Abstossen, Schleifen and das algemeine Fortgehen – thoughts on clavichord touch

Paul Simmonds, Brighton, September 1997

When making a study of a discipline from a past era it is important to start with clearly established  facts from which any developments or divergences can be clearly related. Allowing oneself to commence with an assumption will only lead to unexplainable illogicalities as ones study progresses. For example, when performing old music, I think we would all here accept that the ideal tool, in our case the clavichord, would be either an old instrument in as near original condition as possible, or a close copy. Only then can we be as sure as we can that the instrument is responding as a player of the past would expect it to. If , for whatever reason, a present day builder alters something or redesigns a clavichord without fully realising the implications of what he or she is doing this could have far-reaching consequences for the player interpreting the music.

For at least 150 years the seamless legato has been the basic assumption when performing a series of notes from a printed page, unless staccato or other marks of articulation indicate something else. So that we are all quite clear what is meant by the term legato I quote Robert Donnington’s definition in the 1994 printing of the New Grove: ‘(It.: ‘bound’; Fr. lié; Ger. gebunden). A term meaning connected smoothly , with neither a perceptible break in the sound nor (ordinarily) special emphasis (the antonym of Staccato).’[i]

I am of the opinion that legato assumed as a basic touch for music of the 18th century and earlier deprives the music of much that is important or indeed vital, and makes the interpretation of some signs and indications difficult if not nonsensical. My purpose in this talk is to demonstrate that the assumed basic progression from one note to the other should be one of slight separation, with each note released momentarily before the next note is sounded. Although our primary concern here is the clavichord, the arguments will be seen to apply to all keyboard instruments and indeed, to greater or lesser extent, to other instruments and the human voice as well.

Often quoted is J.S. Bach’s introduction in the 1723 manuscript of the Inventions and Sinfonias where he says they are ‘not only to obtain good ideas but to develop them well; above all, however, to achieve a cantabile style in playing’ (‘am allermeisten … eine kantabile Art im Spielen zu erlangen’). A number of people have taken this to be a clear endorsement of legato playing as the norm. One only has to look at other sources from Bach’s time and earlier to realise that this was not the meaning of cantabile.

Isaac Vossius writing in 1673 says: ‘The musicians learned a three-fold movement of the voice: an unbroken progression, a decidedly separated style, and one which takes the middle ground between these two. The unbroken manner of delivery is like someone stringing words together without any noticeable gaps. It would be unpleasant if a speaker presented each word separately like drops. The voice of a singer should observe the spaces in between, not only with regard to the notes, but also the metre, so that every section and joint can be distinguished.’[ii]This is my translation of a German version by Forkel, 1779.

The 17th century musician J. van der Elst writes about performing vocal polyphony in a resonant building: ‘When several notes are set to a single syllable, each must be articulated distinctly and with attention to detail, lest to anyone hearing them at a distance they seem blurred …. Melodies sung in a clear style, with this well defined separation of the notes, reach the ear of the listener some way away in a smooth distinct flow’.[iii]

Daniel Speer, in 1697, gives technical advice to singers as follows: ‘When fast passages occur in a song they shouldn’t just happen by breathing out, but should be clearly expressed through a clear articulation’.[iv]

That legato and staccato were considered in the realms of ornamentation is clear from an anonymous : Advice to the composers and performers of vocal music,[v]‘The binding together or stringing the notes firm and distinct with the voice, which the Italians express by the terms (Legareand Staccare la Voce) are graces equally agreeable, although contrary to each other; and nothing but good judgement can direct the singer how to use them properly according to the nature and design of the Composition’.

The ideal for instrumentalists was to imitate the human voice. In his ‘Essay’ C.P.E. Bach advises us strongly to ‘listen to accomplished musicians … Above all, lose no opportunity to hear artistic singing. In doing so, the keyboardist will learn to think in terms of song’.[vi]

The cantabilein instrumental terms is described by Heinrich Christoph Koch in his Musikalisches Lexikon. He describes three types of bowing. About the second, which he distinguishes from detached and legato (joined), he says: ‘With this type of bowing all notes in the cantabile (Gesange) are played, which should be neither joined together nor detached’.[vii]

What interests us primarily are, of course, what the surviving keyboard tutors say on the subject. One of the most detailed is also one of the earliest, namely, ‘The elegant art of playing at the keyboard” by Fray Tomas de Santa Maria, published in 1565. Under the heading ‘on playing cleanly and distinctly’ he writes: ‘.. as the fingers strike the keys, the finger which plays first should be raised before the one immediately following it plays, both ascending and descending. Always proceed in this way, for otherwise one finger would catch up with the next, and when one finger catches up it follows that one note overlaps and obscures the next’[viii]. This is the translation by Barry Ife and Barbara Sachs published in 1981. In an earlier German edition the Spanish word  atapar has been translated as ‘to tread on the heels of’[ix]. I understand this is a very free paraphrase, but which conveys the meaning well.

Girolamo Diruta in his Il Transylvano(1593) agrees with this. He writes: ‘Let the fingers play cleanly, that is, not pressing a key down before the finger is lifted from the previous one, moving up and down at the same time.’[x]

Johann David Heinichen’s Der generalbaß in der Kompositionof 1728 gives us the following aside: ‘By the way, one should give considerable care to play all passages cleanly and distinctly without joining the notes (ohne schleiffigtes Wesen), because herein lies the key to brilliant playing, which technique is better heard than described’[xi].

I am often frustrated by the imprecise way C.P.E. Bach expresses himself in his ‘Essay’, and in this matter he runs true to form. In the chapter on performance he writes: ‘There are many who play stickily, as if they had glue between their fingers. Their touch is lethargic; they hold notes too long. Others, in an attempt to correct this, leave the keys too soon, as if they burned. Both are wrong. Midway between these extremes is best. Here again I speak in general, for every kind of touch has its use.’[xii]A few paragraphs further on he writes the following: ‘Notes which are neither detached, connected nor fully held are sounded for half their value, unless the abbreviation tenis written over them, in which case they must be held fully. These sort of notes are usually the quarters and eighth notes in moderate and slow measures.’[xiii] Bach is clearly defining three separate touches, legato, detached and one which is neither of these and which he implies is used, as it were, by default when nothing else is indicated. By defining this as half the written value he reduces the contrast between this touch and the détaché, unless one takes very seriously the technique for playing détachédescribed by Ernst Wilhelm Wolf in his Anleitungof 1785, which would give an added aural dimension to what would otherwise be just a shortening of the sounding length of a note. I quote this description in full, as it is specifically for the clavichord and is as precise as Bach is vague: ‘The best way of playing this type of figure at the clavichord, one which sufficiently differentiates the détachéfrom the normal style of articulation, and one which produces the best tone, is this: one strikes the key with a stiff finger (as when playing a syncopation), and then immediately withdraws the finger back towards the player so that it slides off the front, and the key quickly springs back up. The tone, when thus struck on good clavichords sounds rather as if the consonants ‘t’nt!’ were sounding along with it; this ‘t’nt!’ has a better effect than the ‘t’t’ one gets when the finger releases the key without the slide-off’[xiv]. Note Wolf’s use of the term ‘ the normal style of articulation’. Towards the end of his Anleitung Wolf refers to ‘..the natural touch, neither détachénor slurred, but rather clear and articulate.’[xv]What Bach is referring to has more in common with what Wolf describes as follows: ‘Another kind of détachéoccurs in an accompanying bass line, either in andante when the bass moves in simple eighth-notes, or in allegro when it moves in quarter notes, e.g.:

Example 1:


These notes are sounded for only half their written length, with the finger being retracted from the key as in the détaché, without sounding contrived (unless the bass itself is a singable melody, in which case there would be no détachéor slide-off)’[xvi].

I am not alone in finding Bach’s description confusing. Daniel Gottlob Türk takes him to task on precisely this point in his Clavierschuleof 1789. He quotes the passage and writes: ‘Taken in general this kind of playing does not seem to me to be the best. For (1) the character of a composition necessitates a variety of restrictions in this respect; (2) the distinction between the tone which is usually detached and that which is played in the customary manner is practically abolished; and (3) the execution would probably become too short (choppy) if every note not slurred was held for only half its value and the second half would be a rest.’[xvii] However one chooses to resolve this particular argument, it is increasingly clear that for them all a touch existed between détaché and legato, what Türk describes as the ‘customary manner, that is, neither detached nor slurred’[xviii].

A clear description is given by Friedrich Wilhelm Marpurg in his Anleitung zum Clavierspielen, second edition 1765: ‘The normal way of playing (ordentliche Fortgehen) is in contrast to legato as well as the détachéway of playing, and means that, just before one plays the next note, one lifts the finger from the preceding note. This normal way of playing, because it is always assumed, is never indicated’[xix].

So how much is ‘just before’ the next note? I am of the opinion that the degree is unnotatable, but at least two people have tried. The writer Christian Gottlieb Tubel gives the following example when explaining the equality of triplets[xx]:

Example 2:


The fact that he doesn’t see the necessity of explaining the rests suggests to me that he takes this way of playing for granted. Returning to that widely known and little read source, the Clavierschuleby Türk, we find the following example with the accompanying explanation: ‘For notes which are to be played in the customary fashion (that is, neither detached nor slurred) the finger is lifted a little earlier from the key than is required by the duration of the note. Consequently the notes in aare played approximately as in bor c, depending on the circumstances. If there are some notes intermingled which are should be held out for their full value, then ten. or tenutois written over them.[xxi]

Example 3:


Here we have not only valuable information as to the degree of detachment, but the indication tenuto illustrated in its correct context. If legato were the assumed normal touch, this sign would be redundant.

So much for the historical sources.

After attempting to digest all this in theory I approached my keyboard instruments with the following questions in my mind: (a) What actually happens aurally when one attempts this ‘ordinary touch’ on, for example, a good reproduction clavichord, (b) what does it do for the music which a legato touch doesn’t achieve and (c) could one formulate some ideas as to when this style of touch ceased to be advocated and why.

If one were to depict the volume of a struck or plucked note graphically, the moment of attack would indicate a high point on the graph. This is as we would expect. What is interesting to analyse is the ending of a note. When the finger is released from the key the note doesn’t just stop, but has an audible finish to it. On a harpsichord this is caused by the jack with its damper falling back onto the string. On a clavichord with historical string tensions the moment the tangent leaves the string and the listing comes into play there is a noticeable emphasis given to the end of the note. On a lightly strung ‘modern’ clavichord this ending is considerably less noticeable. On the organ this marking occurs as the palette shuts. Obviously this ending would not be so clearly defined on our graph as the beginning of the note, but it is audible. It is however not audible if a succession of notes is played using legato touch, for then the ending of the first note coincides with the start of the second, the result being that one hears only beginnings and no ends of notes. A comparable effect occurs in speech when, for example, two words, one ending in ‘t’ and the other starting with ‘t’ are so enunciated that one ‘t’ serves both functions, as in ‘fat toad’ or, another example with ‘g’s, ‘big gate’. May I remind you at this point of the many analogies in early writings of music with speech and the constant emphasis on clarity. The word ‘articulation’ is an obvious association with speech.

So what about the degree of separation? How big should it be? As we’ve seen Türk for one suggests that this will vary with circumstances. External acoustic is an obvious factor, particularly in a resonant church. Another factor is the resonance of the instrument itself. The damping on a harpsichord or clavichord is never so efficient that the sound of a note is cut off dead at the point of release. There is always a second or so’s after-bloom to bridge the gap between one note and another, so that the separation may not even be noticed by a listener. What should be noticeable, and I’m convinced that it is, is the difference between a passage played legato and the same passage played with the ‘normal way of playing’ of the 18th century and before. I am also convinced, by the way, that this way of playing did not come about as a result of the problem of playing legato on a fretted clavichord. Most of the sources are not for the clavichord exclusively. I have already mentioned the clarification given to the term tenuto if it is considered in the context of this touch. Other musical benefits are the sense given to the few legato markings we do have which would lose considerable impact if everything were played legato. Passages like the following can become indistinguishable from one another when legato is assumed, but retain their individual meanings when the ‘normal touch’ is used :

Example 4:


(Ornamentation clusters as a rule tend to be played legato, as they are a comment on one note)

If after all this, I may be allowed to assume that this way of playing was the norm, it only remains for me to speculate when and why it ceased to be relevant, and gave way to legato, the touch accepted as default today. An examination of the works and writings of Rameau and François Couperin suggest to me that legato was being advocated in France quite early in the 18th century. Georg Friedrich Wolf advocates it in his Klavierschuleof 1789[xxii].  There is a particularly interesting comment in the German translation of Tosi’s singing tutor made by the Bach student Johann Friedrich Agricola, published some 35 years after the original in 1757, where he laments ‘the mistake of joining notes which should be separated, which is particularly prevalent in Italy (Wälschland) and singers of the newest school’[xxiii]. Changes of this sort do not occur immediately but I think we can conclude that the gradual ascendancy of legato as the primary touch coincided with the marked changes in musical aesthetics towards the end of the 18th century and the falling out of favour of many instruments associated with the older style, including the harpsichord and clavichord.

If the recording industry had come into being a hundred years earlier than it did much of what is still argued about today would have more substantial, if not irrefutable, aural evidence to back it up. The closest we have to the document of a performance are certain mechanical musical instruments of the 18th century. I would have liked to draw on the apparently extremely exact engravings of Dom Bedos dealing with the manufacture of barrel organs, but I was unable to have a look at the souce. However, in a valuable book on keyboard articulation the author, Ludger Lohmann, claims that the minute separations of notes, the ‘normal touch’, are indicated in the descriptions for the construction of the barrels and in a notated piece apparently supervised by the keyboard player Claude Balbastre[xxiv]. We are fortunate that an important barrel organ from the 18th century has survived and is now housed in the Colt collection in Kent, England. The organ was made by an English organ maker known only by his second name, Holland. The instrument’s surviving sixteen barrels of music were all notated by the German born John Christopher Smith the younger (1712 – 1795). Smith was a pupil of Handel’s and his successor as organist at the Foundling’s Hospital. Smith was very close to Handel in his last years, and helped him with the notation of his works when Handel was suffering from the increasing onset of blindness. The barrels contain a number of works by Handel himself, and, in my opinion, provide us with an invaluable link to the composer’s performance style, with particular regard to touch and ornamentation. Again, I would like to have examined the barrels myself before giving this talk, but instead I would like to end by playing you a recording of the organ in action. It was an enlightened decision of the record company Erato to issue, in 1986, a disc featuring music from this organ[xxv], and from this I have selected Let the Bright Seraphimfrom Handel’s Samson. I would ask you to listen carefully, particularly in the passages where what would be the right hand is playing alone and to the bass line. Apart from the degree of additional ornamentation, which may sound excessive to out ears but which ties in perfectly with various studies on this subject, the algemeines Fortgehenis audible, highlighting the groups of notes which are played legato and in contrast to thedétachéwhich one hears in the trumpet like repeated notes. I should add that this approach to touch is the same on every piece on this disc, irrespective of tempo or mood.




In this talk I am attempting to demonstrate that a slightly separated touch was the norm in keyboard music prior to the 19th century, as opposed to legato which today is accepted as default. Various historical sources, ranging from Santa Maria and Diruta in the 16th. century to C.P.E. Bach. D.G. Türk, F.W. Marpurg and E.W. Wolf in the 18th century can be cited to show that three touches were distinguished; legato, (Schleifen), detached (détaché) and what is described as the normal touch or ‘algemeines Fortgehen’, which was never indicated in the music as it was always assumed. This way of playing was also associated with the term ‘cantabile’, as referred to by J.S. Bach in his Inventions and Sinfonias. The meaning of this word changed, in turn, during the course of the 19th century. The sources disagree on the degree of detachment. C.P.E. Bach seems to suggest as much as half the value of the written note, but this is disputed by Türk, who points out that this would makedétachéand the ‘normal touch’ indistinguishable from one another. After giving the historical theoretical evidence for this touch, I turned to the clavichord with the questions (a) what happens aurally when this touch is applied on the clavichord, harpsichord or organ, (b) what effect does this have on the music which is not achieved by legato, and (c) when did this style of playing die out. I draw attention to the sound made by all early keyboard instruments when a note is released, and compare it with a final consonant in a word in speech. When the latter is liased with an identical consonant in a subsequent word one consonant serves as ending and beginning – the musical equivalent of this is legato. I emphasise the numerous associations of early music with speech, and the insistence of the early writings on clarity. Using the ‘normal touch’ each note is heard to have a beginning and an end, although the overall effect should not be one of detachment. Passages such as the two extracts in example 4 would, and often do, sound identical if legato is employed as the basic touch, but are clearly differentiated if the ‘normal touch’ is accepted (appoggiaturas, as ornaments, would be played legato). As to when this basic touch gave way to legato, the evidence would point to a gradual change throughout the 18th century with legato superseding the ‘normal touch’ with the stylistic changes in music at the start of the 19th century. I concluded my presentation by playing a recording of an 18th century barrel organ, now housed in the Colt Collection in England. The preparation of the barrels was supervised by John Christopher Smith, a pupil and assistant to Handel in the latter’s last years, and displays considerable attention to detail. I maintain this is nearest we come to direct aural evidence from the 18th century of the ‘normal touch’, or ‘algemeines Fortgehen’.


[i]  Robert Donnington “Legato” in The New Grove Dictionary of Music and Musicians, ed. Stanley Sadie (Macmillan, 1994) vol. 10, pg. 610.

[ii] “Daher lehrten auch die alten Tonkünstler eine dreyfache Bewegung der Stimme: eine ununterbrochen fortgehende; eine merklich abgetheilte; und eine, welche zwischen diesen beyden das Mittel hält. Ununterbrochen fortgehend, ist die Stimme eines Redenten, wo die Worte so aneinander hängen, daß man keinen merklichen Zwischenraum hört. Denn es würde sehr unangenehm seyn, dem Zuhörer die Wörte einzeln zuzählen, und gleichsam tropfenweise reden zu wollen. Die Stimme eines Singenden muß hingegen Zwischenräume beobachten, nicht nur in Ansehung der Töne, sondern auch des Zeitmaaßes, so, daß alle Füße und Glieder derselben unterschieden werden können.” Isaac Vossius, De poematum cantu et viribus rhythmi, Oxford, 1673, quoted in Ludger Lohmann, Studien zu Artikulationsproblemen bei den Tasteninstrumenten des 16. – 18. Jahrhunderts, Regensburg: Gustav Bosse Verlag, 1986

[iii]  J. van der Elst, Notae Augustianae , Gent, 1657, quoted in Thurston Dart, The Interpretation of Music, London: Hutchinson & Co, 1967.

[iv] “So fern geschwinde Läuffe in den Noten bey einem Gesang sich ereigen sollen solche micht bloß durch des Atems hauchen sondern durch deutlichers anschlagen des Athems soll jeder Note Thon exprimirt werden.” Daniel Speer, Grund-richtiger Kurtz-Leicht und Nöthiger jetzt Wol-vermehrter Unterricht der Musicalischen Kunst, Oder Vierfaches Musicalisches Kleeblatt, Ulm 1697, quoted in Lohmann (my translation).

[v]  Anon, Italian, London 1727, p. 14, quoted in R. Donnington, The Interpretation of Early Music, London: Faber, 1963, p.412

[vi] “Wir fügen allhier noch hinzu, daß man keine Gelegenheit veransäumen müsse, geschickte Sänger besonderszu hören; Man lernet dadurch singend denken..” C.P.E. Bach, Essay on the true Art of Playing Keyboard Instruments, trans. William Mitchell, London: Eulenburg, 1974, pg. 151

[vii] “Mit diesem Bogenstrich werden alle Noten im cantabeln Gesange vorgetragen, die weder zusammengeschleift, noch abgestoßen werden sollen.” Heinrich Christoph Koch, Musikalisches Lexikon, Offenbach 1802, p. 265, quoted in Lohmann (my translation).

[viii] “La primera y principal, es que al herir de los dedos en las teclas, siempre el dedo q hiriere primero se leuante antes que hiera el otro que immediatamente se siguiere tras el, assi al subir como al baxar, Y desta manera siempre procediendo, porque de otra manera se alcançaria vn dedo a otro, Y de alcançarse  vn dedo a otro, se sigue alcançarse y ataparse vna boz a otra se sigue yr fuzio y estropajoso lo q se tañe, y no lleuar lipieza ni distinctiõ de bozes.” Fray Tomás de Santa Maria, Arte de Tañer fantasia, Valladolid 1565, fol. 36b and 38b, inAnthology of Early Keyboard Methods, trans. Barbara Sachs and Barry Ife, Cambridge: Gamut Publications, 1981

[ix]Otto Kinderley, Orgel und Klavier in der Musik des 16. Jahrhunderts, Leipzig 1910.

[x]“… che le dita spiccano bene li Tasti, cioè, che non si batta l’altro Tasto per insino, che non è leuato ol dito dall’altro. & che à vn’medesmo tempo si lieuano, e pongano.” Girolamo Diruta, Il Transylvano, 1593, fol. 8a., English translation by Sachs and Ife, op. cit..

[xi] “Ubrigens befleißige man sich, alle Passaggien rein, distinct, und ohne schleiffigtes Wesen anzuschlagen, weil hierinnen ein grosses Brillant des Spielens stecket, welches sich besser hören, als beschreiben lässet”.J. D. Heinichen, Der Generalbaß in der Komposition, Dresden 1728, facs. Hildesheim: Georg Olms Verlag, 1994, p. 552, my translation.

[xii] “Einige Personen spielen klebericht, als wenn sie Leim zwischen den Fingern hätten. Ihr Anschlag ist zu lang, indem sie die Noten über die Zeit liegen lassen. Andere haben es verbessern wollen, und spielen zu kurz; als wenn die Tasten glühend wären. Es thut aber auch schlecht. Die Mittelstraße ist die beste; ich rede hiervon überhaupt; alle Arten des Anschlages sind zur rechten Zeit gut.” Op. Cit., p.149

[xiii] “Die Noten, welche weder gestossen noch geschleifft noch ausgehalten werden, unterhält man so lange als ihre Hälfte beträgt; es sey denn, daß das Wörtlein Ten: (gehalten) darüber steht, in welchem Falle man sie aushalten muß. Diese Art Noten sind gemeiniglich die Achttheile und Viertheile in gemäßigter und langsamer Zeit-Maasse …” Op. Cit., p.157

[xiv]Das Mittel, welches man auf dem Claviere bey Ausübung dieser Art melodischer Figuren anwendet, um das Abstossen von dem natürlichen Anschlage der Tasten hinlänglich zu unterscheiden, und auch zugleich den besten Ton herfürzubringen, ist dieses: man schlägt die Taste, wie bey der Syncope, mit straffen Finger an, und zieht sogleich denselben wieder davon ab, sodaß er vorwärts, das heist, nach dem Clavierspieler zu, von derselben abglitscht, wovon die Taste forne schnell in die Höhe springt. Von solchem Anschlage klingt auf guten Klavieren der Ton ohngefehr, als wenn die Mitlauter, t’nt! mit hineingefungen würden; und dieses t’nt! thut bessere Würkung, als das t’t, welches man erhält, wenn der Finger ohne Abglitchen von der Taste genommen wird.” E. W. Wolf, Anleitung zum guten Vortrag beym Clavierspielen, Leipzig 1785, English translation by Christopher Hogwood in C.P.E. Bach Studies, ed. Stephen L. Clark, Oxford: Clarendon, 1988, p. 147.

[xv]“ die Passagen (werden) durch den natürlichen Anschlag, der weder Stossen noch Schleifen, sondern Deutlichkeit und richtige Articulation der Töne zum Grunde hat, behandelt.” Op. Cit., p.153.

[xvi] “Noch eine Art von Abstossen kömmt beym Accompagnement im Basse vor, wenn sich nehmlich beym Andante der Baß blos in Achtheilen, oder beym Allegro in Viertheilen bewegt, als: (my Ex. 1) so werden die Noten so vorgetragen, daß nur die Hälfte ihrer Dauer fortklinget, und bey der andern Hälfte wird der Finger nach Art des Anstossens, (aber alles ohne Affectation,) von der Taste abgezogen, es sey dann, daß im Basse selbst eine gesangmäßige Stelle vorkäme, wobey dergleichen Abziehen, oder Abglitchen der Finger nicht Statt findet.” Op. Cit., p.147.

[xvii] “Allein im Ganzen genommen scheint mir diese Spielart doch nicht die beste zu seyn. Den 1) macht der Charakter eines Tonstückes hierbey verschiedene Einschränkungen nothwendig; 2) würde dadurch der Unterschied zwischen den wirklich abzustoßenden und nur auf die gewöhnliche Art zu spielenden Noten beynahe ganz aufgehoben; 3) möchte der Vortrag doch wohl zu kurz (hackend) werden, wenn man jeden nicht zu schleifenden ec. Ton nur die Hälfte seiner Dauer aushielte, und folglich die zweyte Hälfte pausierte…” Daniel Gottlob Türk, Klavierschule, Leipzig/Halle 1789, English translation by Raymond Haggh, School of Clavier Playing, University of Nebraska Press, London/Lincoln 1982, pp.345-346.

[xviii] “… die gewöhnliche Art d.h. weder gestoßen noch geschleift …” Op. Cit., p.345.

[xix] “Sowohl dem schleiffen als Abstossen ist das ordentliche Fortgehen entgegen gesetzet, welches darinnen besteht, daß man ganz hurtig kurz vorher, ehe man die folgende Note berühret, den Finger von der vorhergehenden Taste aufhebet. Dieses ordentliche Fortgehen wird, weil es allezeit vorausgesetzet wird, niemahls angezeiget.” Friedrich Wilhelm Marpurg, Anleitung zum Clavierspielen, Berlin 1755, 2nd. Edition 1765, quoted in Lohmann, op. cit., p.246, my translation.

[xx] Christian Gottlieb Tubel, Kurzer Unterricht von der Musik, Amsterdam 1766, quoted in Lohmann, op. cit., p. 246.

[xxi] “Bey den Tönen, welche auf die gewöhnliche Art d.h. weder gestoßen noch geschleift, vorgetragen werden sollen, hebt man den Finger ein wenig früher, als es die Dauer der Note erfordert, von den Tasten. Folglich werden die Noten bey a) nach Umständen ungefähr wie bey b) oder c) gespielt. Sollen einzelne untermischte Töne völlig ausgehalten werden, so schreibt man ten.oder tenutoüber die Noten.” Op. Cit., p.345.

[xxii] Georg Friedrich Wolf, Kurzer aber deutlicher Unterricht im Klavierspielen, Halle 1789, p.21.

[xxiii] “Der Fehler des Schleifens der Passagien, wo man sie stoßen sollte, ist ansonderlich in Wälschland, zu itzigen Zeiten, bey vielen Sängern aus den neuesten Schulen sehr eingerissen.” Pierfrancesco Tosi, Anleitung zur Singkunst, trans. from Opinioni di cantori antichi, Bologna 1723, by Johann Friederich Agricola, Berlin 1757, quoted in Lohmann, op. cit., p.237

[xxiv] Ludger Lohmann, Studien zu Artikulationsproblemen bei den Tasteninstrumenten des 16. – 18. Jahrhunderts, Regensburg: Gustav Bosse Verlag, 1986, pp. 232 ff.

[xxv] Haendel, un enregistrement d’epoche, Olivier Roux, orgue méchanique, Erato ERA 9274 (1986), regrettably now deleted.