Terraced’ dynamics

Continue of discussion from: Matthäus-Passion BWV 244 – conducted by Günther Ramin

Bradley Lehman wrote (May 1, 2003):
< Thomas Braatz wrote: ‘Terrassed’ dynamics, although now frowned upon by many performers, represent what is commonly seen in Bach’s scores and original parts. Bach will mark entire sections ‘piano’ or ‘forte’ which means that there is a generally ‘static’ level of volume (this does not and should not preclude some variations within these groupings.) Often, however, these are now overlooked by many conductors. Somehow the sense of ‘stepping up or down’ in volume level and maintaining it until the next dynamic marking occurs should be evident, but often it is not. If it is true that Ramin “set the initial parameters at the beginning of a movement” and then did not change noticeably the ‘terrassed’ levels of the dynamics as indicated by Bach, then he would fall into the same category as many current conductors who likewise disregard these markings in the Urtext. >
"Terraced dynamics" are frowned upon because they were invented by Busoni and Stravinsky, and then repeated ad nauseum by textbook writers and teachers who did not do their homework.

(Sure, there are occurrences of "subito piano" in Beethoven &c, and special effects in Bach such as the echo last movement of the B minor keyboard partita...but they are special effects, not indicative of an overall style for all pieces.)

Or do you have hard evidence that "terraced dynamics" (as such) existed before 1900 as a general practice? I could be wrong on this, but I'm pretty sure it's an early 20th century invention: from a too-literal (mis)reading of Baroque scores, and an ignorance of performance practice documentation.

Now, if we're talking about Stravinsky's or Hindemith's "neo-Baroque" compositions, terraced dynamics are the ticket. But Bach isn't Hindemith.

Continue of this part of the discussion, see: Dynamics

Hugo Saldias wrote (May 1, 2003):
[To Bradley Lehman] Some conductors take more freedom than others.Is not up to anybody to put the limits on this freedom. If Ramin does that that is his way. The same as with Von Karajan, I do not have the book here with me but I will be very happy to give you the part he talks about this.

Thomas Braatz wrote (May 2, 2003):
Brad stated: >>"Terraced dynamics" are frowned upon because they were invented by Busoni and Stravinsky, and then repeated _ad nauseum_ by textbook writers and teachers who did not do their homework.<<
MGG (vol. 1, p. 801): "Für die ganze Barockzeit gilt das Gesetz der Terrassendynamik, also scharfer Gegensatz von f und p, besonders beliebt als Echomanier. Aber auch jedes Anschwellen oder Abschwellen geschah stufenweise, nicht allmählich, wofür der letzte Chorsatz der Bachschen Matthäuspassion eine gute Anschauung vermittelt." [Throughout the entire Baroque period the rule of terraced dynamics prevailed. This involved a sharp contrast of forte and piano, particularly popular as an imitation of an echo. Every crescendo or decrescendo occurred in a stepwise fashion, not gradually. The last mvt. of the Bach's SMP offers a good insight into this matter.]

"Aber die Sache ist auch schon bei Orchester-Werken noch des ausgehenden 17. Jahrhunderts spürbar. Hinsichtlich der Dynamik sagte der Kapellmeister die Einrichtung von Echo- und Terrassendynamik beim Orchester nach Vorbild des Manualwechsels auf der Orgel an, ebenso den durch Forte = Tutti, p = nur Streicher, pp = »nur die ersten Pulte« nuancierten Stärkegrad, aus dem nach G. Muffats Zeugnis Lully zum Urahn des Concerto grosso-Concertinos vor Stradella und Corelli geworden ist. Die Crescendo- und Diminuendoeffekte dagegen, durch die Jommelli 1749 in Rom (d.h. vor den Mannheimern) Sensation erregte, werden ihrem Wesen nach eher dem KonzertMeister zu Buch zu schreiben sein (sie schließen das »automatische« Wachsen der Tonstärke bei Sequenzen und den Schwellton seit langem nicht aus). [This matter (refers back to some liberties that may be taken by a conductor) already becomes noticeable at the end of the 17th century. In regard to dynamics, the conductor indicated for the orchestra the terraced dynamics following the model established by the change of manuals on an organ: Forte (f) = every musician played, Piano (p) = only the strings, and Pianissimo (pp) = only the 1st chair players; this according to G. Muffat's evidence which makes Lully the first to use the Concerto grosso-Concertino contrast before Stradella and Corelli did. A different matter, in contrast, are the crescendo-diminuendo effects used by Jommelli in Rome in 1749 (this was before the Mannheim school established this new tradition.) These caused a sensation and were probably more due to the efforts of the concertmaster (rather than the conductor.) (These do not exclude the possibility that an automatic crescendo (increase in volume) occurred during sequences, nor do they exclude the increase of volume on a single tone -- in other words, these things existed alongside of the terraced dynamics.)]

"Dieses Streben nach räumlicher Tiefenwirkung (das in seiner »Terrassendynamik« und registerartigen Klanggruppengliederung auf die Herkunft des Komp. von der Orgel weist) erfährt noch eine weitere Differenzierung durch Bruckners Technik der Klangschichtung, bzw. Kombination verschiedener Klangflächen."

This is a reference to Bruckner's use of terraced dynamics which points back to his previous experience as an organist.

Hugo Saldias wrote (May 2, 2003):
[To Thomas Braatz] In Bach organ music there is a 2 type of "focus" or interpretation of the score.And this is very important in the case of large fugues like the Dorian or the Trinity Fugue (that closes the Klavieruebung 3 part):

1.Forte - Piano contrasts
or
2. Art of registration: changing colors (stops).

In other words:

1.Changing volume
2.Changing color.

Both are valid.
The first reminds of the concertino and ripieno case of the concerto grosso,or brandenburg ones
The second is more orchestral type of focus.

Both are valid.

This similar concept goes to the cantatas SOMETIMES.

Bradley Lehman wrote (May 2, 2003):
[To Thomas Braatz] But the word »Terrassendynamik« itself was invented by Busoni (according to Slonimsky in Lectionary of Music).

And the MGG citation here is not 17th or 18th century evidence; it merely repeats turn-of-20th-century opinions, does it not? Other than adding/subtracting players, where is there any evidence that the appearance of a "forte" or a "piano" in a 17th/18th century score is supposed to indicate a sudden change of dynamics, always, as normal practice?

Think about it this way: you're driving in a 35 mph zone, and then you pass a sign that says the speed limit is now 45. Do you SUDDENLY change speed from 35 to 45 (if that is even physically possible)? Do you accelerate as soon as you see the sign, so you're going 45 when you pass the sign? Do you begin to accelerate only after you pass the sign? Do you simply ignore the sign if you don't feel like going as fast as 45? :) [This example from the highway isn't evidence about the music, either, but it does say some things about the laws of physics; and is music not supposed to sound natural most of the time?]

Bradley Lehman wrote (May 2, 2003):
And here's a fun bit from Kevin Bazzana's book Glenn Gould: The Performer in the Work:

"We can begin to see certain recurring features in Gould's treatment of dynamics: the reduction of detailed dynamic inflection; the pointing up of important structural (usually tonal) events at a background level; the use of dynamic continuity to set off major units of structure. Taken together, they make it plain that he was influenced by the old convention of terraced dynamics, modelled on the registrations of the harpsichord and organ and on the principle of the concerto grosso. Terracing was already common in the 1950s, at the beginning of Gould's professional career (and his musicology never extended much beyond the standards of that day); yet the term
'terraceddynamics' is no more than a century old, and today the practice is somewhat discredited. At least as early as the 1930s, considerable historical evidence had been discovered revealing detailed dynamic indications and graded dynamics in seventeenth- and eighteenth-century music, even in places where stark dynamic contrasts might seem to be indicated [17]. But terracing of dynamics was advocated in the early post-war historical performances of Bach that Gould admired, in the 1938 Kirkpatrick edition of the Goldberg Variations that he studied in his youth, and in much of the early Bach scholarship with which he was familiar. Schweitzer, for example, whose views on Bach and Bach performance seem to have had a significant influence on him, advocated dynamic terracing at the turn of the century, and many of his recommendations, both general and specific, are apparent in Gould's performances: to play some whole works or movements (for example, some preludes and dance movements) at one continuous dynamic level throughout, where no justification for dynamic nuances exists; not to use crescendos and decrescendos to make transitions between dynamic terraces; and not to accent the subject throughout a fugue at the expense of other voices. (...)"

"[17] See e.g. the historical and modern sources cited in Boyden (which dates from 1957)."

That reference is to: Boyden, David D. "Dynamics in Seventeenth- and Eighteenth-Century Music", in Essays on Music in Honor of Archibald Thompson Davison by his Associates (Harvard, 1957).
(Bazzana, p 211)

Bradley Lehman wrote (May 2, 2003):
< Thomas Braatz wrote: ‘Terrassed’ dynamics, although now frowned upon by many performers, represent what is commonly seen in Bach’s scores and original parts. Bach will mark entire sections ‘piano’ or ‘forte’ which means that there is a generally ‘static’ level of volume (this does not and should not preclude some variations within these groupings.) Often, however, these are now overlooked by many conductors. Somehow the sense of ‘stepping up or down’ in volume level and maintaining it until the next dynamic marking occurs should be evident, but often it is not. If it is true that Ramin “set the initial parameters at the beginning of a movement” and then did not change noticeably the ‘terrassed’ levels of the dynamics as indicated by Bach, then he would fall into the same category as many current conductors who likewise disregard these markings in the Urtext. >
A brief inspection of the music will show how wrong it is to have "a generally static level of volume" and a general principle where "stepping up or down...should be evident."

First, as I mentioned earlier, there are some pieces where a sudden change is indeed called for, as the composer is making an "echo" effect. Examples: "In our deep vaulted cell" chorus of Purcell's "Dido and Aeneas"; the final bars of Purcell's "Chacony" for strings; the final chorus of Bach's SMP; the final movement of Bach's B minor partita for keyboard; Bach's E-flat prelude for organ (Klavieruebung III); the aria "Floesst, mein Heiland" from the XmasO; second movement of Brandenburg 4; "He was despised" aria in Handel's "Messiah." In all of these, the echo parts are obvious: very short, and a repetition of the same notes (or similar). This is a special effect, and one reason (among many others) of writing "forte" or "piano" (or their equivalents) into the score and parts.

But there are other usages as well, not having anything to do with "terraced dynamics" (which is a term coined much later, anyway). Here are some examples:

- "Why do the nations" bass aria from Handel's "Messiah": when the bass singer enters, all the string parts have "piano" cautions...simply meaning "stay out of the way so we can hear him"...and then "forte" as soon as he finishes singing. Basically, "forte" means "spotlight is on you" and "piano" means "somebody else has the main theme here."

- "Rejoice greatly" soprano aria: some of the "piano" markings here indicate ornamented echoes, and others simply tell the players the singer's coming in. Context! Handel is also bringing the bassoon in and out of the continuo every few bars, varying the tone color of the accompaniment....

- "Glory to God" chorus: at the end of the movement the orchestra was playing forte (with the chorus) and then encounters "piano" and "pianissimo"...obviously a graded decrescendo here, rather than a series of stepwise changes: the music is going off into the distance as the angels recede into heaven.

- "But who may abide" aria, middle section "For he is like a refiner's fire": the continuo has "un poco piano" telling the players that somebody else has an important effect above. That effect is the first violins playing downward arpeggios: every one of them beginning "piano" and ending on a note marked "forte". Does this mean four notes at a constantly quiet level and a sudden loud conclusion? No! Every one of these is obviously a crescendo all the way through the five-note figure. And the second violins and violas participate in this same figure by playing only the first and last notes, similarly alternating "p" and "f". At the end of that passage, Handel cautions all these players that the last two bars are not the same pattern: he marks "p" over every note!

- Brandenburg 3, first and last movements: Bach uses "forte" and "piano" for different parts of the ensemble at the same time. Obviously it means "this is a main theme, bring it out" and "this is not." It doesn't mean the whole ensemble should suddenly lurch stepwise into loud and quiet blocks! Rather, it simply alerts the players to when they have soloistic material and when they do not.

- Brandenburg 4, first movement: the orchestral violin parts have "pianissimo" at several places (along with bowing slurs) cautioning them to stay quiet. In fact, they are playing extremely close canons after the solo violinist. At the end of this first passage (bar 241) they then encounter "forte" after the canons have already petered out in bar 240. If a conductor has bothered to think about this, he'll have them play a crescendo there in 240 to make the transition, rather than a suddenly loud series of strokes. Then a similar passage happens again in 251, again "pianissimo" in those parts, but this time Bach has five bars of rest for them at the end, and he forgets to tell them to come back in "forte" on the next theme. Does this mean they should stay "pianissimo" for the remainder of the movement, since it's not marked otherwise? Only if the conductor and players are literal-minded automatons instead of thinking!

- Brandenburg 5, last movement: there are markings of "forte", "piano", and "pianissimo" throughout this movement to alert the players to balance among themselves...not to make block-like effects where dynamics shift suddenly. The main theme is marked "forte" and the accompanying material is marked "piano". At one point (bar 99) he marks this accompanying material "pianissimo" and "Solo"...reducing it to a single player there. This is a case where "Solo" means "DON'T bring this out!"

=====

The point of all this is: "forte" and "piano" markings don't automatically mean anything in particular...the context must be examined to see the intended effect. And they don't mean (as a rule) "stay at this dynamic level until told otherwise, and then shift suddenly to the new one!" That would be wrong not because the dynamic level is inappropriate in itself, but because it's a failure to think, and because a long unmodulated stretch of _anything_ is boring. Dynamic markings in a score and parts are there to alert the players that something different is happening, and the musicians should find out what it is (from context), and play accordingly.

In my opinion, "terraced dynamics" are to music what "paint-by-numbers" kits are to painting. Y'know, the kits where they give you two dozen colors which you must never mix, but simply fill in the prescribed shapes with the prescribed colors...no blending across those boundaries of the shapes! Stepwise color transitions. Does anyone think the result of a paint-by-numbers kit is art? Is art merely an abilityto follow instructions?

(And there's the anecdote about the pianist who refused to make piano rolls: he was told that the piano rolls can reproduce "a dozen different dynamic levels" and he replied, "But my playing has thirteen levels in it!" [Or whatever the numbers were: he simply added one to whatever was said. I think this was Rubinstein, if I remember correctly. Anybody know?])

By the way: "terraced tempo" readings are wrong for the same reasons. But, how often we still hear SUDDEN tempo shifts at places such as Brandenburg 1's third movement, or the end of the organ chorale "O Mensch, bewein..."! (Thank you very much, Dr Schweitzer.)

My box of Wheat Thins tells me that the serving size is 16 crackers. I'd better go eat exactly 16 of them now.

Uri Golomb wrote (May 2, 2003):
< Bradley Lehman wrote: By the way: "terraced tempo" readings are wrong for the same reasons. But, how often we still hear SUDDEN tempo shifts at places such as Brandenburg 1's third movement, or the end of the organ chorale "O Mensch, bewein..."! (Thank you very much, Dr Schweitzer.) >
I agree completely with this point -- but why blame Schweitzer for this? I do not know his performance of this chorale-prelude (in fact, I have to check whether he recorded it at all); but I find little evidence in his book that he favoured terraced tempi -- whether in this piece or in general.In fact, he generally stated that "Bach demands great elasticity in the tempo [my emphasis -- U.G.], - the unity of which, however, must not be impaired. If the hearer perceives the nuances too clearly, he at once gets an impression of unsteadiness and a lack of rhythm" (vol. II, p. 401). Tempo changes "should be applied circumspectly and with moderation" (II: 402), with due attention to maintaining ensemble togetherness -- and to "The art of transition [my emphasis]"; the aim must be to maintain the basic pulse. Abrupt changes -- i.e., terraced tempi -- must be avoided in most cases. He does make an exception to this: "Where Bach has sharply contrasted two motives in obedience to the poetic idea, it would be wrong to tone down the antithesis by passing gradually from the one tempo to another" (II: 405). His example is "Friede sei mit euch", the bass-aria-with-chorus in Cantata BWV 67: "a rallentando in the concluding bars of teh passages that are constructed on the 'tumult' motive must necessarily weaken the effect of the entry of the 'peace' motive in the wind instruments; even a diminuendo is out of place here; the more abrupt the transition the better; the instrumental basses must give out the last bar of their 'tumult' motive forte even after the wind have entered piano. In the antiphonal cries, again, of the chorus, "Wohl uns" and "O Herr", there should be no gradual mediation between the two tempi" (II: 405). Whether or not you agree with this prescription -- and I am undecided -- it must be emphasised that this is, for Schweitzer, an exception, not a rule. He advocates terraced dynamics and terraced tempi in this particular case (and, admittedly, a few others) -- but not as a general rule. In an earlier message, I already noted that his attitude towards dynamics was similarly ambivalent: he certainly did not support the harsh, constant and unremitting application of terraced dynamics.

However, it's been a long while since I read his book (in this message, I was relying on notes I made a while ago -- though I did re-check the quotes and paraphrases I noted down at the time); so if you can show that he did advocate terraces as a general rule, enlighten me... It doesn't matter much for Bach performance today, I suppose, but as someone who studies the history of Bach performance, it does interest me.

PS: Here is what he says on the chorale-prelude (the context is a discussion of the chorus "O Mensch" in the SMP): "How the end of the [SMP] movement should be rendered is doubtful. The _adagiosissimo_ that Bach has noted in the last bars of the chorale prelude O Menshc bewein' dein Sünde gross [....] leads us to think that chorus and orchestra should here also die away softly in a slow rallentando" (vol. II, p. 212). Now, I admit this is not entirely clear (maybe the German original is clearer) -- but to me, that sounds like advocating a gradual change in tempo (not just a rallentando, but a slow rallentando) -- not terraced tempi!

Bradley Lehman wrote (May 2, 2003):
[To Uri Golomb] I haven't looked at Schweitzer's writings about this, recently. But this morning I listened to Columbia ML 4601 which has two different Schweitzer performances of this one. (He's playing from two different versions in his own edition.) Both are recorded at the Parish Church, Gunsbach, Alsace.

In both performances he has a slow tempo to begin with, and decent flexibility. And then when he gets to the middle of bar 23 (the pedal has been ascending chromatically and reaches C, and the word "adagissimo" appears above the score) he suddenly shifts down to half speed (almost exactly) and then drags it even more slowly. It's the sudden shift to half speed that troubles me, the doubling of all the note values. (The same way it drives me crazy in the Scherchen performances of the Art of Fugue, those sudden "gear shifts" to half speed. AAAUUUUGGGGHH! And Glenn Gould in Gibbons' "Salisbury Pavan", same problem, but that's because he learned it from an edition that really has the note values doubled, as some do there.)

I don't know where exactly the word "adagissimo" shows up in Schweitzer's edition, but in Lohmann's (Breitkopf) it's over the G-flat...which is later than the place where Schweitzer plays his terraced tempo shift. Again, the exact spot of that doesn't bother me so much as the fact that it happens at all. Isn't Bach simply asking for an especially relaxed treatment of the ending, giving that surprising G-flat (and the C-flat major chord) the time to make its point? "Milk it a little." That can be done very nicely without any sudden changes...and certainly without halving the prevailing tempo.

Brad Lehman
...Been playing this piece for 18 years with a gradual slackening there, given that the piece is already "very much at ease" (that's what adagio assai means, literally)...let it go down to having hardly any perceptible beat at all anymore, but still at a similar speed...

Hugo Saldias wrote (May 3, 2003):
[To Bradley Lehman] Regarding the matter below you talk about:Dr Albert Schweitzer and Bach's BWV 622:

In Amazon.com there are 3 different recordings and in one of them you can hear the samples:

1.Pearl 9992.
2.Magic Talent 48045
3.Grammofono 2000 # 78692 on track # 5 is the audio
example to listen to.The cover says:
The LEgendary 1935 and 1936 HMV Recordings and has his picture showing the console on the left side.

Change in Dynamics in music history

Hugo Saldias wrote (May 5, 2003):
Even this is Bach related group I did read a very interesting comparison of the same piece of music played by the Berlin Philarmonic by 3 of the last conductors:
Artur Nikisch: AN here
Wilhelm Furtwaengler: WF here
Von Karajan: VK here

This will help understand how very prominent musicians can focus the same work in a complete different way and then apply this criteria to Bach Cantatas:

Taken from the book: The art of the conductor by Paul Robinson , page 33:

Since all 3 conductors of the Berlin Philarmonic in the XX century - AN WF VK - recorded Beethoven's Fifth Symph. with that orchestra,a comparison of their interpretations offers a unique insight...

AN DG 2721070
WF Heliodor HS 25078
VK DG 138804

The sound available to AN in 1913, and even to WF in 1946 is far inferior to that afforded to VK in 1962. There is much to be learned.

AN: It omits the repetition of the exposition or first part of the movement...Apart from that the basic tempo is quite deliberate; rather slow, in fact.He speeds up and slows down from time to time...

WF: Conducts the various tempo and dynamics markings in the score rather more freely than we are accustomed to today.He employs 3 basic tempos for the movement disrupting the flow of the music as few conductors have dared to do before or since. Htakes all these liberties; yet Beethoven has indicated only one tempo for the movement ALLEGRO CON BRIO. But, he produces a performance of astonishing spontaneity and intensity wich leads us to suggest that he was perhaps violating the letter of the score in order to get at the spirit of it. Also worth noting in this first movement is the very long silent pause near the end,after the great climax...There is no justification for it in the score because Beethoven HAS NOT indicated any rests or
silent beats. He makes an epic mini-drama of this piece.

VK: He was able to benefit from the most sophisticated editing techniques,which can eliminate the slightest imperfection in the performance.It follows every marking in the score. He maintains one basic tempo. There are not exaggerated holds or pauses.It is, in short, a literal performance of waht Beethoven wrote. He makes a musical tour de force.

So as you see some musicians take lots of liberties. Please have a flexible mind when making a review of a cantata. Anybody has the slow SMPassion by WF? May be is not what Bach wrote, but...

Think about it

Donington and Quantz
The problem with Quantz

Continue of discussion from: Cantata BWV 129 - Discussions

Bradley Lehman wrote (May 27, 2003):
Donington and Quantz

Quoth Braatz: < Robert Donington “Baroque Music: Style and Performance: A Handbook” (New York, London, 1982) (...) also on p. 32:

Quantz in 1752 advised us that ‘light and shade must be continuously introduced’ – on the performer’s initiative – by ‘the incessant interchange of loud and soft’; but here he was presumably thinking of the latest ‘galant’ idioms, rather than of the typically baroque idioms, where much steadier though by no means rigid dynamic levels are ordinarily implied. >
Donington's next paragraph on that p32, so conveniently omitted in Braatz' citation (perhaps because it blows Braatz' thesis out of the water), tells us why dynamics were not completely notated by the composers:

"Quantz actually wanted the volume of each individual chord increased or diminished to match its degree of dissonance or consonance, though he granted that 'good judgement and sensitiveness of soul must also play a part.' C. P. E. Bach replied in 1753 that 'it is impossible to describe the contexts suitable to the forte or the piano' while agreeing that 'it is broadly true that discords are performed loud and concords
soft.' This principle in itself is valid, and of the widest relevance. Thus a preparation, being a concord, will ordinarily be softer than the clash of the ensuing discord itself (whether suspended or freshly struck), and the resolution, being also a concord, will ordinarily be yet softer, thus giving the natural rise and fall of the progression. But we are reaching here the finest nuances, which it is quite certainly undesirable to try to notate into the parts. If they will not come out right by nature, they will not come out right by artifice."

That is: it's no use trying to mark every nuance into the parts: an ignoramus is still going to sound bad because he lacks understanding of the musical grammar, and a good performer will do the right things even if they're not marked. The clutter is not going to help. The performer must be trusted to know what he's doing, to put the music across with appropriate emphasis, despite what little an Urtext edition says. A person who lacks this performance understanding, reading an Urtext and going by only what he sees, will come to the wrong conclusions about the music.

==========

Here's Quantz in comments about dilettantes and charlatans:

"(...) A musician must not occupy himself with too many other things. Almost every science requires the whole man. My meaning here, however, is by no means that it is impossible to excel in more than one science at the same time, but that this requires a quite extraordinary talent, of a kind that nature seldom produces. Many people make this mistake. Some want to learn everything, and, because of the changeability of their temperaments, turn from one matter to another, now to this or that instrument, now to composition, then to something other than music; and because of their inconstancy, they learn nothing thoroughly.

"Some who devote themselves to one of the higher sciences begin by treating music as an avocation for many years. They cannot devote to music the time that it requires, and have neither the opportunity nor the means to retain a good master, or to hear good things. Frequently they have learned no more than to read notes, and to humbug their listeners with some difficult things poorly executed and in poor taste; and if by chance they have the good fortune to become one-eyed kings in the land of the blind and to receive some applause, their lack of knowledge deludes them into thinking that, because of their other skills, they merit preference over other musicians who, though not trained at universities, really know more about music.

"Some practise music simply because they need a livelihood, without having the slightest pleasure in it. Others have learned music in their youth more through their own practice than through correct principles, and in later years are ashamed to receive instruction, or believe they have no further need of it. Disliking correction, they prefer to win praise disguised as 'amateurs'. They are forced to turn to music because fate has denied them success through their other skills; but just as they were formerly only half scholars, they now remain only half musicians, because of the time lost in applying themselves to other sciences. Their talent, which was insufficient for other fields of endeavour, is still less adequate for music, and their prejudice and conceit make them unwilling to endure any correction from others.

"He who does not possess sufficient natural gifts for academic study probably has even fewer gifts for music. Yet if someone who gives himself to academic studies has sufficient talent for music, and devotes just as much industry to it as to the former, he not only has an advantage over other musicians, but also can be of greater service to music in general than others, as can be demonstrated with many examples. Whoever is aware of how much influence mathematics and the other related sciences, such as philosophy, poetry, and oratory, have upon music, will have to own not only that music has a greater compass than many imagine, but also that the evident lack of knowledge about the above-mentioned sciences among the majority of professional musicians is a great obstacle to their further advancement, and the reason why music has not yet been brought to a more perfect state. (...)

"In the beginning we usually please ourselves more than others. We are satisfied if perhaps we can merely double a part on occasion. Then we allow ourselves to be deluded by untimely and excessive praise, and come to take it for a merited recompense. We do not wish to tolerate any contradiction, any admonition or correction. Should somebody undertake something of this sort from necessity, or with good intent, the rash fellow is immediately considered an enemy. Some persons with very little knowledge frequently flatter themselves that they know a great deal, and seek to elevate themselves above those from whom they could still learn. Indeed, from jealousy, envy, and malice, they even go so far as to scorn the latter. But if this pretended knowledge is carefully investigated, in many cases it will be found to be nothing but quackery: these persons have memorized a few technical terms from theoretical writings, or they are able to talk about musical artifices a little, but do not know how to produce them. In this fashion, it is true, they may gain some authority among the ignorant, but they also run the risk of making themselves ridiculous among connoisseurs, since they resemble those artisans who know how to name their tools, but use them poorly. (...)"

- Reilly's translation, pp 24-26.

Brad Lehman
(a performer who was also fairly decent at academics; Curriculum Vitae available at: http://www-personal.umich.edu/~bpl/resume-m.html )

Thomas Braatz wrote (May 28, 2003):
The problem with Quantz

Once again, Brad Lehman, ‘a performer who was also fairly decent at academics,’ avoids staying specifically on topic and sets up ‘red herrings’ (long quotes from Quantz’ ‘Versuch’ on ideas of what amateurs are) as he leads his critics on a ‘wild goose chase.’

He does this when he has no reasonably valid answers to my questions, as, for instance, why it is that ‘terraced dynamics’ as applied to the Bach cantatas can not simply be dismissed as a device to let the instrumentalists in a Bach aria know when the vocalist begins or stops singing, as if this were not already evident to any instrumentalist that is listening while also playing. Terraced dynamics, as I have indicated, are part and parcel of the Baroque aesthetic.

Quantz’ “Versuch….” (1752), published a quarter century after Bach had composed the majority of his cantatas, is, according to Edward R. Reilly (the translator?) in his article for the New Grove, divided into 18 chapters (only 5 of which are directed toward professionals who play or want to play the flute, the remainder address general issues of interest to amateur instrumentalists. “His views cannot be considered absolute guides for the performance of late Baroque music.”

A true test of ‘where Quantz is coming from’ would be to examine his flute sonatas composed c. 20 years before the publication of his book. These are typically and most evidently examples of the ‘galant’ style emphasizing simple melodic writing and thematic variety, but renouncing contrapuntal complexities. Quantz represents a style of composition and performance practice which is rarely encountered in Bach’s cantatas.

Daniel Heartz (New Grove) describes Quantz as being ‘more preoccupied than any of his contemporaries with defining the new style [‘galant’] (both in his ‘Versuch….’ and his autobiography.) In the galant style “flexibility in dynamic nuance went with rhythmic flexibility, or tempo rubato, in the modern Italian style. [Yes, this definitely sounds like Brad's idea of performing Bach's music.] Schäfke showed that Quantz formulated the galant aesthetic of clarity, pleasingness and naturalness in music on the basis of several earlier theorists, including Mattheson, and that these ideals, typical of the Enlightenment in general, went back to the rationalist philosophy of Descartes (‘clare et distincte percipere’).”

It would appear that J. S. Bach, as a composer of sacred music, was primarily a defender of the ‘old’ contrapuntal virtues and only cautiously and rarely incorporated galant-style mannerisms or characteristics into his music and his performance style. Quantz, on the other hand, was writing for the present and the future (he also wanted to sell his books!) and felt no need to uphold the already historical performance practices that concerned Bach a quarter century earlier.
Reading between the lines of the Heartz article, it is possible to view the ‘galant’ style of musical performance and composition as feminine (effeminate) and as Herder characterized a book that tried to give instruction on how to write in the galant style: “It lacked virility.’ According to this perspicacious insight on Herder’s part, and by analogy, Bach’s music is ‘virile,‘ while the style that Quantz defended and used in composing and performing would be ‘effeminate.’
Only those who overlook the considerable differences in performance style that Quantz and Bach represent, will be able to quote verse and chapter from Quantz’ books and pretend that they have uncovered the true source of information on just how Bach wanted his church music to be performed.
Heartz again: “The galant idiom freed composers from the contrapuntal fetters of the church style, to some degree even in the context of church music.”
This does not sound like the Bach that I have come to know.

Bradley Lehman wrote (May 28, 2003):
[To Thomas Braatz] The irony. My compositional gesture (the quotation of Quantz) was so subtle, it flew right through the radar of the man who says things are obvious, that there's no need to gesture. This has now sent him onto the unexpected tangent of Quantz as a composer of girly-mon music, and therefore unworthy to have any relevant opinions about Bach performance. Wild. Maybe I should work on the greater obviousness of my gestures. I guess that's what I'll go do.

Anyway, y'all go read Quantz' book. It's a good one, despite whatever has been said about it by a critic who hasn't read it. And the biographical sketch in the preface (by Reilly, the translator) locates Quantz right in the mainstream of "Baroque" style in Dresden et al, early in his career, working with the big guys of Italy, Germany, and France, getting to know all the parts and parcels of music. He knew his stuff, knew where he got it, and (as a writer) is worth listening to. Plus he was a longtime colleague of CPE Bach, whose only teacher was his father, so he probably picked up some tips there too, things about music that are generally good to know, things that are true and relevant about ALL tonal music. But hey.

Charles Francis wrote (May 28, 2003):
< Thomas Braatz wrote: A true test of ‘where Quantz is coming from’ would be to examine his flute sonatas composed c. 20 years before the publication of his book. These are typically and most evidently examples of the ‘galant’ style emphasizing simple melodic writing and thematic variety, but renouncing contrapuntal complexities. Quantz represents a style of composition and performance practice which is rarely encountered in Bach’s cantatas. >
I agree with these remarks and note this is not the first time a musicologist has dragged in Quantz to justify late Twentieth Century HIP practices. Some time ago I took the trouble to actually listen to Quantz to determine his relevance to Bach's music. My conclusion: there is no relevance! In fact, Quantz' music represents everything Bach was holding out against - a composer of the "Dresden ditties" as he once put it.

Uri Golomb wrote (Maay 28, 2003):
[To Charles Francis] I think there's something of a fallacy here. The relevant question, I believe, is not "did Bach approve of Quantz's compositions", but rather "would Bach have wanted Quantz (or similar-minded players) to play his own music"? Circumstantial evidence, at least, suggests the answer to the latter question is "yes": Bach openly admired the musical establishment in Dresden, at a time when Quantz was second flautist there.

Also, Bach's pattern of study -- his continued insistence on copying, performing, and re-arranging music by his contemporaries, including his Dresden colleagues -- suggests that he was somewhat more broad-minded than Charles suggests here. On the subject of Bach and Quantz, here's an intriguing article
abstract:

Jeanne Swack, "Quantz and the sonata in E-flat major for flute and cembalo, BWV 1031", Early-music , Feb. 1995 Feb; 23(1): 31-53

"The sonata in E-flat major BWV 1031, attributed by some modern scholars to J.S. Bach, is derived from a trio sonata in the same key by Quantz. Detailed comparison of the two pieces and assessment of the sources suggest alternative possibilities: Either Quantz was the composer of both works or else Bach modeled his piece on Quantz's, in emulation of a style used by the Dresden court composer whom he admired." (Abstract from RILM, contributed by the article's author)

Cautionary note: I don't want to place too much credence on this one article -- especially as I have not read it yet. And if the first hypothesis -- "Quantz composed both works" -- is correct, then perhaps there's no relevance to Bach after all. But generally, the idea that Bach absolutely despised and denigrated all galant music from Dresden is questionable, to say the least. Even if this particular article is completely wrong (and it would be extremely rash to declare it as such without reading it first -- as some people on this list are likely to do), the fact remains that Bach did study, perform, re-arrange and emulate music from Dresden, which was an important centre in German music at the time, with a rich international musical establishment. Of c, he could do much better, and he knew it; but this does not meant he considered all such music worthless.

And, again, even if Bach despised Quantz as a composer -- he might still have thought of him highly as a flautist, and he definitely thought highly of Quantz's erstwhile colleagues. Bach wanted his music to be played in Dresden (he wrote the Kyrie and Gloria of the B minor Mass (BWV 232) for that orchestra); in his Entwurff, he held up the Dresden orchestra as an example of the virtues of a truly professional nsemble. If Quantz the flautist was good enough to play Bach's music, then his statements on performance are at least worth taking seriously. On the other hand, the fact that Quantz's music -- whatever its worth -- was in a very different style from Bach's should also not be neglected. A balance should be struck, and an a-priori judgement (either declaring Quantz's book as "gospel", or dismissing it as entirely worthless) is out of place.

And one final thing: if you wish to defend the use of strict terraced dynamics as historically appropriate, it's not enough to dismiss contrary evidence as irrelevant. Even if your opponent's argument is flawed, that does not automatically make you right. You must show positive evidence, from the 18^th century (and perhaps 17th-century, if you believe Bach was perpetauting an older tradition), that terraced dynamics were mandatory or at least highly regarded at the time, and that performers were expected to stay at virtually the same dynamic level until the next dynamic indication came along. Bach's notation isn't evidence enough, not by itself -- you also have to prove that Bach wanted and expected his performers to follow the "anything not openly permitted in the notation is strictly forbidden in performance" injunction; and that's something of a tall order.

Bradley Lehman wrote (May 28, 2003):
[To Uri Golomb] Uri speaks well, folks: both in methods and content. This is the one of the best postings I've seen on this list in a while. Nicely done, Uri!

Uri Golomb wrote (May 28, 2003):
< Thomas Braatz asked: why it is that <terraced dynamics' as applied to the Bach cantatas can not simply be dismissed as a device to let the instrumentalists in a Bach aria know when the vocalist begins or stops singing, as if this were not already evident to any instrumentalist that is listening while also playing. >
I thought Brad had already answered that question! His hypothesis is that the player first got the part for individual practice, on his own; and obviously an oboist couldn't listen to the singers (or the other instrumentalists, for that matter) when doing individual practice, with only his own part in fron of him. The idea was that, by the time he reached the rehearsal, the player would already know when to expect the singer(s) to come in, and will have been prepared for changes in the texture around him. OF course he heard the singers coming in when they all played and sung together. But it was useful for him to be given prior warning when preparing in advance. Given the very tight schedule all singers and players were working on -- was there ever more than one or two rehearsals, for what was then new music and will always remain complex music? -- such a step seems quite reasonable.

(Apologies to Brad if I mis-represented his ideas)

Bradley Lehman wrote (May 28, 2003):
[To Uri Golomb]Uri, you represent them perfectly here. Thank you.

Thomas Braatz wrote (May 28, 2003):
Thomas Braatz asked: >>why it is that <terraced dynamics’ as applied to the Bach cantatas can not simply be dismissed as a device to let the instrumentalists in a Bach aria know when the vocalist begins or stops singing, as if this were not already evident to any instrumentalist that is listening while also playing. <<

Uri Golomb replied: >>I thought Brad had already answered that question! His hypothesis is that the player first got the part for individual practice, on his own; and obviously an oboist couldn't listen to the singers (or the other instrumentalists, for that matter) when doing individual practice, with only his own part in front of him. The idea was that, by the time he reached the rehearsal, the player would already know when to expect the singer(s) to come in, and will have been prepared for changes in the texture around him.

Of course he heard the singers coming in when they all played and sung together. But it was useful for him to be given prior warning when preparing in advance. Given the very tight schedule all singers and players were working on -- was there ever more than one or two rehearsals, for what was then new music and will always remain complex music? -- such a step seems quite reasonable.

(Apologies to Brad if I mis-represented his ideas)<<

Brad Lehman replied: >>Uri, you represent them perfectly here. Thank you.

Uri speaks well, folks: both in methods and content. This is the one of the best postings I've seen on this list in a while. Nicely done, Uri!<<

Uri’s advice: >>And one final thing: if you wish to defend the use of strict terraced dynamics as historically appropriate, it's not enough to dismiss contrary evidence as irrelevant. Even if your opponent's argument is flawed, that does not automatically make you right. You must show positive evidence, from the 18th-century (and perhaps 17^th century, if you believe Bach was perpetuating an older tradition), that terraced dynamics were mandatory or at least highly regarded at the time, and that performers were expected to stay at virtually the same dynamic level until the next dynamic indication came along. Bach's notation isn't evidence enough, not by itself -- you also have to prove that Bach wanted and expected his performers to follow the "anything not openly permitted in the notation is strictly forbidden in performance" injunction; and that's something of a tall order.<<

To show positive evidence, I supplied numerous sources to prove that ‘crescendo’ and ‘decrescendo’ were only applied to long, held notes (messa di voce) in a vocal or an instrumental part as a type of decoration. The terms ‘p’ and ‘f’, as used in Bach’s scores and parts, do not have anything to do with ‘piano’ to ‘forte’ 'crescendo' or ‘forte’ to ‘piano’ 'decrescendo’ until the latter half of the 18th century. Although there may have been isolated attempts at such a thing in this or that country of Europe in the Baroque period or before, it is quite clear to me from the examples that I had cited that the heritage of terraced dynamics which flowed from the Renaissance into the Baroque simply did not disappear over night. Within a single terrace of dynamics there could still be reasonable and natural variations in volume, but essentially the ‘terrace’ of ‘p’ or ‘f’ was maintained throughout the section until another dynamic marking occurred. Remember that there were 17 such markings in a single mvt. of BWV 129 (about 8 or 9 in each part – oboe d’amore or continuo.)

If, as Uri stated "it was useful for him [the player] to be given prior warning when preparing in advance. Given the very tight schedule all singers and players were working on," why would Bach include these same markings in the Dresden parts of the B Minor Mass for some of the best musicians in Europe?

At the risk of repeating myself, I suggest looking once again at Johann Gottfried Walther „Musicalisches Lexicon oder Musicalische Bibliothec“ (Leipzig, 1732) :

[This musical dictionary was compiled at a time when Quantz is only listed as a musician who plays the flute in the court orchestra at Dresden. If there ever was a book to indicate that ‘p’ and ‘f’ were markings only to indicate where other parts were playing or singing, this would be it. Look at this again carefully.

„Forte (ital.) fort, fortement, (gall.) starck, hefftig, jedoch auf eine natürlich Art, ohne die Stimme, oder das Instrument gar zu sehr zwingen.

Piano…p. ist soviel als leise, daß man nehmlich die Stärcke der Stimme oder des Instruments dermaßen lieblich machen, oder mindern soll, daß es wie ein ‚Echo’ lasse.

Echo – [in addition to echo effects with voices or instruments] Das Wort ‚Ecco’wird auch manchmahl an statt ‚piano’ gebraucht, um anzuzeigen, daß der Stimm- oder Instrumenten-Klang ‚moderirt’ und schwächlich gehen soll, gleich als wollte man ein ‚Echo’ machen.“

[‚f’ signifies ‚strong,’ ‚intense,’ but in a natural manner without forcing the voice or instrument too much; ‚p’ means soft, which means specifically that you have to make the volume of the voice sound gentle/sweet, or let it become like an echo. ‘Echo’ is sometimes used in place of ‘p’ in order to indicate that the sound of the voice or instrument should become moderate and weak, as if you wanted to create an echo.]

Uri, where are the documents, books, or references from Bach’s time (until about 1735 to 1740) that can lend even the slightest credence in Brad’s hypothesis?

Think of the Baroque world: everything was stratified into levels. Why was there such a fascination with echo effects? Is there a sociological implication involved here? The lower strata ‘echoing’ the ‘commands’ of the upper strata, mankind ‘echoing’ the will of God? Sometimes things become so common that it takes a later period to coin a term which describes a phenomenon that everyone took for granted at that earlier time. Remember also that Bach never used or knew of the term ‘secco’ as in ‘secco’ recitative, so what is all the fuss about the late coinage of the term, ‘terraced dynamics?’ He used them without having to document the term (nor did anyone else until much later, as it seems.)

But where Bach did include markings, they carried a meaning which would have to be close to what Walther describes in his musical dictionary. And since he was fastidious in documenting many details that might seem unimportant to us today, he should have made mention of such a convention which not only Bach, but many other cantors in Germany would have used and needed in order to signal to the musicians the entrance and exits of the vocal part.

Let's look at some of Bach's scores:

In BWV 248/62, oboi d’amore and continuo have these markings without any voice entrance being marked. In BWV 248/51, there are instances where the voice enters, but no ‘p’s or ‘f’s occur in the accompanying instruments. In the ‘echo’ aria BWV 248/39 (Flößt, mein Heiland), there are ‘pp’s in addition to ‘p’s and ‘f’s , where the Brad’s hypothesis just about collapses completely under its own weight: the instruments are no longer ‘warned’ that the voice was about to reenter after 7-8 ms. of rest. Bach continues to mark the ‘pp’s, ‘p’s and ‘f’s all the way through the mvt. In the midst of this overflow of dynamic markings, it would be all the more important for Bach to mark the instrumental parts at this point where the voice reenters. They (the instrumentalists) are given no prior warning and now they ‘mess up’ during a Christmas performance. No this can not be what Bach meant with these markings!

I only looked for a few minutes to find these examples. I am almost certain, but I don’t want to spend time on this project, that I would find ‘p’s and ‘f’s in the solo vocal parts as well. What is this suddenly supposed to mean? “Watch out! The instruments are going to stop playing or will begin playing (they have been playing practically all the time, so this would be complete nonsense.) Oh, so now, in the case of vocalists, Bach makes these markings mean something else? I don't think so. A ‘p’ or an ‘f’ is what Walther indicated in his dictionary and nothing else. Would that most of the conductors (the worst violators of this rule of reading these indications with the proper meaning are the HIP conductors) adhered to Bach’s indications of dynamics! By simply stepping down or up with terraced dynamics, they could already avoid the boring sameness of keeping everything on one level. Otherwise more extreme gesturing (not advisable) becomes necessary to overcome this uninteresting method of performance (without ‘terraced dynamics.’)

The best answer to this problem is use of ‘terraced dynamics’ to block out sections of the composition as being generally softer or louder. Within these sections, some reasonable (not exaggerated) variation of volume is possible within the designated volume level.

The application of ‘glissando dynamics’ within these ‘terraced dynamics’ sections, as is evident in both non-HIP and HIP recordings is not verifiable for Bach’s time and place, but the ‘terraced dynamics’ performance style is.

The Problem with Rigid Terraces

Uri Golomb wrote (May 29, 2003):
< Thomas Braatz wrote: The best answer to this problem is use of <terraced dynamics' to block out sections of the composition as being generally softer or louder. Within these sections, some reasonable (not exaggerated) variation of volume is possible within the designated volume level. >
That's actually fine (by me, at least) as far as it goes. (*) But whose definition is this? This is not what terraced dyanmics meant in 20th century practice! In too many Bach recordings, especially from the 50s-70s, terraced dynamcis sounded like: "blocking out sections of the compositions as being entirely softer or louder. Within these sections, an unreasonable (exaggerated) lack of variation of volume is sought out within the designated volume level. The ideal is: maximum contrast between sections, but no dynamic inflection whatsoever within sections. Not even the slightest crescendo towards a melodic peak, or the slightest diminuendo towards a cadence, is permissible, because there are no positive indications for it in the notation. Patterns of harmonic tension and resolution, distinctions between consonants and dissonances, the difference between strong and weak beats in a bar -- none of these things should matter in the slightest. Don't change anything unless you're explicitly permitted to do so by the notation. If the passage is marked "forte", don't do anything that might sound remotely like "piano" at any point during the entire passage; and vice versa". That is what I find unmusical and objetionable; and this is the practice that has much to do with 20th century aesthetics, and very little to do with anything else.

Here is an example -- not the most extreme, but still telling: Take Richter's and Ramin's recordings of the duet "So geh ich mit beherzten Schritten" in Cantata BWV 111 (where, admittedly, Bach's "forte" and "piano" markings are slightly out of synch with vocal entries and exits -- assuming the BG edition, which is the only one I have access to at the moment, is accurate -- because the ritornelli sometimes begin while the voices are still singing theirt final, cadential notes). Ramin and Richter don't interpret the term "forte" in exactly the same way -- Ramin is, in this case, more flexible than Richter (though usually, when I compare them directly, it's the other way
around).

But here's the rub: both Ramin and Richter understand the same thing by piano:
they believe -- to judge by what their performances sound like -- that by "piano", Bach meant a uniform, unchanging, subdued piano, in which the voices are given an unremitting spotlight and the orchestra is shoved in the background, and remains nearly inaudible throughout.

I got to know this cantata through Richter's recording; and I actually quite enjoyed his performance. But this bothered me from the very first hearing: I could hear just enough from the orchestra to realise that I wasn't hearing nearly enough. The orchestral material in the vocal passages is simply far too interesting and eventful for me to believe that Bach meant it to be virtually inaudible. No: it makes much more sense to believe that, by "piano", Bach simply meant to tell the players: "Singers are taking part in this episode -- make sure you don't drown them. So play more quietly than in the ritornelli, and don't play so loud that you drown out the singers" (and this interpretation would be perfeclty consistent, both with Brad's hypothesis and with Thomas's definition of terraced dynamics). On strictly musical grounds, I cannot believe he meant: "This entire orchestral passage is an echo to the preceding ritornello; play it as quietly as possible and avoid dynamic inflectionas much as you can". (I'll go even further: if Bach did want the orchestra to be nearly inaudible at this point, I for one don't want to follow his wishes. The orchestral material is far too interesting and important to be treated in this way).

Even in Gardiner's recording, you don't get enough of the orchestra in vocal episode -- but at least you get to hear something without straining your ears.

Now, this is really difficult -- because, in this particular duet, I prefer Ramin and Richter to Gardiner! Whatever the historical evidence, I find their slower tempo and more expansive articulation much more convincing; Gardiner sounds clipped and rushed. I like the fact that he employs a wider range of dynamics and articulation, but I don't like the particular way he goes about it. (I hasten to add: this is not a general judgement of Gardiner -- in Cantata BWV 105, for example, he displays all the profundity and expressiveness missing from this duet. I also think a case could be made for something in between -- faster and more incisive than Ramin and Richter, slower and more dignified than Gardiner).

So, even though I admire much in Ramin's and Richter's recordings of this duet (and I do -- they convey a sense of grandeur which I find entirely appropriate), and would always prefer it to Gardiner's (the only alternative I have at the moment -- though I do remember that I enjoyed Harnoncourt's version when I got to hear it) -- I still think they made a mistake by adhering to slavishly to these _piano_ markings in the score (and Richter also took the _forte_ too seriously: his forte is not entirely strict and rigid, but it's more rigid than Ramin's; it has a statuesque quality I don't really warm to, especially after hearing Ramin).

Often enough, Richter did know better than that -- he does realise that the orchestra, even when marked "piano", has a valuable contribution to make and shapes it accordingly. But his best performances are the ones where he introduces dynamic and articulatory inflections that are not present in Bach's notation as he knew it -- when he dares to go beyond literalism. And that's probably true of everyone: When the "don't change anything until you're explicitly instructed to in the notation" rule is applied, the results are almost always unmusical. There's plenty of evidence (not so much in writing as on recoridngs) that performers at least tried to apply this rule in the 20th century; I have yet to hear of any evidence that this was required at any time prior to the 20th century.

Uri.

(*) Nor is it inconsistent with the type of dynamic use Brad describes: obviously, a player who is taking care not to cover a voice will play more quietly than when he doesn't need to be careful! Nor did Brad claim that this the only thing "forte" and "piano" ever meant -- only that this is what it meant in particular contexts (e.g., arias with obbligato orchestral parts, or orchestral parts in a concerto). There was no need to codify this practice: "forte" still meant "generally loud", and piano still meant "generally soft" -- but the reason for these markings' presence in that particular context was to insure proper balance. They didn't mean "loud with no inflection" or "soft with no infection". And they were included in the Dresden parts (a) because Bach wasn't there to direct the performance (which probably didn't take palce anyhow, but that's a different story), (b) because even "the finest players in Europe" will have been working on a tight schedule on an entirely new piece, and (c) even they couldn't hear what someone else was doing when they were practicing a new piece by themselvs without a score, before the general rehearsals. If the markings were intended, among other things, to help players in solitary practice, the players' professional level is irrelevant for that particular consideration.

The only questions I would ask are: were players given parts in advance of a rehearsal? and Were they allowed to keep them betwen rehearsals? If the answer to at least one of these questions is "yes", that is enough to make Brad's hypothesis plausible (even if it doesn't constitue ultimate proof). I admit that I don't know the answers -- but maybe someone here does...

This relates to a more general point -- that the same dynamic indication, from the same composer, even within the same set of parts, could mean slightly different things in different musical contexts, and that often common sense coupled with experience will enable one to make the distinctions. This should be kept in mind. Dictionary definitions usually don't cover the full use of a word, term or sign in practice -- they're a good place to start, not a good place to finish.

And while Walther indicates that "Echo" and "piano" are sometimes synonomous, that doesn't mean that they're interchangeable in any and all circumstances (in fact, he says that "Echo" can replace "piano" -- but not the other way around; and I already described, above, a case where interpreting "piano" as intercheangable with "Echo" seems utterly wrong).

Thomas Braatz wrote (May 29, 2003):
Uri Golomb stated: >> And while Walther indicates that "Echo" and "piano" are sometimes synonymous, that doesn't mean that they're interchangeable in any and all circumstances (in fact, he says that "Echo" can replace "piano" -- but not the other way around; and I already described, above, a case where interpreting "piano" as interchangeable with "Echo" seems utterly wrong).<<
But Walther is really say that they are interchangeable:

Piano…p. ist soviel als leise, daß man nehmlich die Stärcke der Stimme oder des Instruments dermaßen lieblich machen, oder mindern soll, daß es wie ein ‚Echo’ lasse.

Echo – [in addition to echo effects with voices or instruments] Das Wort ‚Ecco’ wird auch manchmahl an statt ‚piano’ gebraucht, um anzuzeigen, daß der Stimm- oder Instrumenten-Klang ‚moderirt’ und schwächlich gehen soll, gleich als wollte man ein ‚Echo’ machen.

I originally translated:

[‚p’ means soft, which means specifically that you have to make the volume of the voice sound gentle/sweet, or let it become like an echo. ‘Echo’ is sometimes used in place of ‘p’ in order to indicate that the sound of the voice or instrument should become moderate and weak, as if you wanted to create an echo.]

but note:

Walther states: „ p ist soviel als leise, daß man nehmlich die Stärcke der Stimme oder des Instruments…mindern soll, daß es wie ein ‚Echo’ lasse.“

should really read:

‚p’ is the equivalent ‚soft’, which specifically means that you have to reduce the strength/power/volume of the voice or of the instrument, so that it sounds like an ‘echo.’

Sorry about not translating this passage completely/correctly.

Also:

“Das Wort ‚Ecco’ wird auch manchmahl an statt ‚piano’ gebraucht.“ = [The word ‚ecco’ is sometimes used in place of ‚piano.’]

In other words, ‘ecco’ can replace ‘piano,’ but does not have to – it is simply a special instance where there is an imitation of an ‘echo’ (usually only very short musical phrases.) When ‘ecco’ appears such short phrases, it means the same as ‘piano’ which is the word which is used more generally for all situations where a reduction in volume is called for including ‘ecco’ as well as longer musical phrases and for terraced dynamics.] Sometimes in Bach (in the echo arias), ‘p’s are used in place of ‘ecco’ when Bach forgets to mark them that way. They are completely interchangeable and synonymous with the provision that the marking ‘ecco’ is infrequently used.

Uri stated: >>And they were included in the Dresden parts (a) because Bach wasn't there to direct the performance (which probably didn't take place anyhow, but that's a different story), (b) because even "the finest players in Europe" will have been working on a tight schedule on an entirely new piece, and (c) even they couldn't hear what someone else was doing when they were practicing a new piece by themselves without a score, before the general rehearsals. If the markings were intended, among other things, to help players in solitary practice, the players' professional level is irrelevantfor that particular consideration.<<
If you read Bach’s assessment of the musical capabilities of his students, you will discover that Bach will say that he sings or plays ‘vom Blat fertig’ [which means, as in an instance I have before me, a teenager is able to sight-read everything that is placed before him.] Now take the best instrumentalists in Europe (Dresden in Bach’s time), do you seriously think that they would have to practice a new piece before playing it the way that Bach would like to hear it? In auditions for choirs or symphony orchestras today, aren’t these candidates asked to sing or play ‘vom Blat fertig’ as a final test of their true capabilities? I think you have underestimated the musical abilities that Bach sought and obtained despite all the difficulties that Bach reported to the authorities regarding his students (both from the St. Thomas School and from the university.)

Uri stated: >>Nor is it inconsistent with the type of dynamic use Brad describes: obviously, a player who is taking care not to cover a voice will play more quietly than when he doesn't need to be careful! Nor did Brad claim that this is the only thing "forte" and "piano" ever meant -- only that this is what it meant in particular contexts (e.g., arias with obbligato orchestral parts, or orchestral parts in a concerto).<<
Stating “a player who is taking care not to cover a voice will play more quietly than when he doesn’t need to be careful” means that the instrumentalist is and should be reducing the volume level of his playing, just as Bach indicates, to the lower step in terraced dynamics. This will ensure that he will not cover the voice of the vocal soloist. In this particular context, arias with obbligato instruments, Brad's hypothesis [that 'p's and 'f's alert the player to the entrance and exit of the singer] is superfluous/redundant since reducing the volume level is what is naturally done by an instrumentalist in such a situation. By placing a ‘p’ marking into the instrumentalist’s part in such a situation, Bach is saying “play this soft and sweetly [‘lieblich’ is part of Walther’s definition.] This is a ‘terraced’ dynamic level change from the ‘forte’ which precedes it. What Brad was trying to say about BWV 129/4 in the terraced dynamic section marked ‘p’ where all the parts unite in a unison was “here is a special place which does not mean ‘p’ but rather is only an indication that the vocal soloist is singing here, and since there is a special place (not marked by Bach in any way) where all the performers come together, we will have to create very noticeable gestures here so that the listeners will also be able to understand what is going on here that is so special.” This may be just the opposite of what Bach had in mind here. Michael Marissen in his “The Social and Religious Designs of J. S. Bach’s Brandenburg Concertos (Princeton, 1995) in a footnote on p. 68 states: “We know from recent studies of the theological traditions behind 17th and 18th German poetry that the (ostensibly weak) echo answering the soprano’s prayers [the ‘echo’ aria from the Christmas Oratorio] has to be understood to be the voice of (the all-powerful) Christ. [This according to Ernst Koch in the Bach-Jahrbuch 75 (1989) pp. 203-11, in his article “Tröstendes Echo: Zur theologischen Deutung der Echo-Arie im IV. Teil des Weichnachts-Oratoriums von Johann Sebastian Bach.” Accordingly the aforementioned ‘piano’ section where the Trinity appears as a unison of the 3 parts need have no special rhetorical gesturing, on the contrary, it would be completely out of place here because the ‘piano’ terraced dynamic level makes it become even more powerful. Any attempt to emphasize this special passage with even the slightest increase in dynamics (or gesturing) would be completely out of place as far as Bach was concerned. He knew this and marked this section accordingly: ‘p’ in both the oboe d’amore and continuo parts while the vocal part remained unmarked. Had Bach wanted this otherwise, he would have certainly marked it since he had already included 17 dynamic markings in this single aria.

Uri, thanks for sharing your thoughts and observations on the Ramin, Richter, and Gardiner recordings of BWV 111. If I get a chance, I will try to listen to them and see if I can hear the distinctions that you have pointed out.

As I have already pointed out, you would be amazed at the number of times that Harnoncourt/Leonhardt, Leusink, and some other HIP recordings disregard completely Bach’s dynamic markings in the arias. Perhaps Brad based his hypothesis on what he heard in these recordings, but the reality of terraced dynamics can not simply be dismissed upon the evidence from these recordings. Many of the instrumentalists playing their reconstructions of original instruments lacked the ability to play on more than one level (usually quite loud (reeds) or only soft and squeaky (upper strings.) It was possibly more convenient to concoct the hypothesis that ‘p’ usually did not mean ‘p’ nor did ‘f’ have any special meaning other than to note that the singer has stopped singing for a few bars, because everything remained on the same dynamic level as far as these instruments were concerned. I know this is an overgeneralization (when Harnoncourt, for instance, played the bc alone, you can hear true musicianship which takes into account reducing volume when a soloist is singing), but, in the main, this is what I have heard in many of Bach’s arias. I have long abandoned trying to indicated these omissions in my reports because there are other things that also need to be noted.

Now we have seen that a 'p' = 'piano' can have a profound significance in Bach's music, but this is only possible if the conductors insist upon terraced dynamics as marked by Bach. [There is still room for some reasonable, contained flexibility within these sections of terraced dynamics.]

The Problem with Hearsay, Secondary Sources, and Crystal-Ball Gazing

Bradley Lehman wrote (May 29, 2003):
ON GESTURES:

- Many musical gestures have nothing to do with dynamics. There are myriad ways to make a musical point.

- Among dynamic gestures, a whisper is as strong a gesture as a shout is. I've already mentioned this. Dynamic gestures in music are not limited to accents.

- A good example of a gestural performance is forthcoming. It should (I hope) put to rest the outstanding misconceptions about this topic of gestures.

=====

ON QUOTATION:

- To misrepresent an opponent's position by misquoting or paraphrasing him, and then to shoot it down with further interpretation: that's a gesture. A rude one, to say the least. Why not simply supply the URL back to the source posting in the Yahoo archives (whether or not there is also a small bit of quotation for context)? It would save time, space, and confusion. And, it would be more respectful to the person being quoted, and more respectful to readers' intelligence.

=====

ON SECONDARY SOURCES, AND ANALYSIS, AND CRITICISM:

- It is important to analyze Bach's music on paper, without using any recordings as direct "evidence" about the music. I've already mentioned this, too. In music that was written down before the advent of recording technology, all recordings are evidence of a performer's interpretation, not evidence of the composition itself. They might be very informative, even inspiring, but recordings are secondary sources (unless of course the topic is the recorded performance itself). The best thing a recording can do is to send the curious listener scurrying back to the source (a score plus any other documentation) to see how or why a particular effect has come out, to see what the composer and performer did. That's the case with ANY secondary source: it should spark direct examination of the primary source to see better context, rather than being taken as an end in itself!

[This is rebuttal to the insulting speculation: "(...) you would be amazed at the number of times that Harnoncourt/Leonhardt, Leusink, and some other HIP recordings disregard completely Bach's dynamic markings in the arias. PerhaBrad based his hypothesis on what he heard in these recordings, but the reality of terraced dynamics can not simply be dismissed upon the evidence from these recordings."
http://groups.yahoo.com/group/BachCantatas/message/5220
Perhaps nothing. My hypothesis about the _forte_ and _piano_ markings, written at
http://groups.yahoo.com/group/BachCantatas/message/5189 ,
is based directly on analysis of the music, and on thinking in a practical manner as a composer/performer thinks, and familiarity with 16th-18th century notational conventions, and practical experience in ensembles. In fact, I have not listened through any recording of that aria!]

- As for "HIP recordings" and/or "HIP conductors" "disregard[ing] completely Bach's dynamic markings...": is it not possible that these musicians actually know what they're doing, and know quite a bit more about the topic than the critic does?

- Quantz has a chapter 18, "How a Musician and a Musical Composition Are to Be Judged"--47 pages in the English translation, 89 sub-points about music criticism! He goes through a history of style (about 150 years), and compares the strengths and weaknesses of three national styles (Italian, French, and German). And he describes the differences between connoisseurs (those of equitable and balanced judgment) and incompetent critics (who are swayed by ignorance, prejudices, passions, too much reliance on a composer's or performer's reputation rather than examination of the work, and too much reliance on style over content). He even has some things to say about the seriousness of various types of church music, and warns that it must never lapse into pedantry; and he defends church music against the bad performances it often receives. He describes how some kinds of music require much more compositional and performance skill than others. And there's much more; he even has an endorsement of the perfection of JSB's organ playing (the footnote points out that Quantz certainly heard him in Dresden, and possibly also in Potsdam). Quantz knew his business. Would-be critics are advised to read this book.

Brad Lehman
(not to be mistaken for Brad Strawman)

The Problem with Quantz, CPE & Geminiani

Thomas Braatz wrote (May 31, 2003):
It appears that Brad Lehman has now ‚pulled out all his big guns,’ (well, perhaps not yet Leopold Mozart, but that might come too) and all of these, as can be expected, are clearly of a later stylistic period, according to musicologists (New Grove & MGG.) Brad refuses to admit that generational change can affect performance practices considerably and he believes that Bach’s sons are true representatives of the performance style that their father advocated:

Brad stated: “He [CPE Bach] says his father [JS Bach] taught him to know each style (French, Italian, German) well, and to pick the best from each to build one's style: "utilize all that is good, no matter where we may find it." Read Brad’s inference between the lines and you would have to conclude that ‘like father, like son’ would apply to performance practices and compositional styles, but nothing is further from the truth. Anyone who has heard the compositions of CPE and JC Bach would be aware of the major revolution that was taking place in music (and other arts as well - the transition from the Late Baroque to the Rococo = excessively or tastelessly ornate) in the decades after most of JS Bach’s cantatas had been composed and performed (after 1730.) All of this is confirmed by those who have examined these issues much more deeply than Brad has. Add Quantz and Geminiani to this list and you will have some of the most important representatives of the galant style.:

Take both Quantz’ ‘Versuch…’ (Berlin, 1752) and CPE Bach’s ‘Versuch…’(Berlin, 1753) which are not written to give information about performance practices used by JS Bach for his cantatas in the Leipzig churches a quarter of a century earlier, but rather to outline the newly developing styles that were in vogue at the time. Both Quantz and CPE Bach claim that they represent the mixed style based on a combination of Italian, French, and German influences. While JS Bach may have likewise chosen the best of every style available and advocated to others to do likewise, these styles were changing quite quickly and considerably in the middle of the 18th century. The choices made by Quantz and CPE Bach at that point in time regarding which elements of style ‘would be best’ would not be the same as those made by JS Bach in an earlier period.

The MGG states the following regarding the major part of Quantz’ ‘Versuch’ (excluding the few chapters at the beginning that treat the flute specifically): “[Es] beschäftigt sich mit allgemeinen Fragen des musikalischen Geschmacks, der musikalischen Bildung und der Aufführungs-Praxis. Quantz' Anschauungen entsprechen der Musikästhetik des galanten Stils“ [The main portion of the book treats general questions of musical taste, musical education and performance practices. Quantz views are those of the musical aesthetics that belong to the ‘galant’ style.]

Here are additional reasons why these books are not appropriate sources for determining the performance practices of JS Bach’s cantatas in the 1720s:

1) CPE Bach’s book treats almost exclusively keyboard instruments (mainly clavichord with some hopeful comments about the emerging piano). To apply these keyboard techniques and performance advice specifically directed at keyboard playing and to generalize these as if they also gave definitive information on how JS Bach’s cantatas were performed is very much an act of overreaching and wishing things to be as one would like to have them be, and thereby disregarding stylistic evidence to the contrary.

2) The music composed by both CPE Bach and Quantz should serve as further evidence regarding the incongruity of allowing their ideas about music and the performance thereof to be applied to the music of JS Bach. Quantz, with over 305 concertos and 235 sonatas for flute to his name, (candidly now, how many of these have you come to love and listen to repeatedly?) is squarely in the stylistic period known as the ‘galant’ style. This style is based on emphasizing simple melodic lines and thematic variety with frequent use of appoggiaturas and trills, while renouncing contrapuntal complexities. This simply does not sound like the Bach who wrote the great cantatas of the 1720s.

3) In his “Versuch…”, Quantz, in the 1st section speaks of ‘wesentliche Manieren” [the essential mannerisms, mainly French influence] and of ‘willkürliche Veränderungen’ [the freely applied variations, mainly Italian influence], the latter type to be applied only to certain types of ‘adagio’ mvts. This is certainly not the generalized application of ‘sprezzatura’ or any other exaggerated freedoms that Brad claims can be applied to the performance of JS Bach’s cantatas by a ‘musician in the know.’ In the final part of his treatise, Quantz cleverly chooses to focus on ‘taste,’ and in doing so, he allows himself great latitude thereby avoiding detailed issues.

4) A summary of what Quantz stands for [from the MGG]:

„Er fordert den »gemischten Geschmack«, d.h. eine Verschmelzung des italienischen und des französischen Stiles, indem er italienische Verzierungskunst, französische Architektur und Rhythmik mit dem »gearbeiteten Stil« der deutschen Kontrapunktik zu einem neuen Stile verbinden will. Die »gearbeitete Musik« des Barocks erscheint ihm wegen der gleichbleibenden Affekte und instrumentalen Klangfarben langweilig. Er fordert Mannigfaltigkeit des Ausdrucks (18. Hauptstück), Wechsel der Töne und Instrumente, deutliche Stimmungsabgrenzung der Sätze, harmonische Abwechslung durch Dissonanzen, melodische durch Verzierungskunst“

(He [Quantz] demands [a musical style based upon] a ‚mixed taste,’ that is, a fusion of the Italian and French styles which he wants to obtain by combining the Italian art of ornamentation, French ‘architecture’ [the structure and form of music] and [manner of]rhythms with the contrapuntal style of the Germans that is very much Baroque [read ‘old-fashioned’ because it gives the impression of being ‘arduously worked out’] and forming with these a new style [the ‘galant’ style.] This ‘worked out’ [it gives the impression that the music has been composed with great effort] Baroque music seems to Quantz to be ‘boring’ due to the fact that the affects and the instrumental sonorities/timbres remain the same on one level. Quantz demands greater diversity of expression, more frequent changes in the notes [in tonality? harmonic progressions?] and the instruments, more distinct delineation of the changes from one mvt. to another, greater harmonic changes by means of dissonances, and more melodic variety by means of embellishments [the freedom to add things not in the score.])

Had Quantz ever heard any of Bach’s cantatas, he would most likely place them into the category of the ‘gearbeitete’ music or style that was very Germanic and ‘old-fashioned’ Baroque in nature. And such a source as Quantz’ ‘Versuch…’ is called upon to give valid evidence on the performance style that JS Bach must have used!

As far as Quantz having heard Bach play the organ in Dresden and possibly also in Potsdam is concerned, what does this have to do with the performance practices of the Bach’s cantatas in the 1720s?

5) Geminiani also wrote treatises on music in the middle of the 18th century: “Rules for Playing in a True Taste (c. 1748); The Art of Playing on the Violin (1751); etc. These deal with such things as the use of vibrato (‘as often as possible’,) acciaccaturas, trills, mordents, turns, crescendos and diminuendos (this could be something new unless it simply refers to 'messa di voce' applied to solo string instruments,) and how to realize the continuo and dynamic markings. Some of these are quite extreme and in advance of its stylistic period. Geminiani was a pioneer of sorts with a tendency toward extremes and exaggerations in performance practice. One of this major works, ‘The Inchanted Forrest’ (c. 1754-56) is described as being mainly in the ‘galant’ style. A printed review of this pantomime with music (in reality, a number of concerti grossi) was described by Friedrich Melchior Grimm (April 15, 1754) as being accompanied by ‘une mauvaise musique’ (bad music) by M. Geminiani. Here the contrapuntal writing of his earlier works had almost completely disappeared. Yet this proponent of the ‘galant’ style is quoted as further evidence on how Bach’s church cantatas should be performed!

6) From a Geminiani treatise Brad quotes “Indeed: the craftsmanship of a good performance consists in large part of objective knowledge beyond the page, and the following of specific advice above one's own Opinion, because the techniques have been proven empirically to move people.” But obviously his exaggerated, extreme manner of playing caused him to get fired: “And he himself was dismissed from an early job (as a violinist leading an orchestra) because he played with so much rubato (sprezzatura, the rhythmic inflection of the melodic parts) the other players couldn't stay with him.” Certainly Geminiani would not be a good example to follow if one were to perform a Bach cantata properly!

7) Brad stated: “He (CPE Bach) even lists some explicit situations where the performer must IGNORE whatever dynamic marking he sees on the page, and play more firmly: whenever the music modulates (to emphasize the dynamics of harmony!); the cadential progression setting up a soloist's cadenza; the first note after a general pause; whenever the ensemble needs more help to stay together; more. All very practical, and all to help the audience (and the other performers!) "get" the music!” No, this is simply practical advice to keep the performance from falling apart after a fermata. The bc keyboard player, (now probably playing a fortepiano not available generally to JS Bach) who is functioning as a ‘conductor,’ may temporarily disregard the ‘piano’ dynamic that may follow the fermata for only one or two notes if indeed the those notes are marked ‘p’ in the first place. Brad seems to be relying upon a bad translation of CPE Bach, or perhaps he is scraping the bottom of the barrel for evidence that ‘piano’ markings often do not mean what Walther indicates in his lexicon. There is no sense here that the performer must ignore dynamic markings, because there are other solutions to the problem which are more commonly used.

Here is the quotation in the original: “Solten diese nach der Cadenz sich gleich weiter fortbewegenden Grundnoten mit einem Piano bezeichnet seyn, welches aber selten vorkommt, so schläget man wenigstens die ersteren, welche vor dem Eintritte des folgenden Tactes vorkommen, stark an, oder giebet allenfalls den Mitmusicirenden durch eine Bewegung des Cörpers die Tacteintheilung zu erkennen.“ [If these notes, the next ones that follow the cadence (which has a fermata to mark it) – (this implies, as the author pointed out previously, that a ritardando leads into the fermata with a trill) – are designated with a ‚piano’, which rarely occurs after all, then the bc player should either play these notes which lead into the next measure with a stronger dynamic (once again resuming the original tempo) (but this method of disregarding the piano dynamic is not really absolutely necessary), or indicate to the other players with body movements the way the measure is to be divided – this means the keyboard player in an ensemble functions best as a conductor who will get the ensemble moving forward again as a group by means of body movements as they resume playing at the original tempo.]

All of this has nothing to do with terraced dynamics in JS Bach’s arias. CPE Bach has failed to provide evidence in support of Brad’s hypothesis (that is, if CPE Bach, whose musical style and aesthetics can be equated with Quantz, is even to be entertained as providing any kind of valid evidence for the performance of JS Bach’s cantatas in the 1720s.

Brad stated: “When one is trained INSIDE these specific styles, speaking them as a natural rather than a foreign language, the appropriate expression of the musical gestures comes easily.”

It would appear that Brad was trained INSIDE the specific style called ‘galant,’ a language that he speaks easily and naturally because it is in this stylistic period the expression of the musical gestures comes easily: the simple melodic lines and the lack of contrapuntal writing cry out for more embellishment/variation and exaggerated gesturing because the audience would otherwise doze off (Haydn ‘Surprise Symphony,’ etc.)

All of this effort toward implementing the ideals of the 'galant' style is not necessary and not suitable for the performance of Bach’s cantatas; it is best left for the musical style of this later period that Brad best represents.