach may have intended here and not simply accept a tradition from a time which was very different from Bach’s own. Koopman’s rendition is very beautiful (very Herreweghian), but it lacks the strength and conviction that should be present at the end of a cantata. This is a point of strong affirmation and faith which the choir expresses in the manner of the congregation, as a proxy for the congregation, not as pleasant background music for a congregation about to doze off after listening to a boring cantata, but rather a congregation that has participated vicariously in the progression of emotion from the beginning to the end. The end needs to be an exclamation mark to all that has preceded it. Simply being beautifully sung may not be sufficient for a final chorale to be presented properly where words and music need to be treated as equally important.

Bradley Lehman wrote (May 4, 2003):
Thomas Braatz wrote:
< Even with his forever shaky oboi d’amore, Harnoncourt [3] is able to obtain more energetic playing from his orchestra. >
What does the word "shaky" mean to you here? I listened to this movement several times, with the Harnoncourt recording, and there's nothing shaking there except perhaps your own shaky understanding of intonation. His oboe players are playing those notes accurately in a meantone-based temperament. If you don't like it, why don't you blame Bach for [deliberately] writing notes outside the temperament, for this extremely bright effect it makes (having the A#s and D#s etc piespecially high)?

< Mvt. 2. (Alto Aria):
There are three female altos singing this aria, all with more or less operatic-style voices: Schmidt (Richter)
[1]; Watts (Rilling) [2]; and Markert (Koopman) [5]. Schmidt (Richter) provides the most musically satisfying version of the heartrending fear and anguish that Bach wished to project in this aria. Rilling wisely chooses the slowest tempo for this aria. This allows the singer, Watts, to develop the full gamut of feeling. A slow tempo is commensurate with the emotions that are to be projected here. [Watts is the only singer in this group of recordings to use a repeated note pattern in ms. 41 & 42 instead of a vibrato which the wavy line above the dotted half-notes is supposed to represent.] >
Watts uses a trillo there? Really? Nifty; I haven't listened to that recording yet. (Trillo is a vocal ornament that is different from a trill.)

< Harnoncourt [3] hurries Esswood through this aria with the fastest recorded tempo. Combine this with a solo oboe d’amore that sounds as if it came straight from hell and Esswood’s gasping “Ach”’s and ‘shakes’ = coloraturas and you get a version that is dramatic in a very different way. This is drama-lite or a kind of artificial drama, but certainly not the type of musical presentation that moves the soul in its depths with a feeling of almost unspeakable misery and fear. >
To this reader, looking at the words you've chosen, it seems that you were INDEED moved to unspeakable misery and fear by this performance (real drama, not "artificial drama" or "drama-lite"), and you are afraid to speak about it.

Alex Riedlmayer wrote (May 4, 2003):
Bradley Lehman wrote:
< His oboe players are playing those notes accurately in a meantone-based temperament. >
Does that have any meaning at all? Perhaps this "meantone-based temperament" is merely an out-of-tune version of meantone?

< If you don't like it, why don't you blame Bach for [deliberately] writing notes outside the temperament, >
And then I would blame you for equating composer with temperament.

Bradley Lehman wrote (May 4, 2003):
Alex wrote: (quoting me first)
<< If you don't like it, why don't you blame Bach for [deliberately] writing notes outside the temperament, >>
< And then I would blame you for equating composer with temperament. >
My entire sentence (part of which you excised) was:
"If you don't like it, why don't you blame Bach for [deliberately] writing notes outside the temperament, for this extremely bright effect it makes (having the A#s and D#s etc pitched especially high)?"

How is that equating composer with temperament? Bach is using the temperament's features (extra brightness due to the enharmonic differences) as an expressive effect. Bach is doing a clever thing to illustrate the tensions in the text. (And one would assume he's doing it deliberately.) Bach knew the expressive characteristics of the keys in which he wrote his music. How is that equating the composer with the temperament?

Bradley Lehman wrote (May 4, 2003):
Thomas Braatz wrote:
< Mvt. 5 (Alto Recitative):
There are 3 versions of this recitative with female altos, all of which sound operatically dramatic in their treatment of this text: Schmidt (Richter)
[1] and Watts (Rilling [2]) are fully trained voices with ample power to project their voices, while Markert (Koopman) [5] is a demi voix singing the same text primarily sotto voce as if in a chamber music performance. For this reason I would prefer either of the previous two voices even though their voices can be somewhat overbearing at times. Between the male counter tenors Esswood (Harnoncourt) [3] and Buwalda (Leusink) [4] there is no real contest. Buwalda has even less of a voice than Esswood has. >
Again we see an extraordinary prejudice here: in Tom's mind, Merkert, Esswood and Buwalda are not "fully trained voices." Why? These gentlemen are fully qualified professional singers, fully trained to sing this music! Why insinuate that they are not?

< Back to BWV 116, the present cantata:
Leave it to Harnoncourt
[3] to come up with something different (which does not make sense anyhow): Harnoncourt completely rewrites the notation of Bach’s score; for example, where Bach has a fermata over a quarter note, Harnoncourt perversely adds an extra quarter-note rest, so that it almost appears that he has placed the fermata over the rest (which does not exist in the Urtext) and not the actual quarter note. This extra rest sounds strange and this is perhaps the only reason Harnoncourt has chosen to perform it in this fashion. There is less of the usual ‘chop, chop’ effect that Harnoncourt usually applies in a simple Bach chorale. The choir has amazingly poor diction and Harnoncourt simply allows the final consonants on important words such as “Gnad, schad, bist” to slip into oblivion. As an instrumentalist, Harnoncourt does not feel a need for words to be communicated clearly. He is more interested in “Effekthascherei” [“straining artificially for effect” “mainly interested in showiness”] and this is demonstrated in his performance of a simple Bach chorale where such a thing has no place. >
Hmm. Tom, usually you complain that Harnoncourt [3] makes too little of the fermatas in chorales; but this week you're complaining that he makes too much of them! It just looks as if you're predisposed to dislike whatever Harnoncourt does, whatever it is...and then when he does something you approve of (less choppiness) you find another way to castigate him, to make sure your review of him comes out negative.

Have you seen Harnoncourt's performance scores? How do you know if he "rewrote" anything?

You say that "this extra rest sounds strange and this is perhaps the only reason Harnoncourt has chosen to perform it in this fashion." That is, in your mind, Harnoncourt does something ONLY to be strange. That's verrueckt!

Alex Riedlmayer wrote (May 4, 2003):
Bradley Lehman wrote:
< Bach is doing a clever thing to illustrate the tensions in the text. >
But how can you be so sure of Bach's cleverness, when I have denied it? Easily: you have put Bach's mind over your own, and, by extension, over mine. The immortality of the former is confirmed by your use of the present tense.

< (And one would assume he's doing it deliberately.) >
The proposition "Bach knew what he was doing" is credulous and can only be taken on faith. It implies that Bach imbued His music with an exacting comprehensive meaning, one which mortals can never fully discover.

Bradley Lehman wrote (May 4, 2003):
Trillo

<< Mvt. 2. (Alto Aria):
[Watts is the only singer in this group of recordings to use a repeated note pattern in ms. 41 & 42 instead of a vibrato which the wavy line above the dotted half-notes is supposed to represent.] >>
< Watts uses a trillo there? Really? Nifty; I haven't listened to that recording yet. (Trillo is a vocal ornament that is different from a trill.) >
At 1'47" in this movement (in Aryeh's sample on the web site), Watts indeed gives us somebody's idea of trillo. Frankly, it sounds as if somebody (Rilling?) looked it up in a book and decided to use it here, but without understanding it. Watts sings twelve equally-inflected notes per bar: 12 on the A, then 12 on the A# (since the ~~~~~~~ extends over both bars). "A-a-a-a-,a-a-a-a-,a-a-a-a-, a-a-a-a-,a-a-a-a-,a-a-a-a-, -ngst" fitted neatly into the three beats per bar.

For good trilli (that really sound like ornaments, with varied speed and dynamics, rather than like stiff boxy crap like this), listen to Emma Kirkby in Italian music. I remember on the old Nonesuch album where she sings Monteverdi solo madrigals on one side and Sigismondo d'India on the other side, there's a spot ione piece where she's singing about being frozen. And on that word (in Italian) the trillo is a wonderful effect, like a shiver of being frozen.

Hugo Saldias wrote (May 4, 2003):
[To Thomas Braatz] I think you wrote about Harnoncourt [3]:

That he rewrites the score:THE VERB SHOULD BE INTERPRETS THE SCORE. There is freedom to do this in a fermata.You can add rest time,take out rest time to increase the dramatic effect: let us take an example in the famous tocata and fugue in d minor BWV 565. If you are in a large church with a large organ with a lot of reverberation or you may be are in small church using a small organ: The way to play is NOT the same:

First case: Large organ:The first not that is an a with a "mordente"aga" should be short and the rest of the time given to the rest so the reverberation will still PLAY the note people can hear the note still sounding sometimes for 6 seconds (it is play with full organ).

Second case: since the dramatic effect can not be obtained, then, we make the note longer and the rest shorter... There are no rules in how long a note (or the rest after the note should be held) and to make it more evident the composer places the fermata to tell us:do as you wish here according to your taste and location...

Do not try to tell Harnoncour what to do. Do not try to make us think that he is doing something wrong...He is an interpreter that means that is his own personal taste...

Bradley Lehman wrote (May 4, 2003):
< For good trilli (that really sound like ornaments, with varied speed and dynamics, rather than like stiff boxy crap like this), listen to Emma Kirkby in Italian music. I remember on the old Nonesuch album where she sings Monteverdi solo madrigals on one side and Sigismondo d'India on the other side, there's a spot in one piece where she's singing about being frozen. And on that word (in Italian) the trillo is a wonderful effect, like a shiver of being frozen. >
This is the album I was referring to: "Olympia's Lament". It's been both
on Elektra/Asylum/Nonesuch LP (where I heard it) and CD, and on Hyperion.
http://shopping.yahoo.com/shop?d=product&id=1921518735
http://shopping.yahoo.com/shop?d=product&id=1921520662
http://www.amazon.com/exec/obidos/tg/detail/-/B000002ZH8

Aryeh Oron wrote (May 4, 2003):
BWV 116 - The Recordings – Short review of the Aria for Alto (Mvt. 2)

Last week I have been listening to 5 complete recordings of Cantata BWV 116 and 2 recordings of the aria for alto (Mvt. 2):

[1] Karl Richter (1977-1978)
[2] Helmuth Rilling (1980)
[3] Nikolaus Harnoncourt (1981)
[4] Pieter Jan Leusink (1999)
[5] Ton Koopman (2000)
[M-1] Jeannie Tourel (Contralto) with Robert Bloom (Oboe) & Bach Aria Group [Aria for alto only] (1954)
[M-2] Daniel Taylor (Counter-tenor) with Bruce Haynes (Oboe d’amore) [Aria for alto only] (1998)

The background for the review of the recordings of the Aria for alto (Mvt. 2) is based on both Robertson and Young.

This pessimistic and poignant aria is the cry of a soul terrified of the judgement and entirely lacking the confidence heard in the previous movement (opening chorus). The oboe d’amore and the continuo play a grief-motif, which Bach has patterned on the words of the text. The chromatic nature of all the parts betokens acute distress and Bach even put a trill for the voice, two bars long, at ‘Angst’ (anguish). But the soul does cry out to Jesus even though her mind is filled with the terrible vision of God’s anger.

[M-1] The earliest recording of this aria, done by Jeannie Tourel with the legendary Bach Aria Group, is also the slowest. This is not a recording for purists. The continuo is played by cello and piano, and Robert Bloom plays oboe rather than oboe d’amore. To contemporary ears, the singer might sound old-fashioned, with extra vibrato and approach and intonation alien to Bach. To enjoy from this recording we have to do a traversal in time to 50 years ago, imagining ourselves being exposed to this wonderful music for the first time ever. Despite the slow tempo, there is some tension along the aria and also nice interplay between the oboist and the singer. How can we clear our ears from all the recordings that we have heard? Was it the only recording of this aria I heard, I believe that I would have enjoyed it more. But with all the other recordings around, this is actually mission impossible.

[1] Richter has never been dogmatic in choosing the tempo for each movement. Sometimes he is slower than others are; other times he is faster. Here he decides for rather fast tempo and I do not understand why. Trudeliese Schmidt certainly understands what she is singing about and proves that even at relatively fast tempo a great deal of the potential of this aria can be brought out. Her singing can simply be described as heart-rending. One can hear some trembling in the voice production. But it can be understood as a way of conveying the utmost fear of God’s anger.

[2] What we can learn from Tourel’s rendition is that this aria can have a great effect on the listener when it is performed in a slow pace. Rilling indeed chooses such a tempo for his singer, Helen Watts, although not as slow as Bach Aria Group’s. This aria is more demanding emotionally than technically, and Watts, who was technically behind her prime when she recorded this cantata, uses the opportunity to give a moving rendition. She uses all the emotional resources she has to dig deep into the aria and bring out the despair, the distress and the fear. The trill at ‘Angst’ is clearly heard.

[3] The tempo chosen by Harnoncourt for the aria is too fast to do justice with the message of the text and the music, or to enable the singer, Paul Esswood, bringing out the deep emotions embedded in the movement. To be honest, I am not sure that was the tempo slower it would improve the performance, because Esswood’s approach in this case sounds to me rather superficial (BTW, he is much more convincing in the recitative!). The oboist also does not impress, either technically or emotionally.

[4] In direct comparisons of the aria for alto, I cannot skip Buwalda, although I would like doing it. His technical limitations are not so evident here, although Leusink follows Harnoncourt’s footsteps in adopting a similar break-neck tempo. For any singer, as we have heard in Esswood/Harnoncourt case, it would be a difficult task to convey some emotion in such a pace. Buwalda ‘proves’ that it can be done even worse. No feeling is conveyed here. We wonder: does he know what he is singing about? The compensation in this rendition is the delightful and moving playing of Eduard Wesley.

[5] I like the dark timbre of Annette Markert’s voice. Regarding tempo, Koopman’s rendition of the aria for alto is not a major improvement in comparison to Harnoncourt and Leusink. Where are all these HIP conductors running? Are they in a competition with each other to break a world record? Even Herreweghe was infected by this disease according to his latest CD of Bach Cantatas. I do not know if this is historically justified, and to be honest, I do not really care! Anyone with good ears can hear that this sombre aria only gains with slower tempo. Back to Koopman, Markert has done better, much better, in other cantatas. Not only does she not expose too much of her inner feelings, I hear some instability in her voice production. A bad day perhaps?

[M-2] One have to hear Daniel Taylor to realise how good a counter-tenor voice can be. He has a strong and stable voice along the whole range and impressive expressive abilities. Unlike the other HIP recordings, he chooses wisely a slow tempo, and gives a memorable account of the aria. He users the slow tempo to reveal many hidden corners, almost unheard in some of the other HIP renditions.

Conclusion

Preferred recordings of the Aria for Alto: Schmidt/Richter [1], Watts/Rilling [2], and Taylor/Haynes [M-2].

All these recording of the aria for alto can be heard through the page of Music Examples: Cantata BWV 116 – Music Examples

You can listen to these examples and come to your own conclusions. I would really like to hear other opinions regarding the recordings of this aria.

Hugo Saldias wrote (Maay 5, 2003):
[To Aryeh Oron] I will answer you and comply with your request about the aria you request us to talk about.I will be positive and no negative:
In order of preference:

[1] Richter: I like Schmidt , the balance of instruments is perfect:you hear them all.

[M-2] Taylor: All play legato the upper voice but in this version the descending thirds are separated:see the score: The first 7 notes go up, but then we have 3 notes in thirds coming down it is like falling and this 3 notes are phrased like this: there is a separation (non legato)between the 2nd and 3rd note: see second bar: I mean between the E and the C. This nice phrasing is also maintained on the singing:see word: SPRECHLICH...

[5] Koopman: Is very peotic and the oboe sings too:could be a duo of alto and soprano the ways he sees this. See the same situation with the 3 notes G E C descending: there is a small rallentando between the E and the C.. This gives a great feeling...

[M-1] Tourel: I can meditate more because it is slow. Also the cello bas is played with lots of feeling...

[4] Leusink: It is average.

[M-2] Taylor: It is average.

Neil Halliday wrote (May 5, 2003):
[To Aryeh Oron] Thanks for these 2 extra examples of BWV 116's first aria (Mvt. 2).

[M-1] The Bach Aria Group's continuo realisation (which employs a piano instead of a chamber organ) is something of a revelation to me. It comes very close to what I am seeking for the continuo accompaniment, as stated in my post in which I expressed
disappointment with all the available examples.

The piano shows it has the ability to add 'bite' and musical substance to the accompaniment, without overwhelming the other players.

I noted the insubstantiality of the chamber organ part in all the examples, realising of-course, that if the organ with its unvaried sustaining quality, could actually be heard in a recording, it would likely drown out other instruments as well as the voice, so conductors have little choice, other than to settle for this small, pitchless background colour of the organ. (Hence the soft stops and detached chords).

The piano, by contrast, can strongly state the harmony in a chord in an instant, and then allow the other players with the sustaining characteristics (voice, oboe, etc.) to carry on with their own timbres. The piano is also superior to a harpsichord in this regard - harpsichords as recorded in ensemble often become inaudible as as far as clarity of pitch is concerned, and audibilty of pitch is what is needed if this movement is to become something other than what sounds like a simple trio for alto, oboe, and cello.

The old recorded quality of the Bach Aria Group is unfortunately poor, the singer's vibrato is too stong, and the piano is not well balanced in this recording (it could have played a more substantial role, for the reasons I have outlined above) but we get enough for me, at least, to conclude that JSB would have jumped at this instrument for continuo use, if it had been available.

Neil Halliday wrote (Maay 5, 2003):
"The piano, by contrast, can strongly state the harmony in a chord in an instant"; but I don't mean it should be played staccato!. Rather, the instant after the piano chord has been struck, the other players will be clearly heard.

Notice that the Talyor/Haynes example [M-2], while expressive, is of the 'trio' (alto, oboe, cello) type I spoke of, (with the addition of a soft organ stop which adds little to the harmonic structure of the music), and the interest is achieved by the strongly articulated presentation of these three voices.

I prefer musical interest to come from the harmonic structure and timbre of the music first, and its articulation second, rather than the other way round, as in this example (Taylor/Haynes).

Johan van Veen wrote (May 5, 2003):
[To Bradley Lehman] Considering the way Mr Braatz comments performances by interpreters he dislikes, which shows a complete lack of respect, sincerity and integrity I wonder why you still bother to discuss with him.

There is nothing wrong with stern debate, but that only makes sense when all participants show a minimum of integrity. In Mr Braatz' case there is no sign of that.

Johan van Veen wrote (May 5, 2003):
Thomas Braatz wrote:
<< Mvt. 2. (Alto Aria):
There are three female altos singing this aria, all with more or less operatic-style voices: Schmidt (Richter)
[1]; Watts (Rilling) [2]; and Markert (Koopman) [5]. Schmidt (Richter) provides the most musically satisfying version of the heartrending fear and anguish that Bach wished to project in this aria. Rilling wisely chooses the slowest tempo for this aria. This allows the singer, Watts, to develop the full gamut of feeling. A slow tempo is commensurate with the emotions that are to be projected here. [Watts is the only singer in this group of recordings to use a repeated note pattern in ms. 41 & 42 instead of a vibrato which the wavy line above the dotted half-notes is supposed to represent.] >>
Bradley Lehman wrote:
< Watts uses a trillo there? Really? Nifty; I haven't listened to that recording yet. (Trillo is a vocal ornament that is different from a trill.) >
The 'trillo' is something one associates with Italian music of the early 17th century. The most famous example is in Monteverdi's Vespers, in the concerto 'Duo seraphim'. But is this ornament still en vogue in Germany in the 18th century? If a singer would use it in Bach I would be inclined to consider that an anachronism.

Philippe Bareille wrote (May 5, 2003):
Thomas Braatz wrote:
< Leave it to Harnoncourt [3] to come up with something different (which does not make sense anyhow): Harnoncourt completely rewrites the notation of Bach’s score; for example, where Bach has a fermata over a quarter note, Harnoncourt perversely adds an extra quarter-note rest, so that it almost appears that he has placed the fermata over the rest (which does not exist in the Urtext) and not the actual quarter note. This extra rest sounds strange and this is perhaps the only reason Harnoncourt has chosen to perform it in this fashion. There is less of the usual ‘chop, chop’ effect that Harnoncourt usually applies in a simple Bach chorale. The choir has amazingly poor diction and Harnoncourt simply allows the final consonants on important words such as “Gnad, schad, bist” to slip into oblivion. As an instrumentalist, Harnoncourt does not feel a need for words to be communicated clearly. He is more interested in “Effekthascherei” [“straining artificially for effect” “mainly interested in showiness”] and this is demonstrated in his performance of a simple Bach chorale where such a thing has no place. >
It is true that the is slightly muffled. Yet it is wrong to say that the choir has poor diction. Do you now suggest that those German boys cannot even pronounce their mother tongue properly! The training of this choir is geared to the text (read "Bach's Musical "Rhetoric" and its consequence for the performance of His oratorios" by Wolfgang Wemer (DHM, booklet of presentation of the Weihnachtsoratorium, Gerhard Schmidt-Gaden). To claim that "Harnoncourt does not feel a need for words to be communicated clearly" is merely an inversion of the truth.

Neil Halliday wrote (May 5, 2003):
Philippe Bareille wrote:
[3] < "To claim that "Harnoncourt does not feel a need for words to be communicated clearly" is merely an inversion of the truth".>
I'm sure Tom means: "It's as if Harnoncourt does not....etc.", from his hearing of Harnoncourt's performance.

Continue on Part 3

Cantata BWV 116: Details & Complete Recordings | Recordings of Individual Movements | Discussions: Part 1 | Part 2 | Part 3 | Part 4

Recordings & Discussions of Cantatas: Cantatas BWV 1-50 | Cantatas BWV 51-100 | Cantatas BWV 101-150 | Cantatas BWV 151-200 | Cantatas BWV 201-224 | Cantatas BWV Anh | Order of Discussion

Introduction | Cantatas | Other Vocal | Instrumental | Performers | General Topics | Articles | Books | Movies | New
Biographies | Texts & Translations | Scores | References | Commentaries | Music | Concerts | Festivals | Tour | Art & Memorabilia
Chorale Texts | Chorale Melodies | Lutheran Church Year | Readings | Poets & Composers | Arrangements & Transcriptions
Search Website | Search Works/Movements | Terms & Abbreviations | Copyright | How to contribute | Sitemap | Links



 

Back to the Top


Last update: ýDecember 29, 2012 ý00:44:21