30 MYTHS ON BAROQUE HARPSICHORD PLAYING
by Claudio Di Veroli
This document is copyrighted © 2011 by Claudio Di Veroli
1st edition, Bray Baroque, Bray, Ireland, Winter of 2011
To Franz Silvestri
This text is a free companion to the eBook Playing the Baroque Harpsichord (Di Veroli 2010) from now on simply cited as "PBH". Find below, arranged in a logical sequence, a catalogue of highly debatable statements often found in present-day writings related to the harpsichord. The clarifications I provide are based on ideas agreed upon by current musicology and/or demonstrable by evidence. Most of these topics are fully treated in the PBH eBook, which includes the necessary references to primary sources.
For readers of PBH and professional harpsichordists, this text can be seen as a simple checklist or entertainment. For readers unfamiliar with PBH, the 30 Myths are expected to whet their appetite for more information about the instrument and related technique and interpretation matters. Hopefully, they will find that information in the comprehensive PBH book.
Notation: as customary in English texts, the notes of the diatonic scales are identified as C, D, E, F, G, A and B. The different octaves in the harpsichord range are identified thus: CC-BB (CC=lowest C in the piano), C-B (C=lowest open cello string), c-b (c=lowest open viola string), c'-b' (c'=lowest C in the oboe), c"-b" (treble-most complete octave in early keyboards) and c"'-b"' (which includes the treble-most notes in Baroque keyboards).
MYTHS ON . . .
French double harpsichord dated c.1700 extant in Rome, Italy
MYTHS ON THE HARPSICHORDS AND THEIR ACTIONS
1. We have to be thankful to Wanda Landowska for the modern revival of the harpsichord
Landowska did not revive the harpsichord. Exactly a century ago, she commissioned the Pleyel piano factory to build for her a piano with plucking action, based on acoustical and mechanical principles unlike any historical harpsichord. To make matters worse, throughout her very long life she played the very-feeble sounding contraption with a finger technique and interpretation manners unlike anything described in ancient sources. It took decades for organologists, musicologists and musicians to dispel the endless list of Landowska's erroneous concepts, which marred generations of younger musicians. Only in the 1960's true harpsichords started to appear in recitals, and it would take even longer to reach a historically informed performance. (PBH pp. 8, 10 and 18)
2. The ideal instrument for the modern player is the five-octave double, on which all harpsichord music can be played
This is a common, yet misleading, belief. It is true that the entire harpsichord repertoire can be played on such an instrument, typical of mid-18th century France. It is also true that two keyboards were specifically required by J.S. Bach for his large-scale harpsichord works. For other repertoires, however, the once ubiquitous single-manual Flemish or Italian harpsichords—and virginals—provide a better fit, both in sound and in the different feeling of the action. (PBH pp. 14ff.)
3. A five-octave-plus range is needed by the harpsichord literature
Domenico Scarlatti and Soler wrote for a GG-g''' keyboard range. Some French and German composers scored a few FF in very late harpsichord pieces (after 1740). It could therefore be argued that a "universal" harpsichord range is FF-g''' (63 keys). However important, though, these pieces are a tiny proportion of the huge corpus of harpsichord music. All the other works in the repertoire are playable on a GG/AA-d''' range (55 keys), implying an instrument which can be tuned faster, is smaller, cheaper and easier to carry around. (PBH pp. 21ff.)
4. The music of J.S. Bach requires an instrument with a 16' stop
This belief originated in a purported Bach instrument with a peculiar stop disposition, which was the basis for the huge output of German harpsichord factories c.1910-1970. It was eventually found that the instrument was unrelated to Bach and that its 16' stop had been a 19th century addition. However, a handful of extant ancient harpsichords—German as well as Flemish and Italian—do have an original 16' stop. On a recent online forum, the possible harpsichord models in Zimmermann's coffee house (where Bach would have his chamber music played) have been discussed, with arguments put forward for very different types of instruments. One thing is obvious, however: Bach wrote important works for two or more harpsichords, and even if one of them had the uncommon 16', the stop would hardly be used in a performance with other harpsichords that did not have it. Leading makers feel that the addition of a 16' stop is significantly detrimental to the overall sound quality. Finally, the use of the 16' stop has the unwelcome effect of lowering the perceived pitch by an octave. The 16' stop is virtually absent from present-day harpsichord making. (PBH pp. 17 and 19)
5. Building a harpsichord from a "kit" produces a second-rate instrument, if not an unplayable one
This is certainly so for the ubiquitous kits assembled by careless amateurs. However, a good-quality kit is a recommended project for somebody who intends to make one or two harpsichords in his/her lifetime: the kit saves not just money but also lots of trouble in procuring plans, materials and directions. Very importantly, the basic harpsichord-making skills (in woodworking, action assembly and regulation) are always needed. Many kits carefully built by knowledgeable and suitably trained amateurs have resulted in excellent instruments. (PBH p. 20)
6. All the modern changes to the traditional harpsichord jack are detrimental
Most makers today prefer to produce perfect replicas of traditional jacks. This is because most of the modern "improvements"—devised for the so-called "modern harpsichords" during the 1st half of the 20th century—are detrimental, and sometimes even ridiculous. However, a few of those improvements make regulation much easier, and can be fitted without producing any noticeable change in the action's weight, functionality, reliability and feeling from the performer's point of view. (PBH p. 20)
7. The harpsichord jack works in a few consecutive and distinctive stages
Although just a handful of stages—typically only four—are illustrated in most modern works on the subject, actually the jack action has no less than eleven consecutive and clearly differentiated stages. Understanding them is of primordial importance for the harpsichordist. (PBH p. 26)
8. Traditional dampers never damp strings when a stop is disengaged: flag dampers do, and further they "overdamp"
There is nothing more dangerous than a partial truth. Indeed, ancient harpsichords had mostly the strings undamped when a stop was not in use: but this was not always the case. As for flag dampers, if made with the proper material they do not overdamp at all. For a simple and small piece of cloth, the harpsichord damper's history and workings are incredibly complex: a comprehensive explanation is now available. (PBH p. 110-113)
Keyboard of virginal by de Rossi, Milan 1577
MYTHS ON FINGER MECHANICS, ARTICULATION AND ORNAMENTATION
9. The typical Baroque technique is based on withdrawing every fingering curving it towards the palm of the hand, the so-called "snap"
The snap was first described in post-Baroque writings. Its attribution to J.S. Bach, no less, is late and dubious. Hardly compatible with the traditional harpsichord technique, there is no justification to use the snap for the performance of the Baroque repertoire. (PBH pp. 29-30)
10. The basic or default harpsichord articulation is the legato
Virtually all the scholars agree today that this was never the case. Throughout the harpsichord era, one finds plenty of unambiguous and uniform evidence against the legato as basic keyboard articulation. A thorough historical survey is now available. (PBH pp. 31-34)
11. Under some circumstances, Baroque trills and mordents were meant to be played before the beat
Again, virtually all the scholars agree today that this was never the case. This myth is a modern concoction, and it can be proven false by checking the ancient sources in texts and scores. (PBH pp. 39-49)
12. Most harpsichord music—especially chords—was meant to be played with different degrees of audible arpeggio
This is yet another groundless belief: however common, the arpeggio was never a default performance feature. A full scrutiny is now available. (PBH pp. 54-55)
Title page and embellishments from Roger’s edition of Corelli’s Violin Sonatas Op. V, Amsterdam 1710
MYTHS ON NOTES INÉGALES AND RHYTHMICAL ALTERATIONS
13. The practice of notes inégales was flexible and vaguely—or diversely—documented: it is impossible to lay down strict rules for it
On the contrary, collating the ancient sources, a very clear picture emerges of a remarkably uniform practice, with neat and sensible rules about the performance of inégales: rhythm, articulation, when to apply them and further details. (PBH pp. 56ff.)
14. In French Baroque music, notes slurred in pairs imply inégales
It is not uncommon to hear recorded performances played in this way. This is unfortunate, because the sources attests that slurred pairs—on any instrument not just keyboards—always implied égales. (PBH pp. 61)
15. In Rameau' music, there is evidence of slurred pairs being inégales
Related to the previous Myth, this one is also—demonstrably—a misconception, arising from misleading comparisons between Rameau's harpsichord works and the orchestral versions he composed in much later times. (PBH pp. 67-73)
16. In German Baroque pieces—including J.S. Bach—when duplets coexisted with triplets in different voices, the duplets were always assimilated to the triplets
This was indeed a general rule, followed also by non-German Baroque composers, but it had important exceptions. Bach, in particular, left overwhelming evidence that he intended the duplets to be played as written, following the modern sesquialter fashion. (PBH pp. 75-78)
17. Notes inégales apply only to French music, not to Baroque music written in French style but outside France, even less so to J.S. Bach
The—admittedly scarce—evidence strongly suggests that both German and English musicians would indeed play inégales as matter-of-fact in their French-style works. It is most likely that Bach intended a significant number of the pieces in his Suites and Partitas to be performed with inégales. (PBH pp. 79-84)
Jean-Philippe Rameau, engraving made in 1762 after an earlier bust by Caffieri
MYTHS ON TEMPI FOR HARPSICHORD PIECES
18. Most harpsichord music was intended to be played with continuous and very ostensible rubato
While many well-respected modern performers refrain from these excesses, others play in such a way that all sense of rhythm is lost. Yet, let me quote:
"Baroque evidence for rubato is limited to a few suggestions in a few documents by a few authors for a few passages in a few types of pieces." (PBH pp. 89-91)
19. For most Baroque pieces, it is very difficult to establish, with any degree of certainty, the tempo intended by the composer
Even if the metronome only appeared in the Romantic era, at least three general principles have been used since antiquity to determine the most adequate speed for a piece (PBH pp. 91-92). Additional information is available for the many pieces that follow standard forms such as particular French Suite dances or Italian Concerto movements. (PBH pp. 95ff.)
20. Many of Bach's Goldberg Variations are virtuoso displays, obviously intended to be played as fast as possible
In present-day public performances of the Goldberg Variations, harpsichordists play most of them very fast. There are many reasons why they do this. Let me quote from PBH, p. 100:
"1. To show off what they can do, which, being harpsichord virtuosi, is precisely what they are expected to do.
2. They can do it because they had to develop a virtuoso technique in order to become professional harpsichordists in today’s very competitive world.
3. They can happily play the Goldbergs with modern fingerings and finger movements.
4. They have practised assiduously for a neat public performance avoiding wrong notes.
5. They are playing all the Variations in a single performance: slower speeds might turn the recital or recording into a lengthy affair for today’s busy world.
6. Quite a few in the audience, accustomed to the usual fast tempi, may find a slower performance boring. Some may even become sleepy:
after all, legend has it that the Variations were written to help a nobleman with insomnia!" (yet another myth ...)
It makes sense, doesn't it? Yet not a single one of the above reasons was valid in Bach's time! A thorough analysis and reconstruction of historically-likely tempi is now available. (PBH pp. 100-103)
Wittner wooden metronome, c.1970
MYTHS ON HARPSICHORD VOICING AND REGISTRATION
21. The upper 8' in two-manual instruments was voiced much softer than the lower 8' stop: the upper 8' was meant for pianissimo or echo effects only
This practice is not uncommon nowadays, but is unwarranted: it runs counter to some concrete facts and significant indirect evidence. (PBH p. 116)
22. In instruments having on the lower—or only—manual the disposition 8' and 4', the latter should be voiced very much softer than the 8'
There is no historical evidence for this common modern practice, which prevents the use of the 4' as a solo stop. Yet this was specifically mentioned as a possibility in French sources. A very soft 4' has also the disadvantage of not having any audible effect when added to the two 8' stops for a fortissimo. (PBH p. 116)
23. The buff stop provides an ugly pizzicato effect and was almost never used
Modern recordings hardly feature the buff stop at all. Yet Flemish harpsichords were very popular throughout Europe in early and middle Baroque times, and they had buff stops. Most Ruckers harpsichords had their buff battens divided in independent bass and a treble sections: this shows that the buff was considered an important feature. Unfortunately, in most antique harpsichords, the buff pads have not survived, and we lack any information about their material, size and position. A buff stop can sound very blunt, like a pizzicato violin, when the pads are soft, large and not near enough to the nut. However, small leather pads located very near to the nut pins produce a distinctive and beautiful lute-like sound. (PBH p. 117)
24. Registration was deemed irrelevant in the harpsichord era, therefore we should follow that philosophy
This belief may be at the root of so many modern-day recitals in which most pieces are played on the 2x8' combination, even in doubles that could provide a significant variety of sounds with the different combinations of their three choirs plus the buff. Ancient documents attest that the variety of stops and their combinations was considered a very positive feature of harpsichords. It is true that most music scores had no registration directions, but this was so because they would discourage many prospective buyers, owners of simpler instruments with hardly any possibility of stop changes. However, quite a few leading composers included in their harpsichord music directions about loudness or manuals, thus showing their interest in the expressive possibilities available via registration and/or the use of the two manuals in doubles. (PBH pp. 120 ff.)
25. In ancient times double harpsichords were as a rule played with coupled manuals—2x8'—and often with the 4' engaged as well
The evidence claimed for this consists of a few indirect statements that are unlikely to have the purported meaning at all. The implications have been fully investigated. (PBH p. 121)
26. In ancient times, harpsichordists would play as a rule on single stops
This is the opposite of the previous Myth, and some evidence has been show for it as well. However, all things considered, I seems most likely that there was no "default" registration. Harpsichords with two or three movable stops would often be played either with a solo stop or with a combination. This allowed getting different effects and loudnesses, which are indeed asked for in 18th century sources. (PBH p. 124 ff.)
27. Registration indications are so scarce in the harpsichord literature, that there is no hope to formulate rules about the authentic use of harpsichord stops
It can be shown that J.S. Bach used contrasting registration indications ("piano" and "forte") for six clearly differentiated musical effects (PBH pp. 128ff.). Many other authors prescribed some of those effects as well. Surely, only a small percentage of the harpsichord pieces bear registration indications: however, they were written by leading composers and allow—via a thorough analysis and comparisons—reconstructing a very complete set of registration rules for the different Baroque harpsichord models. (PBH 120 ff.)
Title page of Bach's Goldberg Variations, 1741
MYTHS ON REPERTOIRE ISSUES: HARPSICHORD VS. OTHER KEYBOARDS
28. Renaissance and Early Baroque keyboard music was not specific for any keyboard
Some was and some was not. For centuries, clavichords were small portable instruments, and the music played on them would often have been published for either the organ or the harpsichord. Music published for "organ" often included pieces with no pedals, some of them with tell-tale signs of having been conceived for a stringed keyboard. Vice versa, some music published for "keyboard" included pieces that required a pedal organ. Even in works where both harpsichord and organ are mentioned (e.g. Frescobaldi) it is often possible to identify some pieces as obviously written for the organ, and some others obviously conceived for the harpsichord. (PBH pp. 145ff.)
29. At least we all agree that J.S. Bach's WTC was all conceived for the harpsichord (or was it the clavichord, or the organ?)
The WTC score was written to be compatible with any type of keyboard instrument. Attempts to demonstrate that the WTC was primarily conceived for a particular instrument have been shown to incur in unavoidable contradictions. It is possible, however, to identify clearly some pieces of the WTC vol. I as conceived for the harpsichord, and some others for the organ. The same applies to the much later WTC vol. II, where quite a few pieces show the tell-tale marks of music composed for the then-fashionable unfretted clavichord. (PBH p. 151)
30. Bach wrote the Toccata and Fugue in d minor BWV565 for organ, and there is no proof to the contrary
Williams and Humphreys in their well-known papers of 1981 and 1982 showed that the hypothesis of Bach's authorship leads to factual contradictions. They did not claim to have proved BWV565 to be a transcription and spurious. Nevertheless, their arguments can be shown to prove conclusively that the chances of Bach being the author are negligible. It is even possible to state with relative certainty the true author, the approximate date of composition and that the work was originally conceived for the harpsichord or solo violin. (PBH pp. 152-153)
USE OF THIS TEXT, TERMS AND CONDITIONS
Quoting text: Copying text from this webpage and inserting it in other publications or webpages is automatically allowed provided the following conditions are met:
a) As customary, the text copied must be included between quotation marks and left unmodified except for deletions. The latter should not change the meaning in any way and should be identified with ellipses.
b) The source must be quoted as "Myth NN from Claudio Di Veroli, 30 Myths on Baroque Harpsichord Playing, Bray Baroque, Bray, Ireland 2011. http://play.braybaroque.ie/mythsplay.htm".
c) The author should be immediately notified by email: see http:/play.braybaroque.ie for details.
Quotes/copies not adhering to the conditions indicated above will be considered in violation of the author's copyright and legal action may follow.
Disclaimer: The assertions made here about persons and writings are based on evidence from duly acknowledged published material, current at February 2011. The author will not accept responsibility from any offence or damage caused by reading this webpage: its contents is read and used at the user's own risk. For any issues re the above, please feel free to email the author: see http:/play.braybaroque.ie for details.