Cantata BWV 59Wer mich liebet, der wird mein Wort halten [I]
Discussions - Part 1
Discussions in the Week of May 19, 2002
Thomas Braatz wrote (May 25, 2002):
BWV 59 - Provenance:
See: Cantata BWV 59 - Provenance
Aryeh Oron wrote (May 25, 2002):
The subject of this week’s discussion (May 19, 2002), according to Francis Browne’s suggested list, is Solo Cantata for Soprano and Bass, for Whit Sunday (1st Day of Pentecost) BWV 59 ‘Wer mich liebet, der wird mein Wort halten’. This cantata includes chorus for the Chorale, which is Bach’s first version of this work. The second version BWV 74, which had been composed about 1735, was discussed in the BCML a year ago. Some Bach scholars have questioned the date of the first performance, which was thought to be 1716, delaying it until 1723, when it was presented at the Leipzig University Chapel for the Whitsunday service, and stating that Bach had previously composed it while he was in Cöthen. However, Whittaker says that it was performed at the Weimar castle chapel on May 31, 1716. The last scholar in the line, Christoph Wolff, wrote in the liner notes of the Koopman’s recording for Erato  that the cantata was first heard in one or other of the Leipzig’s two principal churches on May 28, 1724, presumably after the sermon. Before the sermon Bach performed his Weimar cantata BWV 172 (I assume that this is based on the NBA). Neumeister was the librettist for the richly orchestrated BWV 59. The work is peculiar because there is no final chorale and there are only four movements. The Gospel, John 14: 23, provides the text for the duet which opens the cantata.
In order to allow the members of the BCML being prepared for the discussion, I compiled a list of the recordings of this cantata, the details of which can be found in the following page of the Bach Cantatas Website: Cantata BWV 59 - Recordings
You should note that the list of soprano soloists in the eight complete recordings of this cantata seems like the best in the crop: Agnes Giebel (with Kurt Thomas), Rotraud Hansmann (with Jaap Schröder), Arleen Augér (with Rilling), Peter Jelosits (with Harnoncourt, perhaps the best boy soprano that the H&L cycle have ever had), Mária Zádori (with Pál Németh), Ruth Ziesak (with Koopman), Magdalena Kožená (with Gardiner), and Ruth Holton (with Leusink).
I hope to see many of you participating in the discussion.
The background below is taken from the following sources:
Alec Robertson: ‘The Church Cantatas of J.S. Bach’ (1972)
W. Murray Young: ‘The Cantatas of J.S. Bach – An Analytical Guide’ (1989)
The English translations are by Francis Browne, a member of the BCML.
See: Cantata BWV 59 - Commentary
Review of the Recordings
 Kurt Thomas (1959)
The couple of Giebel and Adam was made in heaven. The voices of both were in their prime when they did this recording, and intelligence has always characterised their interpretations. This is a lesson not only of voice production, but also of mutual listening. The main fault here is the heavy-handed conducting and the stiff accompaniment. The choir is unsatisfactory, to say the least.
 Schröder (1966-1967)
I hold this LP in my hand, but I cannot listen to it, because I have not given it to my friend in due time to burn it on CD. All the other cantatas on this charming LP (BWV 27, BWV 118, and BWV 158) have already been transferred by Teldec to CD form, but not this one. Based on Warner affair with Koopman, it seems unlikely that it will happen soon.
 Harnoncourt (1976)
Although I would not give up any of the female soprano interpretations of this cantata, I have to admit that Bach must have had somebody like Peter Jelosits in his mind when he wrote this cantata. Jelosits has beautiful voice along all the registers and he seems that he has no technical difficulties. What is more important is the maturity of his performance, rarely to be found with boy sopranos. Ruud van der Meer has an equal part in the success of the opening duet. In the aria for bass he has all the room for himself, to prove what a sensitive and authoritative singer he is. Harnoncourt takes the opening movement very slow, and manages to sound heavy-handed, as much as Kurt Thomas is. Once again the weakest part of this rendition are the fragmented chorales. What a waste of a good choir!
 Rilling (1976-1977)
The playing of the orchestra in the opening duet is somewhat too loud, but this does not prevent the two excellent singers, Auger and Tüller to make the outmost of the vocal parts. Both combine beauty of voice, intelligence and emotion to a performance which is hard to match. Augér is tearing your heart apart in the ensuing recitative and Tüller is doing the same in the aria for bass.
 Néméth (1988)
Last year I heard Mária Zádori singing Psalm 51 BWV 1083 in Israel. She was past her prime and the voice was sometimes unpleasant to hear. Here, 13 years younger, she is doing much better. She can stand proudly with the best of them. Her voice has a unique bright and her singing is flowing ahead smoothly and easily. She also conveys emotion, but some of the other soprano singers are more expressive than she is. The match between her and Polgár is good, both regarding the blending of voices and the way they follow each other in the canon, as two who are enjoying singing with each other. The accompaniment is humble and attentive and does not standing in the way of the singers. Polgár is convincing in the aria for bass with his pleasant low baritone voice. The choir is not bad either.
 Koopman (1997)
The soft playing of the trumpets in the opening duet goes well with charming voices and nice interpretation of Ziesak and Mertens. Their voices are splendidly matched. Ruth Ziesak shows in the ensuing recitative that she has also dramatic sense and her singing is simply heart-rending. The warm singing of choir continues the intimate atmosphere. In the aria for bass Mertens keeps the high standard we are accustomed to hear from him
 Gardiner (1999)
In the opening duet Gardiner takes a brisker tempo than most other conductors do. As could be expected in advance his trumpets glow, and both soloists have high challenge with such circumstances. They pass the test relatively well, but with such fine singers, the movement could benefit from a slower tempo and quieter trumpets. I believe that the beauty of Magdalena Kožená’s voice is already well-known with the members of the BCML. Nevertheless, I feel that she does not reach the depths of emotion that singers such as Augér and Ziesak do in this cantata. Similar things could be said about Harvey in the aria for bass. The choir’s singing is energetic and smooth, yet lacks some warmth.
 Leusink (2000)
Leusink's tempo in the opening duet is faster even than Gardiner’s! I have heard Holton and Ramselaar doing better in other duet movements than they do here. They are trying hard to cope with the accompaniment and with each other, not always with satisfactory results. The ensuing recitative is also done too fast to give Holton enough room for convincing interpretation. The choir here follows Harnoncourt’s path, and I do not like it at all. Furthermore, the singing is not clean and cohesive. Ramselaar’s aria for bass is the most satisfactory movement, although I miss the inner deptthat singers like Adam and Tüller gives to this aria. It seems that the whole rendition lacks from insufficient preparation.
Despite some reservations here and there, I like them all. But if I have to choose…
Preferred soprano singers: Augér  and Ziesak 
Preferred bass singers: Adam  & Tüller 
Preferred choir: Koopman 
As always, I would like to hear other opinions, regarding the above mentioned performances, or other recordings.
Thomas Braatz wrote (May 26, 2002):
Review of the Recordings
BWV 59 Wer mich liebet, der wird mein Wort halten.
This week I listened to the following 6 versions of BWV 59:
Thomas (1959) ; Harnoncourt (1976) ; Rilling (1976-1977) ; Koopman (1997) ; Gardiner (1999) ; Leusink (2000) 
Of these Thomas  and Rilling  are non-HIP and the remainder HIP.
In general the non-HIP recordings are characterized by a larger number of instrumentalists who are playing modern instruments (or some old instruments modified as most of them were to be louder in volume) that are generally louder than the HIP variety, by a larger number of voices singing in the choir, by fully trained voices that have substantial volume as needed, but also vibratos that can become excessive, by a more legato playing and singing style, while
the HIP recordings attempt to use modified versions or copies of instruments used during Bach’s lifetime. Their recordings are characterized by a generally lighter and faster non-legato style of playing and singing, a non-legato, heavily accented style that causes a general feeling of breathlessness and a lack of continuity. The number of singers in the choir is smaller, usually at least two to three singers per part, but not many more than that. The soloists tend to have very small voices that have received a little extra training, but usually lack the ability and range to project all the notes to an audience in a church. They tend to sound more like instruments, use less vibrato than their full-voice counterparts in the non-HIP group and generally lack the ability to add expression to the voice without forcing it and creating an uncomfortable sound to listen to. The HIP vocal soloists include countertenors and high falsettists that are only found in this type of period ensemble.
Comparison of times for Mvt. 1.:
 Thomas (3:38) (Tempo seems just right)
 Harnoncourt (4:15) (Everything threatens to fall apart at this slow tempo without legato)
 Rilling (3:53) (Slower than Thomas and also less joyful)
 Koopman (3:34) (almost the same as Thomas, but a very lite treatment)
 Gardiner (3:14) (a hurried and a light-weight treatment)
 Leusink (3:07) (“Let’s not prolong the agony if we have little to say”)
General Impression of All Recordings:
The Thomas recording seems to be the only one that shows a true understanding of the text by presenting the music in such a way that an exuberant and infectious joy can be felt by the listener. There is a distinct feeling that these musicians truly believe what they are singing, such is the strong affirmation that can be felt in their singing and playing. The other recordings, particularly, those in the HIP category, do not seem to understand the great joy of Pentecost as they either poke around looking for melodies or are more interested in hurrying through these mvts. in a light-weight fashion more conducive to background listening than for seriously engaging the listener. Pardon the oxymoron, but there is a serious, substantial joy that needs to be expressed here, a joy that does not allow the chorales to sound like those extracted from one of Bach’s Passions, nor does this type of joy rush in a dainty, skipping fashion through the duet and the chorales. This is the joy that Christ’s disciples and the rest of humanity experienced with the outpouring of the Holy Spirit, the great comforter. All of this at a point where the disciples had almost given up hope because Christ had been removed from them (the Ascension.) Now that which Christ had promised was actually happening. Listen to the sound of the trumpets and timpani in the Thomas recording as well as the fervor with which the Thomaner Chor sings the words of the chorale. None of the other recordings even come close to matching the feeling they impart to the listener.
Does anyone know why some groups sing ‘Alleluja’ as ‘Halleluja?’ Was this done this way in Bach’s day: you see the word ‘Alleluja’ but you sing it with an added aspiration?
Mvt. 1: Partially due to the well-paced tempo, this festive performance is very joyful, but dignified. Sometimes Adam is somewhat difficult to hear in the low range, despite the fact that he has a large and powerful voice. This is perhaps due to the strong emphasis on the bc and also due to the organ being too loud and distracting with some 4’ and 2’ stops having been drawn. Giebel has a pleasant fully-trained voice that has the power to project the music into a large space with expression. Mvt. 2: Giebel does some modernization of the German (vor=für, alles=allen) which was frequently done in the non-HIP era. Otherwise her pronunciation of German is impeccable. This is normally difficult for sopranos singing in a high range. Her expression of the words is excellent. Mvts. 3 & 5: There is a sustained joy that rests upon the strong conviction that the singers feel and express. Nothing is cut short. Mvt. 4: Adam definitely has an operatic voice, but here it is not yet as objectionable as it became later on. The solo violin has a very wiry sound. Is this due to the recording technique, the slightly excessive vibrato, or both?
Mvt. 1: The extremely slow tempo (perhaps in deference to the boy soprano) brings this mvt. close to the breaking point, threatening to have the entire mvt. fall apart at any moment. This problem is made worse by separating each quarter note from the next one when the strings are playing, thus causing many extra ‘breath’ or ‘separation’ marks in the melody lines. Ruud van der Meer is a would-be Huttenlocher, except for the fact that Huttenlocher has a more powerful voice than Meer. Meer has the same ingenuous quality of singing that, with a fast vibrato, tends to sound ‘funny’ when it needs to sing loud. Jelosits, on the other hand, has a clear, trumpet-like sound in the middle and upper registers, but at the point of this recording had little or nothing in the low range, a common failing of boy sopranos. Jelosits is one of the very few boy sopranos in the Harnoncourt/Leonhardt series whose performances can give us a reasonable approximation of what a boy soprano in Bach’s day might have sounded like. In ms. 43, Jelosits loses some of the low notes. There is some slight blaring of the trumpets who give the impression that they are glad to have made it through without any major mishaps. The slightly out-of-tune quality lends, according to Harnoncourt, credence to believing that this is an ‘authentic’ sound. Some notes in a passage are definitely louder than others. Mvt. 2: Harnoncourt now has the strings play successive half notes with slight pauses between each. This is Harnoncourt’s way of fracturing the accompaniment. He simply can not tolerate the notion that these notes might be connected. Jelosits pronunciation of German is excellent here. Mvt. 3 & 5 The muddychoral sound is complicated by the additional thrusting on individual notes. Instead of holding notes for their full value, Harnoncourt allows them to die out prematurely, as if the voice were also handicapped by the bowing techniques the Harnoncourt claims cause Bach’s string instruments not to be able to play legato. There is a feeling of uncertainty and apprehensiveness among the singers. Their pronunciation is fair. Mvt. 4: Alice Harnoncourt’s violin playing makes this instrument sound very tentative with many notes shortened in value. Her playing sounds thin and squeaky. Here Meer, with his half-voice sings mainly sotto voce with a fast vibrato that makes it impossible to tell the difference between a trilled note and an untrilled one.
Mvt. 1: Rilling’s tempo here is more measured than Thomas, and perhaps, as a result, the mvt. begins to lack some of the exciting joy felt in the Thomas recording. Augér, with her fully-trained voice definitely holds her own against the larger instrumental ensemble, but poor Tüller is mainly too weak. This weakness may in part be due to his fast vibrato that makes his voice sound unclear or even raspy without any round tone to compensate for this. He may be forcing his voice way beyond its natural capabilities. Rilling’s bc is too heavy and the trumpets are less interesting compared to the previous recording. Mvt. 2: Augér has very good diction for a non-native speaker of German. Just listen to her enunciation of the consonants. Compared to Giebel, Augér’s voice is less round and she occasionally gives the impression that she is forcing the voice to overcome some difficult notes (difficult for her to sing). At least the extra effort becomes apparent. Mvt. 3: There is less excitement and joy in this rendition compared to Thomas. There is a soprano voice (or two) which is almost always distracting because it is unsteady and detracts from the firm line that the chorale melody demands. Mvt. 4: Tüller’s disturbing vibrato subtracts from his overall volume in the middle and low ranges. Because of this excessive vibrato, there is no way to tell the difference between a normal quarter note and one with a trill marked on it (ms. 31). They both sound the same to me. This is bad!
Mvt. 1: With Ziesak and Mertens we have two half-voices struggling to be heard even with the light orchestral period instrument sound and the very soft trumpets. Perhaps this is the reason why Koopman holds them back so much. Mertens, for instance, in his middle range coloraturas (ms. 50) drops the ending syllable of a word and also the word, “und.” They can not be heard at all. Mvt. 2: Ziesak has a few pronunciation problems: the beginning of the word, “Vater,” (ms. 4) sounds more like English ‘faw’ and in ms. 11 the ‘s’ at the end of “alles” is dropped. Mvts. 3 & 5: This joyful chorale sounds more like a soft, intimate prayer where one is afraid of being overheard. This subdued approach is completely out of place in a chorale that sings of “brünstig Lieb” which has obviously not yet been “entzünd” [“ignited”] by this conductor. Someone needs to light a fire under Koopman to put him in the proper state of mind and feeling, so that this chorale receives a worthy treatment. For the second chorale at the end of the cantata, Koopman chose to present the chorale that Bach omitted. Mvt. 4: There is excellent balance that is achieved between Mertens and the rest of the ensemble. With his gentle and pleasant-sounding half voice, Mertens is unable to give much expression to the words because he sings mainly sotto voce. The bc is extremely light. If you listen carefully, you might even hear the lute! That gives you an indication of what volume level we are operating at here. In a large church with a full congregation present, there would be very little of this that you could really hear. The solo violinist ‘spices thing up a bit’ by attempting some interesting variation/embellishments when the ritornello reappears exactly as it was in the introduction.
Mvt. 1: With one of the fastest tempi in this set of recordings, Gardiner is no longer able to extract an element of dignity from this music. Now things have to move along more quickly, which implies automatically less of a firm foundation. Kožená would qualify as a full voice, but because of the tempo, she is unable to develop certain vocal sounds more fully. There simply is not time and as a result, she lapses into some sotto voce singing. What a shame, when she could really do much more here if she had been allowed the opportunity. Harvey, her partner in this duet, is unable to hold up his end because he only has a half voice to begin with. Now add speed to the notes and he has even less of a chance to produce significant sounds and expression. He is the weakest link in this mvt. Mvt. 2: Kožená has serious problems with the German language. Now that she is on her own in this recitative, many of her faults become much more apparent. She lisps! She is unable to pronounce the ‘s’ sound properly, a sound that occurs very frequently here. Listen carefully to the phrases, „was sind das”; „worzu uns”; „Jesus der uns“; „so würdig schätzt“; „daß er verheißt“ etc. In this recitative she drops off into nothingness for supposed expression of the words. There are many “Mogeltöne” [“cheating notes”] where she produces almost nothing that carries sound and meaning. She also engages in some of that final thrusting upon the release of a note, a characteristic left over from her native language. In ms. 18 “Ach, daß doch” she sings “noch” in place of “doch.” Without the text or score in front of me, there are stretches in this recitative where I can not understand what she is singing. Mvts. 3 & 5: Gardiner has a nice legato sound, but the chorale is much too subdued for such a festive occasion as Pentecost. In mvt. 5 Gardiner has the choir sing another verse of the same chorale. Mvt. 4: Harvey’s singing is mainly sotto voce and with his funny, fast vibrato, there is simply no way to tell the difference between a note with a trill over it and another note which does not have the trill.
Mvt. 1: At this fast tempo, Leusink can only come up with a lite entertainment version of this mvt. There is separation between all the eighth notes. The double bass in the bc is much too loud for the rest of the ensemble, in particularly for the two half voices singing the duet. Mvt. 2: Holton’s voice is able to produce a few trumpet-like tones (on “O”) that resemble the quality that a boy soprano can achieve. Sometimes her voice is almost not present at all because the sound production capability simply is not there for her. She mispronounced “allerhöchste” by substituting an ‘sh’ for ‘s.’ Mvt. 3: The usual warbling and forced voices stick out of the choir sound at various times. The bass voice does a disappearing act in ms. 19, 20 in the low range, and if the bc were not duplicating these notes, the entire bass foundation for this chorale would be missing. Mvt. 4: Ramselaar has even less voice to produce sound with than Mertens. It sounds as if he is singing sotto voce all the way through the aria.
Mvt. 1 (Duet for Soprano & Bass)
Thomas (Giebel/Adam) ; Rilling (Augér/Tüller) ; Gardiner (Kožená, Harvey) ; Harnoncourt (Jelosits, Meer) ; Koopman (Ziesak, Mertens) ; Leusink (Holton, Ramselaar) 
Mvt. 2 (Recitative for Soprano)
Thomas (Giebel) ; Harnoncourt (Jelosits) ; Rilling (Augér) ; Koopman (Ziesak) ; Gardiner (Kožená) ; Leusink (Holton) 
Mvts. 3 (and 5, if attempt) Chorale
Thomas (Thomaner Chor) ; Rilling (Gächinger Kantorei Stuttgart) ; Gardiner (Monteverdi Choir) ; Koopman (Amsterdam Baroque Choir) ; Leusink (Holland Boys Choir) ; Harnoncourt (Tölzer Knabenchor) 
Mvt. 4 (Bass Aria)
Thomas (Adam) ; Koopman (Mertens) ; Gardiner (Harvey) ; Rilling (Tüller) ; Leusink (Ramselaar) ; Harnoncourt (van der Meer) 
Dick Wursten wrote (May 28, 2002):
[To Aryeh Oron]  In sheer enthusiasm for Magdalena's beautiful voice Aryeh’s computer started to hebraize. At least on my screen the 'beth' appeared in her name.... "Kožená’s"... I don't know whether other computers with different language character sets also tried to nationalize her.. After Pentecost everything in every tongue is possible.
Aryeh Oron wrote (May 30, 2002):
[To Dick Wursten] Altough I naturally read Hebrew, the Hebrewization of Magdalena's last name was done undelibararely. I sent my review late on Saturday night, a minute before the week's end. To save time I sent the mesasage directly from Word instead of using the long procedure I usually do. One would expect a compatabilty between two Microsoft programs: Word and Outlook Express. In future reviews I shall try to get back to the usual procedure. Nevethless, you can read the review on the Bach Cantatas Website in the page:
And regarding La Kozena, I recommend to you looking at the pages:
Continue on Part 2
Cantata BWV 59: Details & Complete Recordings | Recordings of Individual Movements | Discussions: Part 1 | Part 2 | Part 3