Continue from Part 4

Harnoncourt/Leonhardt cycle, some thoughts

Bradley Lehman wrote (March 13, 2003):
< - the employment of boys instead of women
- the employment of singers whom you derogatorily dismiss as "half-voices"
- a tendency to exaggerate contrasts beyond what the acoustics of the hall used for his recordings can bear
- the employment of some players who can't hit all the notes as accurately as you might expect
- the treatment of secco recitative using the commonplace 18^th-century notational conventions, which unfortunately don't agree with what you see on the page reading it with 20^th century expectations
- the treatment of fermatas in chorales
- intonation other than Equal Temperament
- rhythmic treatments that are not as strictly metrical as the notation appears on the page
- a willingness to hierarchize the notes and phrases into strong and weak ("good" and "bad", in 18th century parlance)
- a declamatory style of singing, focused on the sung words as a type of heightened speech, rather than a later ideal of sheerly beautiful tone above all
- a declamatory style of playing, focused on the rhetorical construction of the phrases and musical "paragraphs" rather than on making every note individually beautiful >

In the checklist above about salient features of the Harnoncourt/Leonhardt series of cantatas, I should say that I agree with most (but not all) of those points of the HIP techniques they chose.

Over the years I've collected about 35 of the cantatas from this H/L cycle, and then pretty much stopped. Additionally, I have about half a dozen of their earlier recordings of cantatas on single Telefunken LPs, plus the landmark (and wonderful, IMO) Vanguard LP and CD of Deller singing BWV 54 and BWV 170 with both H & L playing, from 1954. And another 30 or 40 cantatas performed by other people, on LPs from the 1950s-1970s.

Why did I stop collecting the H/L cycle? Simply, I felt I could do better with other more recent recordings that give me more musical satisfaction. There's so much other music I'm interested in, and with a limited budget: generally I can't afford to have more than one or two recordings of each cantata, except for a few favorites. As opportunities come up to buy more of them, I'm working at filling in the holes where I don't have any recordings yet. (With some of Bach's instrumental music, though, I occasionally have as many as 30+ recordings of the same work...a higher priority for me!)

And it's going well, assembling various cantata recordings from Herreweghe, Suzuki, Parrott, Leusink, Kuijken, Gardiner, et al....in almost every case I enjoy what I hear more than I enjoy H/L. The sounds are more immediately comfortable, the balances better, the details more gracefully in place. It's still important to me to listen to H/L, but when I want to hear something just for the pleasure of it, I usually put on somebody else's.

As I said above, I agree with most of the H/L style. The only three things from the checklist below that I don't like are points I disagree with for personal musical reasons, not ideological reasons.
(1) I like to hear the more recent (and usually more accurate/expressive/polished) instrumentalists.
(2) I'd rather hear female sopranos than boys.
(3) Harnoncourt (but not so much Leonhardt) sometimes exaggerates the phrasing or accentuation more than I think is necessary in the acoustic spaces he's in...it comes across as more didactic than naturally expressive.

Those three are the main reasons why I've taken my money elsewhere. With everything else on the list below, I think they were "right on!"

In principle, I think they had very good ideas and a bold approach, still worth hearing. Music doesn't have to be comfortable all the time; a good challenge is sometimes welcome. H/L give us some good challenges.

Johan van Veen wrote (March 13, 2003):
Bradley Lehman wrote: >>In the checklist below about salient features of the Harnoncourt/Leonhardt series of cantatas, I should say that I agree with most (but not all) of those points of the HIP techniques they chose.<<
It is a very useful checklist. The principles laid down in that list can - and in my opinion should - be used to assess recordings of other music from the 17th and 18th centuries. Some of the features are even applicable to some 19th century music

>>And it's going well, assembling various cantata recordings from Herreweghe, Suzuki, Parrott, Leusink, Kuijken, Gardiner, et al....in almost every case I enjoy what I hear more than I enjoy H/L. The sounds are more immediately comfortable, the balances better, the details more gracefully in place. It's still important to me to listen to H/L, but when I want to hear something just for the pleasure of it, I usually put on somebody else's.<<
I have the whole Teldec cycle, and I return to it constantly. That has to do with the fact that I definitely and always - as a matter of principle - prefer the use of boys' voices, but also with the fact that I am annoyed by the fact that many of the features of H&L as listed in the checklist are more or less ignored by other recordings. I am not saying that the others are all bad - most of the earlier recordings by Herreweghe are great, and some of Sigiswald Kuijken as well. I find Leusink unacceptable in many ways, most 'English' performances very unidiomatic, with almost complete ignorance of the rhetorical character of the cantatas and a lack of 'declamation', and Suzuki ist just bland.

What attracts me most in H&L in comparison to more recent recordings is the depth of their interpretation, the fact that the text and spiritual content always come first. Most later recordings - with the exception of those I just mentioned - are relatively superficial.

>>As I said above, I agree with most of the H/L style. The only three things from the checklist below that I don't like are points I disagree with for personal musical reasons, not ideological reasons.
(1) I like to hear the more recent (and usually more accurate/expressive/polished) instrumentalists. <<
I can understand that people want to hear more technically assured instrumental contributions. But it is unfair to compare the players in the H&L recordings with their colleagues of the present. I think what they did was the best which could be achieved at that time.

Since then there has been great progress in the technical command of period instruments, but I feel that the interpretation hasn't necessarily improved as well. I have the impression that sometimes instrumentalists of today don't really know - let alone understand - what the texts are about. The 'declamatory style of singing' often lacks in modern recordings, so does the 'declamatory style of playing'.

>> (3) Harnoncourt (but not so much Leonhardt) sometimes exaggerates the phrasing or accentuation more than I think is necessary in the acoustic spaces he's in...it comes across as more didactic than naturally expressive. <<
One could argue that in many ways Bach's cantatas are 'didactic'. And if the accentuation of some elements in the music helps to get the message across, I don't see the problem.

If I listen to Bach's recitatives I am often reminded of an old-fashioned preacher. In many ways Bach's cantatas can be considered 'sermons on music'. Sermons of the 17th and 18th centuries were often pretty drastic and radical in their language and metaphors (and in many ways Martin Luther set the example). From that point of view it is hard to imagine a performance which is too drastic or too accentuated.

>> Those three are the main reasons why I've taken my money elsewhere. With everything else on the list below, I think they were "right on!"
In principle, I think they had very good ideas and a bold approach, still worth hearing. Music doesn't have to be comfortable all the time <<

Right. Too often modern audiences want music to sound 'beau', even if the text doesn't give any reason for that. "Erbarme dich" from the St Matthew Passion (BWV 244) isn't meant to sound nice, smooth or beautiful - it is a cry for help from someone who's desperate. What's so 'nice' or 'beautiful' about that?

Bradley Lehman wrote (March 14, 2003):
< Johan van Veen wrote: I can understand that people want to hear more technically assured instrumental contributions. But it is unfair to compare the players in the H&L recordings with their colleagues of the present. I think what they did was the best which could be achieved at that time. Since then there has been great progress in the technical command of period instruments, but I feel that the interpretation hasn't necessarily improved as well. I have the impression that sometimes instrumentalists of today don't really know - let alone understand - what the texts are about. The 'declamatory style of singing' often lacks in modern recordings, so does the 'declamatory style of playing'. >
Agreed! A notable exception is Herreweghe, who (perhaps not coincidentally) worked with H & L toward the end of their series. His performances regularly suggest to me that he has his players very well aware of the text, and indeed playing like singers. And, remarkably, that's so whether he is conducting Bach or Beethoven or Faure.

>> In principle, I think they had very good ideas and a bold approach, still worth hearing. Music doesn't have to be comfortable all the time <<
< Right. Too often modern audiences want music to sound 'beautiful', even if the text doesn't give any reason for that. "Erbarme dich" from the St Matthew Passion isn't meant to sound nice, smooth or beautiful - it is a cry for help from someone who's desperate. What's so 'nice' or 'beautiful' about that? >
Here are some similar comments from an article by Robert Marshall, written as a review of the first two volumes of the H/L series when they were new. [Most of this is from a posting of mine to BachRecordings a few weeks ago.]

< Another of my favorite scholars, Robert L Marshall, has this to say in his review/essay of the entire Harnoncourt/Leonhardt series of Bach cantatas. It is in his book The Music of Johann Sebastian Bach: The Sources, the Style, the Significance, published in 1989:
"The artists participating in the Telefunken series demonstrate to a remarkable degree not only their knowledge of the conventions associated with Bach performance but in addition their empathetic grasp of the spirit informing these conventions. In observing, for example, the eighteenth-century practice of sustaining continuo notes in recitatives for only a fraction of their written value, the notes are not shortened uniformly but are held for a longer or shorter time as seems appropriate for the affective content of the text. [And he continues in his endnote:] In other respects as well--the vivid, 'naturalistic' declamation, the reedy, organ-like timbre of the sustained strings in accompagnato settings--the recitatives provide perhaps the most consistently successful feature in the entire cantata series."

(from page 236 and its endnote)
He also has an example where he praises the performers: "This instinct for capturing the essence of a convention is on one occasion so acute that the performers permit themselves to disregard what Bach allegedly wrote down himself (assuming they were aware of it) in order to play a variant that he would surely have approved...." [And then he explicates that example: the bass aria "Empfind' ich Hoellenangst und Pein" from cantata BWV 3, third movement, as recorded by Harnoncourt.]

That is, Marshall lauds Harnoncourt's approach of capturing the spirit of the work, and the unwritten but real conventions, so well that they are able to present performances of artistic merit, going remarkably far beyond any restrictive modern methods of reading notation. >

- He wrote this article for The Musical Quarterly, 1973: reviewing the first two volumes (not the whole series, which didn't exist yet) of the Harnoncourt/Leonhardt cycle of Bach cantatas. When he included it in this 1989 book, he updated only a few points in the text, and his additions are clearly marked with his own [brackets] (while the brackets shown are my own...these portions of his text that I've quoted here are all original to the 1973 version). The main set of changes for this book was his updating of all his footnotes, and inserting a few new footnotes between the existing ones. For example, his new note 20 is a remark that Dreyfus' new book about the Bach vocal works has filled an important void, being "precisely the study called for here."

- That is, Marshall's comments here about Harnoncourt and Leonhardt doing the right thing with recitatives were written 14 years before the publication of Dreyfus' book.

After some specific complaints about a few other aspects of the performances, Marshall's review concludes:

"It would be unfair to conclude on such a disapproving note; for this essay is intended as a strongly favorable review of the Telefunken project. The astounding accomplishments of both the Concentur Music and the Leonhardt Consort in the light of formidable difficulties have been so highly praised so often, however, that it would have been rather patronizing and pointless to issue further mere compliments here. Moreover, I feel compelled rather to append a somewhat embarrassing confession: for pure enjoyment I would prefer to listen to the Bach performances of, say, Karl Richter or Helmuth Rilling, with their straightforward phrasing and articulation, their impeccable intonation and familiar timbres, and their admittedly loose approximation to authenticity. But I must add at once that my delight in the Telefunken performances has grown with each hearing. Like so many new and unfamiliar experiences, they obviously demand a lot of getting used to. There can be little doubt that beauty, its recognition, and its delectation are direct functions of familiarity and that we must be on our guard not to judge the aesthetic validity--much less the historical authenticity--of any rendition of an artwork by our first (or even ultimate) response to it. The notion that 'Beauty is truth, truth beauty' may be a poet's profound wisdom, but it is a seductive, if splendid, fallacy for the historian."

Alex Riedlmayer wrote (March 14, 2003):
Why should I bother with this series? Schaeftlein is playing out of tune half the time. The Vienna chorus has serious problems. Equiluz's contribution is unmoving; at least he's better than Esswood and the anonymous 'trebles'. As for being the first complete recorded cycle, that's not even true (Rilling finished years earlier). I'd rather forget about '18th-century' conventions than listen to the unmusical results produced by their uncritical acceptance.

Peter Bright wrote (March 14, 2003):
< Alex Riedlmayer wrote: anonymous 'trebles'. As for being the first complete recorded cycle, that's not even true (Rilling finished years earlier). I'd rather >
I thought Leonhardt and Harnoncourt began their official 'cantatas project' in 1971 and finished it in 1989 (although they worked together on individual cantatas as early as 1954 - see the bach-cantatas web site). This covers a similar period as the Rilling series, although Rilling did manage to polish off the set in the mid 1980s. L&H was the first complete cycle that was directed with close attention to performance practices of Bach's time (with much subjectivity in place, of course). While some aspects have now been discounted (by some, not others) as not indicative of the period, it is difficult to deny H & L's aims: to produce the first historically informed recorded performances of all the cantatas using period instrumentation.

Rilling's approach is more problematic to me - some of his principles seem guided by by his knowledge of HIP practices, but he seems to pick and choose, and generally favours modern instrumentation. The result is often unsatisfactory from both modern and HIP perspectives. There is also a preference for smooth, somehow anti-dramatic direction which takes the ebb and flow (and thus, the life) away from themusic - particularly in the choruses.

I much prefer more recent offerings, particularly Suzuki, but if I were to choose one set on modern instruments (although incomplete) I would not hesitate in recommending Richter's survey of the cantatas (I am rather ignorant about Werner, but have also always been impressed with what I have heard, too).

Alex Riedlmayer wrote (March 14, 2003):
>> anonymous 'trebles'. As for being the first complete recorded cycle, that's not even true (Rilling finished years earlier). I'd rather <<
Peter Bright wrote: < I thought Leonhardt and Harnoncourt began their official 'cantatas project' in 1971 and finished it in 1989 (although they worked together on individual cantatas as early as 1954 - see the bach-cantatas web site). This covers a similar period as the Rilling series, although Rilling did manage to polish off the set in the mid 1980s. >
All this is true, but doesn't excuse the claim in several of Kirk McElhearn's reviews that Rilling was only the second.

< L&H was the first complete cycle that was directed with close attention to performance practices of Bach's time (with much subjectivity in place, of course). >
Well, they had to be first in something. :-)

< While some aspects have now been discounted (by some, not others) as not indicative of the period, it is difficult to deny H & L's aims: to produce the first historically informed recorded performances of all the cantatas using period instrumentation. >
Those may indeed have been their goals. That they are meaningful goals it is easier to contest.

< Rilling's approach is more problematic to me - some of his principles seem guided by by his knowledge of HIP practices, but he seems to pick and choose, and generally favours modern instrumentation. >
This seems to be consistent with what is said in the note that accompanies the Bachakademie reissues.

< The result is often unsatisfactory from both modern and HIP perspectives. >
These appear to be fictions which I would not share.

< There is also a preference for smooth, somehow anti-dramatic direction which takes the ebb and flow (and thus, the life) away from the music - particularly in the choruses. >
I have noticed this, particularly in the strings and chorus. Rilling's later recordings of secular cantatas, passions, etc. are considerably more lively.

< I much prefer more recent offerings, particularly Suzuki, >
Suzuki's series is overrated; I consider the one release I've heard of no more than fair quality. On the other hand, I can easily perceive why some people with different tastes prefer it.

Harnoncourt family tree

Bradley Lehman wrote (April 9, 2003):
A trivia question (no, I don't know the answer)...are any of Harnoncourt's children or grandchildren musicians?
http://www.ifrance.com/tempo/mefontaine2.htm
http://www.ping.be/~jos81/link/philippides/ph322a.htm

I was wondering that last night while listening to some of the early Alfred Deller records. "Elizabethan and Jacobean Music" (Vanguard 539) has both Nikolaus and Alice (nee Hoffelner, still using her maiden name on this release), plus Eduard Melkus and Gustav Leonhardt, all playing viols. (Yes, Leonhardt playing bass viol
rather than keyboards!)

And, the nonpareil recording of Bach's cantatas BWV 170 and BWV 54 (Vanguard 550), from 1954. The CD reissue says it was "the very first recording on historic instruments made by Gustav Leonhardt and Nikolaus Harnoncourt...." Both Nikolaus and Alice on there, and both Gustav and Marie Leonhardt, along with others...a nice little family project there. And the bonus on there, filling up the end of side 2: Deller and this same gang in the "Agnus Dei" of the B Minor Mass (BWV 232). Wow.

Anyway, the main thing I was trying to look there was whether Alice Hoffelner = Alice Harnoncourt.... Anybody know what the music-making in their house might sound like, if any of the children or grandchildren play anything?

Riccardo Nughes wrote (April 9, 2003):
> A trivia question (no, I don't know the answer)...are any of > Harnoncourt's children or grandchildren musicians? <
Elisabeth is Elisabeth Von Magnus, contralto. She studied recorder before became a singer. She played, obviously, with the Concentus Musicus. In several interviews Harnoncourt said that when he was a child there were family concerts, his father and (most) of his six brothers-sisters playing some instruments. One of his brother (Philipp IIRC) is a Catholic priest and he wrote the liner notes for F.Schmidt "Das buch mit sieben siegeln".

Peter Bright wrote (April 9, 2003):
[To Bradley Lehman] Isn't Elisabeth von Magnus (the mezzo soprano) Harnoncourt's daughter?

Harnoncourt & Wiener Symphoniker

Riccardo Nighes wrote (May 4, 2003):
I have the old Vienna State Opera Orchestra recording of that (Vanguard "Bach Guild" LP), conducted by Mogens Wöldike (1960), with Heiller playing the organ. Does anybody here happen to know: was Nikolaus Harnoncourt playing cello in the orchestra for that performance?

Ditto for Wöldike's SMP (1959), and Scherchen's SMP (1953), and Scherchen's B Minor Mass (1959 - Vienna Symphony), same question. How many of these did Harnoncourt play in?

I'll try to investigate, yours is a question that always comes to my mind when I see some WP recording from the period 1952-1969. I remember, however, that NH spoke in some interviews about his working as a cellist with the WP under Karajan's direction (reheasal, playing concerts, recording) especially in Bach's Mass BWV 232.

(I believe the first recorded set of Brandenburgs that Harnoncourt played in was Prohaska's: he played viola da gamba in #6. I like it.)

Actually it's an obscure set from Supraphon (1950) with Josef Mertin conducting an unidentified orchestral ensemble.

Curiously I discovered the existence of Prohaska's set just some weeks ago: you should have a Vanguard 2-LP stereo set with an orange cover, isn't it? I have another fascinating LP from the 50’s : WP directed by P.Sacher play music by Johann Christian Bach. Soloists in the Concerto In E Flat Major For Harpsichord & String Orchestra, Op. 7, No. 5 is Gustav Leonhardt. Who knows if they (NH & GL) were both there?

Harnoncourt, and voice
Harnoncourt and using scores

Thomas Braatz wrote (May174, 2003):
[To Bradley Lehman] To Brad, if 'imitation is the sincerest form of flattery'….

Nikolaus Harnoncourt (Gustav Leonhardt) is an outstanding instrumentalist with fine technical skills, unquestionably. But he has no clue how to phrase properly five consecutive notes of a choral or vocal line and make them truly expressive. His conducting simply sounds like what it is: the work of a very well-prepared 20^th century cellist (keyboardist) with reliable cello (keyboard) technique and a few theories on what Bach should sound like.

This is not directly a slam against Harnoncourt/Leonhardt. Most other HIP conductors and performers don’t have 1/10^th of a clue about singing either. The voice is an entirely different instrument, with its own set of techniques and demands such as properly controlled, non-extreme dynamics and clear, concise articulation of the text and precision in attacks and releases.

Send ten fully-trained vocalists or choirs to sing Bach cantata arias or choral movements in a large church setting, and it will sound like ten different singers or choirs: ‘full’ voices know how to bring out the expression of the voice parts, and each finds something different and valid to emphasize in the music. Furthermore, when the choir sings, it will sound resonant and colorful, and mold itself to the wishes of the conductor and when the soloist sings there is a performance that can not fail to move because of the sincere depth and profundity of expression.

Now, for contrast, send ten average HIP soloists (that is, very good singers who have not been able to develop a full-range voice) to sing a Bach aria, and all ten will sound alike: flat, stiff (more like an instrument), and relatively inexpressive due to a laof an extended range demanded by Bach. The voice will seem quiet and bland, like an inferior and hard-to-control copy of whatever instrument they’re more accustomed to copying. Generic singing does NOT translate well to the singing of the great choruses and arias in the Bach cantatas.

I’m not speaking over-dramatically here to make this point; I have heard this through years of careful listening to the Harnoncourt/Leonhardt series with the NBA score in hand, each cantata at least a half dozen times and about half of the entire series at least 10 to 15 times in preparation for the weekly cantata discussions on the BCRML where the comparison with other recordings of the same music is undertaken. It is very easy to recognize the conductors who spend the rest of their time playing their instruments (violoncello and keyboard) – even at the highest professional levels—as distinct from those whose primary musical education was based on the voice: the fluid, cantabile lines produced by the human voice. (The best singers have no strong need to sound like a different person each time they sing the same aria – nature ensures variation because the voice is very sensitive to many influences!) Ordinary choirs and vocal soloists can sing the music, being convincing in a different way, but it’s rare for them to sound really comfortable at it or to display much range.

And I’m not saying anything earth-shattering here. The same is true with any solo voice or choir. When the voices are used well by people who really know what they’re doing, who have learned how to sing in a choir or who have fully-trained their solo voices, vocal music sounds better than it does when sung by people of limited skills. The difference is in both the vocal technique/capacity and expressive powers: knowing how to convey a strong sense of conviction and emotion as the vocalists makes use of the full resources of the most glorious of instruments, the human voice. People of merely generic skill have little clue what the resources of the human voice are!

Brad, if you think the Harnoncourt/Leonhardt cantata recordings are the “top” renditions, I envy you (in a way): there is a vast world of truly moving singing and playing out there that you evidently have never heard or imagined, waiting for you to discover it. If you are willing to step past your satisfaction and listen to the real masters of the voice (choral and solo), you are in for a wonderful treat.

Bradley Lehman wrote (May 17, 2003):
[To Thomas Braatz] Tom, well done! And I agree with most of what you've said in this posting.

Really, at the moment I can think of only two points I disagree with. Consider these:

The first, obviously, is your insinuation here (and your regular statements) that professional singers such as Klaus Mertens and Paul Esswood and Ruth Holton and Kurt Equiluz (some of your standard array of 'demi-voix' victims) are not "fully trained." I won't belabor this one, but merely note it [again].


Second: what would you say about our friend Johann Sebastian Bach himself? He too was an organist, harpsichordist, and string player rather than a singer. And his vocal writing is regularly criticized (not only by people here in this discussion list) as being "too instrumental", too difficult because it doesn't respect normal "beautiful" capabilities of human voices.

That vocal and instrumental writing happens to be the territory we're dealing with here in the cantatas; and the instrumental lines are not less important than the vocal lines are, it's all a package. What is wrong with allowing the vocal parts to sound at least somewhat instrumental, according to their nature (the way Bach wrote them), and allowing the instrumental parts to sound at least somewhat vocal, as they are expressing words?

Yes, every conductor has strengths and weaknesses. And yes, you may feel gypped if you don't hear vocal production that pleases your tastes; but another person may similarly feel gypped by other parts of the various performances. As I've said before, I'm not especially fond of the singing in the H/L series either. But I've found: frequently the H or L performance is the only available recording in which the instrumental (and vocal) contributions even BEGIN to sound gestural enough (that is, taking the musical rhetoric seriously, and projecting it), where everybody else sounds tame and generic.

And, as you are a self-styled expert in the human voice, surely you realize that the human voice is capable of many expressive sounds in addition to singing "fluid, cantabile lines." What makes you think that Bach's compositions should be limited to only a narrow range of vocal expression (sounds that seem "profound" to you)?
- When someone screams or groans in agony at an injury or a loss, is that not profound?
- When a woman shrieks in the pain of childbirth, is that not profound?
- When someone croons a lullaby to a child, is that not profound?
- When someone shouts with joy at triumph over a difficult task, is that not profound?
- When someone snarls or cries at failure in a difficult task, is that not profound?
- When someone wails at the stress of impossible working conditions, is that not profound?
- When someone murmurs scarcely-comprehensible sounds in religious ecstasy, is that not profound?
What makes you think that Bach didn't put this range of expression (and more) into his music...that it must instead be restricted to "fluid, cantabile lines" that sound good to you?

Bradley Lehman wrote (May 17, 2003):
[To Thomas Braatz] What about Harnoncourt's choral recordings outside the Bach cantatas?

I appreciate the way he gets his adult choirs to sing with strong commitment to the meaning of the words, and to the dramatic impact of the music. And, especially, I like the way he makes them aware of principles of strong and weak notes, and gets the singers to differentiate: it makes the phrasing so clear and expressive.

Good examples are his recordings of the Mozart Requiem, the Mozart Vespers, some of the Mozart masses, and Händel's "Messiah." That recording of the Requiem is (IMO) shatteringly intense, and one of the few recorded performances that acknowledges the personal terror of death as part of the work (and then its moments of comfort and beauty are, by contrast, that much more comforting). The Swedish choir in "Messiah" has terrible English diction, as do some of the soloists, but they are responsive to every turn of phrase. Harnoncourt's Beethoven is also pretty strong.

I have most of Harnoncourt's recordings of the Mozart operas, too, but haven't spent enough time with them yet to say anything about them.

Tom, I agree with you that Harnoncourt in the Bach cantatas often seems not to have very good control of the boy choirs. But in other repertoire, adult choirs have no problem giving him exactly what he asks for, being alert to every nuance. Could it be that the fault is with the boys, and/or a lack of rehearsal time, and/or Harnoncourt not being so good with a group of children?

By the way, on children: I noticed in the personnel list of Harnoncourt's recording of Cantata BWV 119 last week, his own daughter Elisabeth played recorder in this before going on to her own career as a solo singer. Does she still play?

Bradley Lehman wrote (May 17, 2003):
< Thomas Braatz wrote: (...) This is not directly a slam against arnoncourt/Leonhardt. Most other HIP conductors and performers donâ?Tt have 1/10th of a clue about singing either. The voice is an entirely different instrument, with its own set of techniques and demands such as properly controlled, non-extreme dynamics and clear, concise articulation of the text and precision in attacks and releases. (...) >
Notwithstanding the misleading and dismissive phrase "HIP conductors" as opposed to simply "conductors":

Where would you put Philippe Herreweghe? I consider him one of the best living conductors of the sacred choral repertoire, from the 14^th to 20^th centuries. He seeks out an appropriate style for all the music he works on. And, in all that repertoire, he also gets his instrumentalists to play like singers, obviously of the words they're accompanying and expressing.

What's the problem?

And, what about the Leonhardt recordings late in the Teldec series, where the choir really was conducted by Herreweghe?

(Or, at the other end, what about the "Harnoncourt" recording of the St John Passion (BWV 245) that really was conducted by Hans Gillesberger, and some of the other early Harnoncourt projects that were conducted by Jürgen Jurgens?)

And what about John Eliot Gardiner and Roger Norrington, both of whom had extensive careers in choral conducting before anybody did anything serious with period instruments? What about Roy Goodman, who was a boychoir soloist (and a very good one) in Cambridge, before turning to strings and keyboards? [There is a fine recording of Allegri's "Miserere" in which Goodman is the singer of all those high C's.] What about Harry Christophers and Peter Phillips?

Once again, Tom, your bigotry against "HIP" performers rears its Medusan head.

Johan van Veen wrote (May 17, 2003):

----- Oorspronkelijk bericht -----
Bradley Lehman wrote: >>>What about Harnoncourt's choral recordings outside the Bach cantatas?
I appreciate the way he gets his adult choirs to sing with strong commitment to the meaning of the words, and to the dramatic impact of the music. And, especially, I like the way he makes them aware of principles of strong and weak notes, and gets the singers to differentiate: it makes the phrasing so clear and expressive.
Good examples are his recordings of the Mozart Requiem, the Mozart Vespers, some of the Mozart masses, and Händel's "Messiah." That recording of the Requiem is (IMO) shatteringly intense, and one of the few recorded performances that acknowledges the personal terror of death as part of the work (and then its moments of comfort and beauty are, by contrast, that much more comforting). The Swedish choir in "Messiah" has terrible English diction, as do some of the soloists, but they are responsive to every turn of phrase. Harnoncourt's Beethoven is also pretty strong.<<<
A couple of weeks ago I listened to the complete recordings of Mozart's sacred works as recorded by Harnoncourt. I found the quality uneven. Some pieces are very well done, others are very heavy and lack clarity. The interpretation of the Requiem as such is impressive, but I'm not very happy with the choir (Wiener Singverein), which uses too much vibrato. The Arnold-Schönberg-Choir is a lot better, even though it is too large in my view.

Tom, I agree with you that Harnoncourt in the Bach cantatas often seems not to have very good control of the boy choirs. But in other repertoire, adult choirs have no problem giving him exactly what he asks for, being alert to every nuance. Could it be that the fault is with the boys, and/or a lack of rehearsal time, and/or Harnoncourt not being so good with a group of children?

As far as I know the choir was prepared by its own conductor. It would have made some difference if Harnoncourt would have worked with the choir from the start.
But I think it is true that working with children is different from working with adults. A first class conductor of mixed choirs isn't automatically a first class conductor of a boys' choir. I think there is also a difference between an 'orchestral' conductor and a 'choral' conductor. In my view Harnoncourt belongs to the first category, Leonhardt to the second.

By the way, on children: I noticed in the personnel list of Harnoncourt's recording of Cantata 119 last week, his own daughter Elisabeth played recorder in this before going on to her own career as a solo singer. Does she still play?

I don't know. Right now - as Elisabeth von Magnus - she concentrates on singing (mezzosoprano), in which quality she regularly works with her father.

Hugo Saldias wrote (May 17, 2003):
[To Johan van Veen] Have you see the last New Year's Concret from Wien? That was January 1^st, 2003? Well it was Mr Harnoncoulrt that conducted the orchestra. Let me give you my first view review:

1. He needed a score (other 's Von Karajan, Abbado,etc) did not need any. Who needs a score for these Waltzes? We all know them.

2. The score was not for guide "just in case" he was reading form it.

3. Cameras do not focus all the time on orchestra and conductor on these broadcasts but the show other things as well... When they focus on him he looked sad, mad, lost... Other conductors smile all the time:Abbado enjoyed the performance more than any one else, you could see his face. Not to mention the native Von Karajan born in Salzburg how he inspired players and audience.

What's to deduct? That Harnoncourt has high musical skills but he does not seem to enjoy waltzes, from what I have seen on TV.

These broadcasts are available through amazon on video

Johan van Veen wrote (May 17, 2003):
< Hugo Saldias wrote: Have you see the last New Year's Concret from Wien? That was January 1^st, 2003? >
No, I didn't. I can't stand waltzes.

< Well it was Mr Harnoncoulrt that conducted the orchestra. Let me give you my first view review:
1. He needed a score (other 's Von Karajan, Abbado,etc) did not need any. Who needs a score for these Waltzes? We all know them. >
I don't know them. So he used a score. So what?

< 2. The score was not for guide "just in case" he was reading form it. >
So what?

< 3. Cameras do not focus all the time on orchestra and conductor on these broadcasts but the show other things as well... When they focus on him he looked sad, mad, lost... Other conductores smile all the time:Abbado enjoyed the performance more than any one else, you could see his face. Not to mention the native Von Karajan born in Salzburg how he inspired players and audience. >
I have never seen Harnoncourt smile. Since when is it an obligation to smile while conducting?

< What's to deduct? That Harnoncourt has high musical skills but he does not seem to enjoy waltzes, from what I have seen on TV >
That is your personal interpretation, not a fact.
But: what have all these things to do with skills as musician or conductor?

No offense, but this seems to me all completely irrelevant.

Bradley Lehman wrote (May 17, 2003):
< Hugo Saldias wrote: Have you see the last New Year's Concret from Wien? That was January 1^st, 2003? Well it was Mr Harnoncoulrt that conducted the orchestra. Let me give you my first view review:
1. He needed a score (other 's Von Karajan, Abbado,etc) did not need any. Who needs a score for these Waltzes? We all know them. <
Hugo, what is the appropriate articulation of the viola part in the second section of "The Blue Danube"?

< 2. The score was not for guide "just in case" he was reading form it. >
What is the scoring of the fifth waltz in Weber's "Invitation to the Dance," and which section(s) of the orchestra will require the most cues at that point?

Without peeking: do you know the 7^th and 12^th digits of your credit card? They are on every receipt you ever sign, and on the card itself, and you probably deal with those almost every day, so you "should" know them instantly from memory, right?

Peter Bright wrote (May 17, 2003):
[To Bradley Lehman] As usual, I find my opinion somewhere in the middle of the polar opposites from which Brad and Tom express their views. I agree with Tom's views on the importance of measured and controlled rerading of scores. For example, I prefer Moroney's Art of Fugue to Robert Hill's because I find the brilliant but relatively varied approach of Hill is like a rush. It sounds wonderful for the first few hearings but I cannot maintain my interest in it over successive hearings. Moroney's is less arresting and more considered but ultimately more satisfying and consistently rewarding (before one of you snorts and reacts with a 'but this is so boring' type response remember that I am not attacking your subjective opinions, but perhaps presenting a different set of subjective opinions).

On the other hand, I agree with Brad's views on a number of other issues, particularly in the choral works. I am far more of a fan of Harnoncourt than either Tom and probably Brad - I greatly enjoy H's most recent St Matthew and I think his recent recording of Dvorak's Slavonic Dances issuperb. I simply cannot understand the reflexive dismissal of Harnoncourt - one of the champions and great interpreters of music from the Baroque period onwards.

Would I take Richter or Parrot for the B Minor Mass? Perahia or Hantai in the Goldbergs? Tureck or Jaccottet in the Well Tempered Clavier? I simply cannot decide between one or the other - it is a shame if we can't appreciate very different styles of playing. I certainly enjoy some of Robert Hill's work in Bach - particularly, vol. 2 of the Hänssler "young J S Bach" series is great but it doesn't prevent me from getting sustenance from Moroney or Gilbert or Suzuki or several other quite different harpsichordists. Similarly, I probably get equal pleasure from the warts and all approach of Karl Richter's late '50s and '60s work as I do from Suzuki, Herreweghe etc. For me you simply cannot beat the sound of Herreweghe's cantatas but Karl Richter had an incisive understanding of many of these works.

So, it's all swings and roundabouts and most approaches have much to offer the listener. Unlike pop music it is usually (although not always) the case that classical music conducters and solo instrumentalists are highly talented, educated in the broad view of the musical styles they work within and present carefully considered performances of the compositions they play. Harnoncourt is surely a great example of a great conducter, who, while perhaps arrogant about the "rightness" of his approach, is very talented and knows how to present works in a valid and involving way.

Hugo Saldias wrote (Maay 17, 2003):
[To Peter Bright] Let me suggest a type of listening(valid when you want to decide between a few versions): go number by numer (this may take long but you will see the results). Make notes on each what you like or not. Then compare giving scores and dividing to obtain an avarage. This with oll works that you have not made any decision yet. It works for me. First I play CD recording them in cassette. Then I play the cassette with the same numbers back to back: Aria (all the versions) Next number (all the versions) etc. May be it will not work for you.

Thomas Braatz wrote (May 18, 2003):
Brad, you stated: >>Really, at the moment I can think of only two points I disagree with. Consider these:

>>The first, obviously, is your insinuation here (and your regular statements) that professional singers such as Klaus Mertens and Paul Esswood and Ruth Holton and Kurt Equiluz (some of your standard array of 'demi-voix' victims) are not "fully trained." I won't belabor this one, but merely note it [again].<<
Thank you, for not exhibiting [again] your penchant for coprolalia!

If you had read my past reviews now on Aryeh’s site, you would note that more often than not, Kurt Equiluz, although not a “Heldentenor” (generally, I really do not really want to hear that type of voice singing Bach) has garnered many of the highest marks from me. His voice may not be quite as strong as Schreier’s et al, but he uses it with great intelligence and sings with an emotional warmth frequently missing in voices of other tenors singing in HIP recordings of Bach’s cantatas. There is not such a noticeable drop off in volume in his low range (there is some, but it is intelligently managed with an impeccable technique and control of the voice), but when Bach demands agitated interval jumps, Equiluz begins to force his voice and the result is usually unpleasant. This is where some aspects of a demi voix become apparent.

Klaus Mertens has one of the most pleasant, easy-to-listen-to bass voices on the HIP scene. After hearing Jesus in the SMP performed by numerous basses with unbelievably bad voices (mainly in recordings that go back decades), I was able to heave a definite sigh of relief when I first heard Mertens sing the part in such a way that I was not distracted by the quality of the voice. It was in the cantata recordings with Koopman, that I became aware of the fact that his [Mertens] solo arias and recitatives were not always completely fulfilling. This may be in part due to his inclination to use frequent sotto voce when Koopman wants a light [lite], fast treatment of the aria. An element of true expressiveness is then lost. This frequent use of sotto voce is a ‘cover-up’ pointing to one of the apparent weaknesses of the demi-voix type voice.

Ruth Holton’s voice is the most instrumental with excellent, almost uncannily good intonation, but with a definite ‘hole’ at the bottom of her range where she is barely able to produce much in the way of any volume. Emulating the sound of a boy soprano, she lacks the inner strength and stamina found in many good boy sopranos. An instrumental-like sameness of tone tends to pervade whatever she sings. The expressive range is radically reduced.

Very similar to Holton’s voice, Paul Esswood’s voice also has a problem with the low range. Add to this the problem that he frequently has in hearing the correct intonation (this has nothing to do with the temperament being used) and generally sings flat. Add also to this the quality of his vibrato (annoying in my ears) and the result for me, at least, is not usually a musically engaging performance, but rather one that he has in common with quite a number of other counter tenors which I do not want to list here.


BL:>>Second: what would you say about our friend Johann Sebastian Bach himself? He too was an organist, harpsichordist, and string player rather than a singer. And his vocal writing is regularly criticized (not only by people here in this discussion list) as being "too instrumental", too difficult because it doesn't respect normal "beautiful" capabilities of human voices.<<
Perhaps you are still relying too much on the stories about Bach’s childhood: his difficult life, copying out the scores of keyboard music by moonlight, etc. You, as a keyboardist may have come away with the impression that he grew up playing primarily keyboard music and also learned a bit how to fiddle on his dad’s violin. If you had read Christoph Wolff’s recent (2000) Bach biography carefully, you would have discovered the following: (my paraphrase of Wolff’s comments):

The sons of Ambrosius Bach, all of whom attended St. George’s School in Eisenach were all members of the “Chorus musicus” and regularly participated in the combined vocal-instrumental performances on each Sunday and holiday in St. George’s Church. This chorus met, during Bach’s time in Eisenach during the week from noon to one o’clock MTThF for choral singing and instruction.

Certainly J. S. Bach would have had frequent opportunities to hear, sing or play the compositions of his uncle, Johann Christoph, whose vocal compositions impressed J. S. Bach greatly.

During Bach’s stay in Ohrdruf, he was a choral scholar (implies that he was a member of the boys’ choir as well) under Cantor Herda. Here Bach earned some money (not much, but at least something), particularly with the ‘Kurrendesingen’ when a small group of select singers sang in the streets of Ohrdruf 3 x a year and received tips. It is very possible that Bach was being paid for his singing as a soloist. Before finishing his school year, Bach left Ohrdruf to go to Lüneburg on March 15, 1700.

The fact that J. S. Bach was immediately offered a position as a choral scholar at St. Michael’s church in Lüneburg points to the fact that he had solid experience in singing not only in a choral setting but also as a soloist (concertist.) Now at age 15, Bach was being praised for “seine ungemein schöne Sopranstimme” (“his extraordinarily beautiful soprano voice.”)

It’s quite evident that you overlooked this important factor in Bach’s life, a factor that is all the more important because it involved his formative early years in a very substantial (life-supporting) way.

BL: >>That vocal and instrumental writing happens to be the territory we're dealing with here in the cantatas; and the instrumental lines are not less important than the vocal lines are, it's all a package. What is wrong with allowing the vocal parts to sound at least somewhat instrumental, according to their nature (the way Bach wrote them), and allowing the iparts to sound at least somewhat vocal, as they are expressing words?<<
How about performances of opera music arranged only for orchestra without any voices being used? It might even sound good that way, judging from some of the terrible operatic voices that are singing or have been recorded.

As a singer (reports are that he had an excellent singing voice even as an adult, but that it was only heard in rehearsals,) Bach certainly knew the difference between vocal and instrumental lines and ‘vive le différence!’

BL: >>But I've found: frequently the H or L performance is the only available recording in which the instrumental (and vocal) contributions even BEGIN to sound gestural enough (that is, taking the musical rhetoric seriously, and projecting it), where everybody else sounds tame and generic.<<
This is where our tastes really differ! Too much ‘gesturing’, too forceful or exaggerated ‘gesturing’ begins to destroy the music for me and places attention on aspects of the musical performance that should be secondary and barely noticeable (not call attention to itself!)

BL: >>And, as you are a self-styled expert in the human voice, surely you realize that the human voice is capable of many expressive sounds in addition to singing "fluid, cantabile lines." What makes you think that Bach's compositions should be limited to only a narrow range of vocal expression (sounds that seem "profound" to you)?
- When someone screams or groans in agony at an injury or a loss, is that not profound?
- When a woman shrieks in the pain of childbirth, is that not profound?
- When someone croons a lullaby to a child, is that not profound?
- When someone shouts with joy at triumph over a difficult task, is that not profound?
- When someone snarls or cries at failure in a difficult task, is that not profound?
- When someone wails at the stress of impossible working conditions, is that not profound?
- When someone murmurs scarcely-comprehensible sounds in religious ecstasy, is that not profound?
What makes you think that Bach didn't put this range of expression (and more) into his music...that it must instead be restricted to "fluid, cantabile lines" that sound good to you?<<
All of these things can be performed in a Bach cantata without turning the cantata into a miniature opera. One thing is quite clear, if you read carefully important texts that might shed direct light on Bach’s performance practices: Despite the borrowing of musical forms from secular sources, Bach elevates them to a higher level when they used in a sacred setting. The performance practices as reported by Mattheson, Agricola and others draw a distinct line between church and non-church performances. This did not keep Bach from elevating some of his own compositions from secular to sacred uses as he did with his numerous parodies, but as far as I know, there are no instances where this process was reversed (that is, when a composition once used in a sacred setting, was reused or modified for a secular one.)

It is possible that Bach’s parodies cause confusion in the mind of many performers who think that this means that he adopted freely all the means available to opera composers and their productions. While Bach was always looking for new inspiration even from listening to performances of operas (Hamburg, Dresden) and could easily have become an opera composer had he wanted to and had the opportunity presented itself, once he dedicated himself to composing and performing primarily church music, he knew that he was obligated to present music suitable to a sacred setting. Constantly an innovator, Bach constantly tried new things (a great variety of unusual instrumentation, and an ennoblement of certain dance forms) as he ‘pushed the envelope,’ but he certainly was not interested in making church music into operatic productions, as much as many today still wish to overemphasize Bach’s music in this way. Nowadays people conveniently forget that Bach’s sacred music (including the passions) was performed from crowded balconies with musicians standing out of view behind each other. The congregation did not really get to see them. Turning around to stare at them while craning one’s neck was simply not acceptable church decorum.

Bach composed his music (and notated it) in such a way that the vocal parts already have most of the expression already written into them. His careful attention to notation came about mainly because he wished to prevent the types of excesses that come through extreme ‘gesturing’ or additional attempts at adding even more feeling or expression to a part than is called for. Bach wished to limit the excesses in embellishments that some singers may have been prone to produce in order to call attention to themselves.

Continue of this part of the discussion, see: Bach and Opera

Christian Panse wrote (May 18, 2003):
< Thomas Braatz wrote: Brad, [...] Thank you, for not exhibiting [again] your penchant for coprolalia! >
What's that for a blinking light? Never saw this button before on my Mail program. "Press now!" it says... well, let's try...

*** P L O N K ***

What was that for a noise? Never heard this before on the BCML. But wait... now the air is somewhat fresher - and where is Mr Braatz suddenly gone? Can't see him anymore. Well, never mind; I can look up a music dictionary on my own when I'm in need.

Thomas Braatz wrote (May 19, 2003):
An additional point about Bach's singing ability after he lost his beautiful and strong soprano voice. This is from a document by C.P.E. Bach about his father's abilities:
{Bach-Dokumente No. 801}

"Er [J.S.Bach] hatte eine gute durchdringende Stimme von großer Weite und gute Singart." ["He had a [singing] voice that carried well with a great range and he had fine manner of singing.]

In the same document, C.P.E. Bach reports that his father conducted best of all using a violin and not from the keyboard (organ or harpsichord.):
"In seiner Jugend bis zum ziemlich herannahenden Alter spielte er die Violine rein und durchdringend und hielt dadurch das Orchester in einer größeren Ordnung, als er mit dem Flügel hätte ausrichten können." ["From his youth until he was getting rather old, he played the violin with precision (clearly in tune without mistakes) and penetratingly {with a deliberate force the carried the sound over distance,} and by doing this, he was able to keep the orchestra playing together better than he could have, had he been playing the harpsichord."]



Continue on Part 6