Cantata BWV 9Es ist das Heil uns kommen her
Discussions - Part 1
Discussions in the Week of July 22, 2001 (1st round)
Aryeh Oron wrote (July 25, 2001):
This is the week of Cantata BWV 9 according to Thomas Braatz’ suggestion, the 9th one in his proposed list of cantatas for discussion (is there any number symbolism in this connection? we might ask Thomas Braatz). As a background to the three movements from this cantata, which I chose to review, I used this time the liner notes to Teldec’s recording, written by Alfred Dürr.
I am aware of 6 complete recordings of this cantata and last week I have been listening to 5 of them. There is also one recording of the concluding chorale only. See: Cantata BWV 9 – Recordings.
(1) Hans Grischkat (Early 1950’s)
I do not have this recording from a conductor, who was one of the renowned authorities in the first generation of Bach Cantatas recordings. In an old book from 1955 by Philip L. Miller, I found a short review of this recording. It says: “The ninth cantata is one of Bach’s finest. Spitta says it ‘gives us perfect satisfaction by its masterly completeness and fullness of form’. The recording shares the characteristics of that of BWV 6 (a recording by the same conductor, choir and orchestra, but with different soloists, about which the same author wrote: ‘In the recording, some concern is apparent over the matter of balance, for the chorus seems to be placed beyond the orchestras. The effect is generally good, though such an arrangement would account for a certain dullness in the chorus tone), though the solo singing is somewhat less impressive.”
(2) Gustav Leonhardt (1972)
(3) Karl Richter (1975-1977)
(4) Helmuth Rilling (1984)
 Sigiswald Kuijken (1999)
(6) Pieter Jan Leusink (2000)
Review of the recordings of 3 Movements
Mvt. 1: Chorus
“The form of the opening chorus is typical of Bach’s chorale cantatas. The melody is presented line by line in the treble, supported by the three imitative lower vocal parts all of which is embedded in thematically independent instrumental writing. The instrumental sound owes its distinctive charm to the use of a flute and oboe d’amore, which at times play in concertato style against the strings and at times, include the first violins in their concertino.”
Leonhardt recorded this cantata before either Rilling or Richter. That is probably the reason why he is slower in this cantata than both of them are. He takes his time (especially in the arias) and the whole performance is benefited from it. Furthermore, he allows himself to be more expressive than his usual self and to expose more feelings. The choir also sings with more unified tone and more clarity than the choirs in many of Leonhardt’s later recordings. Judging by this recording and the exemplary rendering of BWV 7, which is included on the same CD, I come to conclusion that Leonhardt’s is better in his earlier recordings of the cantatas than his later ones, which tend to be dry, dogmatic and fragmented.
Richter is bolder and livelier than Leonhardt is. The choir is somewhat bigger for the demands of this chorus, but somehow Richter succeeds to keep everything well tightened.
Rilling is even more cheerful than Richter is. The choir is relatively small, and therefore the balance between them and the instruments is kept. All the components are integrated to form a very enjoyable and convincing rendition.
For me the main benefit from listening to OVPP recording of a Bach cantata, is that it opens new possibilities, and sometimes exposes hidden corners and minor details in the work, which other (and more conventional recordings) have skipped. But with Kuijken’s rendition  of the opening chorus I am not pleased. The first problem with this rendition is that the introductory ritornello is played too fast. The music must breath and here it sounds short of breath. The second is that it sounds too calculated, and I believe that to keep the vividness some spontaneity must delivered. The third problem is the lack of balance, when sometimes the instruments cover the voices. And the last and the major problem is that this rendition does not reveal any new ground. It simply sounds like unfocused choir, which has been reduced to minimum. And I do not think that this is the singers’ fault.
I believe that some preparation had been put into Leusink’s recording, because the choir and the orchestra sound relatively precise and polished. This is a light and delicate rendition, yet also very vivid. The separation between the vocal lines is good, and the playing of flute and the oboe is also enjoyable.
Mvt. 3: Aria for Tenor
“The first aria is an example of Bach’s pictorial representation of the text: the download striving figures on the violin and syncopated rhythms symbolyze the reeling plunge into the abyss of sin.”
The match between the approaches of the conductor and the singer in Leonhardt’s rendition of the tenor aria is exemplary. The strings prepare the sombre atmosphere and Equiluz singing fits the mood. The tempo, which is slower than that of any other rendition, sounds so right and Equiluz singing is so touching, that when the aria is finished you want to hear it again right away. Some of the other renditions, which are played faster, sound almost boring in comparison. That is the only rendition in which I could almost visually see the depths of the despair.
Peter Schreier (with Richter) is so convincing in his expression of despair, against the strings which sound a little bit too cheerful and play a little bit too fast to my taste. The contradiction between them and the singer intensify in a strange way the mood he wants to convey.
The playing of the instruments in Rilling’s rendition sounds to me more suitable to the demands of the aria than Richter’s accompaniment does, but Kraus does not dig into the despair as deep as Schreier does.
Kuijken’s rendition  puts the focus on the playing of the violin (I assume that this is Mr. Kuijken himself). Knut Schoch’s voice, which we know from Leusink’s cycle, is strong and impressive here, however his interpretation is not very interesting. There is not enough feeling here from either of the participants.
I do not have any problem with Nico van der Meel’s voice, however I do have problem with his interpretation, which does not get below the surface. In certain places I get the impression that he has not made up his mind about what he wants to express. The accompanying instruments maintain the same high level they have shown in the first movement.
Mvt. 5: Aria (Duet) for Soprano & Alto
“The second aria, a duet for soprano and alto with flute, oboe d’amore and basso continuo, is quite different in style. Whereas the basso continuo limits itself to simple supporting harmony, the upper parts develop a series of instrumental canons, as even more complicated double canon arising with the addition of the vocal parts. The middle section is also treated as a canon, though here the instruments merely follow the vocal parts while occasionally decorating the melodic line. What is so amazing about this movement is the ease with which Bach solves all the problems created by this strict counterpoint without the listener’s even aware of the strictness of form.”
I like the singing of the anonymous boy soprano (in Leonhardt’s recording), the sensitivity with which Esswood sings along with him, and the delicacy with which the instruments are interwoven with the singing of the two vocal soloists. From time to time the boy soprano has some difficulties to hold the long lines, but the charm of this whole rendering compensates for any minor fault.
The duo of Edith Mathis and Julia Hamari sounds so dfrom the team of Leonhardt. We have here two mature women against boy soprano and counter-tenor. Hearing the two women immediately after Leonhardt’s team causes their approach to sound improper – almost operatic, too much vibrato, over-expressive. Although every detail can be clearly heard in this rendition too, it lacks delicacy.
With Rilling we have another feminine couple. Sonntag’s voice is in higher range than that of Mathis, and her vibrato can be heard, she has almost a boyish timbre. Although the sound more restrained and therefore more appropriate than Richter’s women do, I still have the feeling that this is not the right staffing for this duet.
With Kuijken  we return to the feminine couple. We have here two of the most impressive voices among the modern Bach female singers. I hear a kind of competition between them, who we will impress us more. I prefer more empathy between the two singers, like those that we have in some of the other renditions.
The charming instrumental introduction to Leusink’s rendition pulls you in with magical ropes. I am also not disappointed by the team of Strijk and Buwalda, because their voices blend nicely together and there is also beautiful match between their voices and the wind instruments.
Leonhardt’s rendition is the one to which I shall return when I want to hear this cantata again (especially Mvts. 1 & 3). Some of the other renditions have also interesting things to offer. I find the Kuijken’s recording  as the most disappointing.
If I had to choose only one movement from this cantata to take away with me, it would be the duet for soprano and alto. Alas, among the existing recordings of this movement none is fully satisfying. I hope that Koopman and/or Suzuki will do better.
And as always, I would like to hear other opinions, regarding the above mentioned performances, or other recordings.
Thomas Braatz wrote (July 26, 2001):
Because this cantata is not only musically important, but historically as well, I have decided to include in some greater detail the results of musicological scholarship that normally do not interest a listener. In this instance, however, this cantata has had a rather checkered history with many interesting twists and turns that may illustrate what the Bach experts have to contend with. It also gives us some insight into the reasoning involved in arriving at certain conclusions regarding the composer's final intentions. (As we know, these cantatas were 'works in progress,' but sometimes it helps us, when we can see how Bach changed things each time the cantata was performed.)
See: Cantata BWV 9 - Provenance
Insightful Commentary by the Experts/Scholars
[Forkel, Spitta, Schweitzer, Friedrich Smend (1950), Alfred Dürr (1971), Eric Chafe (1991), Chafe (2000)]
See: Cantata BWV 9 - Commentary
I have the following recordings: Leonhardt (1972) (2); Richter (1975-77) (3); Rilling (1984) (4); Leusink (2000) (6).
(6) Leusink: The thumpy bass (Leusink's bass is almost always to loud) causes the flute to be drowned out in the lower range, hence the balance is off. Marion Moonen playing the flute does not have enough of what is necessary to make this important part play the role that it should in this cantata. The alto voice is quite weak and there are passages in the lower range where some notes simply can not be heard. To make up for this deficiency there is a tenor who is overproducing and straining his voice beyond its normal capabilities. He sings with a vibrato (ughh) and presses too hard. The cantus firmus in the soprano voice is limp and definitely lacks a strength and conviction that is required here. The choir's pronunciation of German leaves very much to be desired. When the unaccented last syllable at the end of a line in the chorale ends in '-ten' ("behüten") or '-den' ("worden"), the vowel sound they sing is a schwa, and this is definitely wrong. German requires a bright, short 'e' instead. The same happens with the accented syllable '-tan' ("getan") where a tenseness in the vocal organs is required to create this long 'a' properly (definitely not 'tawn' which is what I hear in this recording.) It is interesting that the Dutch have the same difficulty with this sound that the English do. The classical music announcers on the radio station WFMT (also available on the internet) that I listen to take pride in reproducing the most exotic names with correct pronunciation. As part of their audition they need to announce correctly a Bach cantata with the original German title, but do you think that they can pronounce Bach's name correctly? Most of them can not. It sounds like 'Box SMP' or 'Box KdF' because they are unable to reproduce the correct vowel 'a' which needs to be tense and placed higher in the throat (I have not even mentioned the 'ch' which is a bane to English-speaking announcers.) Why is it that Suzuki outshines almost all other choral groups except the native German ones? Does it take greater effort on his part? Does he have a more musical ear? Is his respect for the text greater than that of others?
(2) Leonhardt: Here the bass is slightly lighter than Leusink's. The oboe d'amore is distracting because of its 'funny' sound, a sound that I had frequently heard before in this series on Teldec. This time I analyzed the sound and came to the conclusion that whoever is playing this instrument (5 players are listed) is attempting to create a vibrato even on the faster moving notes. Quite a feat in itself, but definitely so out of place here in that makes the pitch of the notes insecure. It is as though he is trying to cover up for failing to play the pitch of each note correctly. It undermines the solidity of the performance when I have to guess along with the performer, whether the correct note is being played at the correct pitch. When I watch a figure skater, I do not want to see close-ups revealing how wobbly the feet really are. This only serves to remind me how close an impending disaster might be. The transverse, wooden flute had a much better sound here than in Leusink's version. It was particularly brilliant in the upper range and yet it was not a modern metal flute. I had to look it up in the booklet and discovered that Frans Brüggen was playing a Stanesby jr. (c.1730) original flute and it sounded great, probably because an excellent player was playing an excellent instrument. The balance in the winds was very good too, particularly in the passages when both are playing unison. You can still hear both simultaneously. The cantus firmus was sung more clearly and with conviction, the way it should be, but, what was that? Although I could hear all the parts clearly and in balance with each other, there were slight imperfections in the manner in which they attacked the notes timewise. It was as though the timing was off at times. This was due to another phenomenon that I heard repeated in another cantata mvt. on the same CD: Leonhardt tries to increase the tempo each time the choir enters. This causes the listeners as well as the performers to become insecure about the degree of increase that the conductor wants. Throughout the mvt. there is this see-saw effect going on. In these sections there is the impression of being rushed, which is really Leonhardt's fault due to his inability to maintain a proper tempo. Too bad!
(2) Richter: This version, and Rilling's too, are 1/2 tone higher in pitch. The balance in the woodwinds is excellent with both instruments being heard clearly without being covered up by the other. Here are the vocal parts can be heard clearly, and they definitely sing with conviction. The bass is somewhat heavy at times, even though Richter resorts to a technique that many of Bach cantata conductors use when they have loud string bass players: they ask them to play pizzicato (not indicated in the score!,) an effect that reduces the volume they produce, but also creates a jazz-like, percussive sound when the bass moves swiftly. Sometimes I like this, at other times I don't. One thing is certain, Bach would have marked it in the score, had he wanted this effect. In a case such as this, Bach prefers to indicate the dynamics (even 'pianissimo' as he did in last week's cantata, BWV 88.)
(4) Rilling: Here the string bass does the same thing, but now a bassoon is added. This results in a really punchy bass line. The choir sings with conviction and all the voices can be heard clearly. The balance would be great except that there is a 'sore-thumb' soprano unable to control her vibrato while singing the cantus firmus. Her voice also has a raspiness that makes it stand out over all the others. If there is one place I really do not want to hear a vibrato, it is in the cantus firmus. Otherwise this is a good performance.
Mvt. 7 Chorale (out of sequence since the choirs appear only here and in Mvt. 1)
(2) Leonhardt: This rendition of the chorale is surprisingly good, and this is unusual for this series generally. For some reason, the members of the King's College Choir Cambridge, very infrequently used in this series, were unmoved by Leonhardt (with his mentor, Harnoncourt, hovering in the background)pleas for a 'more genuine' HIP and refused to shift away from their high standard of choral singing in order to accept Harnoncourt's misguided attempt at revolutionizing choral singing "to make it sound more like what Bach had in mind." All I can say is, Hurray! and congratulations to David Willcocks and his choir members for standing up for a higher standard of choral singing and not succumbing to Harnoncourt's obviously incorrect conception of HIP choral performances. Perhaps this was the reason they were no longer used in this series.
(3) Richter: Here we encounter Richter's penchant for overly long fermati, the purpose of which completely eludes me. Richter also pulls out some out-of-tune, shrill organ stops that are simply distracting. However from the standpoint of interpretation this is the best performance as it expresses conviction, and strength of belief solidly placed on a firm foundation. The singers really mean what they are singing about. Listen for Bach's musical word-painting in the harmonic progressions of the final measure, where the choir sings, "grauen" (to cause horror, to be horror-struck.)
(4) Rilling: Here we have a softer, kinder treatment, which is not appropriate for the text. There is less conviction as a result, and no terror on the word, "grauen." The balance is fairly good, but I am disturbed by the lack of clarity in the melody (soprano part.) This may be caused by a number of factors: a little of the raspy soprano's voice referred to in Mvt. 1, and the addition of the violin, oboe d'amore and modern flute all playing the same notes.
(6) Leusink: The fermati are, as usual, foreshortened, which means that the value of the note is prematurely cut off. As a result there is gap and a lack of connection between one line and the next. This is a little bit like a hiccup between the lines of the text. The chorale as a whole becomes rather breathless; one feels like gasping for air. Leusink has somehow achieved just the opposite of Richter's version by overcompensating for Richter's excesses. One is as bad as the other. In the upper voices the yodelers draw attention to themselves and a tenor succeeds in doing the same thing when reaching for high notes.
Mvt. 2, Mvt. 4, Mvt. 6 (Bass recitatives)
(6) Leusink: Ramselaar sings almost everything on one level (sotto voce) with little or no expression. There is not much variation in his voice. He sounds as though he is holding back on purpose, as a result he is unable to achieve much variation in the voice. [It might also be the case that Ramselaar simply does not possess a full voice.]
 Leonhardt: van Egmond has more expression than Ramselaar, but his trembling, fast vibrato creates unclear notes. His voice sounds like the bleating of a sheep.
(4) Rilling: Schöne has a full voice with good expression.
(3) Richter: Fischer-Dieskau has the ultimate expressive power to convey all the words with meaning. Richter uses a tremulant on the organ, not a good choice here, because one voice with vibrato is certainly enough. To hear the tremulant and a vibrato-producing voice together is simply too much of a good thing, and sometimes I seriously wonder if most vibratos that I hear are 'good'!
Mvt. 3 Tenor aria
(2) Leonhardt takes this 12/18 aria at a extremely slow tempo. One would think that the piece would fall apart at this tempo, but not with Equiluz! He makes his reading of the text reveal the sadness and weight of all the sins of mankind. This is a viable interpretation and it works. Just as Harnoncourt has his wife Alice play the solo violin parts, likewise Leonhardt has his Marie do the same when he is conducting. Both women also play a Jakobus Stainer violin (referred to above.) Unfortunately Marie is having some difficulty adjusting to the strong-first-beat-in-the-measure thing that Harnoncourt idolizes, so she only does it part of the time. (I can hear Leonhardt saying to her: "That's okay. You'll learn."). Because she is doing reasonably well, Leonhardt forgets all about asking Marie to follow the dynamic changes that Bach had indicated on his score. Anner Bylsma is on the cello.
(3) Richter: Wow! What a difference in the opening bars when you hear all those violins at once! The tempo is much faster too. Schreier is very expressive. Now the emphasis is on being thrust down violently and emphatically. There is terror as the world threatens to come apart. Listen to the dynamics which Richter observes very carefully (he would even drown out Schreier's voice if he allowed the violins to continue at the volume level that they had established in the opening bars. This is also a very good interpretation, one very different from Leonhardt's, but acceptable on an equal level.
(4) Rilling: Rilling uses a solo violin and does not observe Bach's dynamics. Although the tempo is just as fast as Richter's, there is hardly any terror or threat to be felt here, hence this interpretation is less convincing than Richter's. Everything is musically very clear, but there is less expression. The accompaniment just keeps 'playing away.' Yes, Kraus sings the notes correctly, and even does something special with the word, "dennoch," but this is not enough to put this performance on the level with the previous two.
(6) Leusink: The usual heavy bass and little or no attention being paid to Bach's dynamic marking are typical characteristics of Leusink's performance style. The violin is very tentative and soft. There is absolutely no sensing of the importance of the words. Van der Meel sings many notes correctly, but there is little expression in his voice, except on the word, "dennoch." (Did he pick this up from Kraus?)
Mvt. 5 Duet for Soprano and Alto
Listen for the first note that the soprano sings, and from this you will be able to judge the rest of the performance. The first word is, "Herr" ("Lord") which requires a full voice and must be sung with joy and conviction. This is somewhat like listening for initial "Ach" at the beginning of a recitative. Bach usually makes this a high note. When I hear Kraus sing this first note, I know that the entire experience of listening to the recitative will be difficult. The situation in this duet is similar. This mvt. is very problematical for all the versions that I listened to. Perhaps a future recording will succeed where these have failed.
(2) Leonhardt: From the very first note the unnamed boy soprano (this is one unnamed boy soprano whose real name I do not care to know,) accompanied by a wailing oboe d'amore, sings, it becomes apparent that this will be a disaster. Here is another one of a number of boy soprano prima donnas in this series who has been taught that a vibrato is important for covering up all the other imperfections, such as not being able to sing the correct notes at the correct pitch. I sometimes wonder how many young voices have been over-trained, and yet continue to sing this way because the parents and elders find this charming. The intonation is off and it is very difficult to listen to this voice for any length of time. I become very uncomfortable. It is somewhat like following a car, in which the driver is deeply engrossed in a conversation on the cell phone/handy, while all the while the car weaves and swerves back and forth in traffic. This boy's vibrato is like that because he can not control it properly. Esswood also has his usual intonation problems. He tends to sing flat much of the time. Expression? There is no way I would or could accept the words that they are singing. This duet brings no joy.
(3) Richter: This version is faster, the pitch is a semitone higher than Leonhardt's and Leusink's, and the modern, metallic flute is brighter in sound. Mathis' first note tells me everything that I want to know. Here her voice sounds like a worn-out operetta singer who is past her prime. Perhaps she once had a better voice at some earlier time, but she should not have been singing this at this time. Who knows? Perhaps she was still under contract with the recording company, and Richter was forced to use her here. It is not only her excessive vibrato that is objectionable, she also has intonation problems as well. Poor Hamari often gets lost in the lower range of her voice, that is, she can not be heard clearly because of everything else that is going on around her. What does she do to remedy the situation? She pushes out more volume to the detriment of her vocal production because then her vibrato becomes unpleasant as well. Here are two operatic divas not engaged in unifying and blending their voices to create a wonderful duet, but rather they are engaged in a duel, not a 'war of the bands' but rather a war of the vocalists to see who will win. Whatever message Bach was trying to convey here is lost in the feeling of contention between these two voices.
(4) Rilling: The performance here is slightly slower, but very similar to Richter's in many ways, only that a harpsichord is heard in the continuo. Perhaps Sonntag is slightly better than Mathis who managed to reach an all-time low in her performance. The difference between the two is negligible. Both Hamari and Schreckenbach (altos) are better than their counterparts in the soprano voice. Again the expression suffers because of the nature of these voices. They are not at all happy or joyous about what they are singing. Bach has imposed a task on these voices, and this task, due to their inability to fulfill the requirements thereof, becomes a burden rather than a joy.
(6) Leusink: The tempo here is similar to Rilling's. The sound of the flute is rather nondescript (see comments in Mvt. 1) and weak. Strijk's initial entry on the word, "Herr" ("Lord") is weak and tentative. This sets the tone for everything that follows. Although she is the first in this group to sing with little or no vibrato (what a relief!), her voice is quite limited and not suitable for projecting a true affirmation of joy that this text demands, a joy that follows all the tribulation that mankind has caused. O, for a voice like Strijk's and Holton's, but more developed and capable of projecting an understanding of the text and not simply singing all the notes correctly. Buwalda becomes unbearable as soon as he reaches the higher range of his voice which requires more control as he produces more volume. This voice is not a pleasure to listen to.
Very uneven results (two excellent renditions of the same aria, not a single acceptable duet, etc.) make it impossible to make any general recommendations here.
Kirk McElhearn wrote (July 26, 2001):
[To Thomas Braatz] BTW, great post!
Michael Grover wrote (July 26, 2001):
[To Thomas Braatz]You have outdone yourself again. Thanks for a wonderful and insightful commentary. I only wish I had your knowledge. (And the time to write these summary-reviews!!!)
BWV 9 - Kuijken/La Petite Bande 
Charles Francis wrote (July 29, 2001):
Sigiswald Kuijken, was born in 1944 and is a Professor at the Bruges and Brussels conservatories. Since 1969, he has pioneered the reestablishment of old techniques of violin playing without using a chin or shoulder rest and without holding the instrument with the chin at all. In 1972, he formed 'La Petite Bande' which consists of internationally renowned specialists in the field of early music. Since its foundation Kuijken and Gustav Leonhard have shared direction, with Kuijken as permanent leader.
Regarding 'One Voice Per Part' Kuijken writes:
" These [Rifkin's] conclusions were felt to be "sacrilegious". For those of us who had always adored the repertory and who had followed attentively and passionately stylistic developments in performing styles since the 1950’s, it was initially difficult to envisage and accept this set of conclusions. The use of a good choir had always seemed essential, and so many choral societies had attempted to develop a suitably detailed and sophisticated approach to these major works. Did all this now have to be rejected? This brings us to the very nub of the problem. This is not the place to re-examine all the arguments for and against Rifkin's thesis: readers interested in doing so should consult the relevant specialist literature, including the polemical pieces in "Early Music". All I want to say here is that slowly but surely I have felt myself moving towards a personal understanding that is fully in accord with Rifkin's conclusions (and also with those of his ever growing band of followers).
Having for many years performed Bach's vocal music - his Passions, motets, Magnificats, Masses, cantatas and so on - with the involvement of a traditional choir, I now feel that for the last few years I have been rediscovering these scores in the light of this new working hypothesis. Everything seems to come together as soon as you adopt the formula of one singer to a part, a singer, in short, who sings not only arias and recitatives but also the vocal ensembles or cori. One discovers empirically the extent to which the very style of writing in the great choruses is perfectly compatible with the use of a quartet of soloists: there is an enormous gain in transparency, and vocal feats that seem to be a tour de force for a choir become organic and plausible when performed by no more than a quartet of good singers.
Quite apart from any musicological interest that it may bring with it, such an approach opens up a new artistic dimension for me, affording new pleasure and a new understanding. On a personal, I feel that I am being more true to the original and that I am better equipped to face the demands of this repertory.
Having said this, I have no wish to imply that it would be wrong in future to use a choir instead of four soloists. Even if there are historical arguments that on one level are utterly compelling, we shall always be left with the question of taste (good or less good - everyone has the right to evince even the worst taste). Inreturn this offers us the freedom, after due reflection, to do whatever seems best. There can be no question of any moral obligation to copy something merely for the sake of it. Our first duty as artists is to communicate with our public, and such artistic communication can assume the most varied forms, all of which are valid as long as they are imbued by a creative spirit. "
I love this performance, which is very much in the style of Rifkin himself. The recording uses"authentic" instruments, but the orchestral playing is still able to reach the same excellent level as Rilling, with similar transparency and balance. The big difference, however, is when the cantus firmus enters: in the case of Rilling, Bach's lovely part writing disappears as it is obscured by the sound of the dominant choir - and this despite the use of louder "modernised" instruments with metal strings etc. How different with Sigiswald Kuijken, where he copies Rifkin's approach of treating singers as instruments in perfect balance with other instruments. Here the instrumental writing, including the lovely flute playing of Sigiswald's brother, Barthold, can be clearly heard throughout the cantus firmus. A most enjoyable and compelling approach.
Thomas Braatz wrote (July 29, 2001):
Thanks for sharing with us your impressions of this cantata and also the significant quote by Kuijken on OVPP from which I quote:
“Everything seems to come together as soon as you adopt the formula of one singer to a part, a singer, in short, who sings not only arias and recitatives but also the vocal ensembles or cori. One discovers empirically the extent to which the very style of writing in the great choruses is perfectly compatible with the use of a quartet of soloists: there is an enormous gain in transparency, and vocal feats that seem to be a tour de force for a choir become organic and plausible when performed by no more than a quartet of good singers.”
Kuijken makes a formidable task seem so simple. Perhaps this is the sign of a great artist and skilled musician. Not having heard this recording I can only refer to Rifkin and others whose recordings I do have. I will dare/risk a broad generalization that IMO a disparity exists between the roles of a single voice on a part in the choral sections and the solo work required by the individual soloists. For me, not "everything seems to come together" as easily as Kuijken might suggest. Although a quartet of singers might handle difficult choral sections more easily than a larger choir, there remains the two-way problem: If you have a truly excellent, full-voiced and vocally-trained soloist, the arias might possibly result in good performances, but try to ask this same artist to 'shift gears', listen to the other singers in the quartet, possibly reduce the volume in the voice for this purpose, stop using vibrato, etc. Most vocalists can not and will not comply. The results then begin to sound like the final quartet of vocalists (in almost all the recordings I have heard) in Beethoven's Ninth or Brahms' Liebeslieder Walzer sung by the likes of Edith Mathis, Brigitte Fassbänder, Peter Schreier, and Dietrich Fischer-Dieskau (most of these artists also known for their performances of Bach cantatas). In the latter Deutsche Grammophon recording from 1981, particularly in the quartet mvts of which there are many, it becomes quite apparent what difficulties these great artists have in adjusting (and listening) to each other. Never in my life would I want to hear these great artists singing the choral sections of the Bach cantatas. Many years ago I heard a Robert Shaw recording (LP era) of the Liebeslieder with a choir that was far superior to this big-name soloist-quartet.
The other side of this two-way problem looks like this: On one recent weekend I listened to a 2-CD set of Rifkin's cantatas quite a number of times. Admittedly, I was frequently occupied with something else at the same time. However, the general impression that I received was (this is very similar with the Cantus Cölln recording that I have) that the choral sections with OVPP are extremely effective and in some ways (clarity and balance) better than anything else recorded thus far. The same is true when Suzuki does this with a final chorale. I enjoy these recordings more than the traditional approach, at least I get the feeling that I am discovering something very new in the music. Now try to have these same singers do the arias, this is where they, for the most part, fail to measure up to my expectations of what a trained voice should sound like, even with little or no vibrato. These voices must be able to project the musical sounds and to express the message of the text, not simply sing the notes sotto voce because the conductor is afraid that his 'authentic' instruments might otherwise be covered up. Rifkin's (and all the other similar groups that I have heard so far) solo voices are unable to perform at this higher level of expectation when they are required to sing a Bach aria.
Sometimes I try to imagine a vocal quartet composed of such vocalists as an unnamed boy soprano (not one of the really good ones) Equiluz, Esswood, van Egmond singing the introductory mvt. to a Bach cantata, and I begin to shudder, because I have memories of what happens when they are involved in a duet or a trio in a Bach cantata. Individually they have brought us memorable recordings, but as a group?
Pablo Fagoaga wrote (July 29, 2001):
[To Thomas Braatz] OHH!!! Kuijken!!!
The first time I heard him was with the Petite Bande, in what may be the most delicate Harmonie Messe (Haydn) I've ever heard. Simply outstanding. Off topic, may be, but worth a few lines. An Agnus Dei simply out of this world. To anyone willing to have a wonderful time, it is on Deutsche Harmonia Mundi.
Johnan van Veen wrote (July 29, 2001):
This is exactly the problem with OVPP-recordings. Solo singing and ensemble singing require different qualities, which very few singers have. Many baroque operas close with an "ensemble" which is supposed to be sung by the soloists together. It seldom works.
I remember a performance of Rossini's Petite Messe Solennelle which was sung by 12 singers, just like Rossini intended. But the conductor, Simon Schouten, said in an interview that he had great problems in finding four singers who were able and willing to sing the solo parts and singing as part of his ensemble as well.
I think the OVPP approach will work best with an ensemble of singers who regularly work together like Cantus Cölln. I am more positive than you about the solo contributions of the members of CC which all have solo careers, although I think that some arias could have been done better.
But if this was the way Bach performed his cantatas it could make modern performances more difficult rather than easier. Bach of course had singers who did very little else than singing cantatas every Sunday, and they didn't sing a Bach cantata one week, and a baroque opera the next, as many baroque singers are doing today.
Maybe the best thing to do would be to have a small number of singers who do nothing else than sing Bach cantatas, week after week.
Continue on Part 2
Cantata BWV 9: Details & Complete Recordings | Recordings of Individual Movements
Discussions: Part 1 | Part 2 | Part 3 | Part 4