The nature of Bachs Italian Concerto BWV 971

Federico Garcia
March 2004

Introduction
The genre of concerto was favored by Bach throughout his life. From the transcriptions of Weimar
to those of Leipzig (the former from other composers concertos, the latter of his own), he presents
us with a development of the concertante forms that both culminate and surpass the Baroque
tradition: with his use of motivic counterpoint in the orchestral (accompanying) parts, the ubiquity of ritornello material, textural variety within the ritornello, etc., he challenges the Baroque
conception of the genre, based as it was on the strong oppositions of three basic dichotomies:
ripieno/concertino, tutti/soli, and ritornello/episodes.
Seen in this context, the Italian Concerto emerges as an intriguing exception. All that raises
admiration in Bachs orchestral concertosthe carrying forward of the genre, to give it a name
is wanting here. The tale of historical progress, in which the Brandenburgs and their orchestral
relatives fit so well, all but fails to account for the Italian Concerto in its earthly, unassuming,
direct prosaism.
Perhaps this explains the scarcity of historical/critical discussion on the piece, and, more interestingly, the curious dismissive attitude that emerges when it is addressed. Laurence Dreyfus, for
example, writes in Bach and the Patterns of Invention that the first movement,
despite its size and form pointing to a real concerto movement, does not exploit the ritornello segments as do Bachs most advanced works in the genre. For example. . . , the ritornello formations . . .
are strikingly more primitive than those of the Echo [of the French Overture]. With its two identical
framing ritornellos, the piece seems positively disinterested in exploring the substance and sense of
its segmentation. . . . The conventional recasting of the given materials . . . never takes place. The

lengthy ritornello itself evokes some aspects of the concerto style of Bachs younger contemporaries,
such as Graun, Hasse, and even Scheibe himself, who, proclaiming the slongan of natural melody,
were indifferent to the capability of music to rethink its own materials.1

More direct is David Schulenberg in The Keyboard Music of J. S. Bach:

While one would never want to underestimate a work of Bachs, it is possible that the novelty
of the two pieces [in the Clavier-bung II] exceeds their intrinsic musical worth. . . . There are even
signs of a certain carelessness in the production of the volume, for the indications for the use of
two manualsone of the special features of the publication, announced on the title-pageappear
to have been a late addition and are not always fully integrated into the musical fabric.2

Both Dreyfus and Schulenberg further discredit the work by quoting Johann Adolph Scheibes
famous praise (which I myself, though to different ends, quote below, see page 4). Dreyfus is suspicious of Scheibes applause, interpreting it (with some reason) as a half-hearted flattering aiming at the advance [of Scheibes] own program which the Italian Concerto superficially seemed to
support; in the scholars view, Bach had intended to accommodate himself to his audience.3 In
the same vein, Schulenberg states that the concerto, shorter and in some respects more up-to-date
. . . reflects not so much the style of the early Venetian concertos . . . , nor Bachs own orchestral
concertos . . . , as it does the later works of Vivaldi and the concertos of Bachs younger German
contemporaries like Quantz and the Graun brothers.4 Finding the piece strikingly uninteresting,
Dreyfus and Schulenberg coincide in blaming Bachs (clearly inferior) audience. Gregory Butler
offers an alternative explanation, not totally unrelated, but still distinct. Never quite expressing
doubts about the quality of the work, Butler is nevertheless quick to explain away the pieces
simplicity, pointing to its alleged didactic nature: there is an almost formulaic structural simplicity about this work that sets it apart from its orchestral counterparts. . . . These works [the
Italian Concerto and the French Overture] seem to be nothing less than textbook examples of their
1

Laurence Dreyfus, Bach and the Patterns of Invention (Cambridge, Mass. & London: Harvard University Press,

1996), pp. 2301.

2

David Schulenberg, The Keyboard Music of J. S. Bach (New York: Schirmer Books, 1992), p. 301.

Dreyfus, op. cit., pp. 2234.

Schulenberg, op. cit., p. 302.

genres. . . .5
Of course, what merits comment is not so much the different attempts at justifying the piece,
but the need to justify it in the first place. Apparently, the Italian Concerto is somehow unworthy
of its author, and Bachs reputation has to be defended. (In a truly deconstructivist spirit, it would
be said that the piece, after all, not only is, but also has to be keptconstruedas exceptional.)
In any case, the general dismissal, in the form of either silence or apology, is understandable in
the light of the fact that, in 1735, having already composed The Well-Tempered Clavier, the Partitas, the Brandenburg Concertos, and other such milestones, Bach decides to collect and publish
this simple, superficial piece: it does look like a withdrawal from his most advanced works. But
of course it is our thorough knowledge of and informed admiration toward the latter that create that
illusion (cf. Robert Marshalls warning that his supreme, iconic stature renders us unwilling or
unable to assess him or his music critically6 ). We tend to forget that advance might not always
have beenwas it ever?Bachs primary concern.
In the case of the Italian Concerto, for example, Bach might have been driven by the very
interesting possibilities, the techniques, and the challenges, of playing the orchestra from the
keyboard. We know he had already done that, when at Weimar he reduced concertos by Italian
composers; we know too that the natural following stepthe transcription not of a piece but of
a genrewas being taken, as there was at the time a generalized interest in the unaccompanied
concerto;7 and, finally, we also perceive a similar interest in Bachs own (and earlier) French Overture. The present essay is an inquiry on the Italian Concerto from that perspective. Paying more
attention to the Weimar transcriptions than is usual in connection with the piece, and thus departing from the more common but misleading nonstop comparison to the orchestral genre, I propose
5

Gregory Butler, The Aesthetic and Pedagogical Context of Bachs Concerto nach Italienischem Gusto BWV 971,

Paper presented at symposium Bachs Musik fr Tasteninstrumente, Dortmund, 2002.

6

Robert L. Marshall, Toward a Twenty-First-Century Bach Biography, The Musical Quarterly 23 (2000), p. 500.

See for example Monika Willer, ,Sonderflle, ,Modeerscheinungen und das ,Normale: Konzertsatzkonzep-

tionen in beglieteten un unbegleiteten Klavierkonzerten vor 1750, in Reinmar Emans and Matthias Wendt (eds.),
Beitrge zur Geeschichte des Konzerts: Festschrift Siegfried Kross zum 60. Geburstag (Bonn: Gudrun Schrder Verlag, 1990).

a history of the conception of the piecea biography of the piece, as it weresupported by

archival and, I claim, musical evidence. If I am correct, Bach composed and abandoned the first
movement fairly early on; later, he was reminded of it by his unpublished French Overture, orchestral in character, and then he proceeded to revise and complete with two more movements. The
third movement, especially, shows the mature Bach at work, as it is a compendium of transcriptional techniques: in fact, when the piece is seen from the perspective of the transcriptions, in my
view, it reveals itself as eminently Bachianeverything but an exception to the synthesis that he
means for the history of music.

The literature on the Italian Concertothe piece vs. the orchestral model
From Scheibe to Schulenberg
It is obvious that the Italian Concerto is related to the orchestral genre of concerto in a very
essential way. The first document of the reception of the piecethe mentioned review by Johann Adolph Scheibeis indeed little more than a definition of an unaccompanied concerto in
terms of the orchestral model, followed by the claim that Bach has succeeded at realizing it (my
italics):
Finally I must briefly mention that concertos are also written for one instrument alone, without
any accompaniment by othersespecially clavier concertos or lute concertos. In such pieces the
basic structure is kept the same as in concertos for many instruments. The bass and the middle
voices, which are added now and then to fill out the texture, must represent the subordinate parts.
And those passages which above all form the essence of the concerto must be most clearly differentiated from the rest. This can very well be done if after the principal idea of a fast or slow movement
is concluded with a cadence, new and distinct ideas enter and these in turn give way to the principal
idea in varying keys. By such means, a piece of this sort for one instrument becomes quite similar
to one for many instruments. There are some quite good concertos of this kind, particularly for
clavier. But pre-eminent among published musical works is a clavier concerto of which the author
is the famous Bach in Leipzig and which is in the key of F major. Since this piece is arranged in
the best possible fashion for this kind of work, I believe that it will doubtless be familiar to all great
composers and experienced clavier players, as well as to amateurs of the clavier and music in general. Who is there who will not admit at once that this clavier concerto is to be regarded as a perfect
model of a well-designed solo concerto? But at the present time we shall be able to name as yet
very few or practically no concertos of such excellent qualities and such well-designed execution.

It would take as great a master of music as Mr. Bach who has almost alone taken possession of the
clavier, and with whom we can certainly defy foreign nations to provide us with such a piece in this
form of compositiona piece which deserves emulation by all our great composers and which will
be imitated all in vain by Foreigners.8

Who is there who will not admit at once that this clavier concerto is to be regarded as a perfect
model of a well-designed solo concerto? Some 250 years have passed, and todays refined eye is
more skeptical of particular instantiations of ideal definitions. Schulenberg, for example, finds that
the association between the Italian Concerto and the orchestral genre is, at best, not systematic.
He points out, for example, that the piano-forte contrast (which at first sight is the main means
of depiction of orchestral roles in the two-manual harpsichord) does not quite match with the
ritornello-episodes scheme:
In the first movement of the concerto the episodes are scored mainly for a forte upper line with
softer accompaniment in the lower voices. But there is also a piano echo within a ritornello (bars
6768), and in the last movement both hands have forte passages within solo episodes.

It is indeed not true, as Scheibe would want it, that the bass is added now and then to fill out
the texture or that it must represent the subordinate parts: rather, we see it actively competing
for the leading role in the third movement. The echos, in addition, render the ritornelli less than
most clearly differentiated from the rest. And these are only some of the inconsistencies that
should be accounted for to maintain the assertion that Bachs Italian Concerto is a direct mapping
of the orchestral genre. Scheibe, of course, does not account for them; Schulenberg, on his part,
takes the inconsistencies to refute the assertion:
it is probably a mistake to regard either work [the Italian Concerto or the French Overture] as a
literal adaptation of an orchestral genre or the changes of manual as direct imitations of ensemble
scoring.9
8

Johann Adolph Scheibe, Critischer Musikus (Leipzig, 1745), pp. 6378.

Schulenberg, op. cit., p. 301.

Rthlisbergers insight
Half-way between Scheibe and Schulenberg there is an interesting piece of criticism by the Swiss
musicologist Edomnd Rthlisberger (18581919),10 who wrote that
The view to be followed, at least the one I have followed to arrive at these reflections, is: the
point of departure for the Italian concerto is to be found in the original scores of Vivaldis concertos.
...
Compare [Bachs] transcription of Vivaldis concerto [in G Major] with the Italian concerto, and
you will be convinced that both are exactly of the same essence and form. It follows that the Italian
concerto is in principle a concerto for one solo instrument and orchestra (in the genre of those by
Vivaldi), whereof Bach wrote no more than a reduction for solo harpsichord, a kind of transcription
much paler than the original conception.11

Two things are to be noted here. First, the linking of the Italian Concerto to the transcriptions, rather than the orchestral genre. The consequence of this shift of focus is potentially very
important: it opens a new door for the explanation of those inconsistencies that Schulenberg, for
example, has pointed out. There are passages in the Italian Concerto that contradict the conventions of an orchestral concertothis much is granted. But that does not automatically mean that
they contradict also what happens in the transcriptions. In fact, it is the pursuit of this consequence
that will lead me to my proposed reassessment of the nature of the Italian Concerto. Why did not
Rthlisberger himself advance along these lines? To begin with, it must be recalled that when he
was writing (his book was published in 1920, a year after his death), only one of Vivaldis originals,
the one he refers to, had been rediscovered and identified. The revolution caused by discoveries
such as the fact that Bachs four-harpsichord concerto was actually by Vivaldi for four violins lay
10

I have not been able to find more information about Rthlisberger. I run into his text almost by chance at the

Library of Congress; as it will be clear, it proved to be inspiring for my own take on the Italian Concerto.
11

La voie suivre, du moins celle que jai suivie pour aboutir ces rflexions, la voici : Le point de dpart du

concerto italien se trouve dans les partitions originales des concertos de Vivaldi. . . .
Que lon compare maintenant cette transcription du concerto de Vivaldi avec le concerto italien et lon se convainra que tous deux sont exactement de mme essence et de mme forme. Il sen suit que le concerto italien est
en principe un concerto pour un instrument solo et orchestre (dans le genre de ceux de Vivaldi), dont Bach na crit
quune rduction pour clavecin seul, soit une sorte de transcription beaucoup plus ple que la conception originelle.
Edmond Rffthlisberger, Le Clavecin dans luvre de J.-S. Bach (Genve: dition Henn, 1920), p. 83.

still in the future. There was little material for Rthlisberger even to imagine the possibilities of
his insight.
The second thing to be noted in Rthlisbergers text, on the other hand, is that he deems the
Italian Concerto much paler than a true concerto. While in Vivaldi the contrasts between tutti
and soli are obvious, in Bachs transcription, on the contrary, everything appears much duller, and
the general effect remains vague, if one does not realize that Bach, when he played his transcription himself, evidently sought to render the effect of the originalthat floated in his thought
reinforced the tutti by means of coupling registers, and thus attained the characteristic contrasts
of the orchestral version.12 Thus, by appealing to the Masters intentions (something that today
we find amusing), Rthlisberger finds a rationale for the use of the two manuals (i.e., the coupling registers) of the harpsichord in the Italian Concerto. In fact, for him, it is impossible to
make complete justice to the piece in the piano, unless perhaps transcribing it to simulate the coupling registers by octave doublingsa task for which he diligently gives directions. (Interestingly,
Schulenberg is of the exact opposite opinion: the concerto has [in spite of not having the splendor
of the D-Major Partita or the depth of the Partita in E minor] nevertheless gained deserved popularity among pianists, for whose instrument it is better suited than most of Bachs harpsichord
works.13 ) In any case, Rthlisberger is too bold in ruling out a piano performance. But he is
right when he realizes the very important fact that a transcription implies, foremost, a loss. The
following section explores this notion.
What is it to transcribe?
The general view of Bachs transcriptions as a whole is that Bach improved the originals. This,
for example, is what Pippa Drummond says:
. . . occasionally he imposes his own ideas on the originalenriching the harmony with suspensions, fitting in snippets of imitation, extending contrary motion scale passages, and adding
12

Dans la transcription de Bach, par contre, tout apparat beaucoup plus terne et leffet gnral rest vague si lon

ne ralise pas que Bach, losrquil jouait lui-mme sa transcription, cherchait videmment rendre leffet de loriginal
qui flottait dans sa pense, renforait les tutti au moyen des registres de copulation et obtenait ainsi les contrastes
caractristiques de la version orchestrale. Idem.
13

Schulenberg, op. cit., p. 301.

figuration to static bass lines. [In a passage, the suspensions he added] enliven the harmonic scheme
and make the texture more varied.
. . . Bachs additions make the original more complex.
. . . it may be said that he imposed something of his personal style on the transcriptions. Certainly
the arrangements sound utterly unlike the originals; nor is the change of medium wholly responsible. In the course of transcription, Bach has created orderly, artificial works which have more in
common with the solid traditions of German keyboard music than with the colourful vivacity of the
orchestral concerto.14

This and other accounts, such as Dreyfuss ironical analysis of the transcription by Bach of
a work by Prince Johann Ernst (which, while charming, cannot be taken seriously as anything
more than a flattering attempt to cover up the transparent harmonic poverty endemic to the original
concerto,15 ) leave the impression that Bach approached the Italian concertos, yes, with curiosity,
but above all with a good deal of paternalism.
When Bach transcribes at Weimar, however, he is not helping out his contemporaries; nor,
for that matter, mainly trying to learn the good things from their works in order to familiarize
himself with an important genre. He is simply following a very long tradition, that stems from the
ancient intabulation, in a way that, if perhaps arguably idiosyncratic, is by no means revolutionary.
An anonymous fifteenth-century arrangement analyzed by Theodor Gllner shows every feature
pointed out by Drummond in Bachs transcriptions, except for the suspensions.16 Bach transcribed
also his own concertos, and in doing so he appealed to basically the same resources and techniques.
The tradition, needless to say, would continue alive, as in Liszts transcriptions of Beethoven, or
in Weberns Ricercariin these cases, however, we are less compelled to use expressions such as
enrich, enliven, and create order.
The transcribers enhancements respond not to a quest for complexity, order, or unity, but to
the violence done to a work when its medium is changed. The comparison with translation from a
14

Pippa Drummond, The German Concerto: Five Eighteenth-Century Studies (London: Clarendon Press, 1980),

pp. 1417.
15

Dreyfus, op. cit., p. 46.

16

Theodor G-ollner, J. S. Bach and the Tradition of Keyboard Transcriptions, in H. C. Robbins Landon and

Roger E. Chapman (eds.), Studies in Eighteenth-Century Music: A Tribute to Karl Geilinger on his Seventieth Birthday
(London: Gorge Allen and Unwin Ltd., 1970), p. 245.

language to another, though somewhat commonplace, is nevertheless accurate. Something is lost,

not gained, when a work is translated or transcribedthe relative quality of writer and translator
notwithstanding. The least we can expect from the translator is that he provides compensation
for the loss, extracting it from the essential properties of the new medium. Thisnot a critical
commentary of the pieces at handis what Bach is after at Weimar. (Fortunately, Drummonds
opinion that the arrangements sound utterly unlike the originals is completely untenableshould
it be true, it would testify to Bachs failure, not to his success.)
Rthlisbergers position regarding the rendering of orchestral contrasts in the keyboard shows
how he recognizes that the transcription implies a loss. But then his longing for the coupling
registers to directly simulate the density of the orchestral tutti is little less than a longing for the
orchestra itself, and therefore a failure to acknowledge that an essentially different medium is now
in charge of the genre. Just as Scheibe before and Schulenberg after, he too is misled by granting
to the orchestral genre too consequential a role in his account of the Italian Concerto.
A recent discovery
In 1995, Kirsten Beiwenger published an article on an enigmatic copy of the first movement of
the Italian Concerto, discovered in the manuscript collection of Ludwig Scholz.17 The movement
appears by itself, and with no piano/forte indications. Moreover, as Beiwenger argues, the differences between this and the published version are more substantial than what Scholzs prudent
simplifying habits (as illustrated by the other pieces in his collection) lead to expect. It is always
dangerous to judge the quality or finalness of an early version from the perspective of the definitive one, but in this case it seems safe to conclude that perhaps we have the movement in a first,
discarded draft.18 The piece, at the time that Scholz made this copy, was unfinished.
There are no piano/forte markings in the Scholz version. But the tutti/solo differentiation is,
as a rule, more clear than in the printed version: it is as though an originally transparent depiction
of the relationship between soloist and orchestra had been obscured by Bachs later additions and
17

Kirsten Beiwenger, An Early Version of the First Movement of the Italian Concerto BWV 971 from the Scholz

Collection?, in Daniel R. Melamed (ed.), Bach Studies 2 (Cambridge: Cambridge University Press, 1995).
18

Idem, p. 17.

modifications, which include, naturally, the piano and forte indications. (At first sight, these indications are the main means of depiction of orchestral contrasts, and it is in a way surprising to
find that they were actually not part of the original conception, and that in fact they obscure it.) As
has been said, it is in part the inconsistencies in the piano/forte markings that raised Schulenbergs
doubts about the Italian Concerto being an adaptation of the orchestral genre. A similar kind of
doubt is identifiable in Werner Breigs opinionreported by Beiwengerthat the piece cannot
derive from an orchestral original because the structure of the outer movements is substantially
different from that of Bachs other keyboard concertos: it is not the solo/ripieno structure that determines the disposition of the material, but rather the performing medium.19 But the Scholz version
dispels all these doubts in a single stroke, showing that the piano/forte markings, the performing
medium, are secondary to the nature of the piece.
But if these doubts are dispelled, and in addition this earlier version features a more clear
representation of the orchestral contrasts, it is again possible to think that the movement is, after
all, a literal adaptation of an orchestral genre. Beiwenger herself perceives this, and to that
respect writes:
The more conspicuous tutti/solo contrast in the Scholz version could lend credibility to the
possibility that Bach reworked a solo concerto, if it were not that other passages decidedly contradict
this hypothesis (for example, mm. 10521, with their sixteenth-note figuration in the bass).20

I shall return later on to this passage, to explore Beiwengers sixteenth-notes argument. But,
before leaving the Scholz version of the first movement, let me highlight the two most important
ideas that its existence and analysis have brought to the discussion. Firstly, this copy suggests that
the first movement of the Italian Concerto is independent and different in nature from the other
two, and in fact was probably composed fairly long before them. Secondly, it shows that this first
movement is close indeed in nature to a transcriptioncloser, in any case, than one might be led
to think by the final version (with its piano/forte indications and the other modifications). As we
shall see, these two factors play an important role in the thesis about the piece that I introduce in
the following sections of this paper.
19

Idem, p. 18.

20

Idem, pp. 1718.

10

The history and the nature of the Italian Concerto

So far, I have separately addressed several topics of the discussion on the Italian Concerto: different attitudes toward and arguments around the Italian Concerto, various opinions on its relationship
to the orchestral concerto as a genre, the problems and goals of transcribing, and the suggestions of
archival evidence. This has set the stage for what follows, where I propose a thesis about piece
more precisely, about how the piece was conceived of, which is at the same time a thesis on its
history and on its nature.
To do that, however, some elements are necessary that have not yet been treated. Two are
particularly important: the French Overture BWV 831 and the actual techniques and usages of
transcription. The former provides the circumstantial side of my thesis; the latter gives musical
evidence in its support, and ultimately rounds off my description of the nature of the Italian Concerto.
A French digression
Bach published his op. 1, the Clavier-bung, in 1731. The volume contained the six Partitas;
excluded, however, was an additional C-minor suite that had been ready since around 1730, but
which was too removed in nature from the other six: the piece was French in character, and called
to that end for a two-manual harpsichord. It ended uniquely, not with the gigue, but with an Echo,
clearly composed to expound and exploit the contrast between manuals. This seventh suite is of
course the piece that was eventually to be published as the French Ouverture.21
As quoted above, when Schulenberg discovers that the use of harpsichord registration in the
Italian Concerto does not follow the orchestral conventions, he extends this judgement to the
French Overture. But in fact the piano and forte markings in the suite are not only completely congruent as a system, but above all follow the orchestral uses rather directly. Schulenbergs constant
comparison between the French Overture and the six Partitas, on the other hand, is misleading:
it is not clear why it should be compared precisely to them, and the sole fact that all of them
are keyboard suites does not seem enough to make the comparison relevant. For example, Chris21

The final B-minor version is BWV 831; the original C minor suite, part of the Bchlein fr Anna Magdalenna, has

been assigned the number 831a.

11

tian Wolff writes that the ouverture (suite) has no direct counterpart in Bachs previous work in
its imaginative exhibition of French manners of genre and style.22 Interestingly, it is again Edmond Rthlisberger who brings an important point home, by stressing the inner division of the
genre of suite:
In the times of Bach, there was between the harpsichord suite (or partita) and the orchestral suite
(or partita) a fairly sensible difference.
The harpsichord suite comprised an almost stereotyped grouping of dances, with or without a
prelude.
The orchestral suite included above all a big prelude, the pice de rsistance. After the latter
came the musique dagrment, i.e., dances and small pices de fantaisie. To perceive this difference,
it is enough to compare Bachs orchestral suites (titled Ouvertures ) with his harpsichord suites.23

Where does the French Overture belong, in the light of this distinction? It is of course a
keyboard piece, but it should be understood as a specimen of the orchestral kind of suite. In fact,
its greatest weight lies on the first movement, a feature thatamong othersseparates the piece
from the six keyboard Partitas and relates it closely to the four orchestral Ouvertures. That Bach
was aware of all this is seen in the consistency of his titles: this suite in

minor, still not named

French, is since the beginning called ouverture. In general, it would have been very strange for
this piece to appear alongside the Partitas in the Clavier-bung.
The birth of the Clavier-bung II
If the Partitas were not the best context for this ouverture, what could be a fitting companion?
Well, another orchestra-like keyboard piece was the perfect candidateand, of course, the seed
of such a piece is exactly what that early F-major concerto movement was. Its style, in addition,
22

Christian Wolff, Bach und die italienische Musik, in Bachtage Berlin (Neuhaussen: Hanssler, 1985), p. 376.

23

Au temps de Bach, il y avait entre la suite (ou partite) pour clavecin et la suite (ou partite) orchestrale une

diffrence assez sensible.

La suite pour clavecin comprenait un groupement peu prs strotyp de danses avec ou sans prlude.
La suite orchestrale comprenait avant tout un grand prlude, la pice de rsistance. Aprs celui-ci venait de
la musique dagrment, cest--dire des danses et de petites pices de fantaisie. Pour percevoir cette diffrence, il
suffit de comparer les suites orchestrales de Bach (intitules ouvertures ) avec ses suites pour clavecin. Rthlisberger, op. cit., p. 85.

12

further improved the fit: it complemented the French with the Italian.
This, I propose, Bachs the motivation to revise that movement, and to compose the other two to
complete it. Thus, a sequel for the Partitas, in the form of a volume grouping an Ouvertre nach
franzsischer Art and a Concerto nach italinischem Gusto, came to existence. The national
opposition of styles, that had become a central part of the project, was to be furthered by the
transposition of the suite to B minor (a tritone away from the concerto).
Regarding the Italian Concerto, the most important implication of this biography of the
Clavier-bung II is that the second and third movements, unlike the first, were composed explicitly for this project. Indeed, that they are the result of a different project entirely, one into
which the first movement was incorporated, but for which it had not been originally intended. This
in turn implies a difference in nature between the first movement and the other two, and identifying
such difference is therefore the main form of providing support for the thesis. As we saw, the first
movement stands by itself in the Scholz version, and this can be seen as a kind of archival proof;
but, in my view, much more compelling evidence is to be found in the music itself. The following
sections are intended to show that there is a difference in nature between the first and the other
movements: the former is much closer to the genre of orchestral concerto.
This conclusion will be established through an examination of the outer movements in the
light of Bachs early Weimar transcriptions. But the study of the third movement will in addition
complement the picture by showing another important dimension of the piece: it is atypically
Bachiansynthesis, a compendium of the techniques and possibilities of transcription.
The first movement
The difference in nature between the first and the other movements of the Italian Concerto lies
in the question: could each of the movements have derived from an orchestral original? I shall
claim that, musically, the first movement could be a transcription from an orchestral originalthat
there is no refutation of that possibility in the music itself. But it should not be concluded that I
that the first movement is in fact a transcription. I most definitely think it is not: extra-musical
consideration to be mentioned later convince me that this is not the case. All I claim is that the
music of the first movement could have been the product of a transcriptionwhile that of the other

13

       
2

106

  
  


 

   2

  4
6  


2

Figure 1: Bass line in mm. 105ff. (first movement)

movements could not. In this aspect, the movement is essentially different from the other two, and
that is the sought-for difference that supports the history of the piece offered above. Put in words
that will be more clear after my analysis of the third movement, the difference is that when Bach
composed the first movement he set out to imitate a transcription, and therefore the movement
comes close to actually resembling one (and could in principle make us think that it is one); on the
contrary, when composing the other two, he was thinking of a keyboard piece. In this case he was
inspired by the techniques of transcription, but he was not constrained by the actual reference to
the orchestral genre.
Let us proceed by clarifying where we are regarding the question can the first movement be
derived from an orchestral original? Schulenberg, as seen, does not believe it can. His main arguments, however, are based on the blurring of ritornello-episode contrasts, and on the characteristics
of the piano/forte indications. But we now know that both things, the blurring and the markings,
were later additions to the piece. The Scholz version, a more faithful document of the original
conception of the movement, features a very clear (and clever) implementation of the orchestral
contrasts.
After having made this discovery, Beiwenger had to face the question anew. She does not
think that the movement could be a transcription either, but her arguments are of a new kind: the
sixteenth-note figurations in the bass of, for example, mm. 10521 (see Figure 1).
It would indeed be hard to find this kind of sixteenth-note figurations in the bass of an orchestral
concerto. But the fact is that such bass figurations are found, not in the orchestra, but in the
keyboard transcriptions. Talking about Bachs transcriptions, Schulenberg points out that it is at
first surprising that Bach tended to add embellishment more frequently in the bass than in the upper
part. But in quick movements the solo violin part could receive only occasional elaboration.24
24

Schulenberg, op. cit., p. 92.

14

This is true, but, in my view, there is more to it than that. It is not exactly that Bach, looking
for some voice to embellish, finds the solo violin already too elaborated, and therefore decides to
elaborate on the bass. Rather, Bach uses the sixteenth-notes in the bass as a means of compensation for something that has been lost with the orchestra: the contrast in density.
For instance, the first movement of BWV 976 (from Vivaldis last piece of LEstro Armonico)
has sixteenth-note embellishment of the bass only in episodes. In the orchestral original, in which
both ritornello and episodes feature chordal accompaniment, the contrast between them is achieved
through variation in density. But contrast in density is not a real possibility in a keyboard transcription. It is the use of sixteenth-notes in the bass that stands for it, fulfilling in the transcription the
articulating role that variation in density has in the orchestra.
Similarly, when the bass has been embellished in the ritornello itself, Bach avoids the sixteenthnotes in the episodes, or at least treats them with prudence. This is the case, for example, of
BWV 978, from Vivaldis Op. 3 No. 3. In the first movement of Bachs transcription, the ritornello

has sixteenth-notes in the bass; sixteenth-notes in the bass (except for one passage, mm. 4145)
are accordingly saved only for those moments in which the basso continuo plays in the original.25
Also the third movement of the Concerto in G BWV 973 is illustrative: the orchestral unison in
several ritornelli and tutti passages is complemented by Bach with a very virtuoso (and totally
new) bass line. Of the episodes, on the contrary, only the last one shows sixteenth notes, and this
clearly to the effect of achieving a grandiose ending.
From this point of view, then, the sixteenth-notes of measures 10521 in the first movement
of the Italian Concerto do not preclude the possibility that it reflects an orchestral original.26 The
25

In the exception, the appearance of the sixteenth-notes coincides with a change in the original accompaniment of

violas, which in that moment start a continuo-like figure.

26

One could perhaps make the case that the sixteenth-notes in the bass of the Italian Concerto weaken after all this

possibility, not because of their existence, but because their treatment is non-systematic (i.e., only some of the episodes
have sixteenth-notes), contrary to what tends to happen in the transcriptions. This is not the place to go into details,
but it can be noted that the real transcriptions are systematic partly because the Vivaldian originals are themselves
already systematic in the relationship between episodes and basso continuo. The non-systematic sixteenth-notes in the
Italian Concerto could have derived from a non-systematic (non-Vivaldian, Bachian) orchestral original.

15

same is true of another possible objection: the piano echoes within ritornelli that Schulenberg
singles out. Echoes like these are fairly common in Vivaldis and Bachs orchestral concertos, and
they of course show up in the transcriptions.27
This makes my main point about the first movement: there is nothing in the music that provides strong refutation to the possibility that it derives from an orchestral original. From the score
alone, one would have to answer the question can the first movement be a transcription? in the
affirmative: yes, it can. As I said at the beginning of this section, however, this is different from
answering yes, it is. Like Schulenberg and Beiwenger, I do not think the movement stems from
a lost orchestral concerto. The arguments are extra-musical: to begin with, there is absolutely no
reference of any kind to such an original. It would in all likelihood be by Bach,28 but then it remains to be explained why he transcribed for un-accompanied keyboard (the rest of his concertos,
when transcribed, were transcribed for accompanied harpischord). Why is the piece so different
from the rest of his concertos in the first place? And, finally, why would the transcription (the
Scholz version) be so draft in quality, if it was already a re-working?
The final conclusion is: Bach composed the first movement of (what would be) the Italian Concerto with the idea of imitating, to the last detail, the result of transcribing an orchestral concerto
to the keyboard. This is something, as I argue in the next section, that cannot be said of the third
movement.
The third movement
There are three characteristics in the music of the third movement that are irreconcilable with
orchestral writing, and therefore make it clear that the movement could not have stemmed from
an orchestral original. The first is a general prominence of two-part counterpoint that is at odds
with the scheme of soloist and accompaniment. The second, related characteristic, is the pres27

The ritornello of the already mentioned BWV 976/1, for example, has echoes as a constitutive element. Bach even

writes piano and forte indicationsin this case not referring to changes of manualwhich would in any case prove
an easy target for Schulenbergs charges of incongruence.
28

Beiwenger, op. cit., p. 17.

16

93

      

4  
G 2     6



I
4   
2
6


 


99

>

G 2 >



4



>
>



4

 
4 6    

>

 
      >









4



 4 
2

  

  4     










4  

>

     4 6






  

 4 

  4


>



4

  4     


>

>




4

   

>







4 


9



 




Figure 2: The inverted version of the ritornello in the third movement

ence thematic invertible counterpoint between the hands.29 In the third place, the ritornello of
measures 93104 has the top part transferred to the bass, an orchestrally unthinkable switch (see
Figure 2).
There is no way to conciliate these characteristics with the idea of a transcription as we did
with the problematic passages in the first movement. This in itself is sufficient to support the
biography of the piece that I have been proposing (the first movement composed early, the other
movements added to it for inclusion in the Clavier-bung II). But a further exploration of these
three characteristics, and their relationship with the transcriptions, will reveal another dimension
of the nature of the Italian Concerto. The first movement is close to the purely orchestral in a
way that the third is not. But this should not suffice to conclude that the third movement is purely
keyboardistic. The transcription stands in the middle of the two extremes (the orchestral and the
keyboardistic), and exploring the third movement in relation to it shall prove illuminating.
The first characteristic is the prominence of two-part counterpoint. Looking at the transcriptions, it turns out that this is not at all uncommon in the finales of transcribed concertos. Vivaldi
(and the rest) did not always end with a ritornello-form movement, but often composed a giguelike finale instead. For example, the finale of Bachs BWV 980 (transcribed from Vivaldis Op. 4,
No. 1), is entirely written out in two parts, and in fact it would be hard to guess that it actually
29

Cf. mm. 2532; on the surface, this is what creates the piano-forte incongruence pointed out by Schulenberg.

17

 )

G 

.
I
I

Rckpositiv










 )



>




 )




Oberwerk










 )
.









.















>




Oberwerk

.


 )







Rckpositiv




 )

.


Figure 3: From the organ transcription BWV 595

comes from a concerto (and not, say, from a sonata or a suite).
Two-part counterpoint also arises in another context in the transcriptions: as a result of transcribing concertos for two soloists. These are usually transcribed for organ, each hand taking the
role of one of the soloists on the basis of the basso continuo in the pedal.
But some times Bach transcribed both for clavier and organ. BWV 984/1, from a lost original
by Prince Johann Ernst, is such a case. The organ transcription, BWV 595, behaves as expected
(Figure 3): the continuo is in the pedal while the soloists engage in interplay on the manuals.
Now, what happens in the clavier transcription, where there will be obviously no pedaliter to
play the bass? The answer is simple as it is revealing: the bass part is omitted altogetherand
the passage becomes pure two-part counterpoint (Figure 4). Again, should we not know the do, it
would be hard to believe that these passages come from orchestral originals.
3

G 

   


   
 

  
       

    
    

    

   

  

   

   
       

   

Figure 4: From the clavier transcription BWV 984

The finale of BWV 978 is also illustrative. On the one hand, it too resembles the finale of the
Italian Concerto in being mostly a two-part movement; the texture is denser in the ritornelli, but
in the episodes the focus is on the bass lineit would almost seem to depict a concerto for cello,
not for violin. (The ambiguity comes directly from Vivaldis original, in which the basso continuo
line is virtuoso in itself.)
18

G 3
8

mm. 14

    

I4 3




 

   

 

   

mm. 11922

   

I4




 











  
?


     
     







Figure 5: Inversion of the ritornello in one of Vivaldis concertos (as transcribed by Bach)
The most significant feature of this movement, however, is Vivaldis use of a most unexpected
device: in one of the ritornelli (mm. 11922), he exchanges the roles of the high and the low registers. Figure 5 shows the resulting inversion (in Bachs transcription). Clearly, this is a precedent
for the third characteristic of the Italian Concerto, the most un-concerto-like one.
Instances of two-part and invertible counterpoint are thus documented, even in ritornelli, in
Bachs Weimar transcriptions.30 In this way, precedents are to be found for even the most radically
keyboardistic characteristics of the third movement of the Italian Concerto. What does this mean?
It does not mean that the third movement could derive from an orchestral source: after all, this
movement is not a gigue, the kind of two-part counterpoint is not the normal interplay between
two soloists, and the inversion of the ritornello, unlike that of Vivaldi, creates what would be a
revolutionary continuo line, even by Bachs standards.
Rather, the implication is that the movement is deeply rooted in the Weimar transcriptions in
a very conscious and consistent way: it is a compendium of techniques and possibilities. Bach
30

They are not exactly thematic, but this is natural on the view that baroque concertos in general, and episodes

in particular, are themselves seldom thematic, but rather are composed of small motives and figurations in sequence.
But the transcriptional devices that have created these instances would have easily created thematic counterpoint, had
their originals been thematic. In any case, the clear distinction between the two parts of the counterpoint has been
abandoned, and both lines stand at the same level of melodic importance. From this to real thematic counterpoint there
is only a small step, one that on the other hand is very natural for a more thematic musical thinking like Bachs.

19

appears here preoccupied not with transcription as a keyboard reference to the orchestra, but with
transcription in itself with that which transcription, and only transcription, means to the keyboard.
Conclusion
This is, as I see it, the nature of the Italian Concerto (as a three-movement, 1730s project): an
encyclopedic compendium of the possibilities of the task of transcribing, further realized beyond
what the actual transcription usually calls for. Displacement of the bass, treatment of both hands
as soloists with an imaginary orchestral accompaniment, inversion of the whole textureall these
are devices that could only be created on the keyboard and for the keyboard, but whose embryonic
idea could only have been tasted when transcribing an orchestra.
The resulting synthesis is, on the surface, relatively autonomous from the genre of concerto. At
first sight Schulenberg is right when saying that in the Italian Concerto a fairly realistic evocation
of the orchestral model at the outset is evidently overcome by the urge toward free development.31 As I have tried to show, free development does not overcome, but complements and
completesin fact, developsthe realistic evocation of the first movement. Bachs urge, typical
of his mature period, was toward completeness.
A word on the second movement
My argument has progressed with no reference to the second movement. This last section, almost
an appendix, is simply the incorporation of this movement into the model that emerged in the main
body of the present essay. My general thesis implies that the second movement is more akin to the
third than to the first, in the sense that it was conceived of long after the latter, with the conscious
goal of completing it for publication in the Clavier-bung II. The middle movement does not
provide the kind of evidence that was advanced in connection with the outer movements, not least
because it is not a ritornello form. But on other grounds this movement seems to me to have been
more plausibly conceived directly for the keyboard, rather than as a transcription.
The first thing to note is the absence of the bass from the downbeat in the accompanying pattern
31

Schulenberg, op. cit., p. 303.

20

 (
23
4

? ?
H  


 ( J? ? H 
  
4 (

Figure 6: The accompanying pattern of the second movement.

(that remains consistent throughout the movement; the first two measures are shown in Figure 6). It
is simply unthinkable that a Baroque orchestral bass (with the added weight, not only timbrical but
also harmonic, of the basso continuo) be composed off-beat in such a way. It can be safely assumed
that an orchestral original, if it existed, would affirm the downbeat with the bass. This raises a first
problem: what is there to sound in the second and third eight-notes of the measure (which are
taken by the bass in the keyboard)? The only sensible conjecture is that the bass note would be
repeated throughout the measurelet us say by the viola. So: the continuo has the bass note, the
viola repeats it, and the violins have the top parts. But as the movement advances, this hypothetical
pattern of an original accompaniment becomes unrealistic. The bass itself raises many questions
of register, but in any case the top parts descend well beyond the range of the violin (as much as a
seventh). No strategic change of octave do: the smoothness of the line would be destroyed, and the
accompaniment would cross the soloist.32 In general, the progression of these voices as they stand
in the keyboard version makes so perfect sense, that it seems almost unescapable to conclude that
they were conceived directly as they appear there. The only credible supposition is that Bach set
to produce something closely resembling an orchestral second movement (his intuition suggested
him something very similar to the A-minor Violin Concerto BWV 1041 as a starting point), but
was the whole time composing a keyboard piece. The texture and the particular registration to use
(piano for the left hand, forte for the other) was all that the orchestral model fixedbeyond that,
he was free.
32

In concrete: transposing these voices up an octave from m. 35 to the end would be acceptable for those voices

themselves, but the crossing in mm. 45ff. would be unacceptable. Even transposing just the lower voice, thus creating
a movement in parallel sixths rather than thirds, is not enough to avoid the crossing. In the previous problematic
passage, mm. 1728, there is no good chance to transpose up without doing violence to the melodies.

21

References
Beiwenger, Kirsten. An Early Version of the First Movement of the Italian Concerto BWV 971 from the
Scholz Collection? In Daniel R. Melamed (ed.), Bach Studies 2. Cambridge: Cambridge University
Press, 1995 . Pages 119.
Butler, Gregory. The Aesthetic and Pedagogical Context of Bachs Concerto nach Italienischem Gusto
BWV 971. Paper presented at symposium Bachs Musik fr Tasteninstrumente. Dortmund. 2002.
Dreyfus, Laurence. Bach and the Patterns of Invention. Cambridge, Mass. & London: Harvard University
Press, 1996.
Drummond, Pippa. The German Concerto: Five Eighteenth-Century Studies. London: Clarendon Press,
1980.
Gllner, Theodor. J. S. Bach and the Tradition of Keyboard Transcriptions. In H. C. Robbins Landon
and Roger E. Chapman (eds.), Studies in Eighteenth-Century Music: A Tribute to Karl Geilinger on his
Seventieth Birthday. London: Gorge Allen and Unwin Ltd., 1970 . Pages 253260.
Marshall, Robert L. Toward a Twenty-First-Century Bach Biography. The Musical Quarterly 23 (2000):
499525.
Rthlisberger, Edmond. Le Clavecin dans luvre de J.-S. Bach. Genve: dition Henn, 1920.
Scheibe, Johann Adolph. Critischer Musikus. Leipzig, 1745. Facsimile Hildesheim: Georg Olms Verlag,
1970.
Schulenberg, David. The Keyboard Music of J. S. Bach. New York: Schirmer Books, 1992.
Willer, Monika. ,Sonderflle, ,Modeerscheinungen und das ,Normale: Konzertsatzkonzeptionen in
beglieteten un unbegleiteten Klavierkonzerten vor 1750. In Reinmar Emans and Matthias Wendt (eds.),
Beitrge zur Geeschichte des Konzerts: Festschrift Siegfried Kross zum 60. Geburstag. Bonn: Gudrun
Schrder Verlag, 1990.
Wolff, Christian. Bach und die italienische Musik. In Bachtage Berlin. Neuhaussen: Hanssler, 1985 .
Pages 225234.

22