Sie sind auf Seite 1von 414

This page intentionally left blank


W. Dean Sutcliffe investigates one of the greatest yet least understood repertories
of Western keyboard music: the 555 keyboard sonatas of Domenico Scarlatti.
Scarlatti occupies a position of solitary splendour in musical history. The sources
of his style are often obscure and his immediate inuence is difcult to discern.
Further, the lack of hard documentary evidence of the sort normally taken
for granted when dealing with composers of the last few hundred years has
hindered musicological activity. Dr Sutcliffe offers not just a thorough reconsid-
eration of the historical factors that have contributed to Scarlattis position, but
also sustained engagement with the music, offering both individual readings and
broader commentary of an unprecedented kind. A principal task of this book,
the rst in English on the sonatas for fty years, is to remove the composer
from his critical ghetto (however honourable) and redene his image. In so do-
ing it will reect on the historiographical difculties involved in understanding
eighteenth-century musical style.
w. dean sutcli f f e is University Lecturer at the University of Cambridge and
a Fellow of St Catharines College. He is author of Haydn: String Quartets, Op. 50
(1992) in the Cambridge Music Handbook series and editor of Haydn Studies
(Cambridge 1998). He is also co-editor of the Cambridge journal Eighteenth-
Century Music, the rst issue of which will be published in 2004.
St Catharines College, Cambridge
caxniioci uxiviisir\ iiiss
Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore, So Paulo
Cambridge University Press
The Edinburgh Building, Cambridge cn: :iu, United Kingdom
First published in print format
ISBN-13 978-0-521-48140-3 hardback
ISBN-13 978-0-511-06764-8 eBook (EBL)
Cambridge University Press 2003
Information on this title:
This book is in copyright. Subject to statutory exception and to the provision of
relevant collective licensing agreements, no reproduction of any part may take place
without the written permission of Cambridge University Press.
ISBN-10 0-511-06764-X eBook (EBL)
ISBN-10 0-521-48140-6 hardback
Cambridge University Press has no responsibility for the persistence or accuracy of
uiis for external or third-party internet websites referred to in this book, and does not
guarantee that any content on such websites is, or will remain, accurate or appropriate.
Published in the United States by Cambridge University Press, New York
Preface page vii
1 Scarlatti the Interesting Historical Figure 1
2 Panorama 26
Place and treatment in history 26
The dearth of hard facts 29
Creative environment 32
Real-life personality 34
The panorama tradition 36
Analysis of sonatas 38
Improvisation 40
Pedagogy 41
Chronology 43
Organology 45
Style classication 49
Style sources 54
Inuence 55
Nationalism I 57
Nationalism II 61
Evidence old and new 68
3 Heteroglossia 78
An open invitation to the ear: topic and genre 78
A love-hate relationship? Scarlatti and the galant 95
Iberian inuence 107
Topical opposition 123
4 Syntax 145
Repetition and rationality 145
Phrase rhythm 167
Opening and closure 171
Sequence 181
vi Contents
Kinetics 188
Vamps 196
5 Irritations 217
Der unreine Satz 217
Introduction 217
Voice leading 223
Counterpoint 230
Cluster chords and dirty harmony 236
Rationales 247
Tempo and Scarlattis Andantes 250
Ornamentation 256
Source matters 263
6 Una genuina m usica de tecla 276
Fingermusik and mere virtuosity 276
Keyboard realism 292
Texture and sonority 297
7 Formal dynamic 320
Binary-form blues 320
Thematicism 325
Formal properties and practices 334
Dialect or idiolect? 355
Lyrical breakthrough 358
Pairs 367
Finale 376
Bibliography 381
Index 392
This book deals with one of the greatest but least well understood and covered
repertories of Western keyboard music, the 555 keyboard sonatas of Domenico
Their composer occupies a position of somewhat solitary splendour in
musical history. The sources of his style are often obscure, there are no contempo-
raries of his with whom he can be more than loosely grouped, and his immediate
historical inuence, with the exception of a few composers of the next generation
in Spain, is difcult to discern. Yet enthusiastic testimonials on his behalf have been
provided by many later musicians, whether composers, performers or writers. For
all the acknowledgement of mastery, however, the fact remains that the acknowl-
edgement is usually brief. The extreme lack of hard documentary evidence together
with Scarlattis uneasy historical position has hindered sustained musicological en-
gagement with his music, and this has a ow-on effect into other spheres of musical
life. Nevertheless, there is undoubtedly a wide gap between the general publics
and performers interest in the composer and the amount of writing available to
answer that. Thus my principal task is to remove the composer from his critical
ghetto (however honourable), redene his image, and to place him more rmly in
the context of eighteenth-century musical style. At the same time I would hope to
offer some useful thoughts on just this larger context, and indeed on the concept of
style as well.
An uncertain and sporadic critical tradition has determined my approach to the
task. Reception history and close reading constitute the basic lines of thought. Given
the lack of so many contextual and documentary resources, reception history lls
the gap not just faute de mieux but also as a way of investigating how one constructs
a composer when so many issues are oating. Chapter 2 forms the focus for this,
building on aspects outlined in Chapter 1. In view of the justied charge that
Scarlattian research has been uncoordinated, I wanted here to coordinate as many
views as possible, even at the risk of overloading the discussion. Further, I can hardly
assume a familiarity on the part of the reader with so much far-ung literature, in
many different languages. There is insufcient scholarly momentum for any views to
The often quoted total number of 555 sonatas is in fact something of a fabrication on the part of Ralph Kirkpatrick.
In his determination to produce a memorable gure, he numbered two sonatas K. 204a and K. 204b, for instance,
and allowed to stand as authentic several works that have since been widely regarded as dubious. See Joel Sheveloff,
Tercentenary Frustrations, The Musical Quarterly 71/4 (1985), 433.
viii Preface
be taken as read. Another way in which I have plugged the gap is by incorporating
substantial discussions of recorded performances. This may be an unusual move,
but performances after all represent the business end of any reception history, the
ultimate engagement with the texts offered by a composer. I only regret that, perhaps
inevitably, I am more likely to draw attention to readings and approaches with which
I differ than those with which I am in agreement.
The case for close reading is of course more delicate nowadays. While the larger
issues relating to such interpretation will be answered both by word and deed in
the chapters that follow, there is a particular justication for its employment in the
case of a gure like Scarlatti. It is one thing to problematize close reading when a
composers craft has been established by a long tradition when there is, rightly
or wrongly, some centred notion of how the music goes. With Scarlatti, though,
there has been an almost total absence of detailed analytical writing. It therefore
seemed important to try to establish some credentials for his style, to gain a strong
feeling for the grain of his language. Indeed, many of the most special and radical
aspects of his music only seem to emerge through close attention to detail. I have
certainly missed the existence of such readings that could be used as a means of
sharpening the eld of enquiry. In no other respect has my work felt like such a
leap into the dark. And I should emphasize too that many of the readings, and
the larger arguments to which they give rise, were extraordinarily hard won. They
only arose after endless hours playing the sonatas (with many more dedicated to
playing other keyboard music of the century) and often simply staring at the printed
page, hoping for enlightenment. This process unfolded principally during the years
1993 to 1997. My study is appearing fty years after the last book in English to be
devoted principally to the Scarlatti sonatas, by Ralph Kirkpatrick. Coincidentally, as
I recently discovered, Kirkpatricks systematic stylistic examination of the sonatas
occupied an equivalent period fty years ago, from 1943 to 1947. I hope this is a
good omen.
The relative absence of sharpening material referred to above reects a broader
difculty in approaching my subject the at critical landscape of the Scarlatti
literature. There are no established leading critical issues to which one responds and
which help to create a framework for interpretation, although there are certainly
plenty of specically musicological ones. By critical I mean those ways of thinking
that try to interpret in broad cultural and artistic terms, that are readily accessible
to those who lack detailed musical knowledge. (The lack of critical engagement
is evident in the new entry on Scarlatti in the recent edition of New Grove; it
seems to me to represent a step backwards from its predecessor.) Because of this I
have not specialized within my eld a atter terrain has had to be traversed. In
another world, for instance, I might have devoted the whole study to those issues
of syntax and temporality that are tackled primarily in Chapter 4. On the other
hand, no comprehensive survey of the output is intended. There are many areas
which have been merely glanced at or for which I ran out of room. These include
the history of editions, especially those in the nineteenth century, the history of
Preface ix
arrangements (although there is some material on Avisons concerto arrangements
in Chapter 4), coverage of some of Scarlattis very talented Iberian contemporaries,
and an examination of the various new sonatas that have been unearthed in the
past generation.
There is an advantage, however, to this state of affairs. It has encouraged me to
think big when attempting to place the composer, especially since it was not my
primary concern to advance further some of the acknowledged problems of hard
evidence. The generic and geographical circumstances short keyboard sonatas
written mostly on the Iberian peninsula might not exactly encourage monumental
interpretation, yet, as will I hope be shown, there is plenty to be expansive about.
Another large-scale quantity is style. In engaging with this as a central point of
enquiry, I have had to dance around several nasty issues of denition. These are
engaged with consistently through my text, but several ought to be signalled now.
One concerns the characterization of the popular elements that loom so large in the
world of the sonatas, and the appropriateness of terms such as Spanish, Portuguese,
Iberian, amenco, even Neapolitan. The other relates to those established larger
points of stylistic reference, Baroque and Classical. In the rst case there is the
difculty of whether such terms can be used with any precision, which is addressed
particularly in Chapter 3. In the latter case, the issue concerns the utility of the
terms altogether. What is perhaps most important to note at this stage is that these
are just the kinds of difculty that have discouraged scholarly endeavour, especially
in relation to a gure such as Domenico Scarlatti. They prompt pangs of conscience
that I too have experienced in writing my account; yet they have added to the
fascination of the project.
The rst chapter of my study introduces some of the issues surrounding Scarlatti
and sets up some parameters for interpretation by dealing with four individual
sonatas. After the focus on reception in Chapter 2, Chapter 3 (Heteroglossia)
investigates the types of material found in the sonatas, the ambiguity of their def-
inition and the composers relationship to them. This is followed by the longest
and possibly most important chapter (Syntax), which deals with all the unusual
patternings, shapings and treatments of repetition which promote a sense of syntac-
tical renewal in the sonatas. Then Chapter 5 (Irritations) reveals a number of those
special details that do so much to dene Scarlattian language. These include not just
the well-known irritations of harmony and voice leading, but also apparent incon-
sistencies of ornamentation and tempo designation. An examination of the peculiar
character of many of Scarlattis Andantes follows naturally from this last category.
Following on from all the above is a consideration of the sources, the master category
of irritation. The difculty of the source situation will be evaluated through a num-
ber of case studies. Macario Santiago Kastners phrase una genuina m usica de tecla
(a genuine keyboard writing) is used as a springboard for a discussion of key-
board style in Chapter 6, isolating such characteristics as Scarlattis use of register
and doubling. I also consider the physicality of this keyboard style and how we
might understand the place of unthinking virtuosity. Chapter 7 (Formal dynamic)
x Preface
examines the thematic and formal properties of the sonatas, vital to an understanding
of Scarlattis historical position. The section entitled Dialect or idiolect? reviews a
number of the composers ngerprints and considers their possible historical sources;
this also enables us to return to the problematic notion of originality that has borne
so much weight in the Scarlatti literature. Lyrical breakthrough describes those
moments when suddenly, and generally briey, a sonata unveils more personally
inected melodic material. The nal section, although proceeding from a sceptical
position, investigates possible instances of paired sonatas and considers the status of
such connections.
The primary sources for the Scarlatti sonatas, those copies now held in libraries in
Parma (the Conservatorio Arrigo Boito) and Venice (the Biblioteca Marciana), are
sometimes referred to in the text by means of the abbreviations P and V; the same
holds for the important M unster (M) and Vienna (W) collections. A comprehensive
work list giving full source details for all the sonatas may be found at the end of
the article on Scarlatti in the second edition of New Grove.
Pitch designations
follow the Helmholtz system (c
= middle C) where specic pitches need to be
given; otherwise a neutral capital letter is employed. The sonatas themselves are
referred to according to the established Kirkpatrick numbering, while the sonatas
of Scarlattis Lisbon colleague Seixas are cited according to the separate numberings
given in the 1965 and 1980 Kastner editions. For the collection of thirty Scarlatti
sonatas published in 1739, I have standardized the spelling to the original Essercizi
rather than the modern-day Esercizi. All translations from the literature are mine
unless otherwise attributed.
Musical examples for the sonatas are reproduced by permission of Editions Heugel
et Cie., Paris/United Music Publishers Ltd. The version of the sonata K. 490 given
as Plate 1 is reproduced by permission of the Syndics of the Fitzwilliam Museum,
Cambridge. I am grateful to both. Inevitably in such a wide-ranging undertaking,
not all discussions of sonatas have been illustrated with music examples. Especially
with some of the works covered in greater detail, there is either no example or a
partial one, for reasons of space and economy. Readers will require some access to
editions of sonatas.
I would like to thank, for their help in all sorts of capacities, the following friends
and colleagues: Richard Andrewes, Andrew Bennett, Malcolm Boyd, John Butt,
Jane Clark, Larry Dreyfus, Jonathan Dunsby, Ben Earle, Emilia Fadini, Kenneth
Gilbert, Daniel Grimley, Fiona McAlpine, Roger Parker, Simon Phillippo, Vir-
ginia Pleasants, Linton Powell, Nils Schweckendieck, David Sutherland, Alvaro
Torrente and Ben Walton. I owe a debt to the staff of the Pendlebury Library
of the Faculty of Music and the University Library, Cambridge. I also learnt much
from the Part II undergraduate seminar groups who took my course on Domenico
Scarlatti; their enthusiasm for, and sometimes their incomprehension of, Scarlattis
Roberto Pagano, with Malcolm Boyd, (Giuseppe) Domenico Scarlatti, in The New Grove Dictionary of Music and
Musicians, second edn, vol. 22, ed. Stanley Sadie and John Tyrrell (London: Macmillan, 2001), 398417.
Preface xi
creative practices were enormously stimulating. Many thanks to Penny Souster at
Cambridge University Press, for all her encouragement over the prolonged period
during which I wrestled with Scarlattis demons. Michael Downes copy-edited the
typescript not only with great care but with real sympathy for the project. Finally,
I wish to acknowledge the contributions of friends such as Michael Francis, Rose
Melikan and Julian Philips, my partner Geoff and my parents Pat and Bill, who all
put up with endless progress reports on the odyssey.
Cambridge, July 2002
Domenico Scarlatti does not belong. Whether we ask to whom, to where, or to
what he belongs, and even if we ask the questions with the slight difdence proper
to any such form of historical enquiry, no comfortable answers can be constructed.
The only category into which we may place the composer with any condence,
one especially reserved for such mists, is that of the Interesting Historical Figure.
Thus, although the signicance of the composers work, certainly in the realm of
the keyboard sonata, is generally agreed, just how it is signicant is yet to be happily
established. Most treatments of composers and their music may be divided into two
categories, depending on where they locate the composers image the rationale for
the treatment is either one of reinforcement or one of special pleading, according to
whether the composer lies within or beyond the canon. The normal way of arguing
a case for the inclusion of music that lies outside the canon is to demonstrate its
relevance to or inuence on music that lies on the inside. Until the music or the
composer concerned have crossed the threshold, this is effectively the only mode of
treatment possible.
This may seem far too simple an equation, but one only need bear in mind the
difculty that has always been apparent in treating musical works of art on their
intrinsic merits, as it were. Warren Dwight Allen, after surveying musicological
writings spanning three hundred years, stressed the evolutionary current running
through all of them:
Some idea of progress, it seems, was xed immovably in the ideology of musicology, and this
was true whether musicologists dealt on the broadest scale with the music of widely separated
cultures or on a narrow scale with musical events of a single culture in close chronological
proximity. At every level music was treated in terms of its antecedents and consequents, not
as a thing in itself. Music passed through elementary stages to more advanced ones. What
was more advanced was almost always seen as better.
Given this rather bleak prognosis, now well accepted in principle if not so easily
avoided in practice, it is understandable that the only manoeuvre available to the
special pleaders is to make a case for their subject as an antecedent of or a consequent
This chapter is based on a paper given rst at the University of Auckland in March 1995 and subsequently in
shortened form at the British Musicology Conference, Kings College, London, in April 1996.
Joseph Kerman, Musicology (London: Fontana, 1985), 130. This represents Kermans summary of Allens ndings.
2 The keyboard sonatas of Domenico Scarlatti
to this or that composer, school, style. The reinforcers, on the other hand, are, even
if unconsciously, busy afrming the status of their subject as an advanced stage.
The place of Domenico Scarlatti in such a scheme, as suggested at the outset, is
decidedly tricky. While he does not count as a genuine outsider in the manner of an
Alkan or a Gesualdo, equally he does not t well into any of the habits of thought
through which we could expect to arrive at some construction of his signicance. His
father Alessandro, for instance, has long had a more secure place in history, although
presumably few would claim him to be a better or more signicant composer.
fact, Domenico might be regarded as a unique test case for the nature of musicology
as it has been practised in the last few generations, offering us a chance to reect on
its methodologies and priorities.
The circumstances of this claim to exclusiveness are worth reviewing. In every
conceivable musicological sense, Scarlatti is a problematic gure. For one, we know
remarkably few details regarding his life and views. Especially from the time he left
his native Italy to serve the Princess Mara B arbara as music tutor rst in her native
Portugal, then for the best part of thirty years in Spain until his death in 1757, we
only have the means to put together the most minimal of biographies. More than
one writer has commented that the scarcity of information almost seems to have
been the result of some deliberate conspiracy.
Given the fact that only one single
letter from the composer survives, such remarks are not altogether in jest. Related
to this dearth of hard facts is the lack of external evidence as to the composers
personality. Much has been made in the literature of the composers alleged passion
for gambling, with Mara B arbara at least once having had to pay off his gambling
debts, but even in this instance the verdict must be likely but not proven.
In the absence of information, the sonatas themselves have had to bear a good deal
of such interpretative weight, a happy situation, one would think, in the search for
the signicance of the composers work. In reality, though, the sonatas have often
been used as evidence for personality traits as this bears on the biographical picture
of Scarlatti rather than on the musical one. If we return for a moment to the matter
of comparative ideologies, it is probably fair to say that music has long invested more
capital in biographical portraiture than have the other arts. One rationale for needing
a good control over biographical circumstances has been that it will tell us a great
deal about the music that is the product of the personality the greater the control
over the life, the more acutely can we judge the works.
For Cecil Gray in 1928, however, Domenico was a gure of innitely smaller proportions and artistic signicance
than Alessandro; The History of Music (London: Kegan Paul, Trench and Trubner, 1928), 139. Writing in 1901,
Luigi Villanis stated: We will not nd in [Scarlatti] the profound musician that lived in his father; Domenico
Scarlatti, in Larte del clavicembalo in Italia (Bologna: Forni, 1969; reprint of original edition [Turin, 1901]), 166.
That such verdicts have become less likely in the more recent past tells us more about the decline of Alessandros
reputation than about any change in the critical fortunes of his son.
Malcolm Boyd, for instance, writes that it almost seems as if Domenico Scarlatti employed a cover-up agent
to remove all traces of his career . . . and contemporary diarists and correspondents could hardly have been less
informative if they had entered into a conspiracy of silence about him. Nova Scarlattiana, The Musical Times
126/1712 (1985), 589.
Scarlatti the Interesting Historical Figure 3
Stated thus, this equation also sounds too simple, but it is the best explanation
for the thrust of a good deal of musicological activity, whether applied to Scarlatti
or any other composer. The assumption that music is primarily an expression of
personality, of emotion, that in order to understand the music we must understand
the man and his private circumstances, is historically bound to nineteenth-century
music aesthetics, but it is a notion that has retained much of its strength through
to the present day. And it is one that colours our approach to all the art music of
at least the last few hundred years. Indeed, the notion has in the present scholarly
climate received a new lease of life, if in rather different intellectual conditions. With
the current emphasis on the situatedness of music, an engagement with its public,
social and political dimensions, the personal and emotional have been recovered for
inspection. Thus any sense of an ideally strict separation between artist and work,
or even person and persona, might be frowned upon as a species of puritanical
modernism. If investigation of the perceived historical personality of the composer
has to an extent been reclaimed as a legitimate object of study, it will naturally take
a more ideologically contingent slant than the great man approach of yesteryear.
Such interpretations must still rely, however, on an abundance of the sorts of data
which are in Scarlattis case simply not there. Given the paucity of biographical
information on Scarlatti, there has instead been the opportunity to grasp the music
in all its glory the sonatas constitute the only substantial hard facts that we have.
That opportunity has not been taken.
If this failure is due to the lack of evidence impeding the customary ow chart of
musicological procedure, it must not be construed that the holes are only biograph-
ical even more distressing is the impossibility of achieving good bibliographical
control over the composers works. The central problem is the complete absence
of autographs. The two principal sources for the sonatas are the volumes, almost
all copied by the same scribe, which are now housed in libraries in Parma and
Venice (hereafter generally referred to as P and V). Neither contains the full number
of about 550 authenticated sonatas, they contain the works in somewhat different
orders, and there is no agreement about which of the two copies is generally the
more authoritative. We cannot even be certain that the copies were prepared under
the direct supervision of the composer, although at least some input from Scarlatti
seems very likely. This lack of autographs means that no chronology for the sonatas
can be established. We can distinguish only two layers
amongst all the works
the rst 138 of the sonatas in the Kirkpatrick numbering
were copied into V or
published by 1749, thus xing a latest possible date for composition, and the rest,
copied between 1752 and 1757, may have been written earlier and/or later than
Joel Sheveloffs term in The Keyboard Music of Domenico Scarlatti: A Re-evaluation of the Present State of
Knowledge in the Light of the Sources (Ph.D. dissertation, Brandeis University, 1970), 196, where he avers that
the two groups of sources represent two denite though not completely separate layers of compositional activity.
This was rst contained in the Catalogue of Scarlatti Sonatas; and Table of Principal Sources in Approximately
Chronological Order near the end of Kirkpatricks seminal Domenico Scarlatti (Princeton: Princeton University
Press, 1953), 44256.
4 The keyboard sonatas of Domenico Scarlatti
this. Following Kirkpatricks lead, a chronology has often been assumed that runs
more or less in tandem with the sequence of copying of the works.
Much ink,
though, has been spilt lamenting the impossibility of truly determining the order of
composition of this vast corpus.
One might ask, though, just why it is so important to establish a chronology. The
standard answer must be so that we can trace the stylistic and creative development of
the sonatas. It is at this point that we must reect on Warren Dwight Allens ideology
of progress that underlies much musicological discourse. The lack of any chronology
for the Domenico Scarlatti sonatas means that they cannot be tted into the narrative
pattern whereby earlier, immature works lead to more rened and masterful ones,
whereby certain stylistic and creative elements gradually evolve while others fade
away, where, in other words, the individual works are made to tell a story in which
they function merely as pieces of evidence. A simple example of how chronology
may be used as a prop can be found in the case of Mozarts Piano Sonata in B at,
K. 333. It was regarded as a comparatively immature and unremarkable work when
its provenance was thought to be about 1778, its signicance perhaps residing in the
hints it gave of future work, but Alan Tysons study of paper types has not so long ago
established that its date of composition was in fact late 1783.
Since then the work has
been credited with previously unsuspected qualities and now reects the concerns of
the mature piano concertos that were about to be written. From this perspective,
one can only hope that no dated Scarlatti sonata autographs ever come to light, since
a knowledge of their chronology can only force a further distortion on this body
of music. (Not that such distortions can be altogether avoided: without attening
out the particulars in a body of information, how can we know anything at all?)
One might have thought, again, that the absence of this information would have
driven scholars into a more direct confrontation with the works themselves, but
by and large there has instead been a good deal of hand-wringing and a retreat
into other problems of documentation, transmission and organology. Admittedly,
these are once more rather intractable. For instance, Scarlatti has traditionally been
regarded as the composer who wrote as idiomatically and comprehensively for the
harpsichord as Chopin did for the piano of his time. However, recent research has
suggested conclusions that sit uncomfortably with the idea of the composers work
representing a nal owering of harpsichord style and technique. Not only are the
majority of the sonatas playable on the pianos owned by Mara B arbara, at least
those accounted for in her will, but there is strong circumstantial evidence linking
Scarlatti with the history and promulgation of the early fortepiano.
Another issue
The dates of the manuscripts prepared by the Queens copyists seem to correspond at least roughly with the
order in which the sonatas were composed. Kirkpatrick, Scarlatti, 144.
See The Date of Mozarts Piano Sonata in B at, K. 333/315c: The Linz Sonata?, in Musik, Edition, Interpre-
tation: Gedenkschrift G unter Henle, ed. Martin Bente (Munich: Henle, 1980), 44754.
See for example David Sutherland, Domenico Scarlatti and the Florentine Piano, Early Music 23/2 (1995),
24356, and Sheveloff, Domenico Scarlatti: Tercentenary Frustrations (Part II), The Musical Quarterly 72/1
(1986), 90101.
Scarlatti the Interesting Historical Figure 5
concerns the possibility that the majority of the sonatas were conceived in same-
key pairs. Naturally enough, amidst the heat generated by this dispute, the question
of the artistic status of the pairings has been insufciently addressed. Occasionally
pairs have been examined for thematic connections of a rudimentary kind, which
barely scratches the surface of the matter. All that the originator of the idea, Ralph
Kirkpatrick, could really offer was the formula that the relationship between pairs
was one of either contrast or complementarity.
This could cover a multitude of
sonatas in the same key.
Another concern, one that Scarlatti research has mostly addressed with a bad
conscience, is the matter of Spanish folk inuence. Some have claimed that certain
sonatas amount to virtual transcriptions of amenco or folk idioms, while others have
tried to minimize its import. Italian writers have often preferred to nd in Scarlatti
an embodiment of Mediterranean light and logic. A typical sentiment comes from
Gian Francesco Malipiero: far more than the Spaniard of the habanera or malague na,
which make their transient apparitions, it is the Neapolitan who predominates with
the typical rhythms of the Italians born at the foot of Vesuvius. Domenico Scarlatti,
in fact, is a worthy son of Parthenope; mindful of Vesuvius, he loves to play with
light and re, but only for the greater joy of humanity.
This is just a variant of a common strain in the literature on all Latinate composers,
from Couperin to Debussy, whose achievements can only be dened in opposition
to the assumed creative habits of the Austro-Germanic mainstream: their music
lives by lightness, delicacy, precision, logic and all the rest. More surprising, on the
surface, is that Spaniards have mostly been reluctant to deal with questions of folk
inuence, and indeed with Domenico Scarlatti at all. Whether this suggests a bad
conscience or not, in a strange way this may be allied with the too easy assumption
by Italian writers that Scarlatti counts rmly as one of their own. The extent of the
Scarlatti literature in Italian is in fact not so great in its own right, suggesting that
nationalistic considerations have played a part here too. In other words, another of
the things that Scarlatti does not belong to is a country. He thus lacks the weight
of an entire culture industry behind him.
Nationalism is of course another of
those properties that we dene in relation to mostly Germanic and nineteenth-
century norms. We are barely aware any more of the nationalist agendas of German
writers past and present, just as it is difcult for us to hear the ethnic accents in
German music, so rmly does it constitute the mainstreamof our musical experience.
Hence when trying to make something of Scarlattis music we are not readily able
to align him, at least as a point of reference, with the art music of a particular
There are various lower-level features to the sonatas that have also proved to be
stumbling blocks in the literature. There is, for instance, a marked inconsistency in the
See Kirkpatrick, Scarlatti, 143.
Domenico Scarlatti, The Musical Quarterly 13/3 (1927), 488.
A comparable eighteenth-century case is that of Zelenka. Michael Talbot notes the cultural problem [of]
ownership of the composer in his review of Jan Dismas Zelenka (16791745): A Bohemian Musician at the
Court of Dresden by Janice B. Stockigt, Music and Letters 83/1 (2002), 115.
6 The keyboard sonatas of Domenico Scarlatti
sources ornamental indications, so frequent that this cannot simply be put down to
scribal error. Performers (and editors) overwhelmingly correct these inconsistencies
so that parallel places contain parallel ornamentation, so tidying up their scripts well
beyond any claims for licence as understood from eighteenth-century performance
practice. Few players seemed to have stopped to consider whether it is precisely our
instinct for such symmetrical tidying that the composer is playing with. All this is by
way of re-emphasizing that almost all the effort in the Scarlatti literature has gone
into problems of evidence which will be amplied in the more detailed survey of
the literature that follows in Chapter 2 and very little into critical interpretation.
The rationale for this is apparent enough, and only reects in extreme form the
customary work habits of musicology as a whole (extreme form because the amount
of evidence that can be dealt with is so comparatively slight). Back in 1949 Curt
Sachs entertained thoughts relevant to our consideration of the nature of Scarlatti
Do not say: Wait! We are not yet ready; we have not yet dug up sufcient details to venture
on such a daring generality. There you are wrong. This argument is already worn out,
although it will none the less be heard a hundred years from now, at a time when specialized
research has lled and overooded our libraries so completely that the librarians will have to
stack the books and journals on the sidewalks outside the buildings. Do not say: Wait! The
nothing-but-specialist nowdoes not, and never will, deemthe time ripe for the interpretation
of his facts. For the refusal of cultural interpretation is . . . conditioned by the temperaments
of individual men, not by the plentifulness or scarcity of materials.
Scarlatti research may thus be seen to have painted itself into something of a corner,
virtually denying the admissibility of critical interpretation until more facts become
But why relive past battles? This questioning of positivistic rigour may seem
no longer necessary; havent we established new contexts for investigation, indeed
new denitions of what knowledge we are after? Yet musicology remains highly
dependent on outside reinforcements for its assumed methodologies and for its sense
of self. A strong allegiance to scientic method has been replaced, at least at the
cutting edge, by a strong allegiance to interdisciplinarity, with particular emphasis
on literary studies. This interest has barely been reciprocated. Also uniting old and
newis the consequent skirting of what Scott Burnhamcalls our fundamental relation
to the materiality of music.
The very notion that the music exists as a self-evident
category for investigation has become highly compromised, of course, but what is
meant here goes beyond the usual considerations of the work concept. It means being
able to x on the corporality of the art the way, through our understanding of its
grammar and feeling for its gesture, that music incites our physical involvement and
so renews a claim to be self-determining and intrinsically meaningful.
There has
Cited in Kerman, Musicology, 127.
Theorists and The Music Itself , Journal of Musicology 15/3 (1997), 325.
Note in this respect the contention of Charles Rosen that in so far as music is an expressive art, it is pre-verbal,
not post-verbal. Its effects are at the level of the nerves and not of the sentiments. The Classical Style: Haydn,
Mozart, Beethoven (London: Faber, 1971), 173.
Scarlatti the Interesting Historical Figure 7
on the whole been a failure in the discipline to address the study of music in this most
concrete sense: we have been so busy problematizing the status and apprehension
of music that we do not square up to its sensuous material impact. The issue of
materiality, indeed, can be raised with particular urgency in the case of Domenico
Scarlatti, given some of the most striking traits of his music.
There is in any case another side of the story that must be conceded. Joel Sheveloff,
the doyen of Scarlatti sonata scholars, has often warned of the need to tread with
great caution, given the many uncertainties surrounding text and transmission.
The details of Scarlattis style remain so comparatively strange to us that the inability
even to establish highly authoritative texts affects our global view of the composer
far more seriously than might normally be the case; our perception of his style, after
all, is dependent on the accumulated impression of a wealth of details. When so
many of these details vary from source to source or simply remain ambiguous, then
particular scholarly care may indeed be in order. Postmodern musicology can afford
to disdain the methods of positivism when so much of the dirty work has already
been done; it still nds uses for much of the material thus created. It is another matter
altogether to launch oneself beyond such concerns when, as is the case with Scarlatti,
there is often the thinnest of documentary bases. With future progress along such
lines looking to be highly unlikely, barring a major breakthrough, it may be time to
gamble a little.
This is the dilemma facing any fresh approach to Scarlatti. Postmodern musicol-
ogy does not necessarily allow much more room for manoeuvre given the state of
knowledge than do the more traditional methods. Indeed, while the type of con-
texts sought may have changed, there is now a stronger sense that music may not
be approached in the raw. This is guided by the conviction that what we call the
music is constructed according to various perceptual and cultural categories and is
not innate; it is not simply there for universal access. Nor can one underestimate
the impact of documentary difculties. Imagine, for example, what the state of play
might be in the literature on Beethovens symphonies or Verdis operas without a
knowledge of chronology and a comforting array of documentation. What could
one write and, indeed, how could one write were all this contextualizing material
This is not to imply that there does not exist a fairly substantial body of commen-
tary on the sonatas themselves. Unfortunately, with hardly any exceptions this has
dealt with the sonatas rather than sonatas, discussed according to a few well-worn
notions. Characteristic features such as the harsh dissonances, the freakish leaps and
all the other technical paraphernalia are accounted for, Spanish elements are men-
tioned, as are other impressionistic
features such as the employment of fanfares,
street cries and processional material, and there is often evidence of a form fetish
occasioned by the use of the term sonata itself for these pieces. Most writings on
See for instance Sheveloff, Frustrations [I], 422 and 428. This article and its successor, cited above in fn 9, will
hereafter be referred to as Frustrations I and Frustrations II respectively.
I borrow this term from Donald Jay Grout, A History of Western Music, rev. edn (New York: Norton, 1973), 456,
without necessarily dissenting from all its implications.
8 The keyboard sonatas of Domenico Scarlatti
the sonatas, however, fail to go much beyond this level of characteristic features and
therefore tell us little about the dynamics of the individual work. Underlying such
approaches may be the subtext that, however splendid the results, the Scarlatti sonatas
are a product of a transitional style and a mannerist aesthetic from which too much
coherence should not be expected. Accordingly the literature emphasizes freedom
and improvisation and variety rather than seeking to investigate the composers sense
of musical argument as conducted in individual works. It takes refuge in evocation.
If we want a deeper understanding of Scarlattis style, though, and of the part his
work plays in the development of eighteenth-century musical language, there is no
substitute for a detailed reading of particular sonatas, informed by a reassessment of
what constitutes a context in the case of Scarlatti.
Reference just now to the development of eighteenth-century musical language
may appear to t uneasily with the earlier dismissal of ideologies of progress, yet
there need be no injury as long as development is not taken to suggest the sort of
inexorable improvement and organic growth of a style that it all too often connotes.
Not only that, but the monsters of evolutionary ideology, labels for musical periods,
are indispensable in attempting to get closer to Scarlattis achievement. That the
composer has one foot in the Baroque and one in the Classical era is one of the
commonplaces in his reception history, and, although this very fact has ensured
marginal status for Scarlatti in all history textbooks since he does not clearly belong
to either period it can be turned to account in a more useful way than suspected.
My contention is that, due to the circumstances of his life, which involved near
incredible changes in environment and professional demands, and obviously even
more due to his creative turn of mind, Scarlatti was acutely conscious of his own
style. This in effect meant being conscious of styles, of various options for musical
conduct. After all, the composer at various points of his career found himself in
positions as different as writing operas for an exiled Polish queen, acting as chapel
master at the Cappella Giulia in the Vatican, and being music tutor within a Spanish
royal family of strange disposition in a strange environment. What these changes
may have promoted, or merely conrmed, was a reluctance on the composers part
to identify himself with any one mode of speech in the keyboard sonatas, to make
a virtue out of not belonging, or not wanting to belong. Of course all composers
are to a greater or lesser extent conscious of their own style, and the eighteenth
century saw many composers addressing the perceived stylistic pluralism of musical
Europe, but what I think makes this a distinguishing mark of Scarlatti is that none
of the styles or modes of utterance of which he avails himself seems to be called
A simple example of this property can be heard in the Sonata in A major, K. 39,
shown in part in Ex. 1.1. This work has the virtue, for present purposes, of corre-
sponding to most listeners idea of a typical piece of Scarlatti. Its stylistic starting point
is undoubtedly the early eighteenth-century toccata of the moto perpetuo type. It is
not hard to understand the way in which writers can lapse into a mode of superlative
evocation when attempting commentary on such music; it seems to invite all the
Scarlatti the Interesting Historical Figure 9
Ex. 1.1 K. 39 bars 617
stock references to vitality and virtuosity. Yet it seems to me that the almost obscene
energy of the piece is harnessed to a particular end, that of taking Baroque motor
rhythms beyond the point where they can sustain their normal function. Instead of
being agents of propulsion, they take over the piece and threaten to strip it of any
other content. Only the references to the repeated-note gure of the opening hold
the piece together. Especially notable is the overlong ascending progression of the
rst half (bars 7
), which seems to represent a nightmare vision of sequences
without end, allowed to run riot.
What is typical about this sonata is its swiftness and athleticism, and for once we
must reverse the claims of stereotyping to make an important observation. There
Sheveloff, Kirkpatrick and Giorgio Pestelli all mention the connection between this sonata and K. 24, to the
detriment of the former. See Sheveloff, Frustrations I, 416; Pestelli, Le sonate di Domenico Scarlatti: proposta di
un ordinamento cronologico (Turin: Giappichelli, 1967), 158; and Kirkpatrick, Scarlatti, 1556. Surely, though, it
is only the openings and closings of the halves that are so similar. Aside from that, K. 39 has an independent
10 The keyboard sonatas of Domenico Scarlatti
can be no doubt that a high proportion of the Scarlatti sonatas are fast and, if
one will, loud. It seems that it is the generally more responsible critics who try
hardest to mollify this fact, stressing the variety of the composers moods, his ability
to write slower and apparently more heartfelt movements as well. A good many
performers also seem conscious of not wanting to play Scarlatti up to his reputation,
and consequently they invest their performances with what seems to me a false
gravitas; by slowing the speed of execution down, they obviously hope to make the
composer sound more serious.
But there is no getting around the fastness of the
majority of Scarlatti sonatas.
What is wrong with speed? Once more the problem lies with our nineteenth-
century ears. Ironically for an age thoroughly associated with the so-called rise of the
virtuoso, the nineteenth century also bequeathed us a suspicion of virtuosity, which
for our purposes may be translated as a suspicion of prolonged displays of virtuosity at
high speed. Only so much may be allowed, the received opinion seems to go, before
there must be a return to real invention: the exposing and development of themes.
One senses a comparable response to the totality of Scarlatti sonatas: fast movements
are all very well, but if only there werent so many of them the composers image
might be more solid. (When Brahms sent a volume of Scarlatti sonatas to his friend
Theodor Billroth, he wrote You will certainly enjoy these as long as you dont
play too many at a time, just measured doses.
Too much unhealthy excitement was
evidently to be avoided.) Unfortunately, our cultural conditioning means that for us
serious is cognate with slow, or at least a moderate speed: thus the Beethoven slow
movement represents the ultimate in depth of communication, the Mahler slow
movement is intrinsically more worthy of contemplation than the Mendelssohn
scherzo. These terms are bound up with a discursive model for composition, the
highest to which instrumental music can aspire in nineteenth-century aesthetics
presumably the reason why speed kills is that it does not readily allow time for
the perception of an unfolding musical plot. While there are many Scarlatti sonatas
which could involve a possible dramatic or narrative sequence, loosely understood,
for many others we will have to nd alternative models that can satisfy us intellectually
and obviate the need to be apologists. If our conditioning suggests to us that the
business of music is above all emotional or mental expression, we can consider as
an alternative the notion of music as bodily expression. In the case of Domenico
Scarlatti, the simplest way of saying this is music as dance.
Dance in this sense is not necessarily meant to call to mind minuets and waltzes, and
not even the various Iberian and Italian forms that may have inspired the composer;
Note Christophe Roussets assumption that the performer preparing a recital will want to include a certain
number of slow movements to allow some air into the programme, where the speed and exuberance of Scarlatti
risk becoming tiring. Approche statistique des sonates, in Domenico Scarlatti: 13 Recherches, proceedings of
conference in Nice on 1115 December 1985 (Nice: Soci et e de musique ancienne de Nice, 1986), 79.
Cited in Eric Sams, Zwei Brahms-R atsel,

Osterreichische Musikzeitschrift 27/12 (1972), 84.
Compare the hypothesis of Ray Jackendoff, also proceeding from the parallel with dance, that musical structures
are placed most directly in correspondence with the level of body representation rather than with conceptual
structure. Consciousness and the Computational Mind (Cambridge, Mass.: MIT Press, 1987), 239.
Scarlatti the Interesting Historical Figure 11
it is simply to suggest that music may function balletically as well as, or instead of,
discursively. Our inclination to place one above the other as an object for study
and contemplation may or may not have an inherent aesthetic justication, but it
seems to me to be another symptom of musics unsure sense of itself: we are happiest
when accommodating those works that suggest literary models or parallels, just as
nineteenth-century musical culture addressed itself constantly to literature.
The D major Sonata, K. 277 (Ex. 1.2), may, as we shall see, contain its own plot,
but I have chosen it for consideration in the rst instance because it will enable
us to focus on the composers awareness of style, indeed, on the construct of style
altogether. To return to Curt Sachs, we may be not yet ready for an approach to
this individual sonata and to the two that follow, but a confrontation in at the deep
end, as it were with some of the music that animates my whole enterprise may
suggest to the reader the urgency and fascination of the task.
The natural lyrical eloquence at the start of K. 277 is a quality that Scarlatti nor-
mally feels the need to shape in some overt way; he is rarely content with an idyll,
preferring to give such pieces a sense of dramatic progression. Temperament be-
comes a foil for the lyricism, with a strong sense of creative intervention in what
can in fact become quite an impersonal mode; witness for example Bachs Air on
the G string. Only in anachronistic nineteenth-century terms can we hear the
lyricism of Bachs movement as involving the expression of personal or individual
emotion. If the Air does indeed express grief or nostalgia, then it must be heard as
collective in its import; note also in this regard the measure of control provided
by the consistent movement of its bass line. Scarlatti is not at all interested in such
means or ends; to invoke our style labels once again, his starting point is the galant
notion of the individual lyrical voice. This is reinforced by many aspects of diction in
the opening material, with its small-scale, detailed inections of melodic writing
the Lombard rhythms, grace notes, appoggiaturas, and Schleifer-type gures.
All these, along with the very indications Cantabile and andantino, are mark-
ers of the galant. Such miniaturism helps to delineate a voice that does not speak
on the basis of collective authority or experience, but as if on behalf of the lone
A more important ingredient for the shaping of the whole work, though, it seems
to me, is folk music, and perhaps Spanish amenco in particular. K. 277 contains
nothing whatever on the surface that suggests this, but the sort of inuence meant is
more profound than the appropriation of various idiomatic features. Contact with
such a folk art seems to have made this composer acutely aware of the gap between
folk idiom and its expressive world and the way art music in contrast behaves. It is a
distinction between distance and control and what is perceived as a musical present
tense. For all that the galant may as a point of departure represent comparative
A Schleifer is normally a gure of three notes covering the interval of a third, the rst two rapidly played to act
as a decoration to the nal one. The classic form of the gure is found at the beginning of bar 12, but there are
many variants to be found, for instance at bars 13
or 8
12 The keyboard sonatas of Domenico Scarlatti
Ex. 1.2 K. 277 bars 140
freedom of action, in the context of the whole work its claims to just that freedom
are undermined. The musical present tense referred to enters when the normal style
of melodic speech disappears, at bar 27; this is particularly marked given the detailed
inections of the previous writing as described before. At bar 27 the melodic voice
seems to stop, to be replaced by undifferentiated rhythmic movement in consistent
four-part crotchet chords, with unpredictable and complex harmonic movement.
The top line does not of course lose all melodic character, but in this context it
seems like a skeleton. The most expressive part of the sonata is therefore the most
Scarlatti the Interesting Historical Figure 13
Ex. 1.2 (cont.)
plain, the least mediated stylistically in the terms of the rest of the piece, it may be
regarded as primitive.
If the harmonic movement from bar 27 is the most striking feature of this passage,
this may protably be compared with the opening. Part of the delicacy of the idiom
here is the lack of decisive bass movement; instead the bass moves in small steps. The
rst two bars express the tonic by means of neighbour-note formations, and indeed
the rst strong perfect cadence does not occur until the end of the rst half. In this
14 The keyboard sonatas of Domenico Scarlatti
respect and in its high tessitura, leaving the conventional bass register largely vacant,
it seems to be formed in deliberate opposition to the solid, continuo-like bass lines
of the Baroque. The rst break to the idyll occurs at bar 16, with the unexpected
repetition of the cadential unit. After the undidactic freedom of organization of the
earlier music, with melodic ideas shifting in and out of focus,
the sudden square
formality of the repetition at 16 arrests our attention. The resumption of the material
of this repeated bar at 20 strengthens the sense of the intervening passage (bars 1719)
as a minore insertion. It casts a shadow without proving too disruptive. That it does
represent a break with the uid galant diction, however, is remarkably conrmed
at the outset of the really signicant interruption. The rst beat of bar 27 picks up
on precisely the pitches that began bar 17, c
, b
and e
, here verticalized into a
thoroughly characteristic dissonance. It is also signicant that the rst beat of bar 17
contains the last Lombard rhythm of the piece.
The opening of the second half may seem reassuring enough, but it is disruptive in
its own way. The answering unit of bar 2 has now become an opening gambit. The
expressive weight of bar 2 is helped in context by the registral isolation of the G-F
progression in the right hand, followed as it is by a jump to a
in bar 3. Bars 256
in fact exploit this feature by their turn to B minor, featuring As. The interrupting
passage then seems to energize the unit beyond its previous manifestations. At bar 31
the melodic range is wider, as is the whole tessitura, and the texture is heavier. After
this the gure is made to settle down until it resumes the likeness of the opening.
Thus bar 33 is identical with bar 2 (and bar 24), but now with a more unequivocal
closing function; in conjunction with this, the c
succession in the right hand
of bar 32 suggests the same pitches as in the very rst bar.
It is almost as if we have turned full circle, although such an expression sug-
gests a satisfying dramatic symmetry that is not present. The rupturing force of the
outburst note especially the crude voice leading of bar 28
, which is so remote
from any notion of galanterie may allow the return of the opening gures, but these
could be understood as remnants. All the most characteristic aspects of the melodic
writing fail to reappear at all, creating a binary form that is very far from being bal-
anced. Instead of such a resumption, frombar 34 we hear continuous melodic triplets
that are a far cry from the rather small-scale diction of the rst half, but this style is
equally remote from the plain crotchets of the interruption. Materially, it takes its
cue from elements in the rst half bars 34 and 37, for instance, allude once more to
bar 3 but the melodic triplets almost seem like a means of regaining equilibrium
after the unexpected outburst.
This stream of song seems to inhabit a different sphere, almost as if it is a com-
mentary on both the preceding vehement expression and the galant gestures of the
rst half. What are we to make of this sonata as a total structure and what can we
compare it with to comprehend it? We hear a succession of three radically different
Note, for example, the parallelism of descending units at 3 (from g
), 8 (f
), 12 (e
), then 18 (from d
, with the
preceding e
functioning in this light as a quasi-appoggiatura). This parallelism does not coincide with structural
or phrase boundaries and hence may be heard as a free association of material, personal in organization.
Scarlatti the Interesting Historical Figure 15
rhythmicmelodic types with barely any interaction between them galant nicety,
plain crotchets that would deny any melodic nesse,
and then an endless melody.
Both latter types are preceded by three bars of the opening gesture repeated, as if
to give a point of comparison. From this perspective, the material of the opening
two bars could be conceived as a kind of frame, a sort of ritornello that provides
the cement for an out-and-out progressive form. Rather than the question mark
provided by this reading of the structure, with the composer reviewing various styles
and forms of expression without committing himself to any of them, a more opti-
mistic interpretation is possible. Bars 34ff. may be heard as a kind of liberation: the
brutal interruption of the galant melodic style, a codied and socially determined
expression of the individual voice, allows for the entry of a purer form of song,
which we are to understand as a more genuinely personal voice. No matter which
interpretation is nally more congenial, one must repeat that the essential genius of
the structure may well owe its provenance to an engagement with folk music, and its
implications for the means chosen by art music. This, I contend, lifted Domenico
Scarlatti right out of all notions of expressive routine and settled styles, encouraging
the sort of fruitful creative schizophrenia on display in K. 277.
In spite of the evidence of this and many another sonata, received opinion is that
Scarlatti was either unconnected with the galant as a style or extremely indifferent
to it. His one surviving personal letter, written to the Duke of Huescar in 1752,
is often cited in support of this contention.
In it he makes a familiar lament on
the poor compositional standards of the younger generation, claiming that few of
them now understand [la] vera legge di scrivere in contrapunto- the true laws of
writing counterpoint.
The letter has always been taken at face value; it seems
somehow indicative that one of the few pieces of hard evidence we have has been
so objectively interpreted in other words, misinterpreted, in my view. Not only
does the musical evidence disprove the notion that Scarlatti was out of sympathy
with or uninterested in newfangled styles like the galant K. 277 cannot be heard
simply as a besting of the idiom but a calm acceptance of the composers ringing
words on counterpoint is contradicted by the reality of the sonata texts themselves.
Such a contradiction can be found in the C minor Sonata, K. 254.
This sonata, written almost entirely in two parts to an extent actually very rare
in Scarlatti, may be thought of as a skit on counterpoint, or an invention gone
wrong. A good many Scarlatti sonatas do in fact begin with imitation between the
hands, but in the majority of cases this has no larger consequences for the texture of
the work. Here, however, the opening, suggesting the learned style in its use of a
In his recording of the work (Deutsche Harmonia Mundi: 05472 77274 2, 1992) Andreas Staier adds a trill at
and splits the right-hand thirds of bar 30
into unfolded quavers, as if uncomfortable with the nakedness
of this passage.
For example by Eveline Andreani, Autour de la musique sacr ee de Domenico Scarlatti, in Domenico Scarlatti:
13 Recherches, 99; Francesco Degrada, Tre Lettere Amorose di Domenico Scarlatti, Il saggiatore musicale 4/2
(1997), 300301; and Sebastiano Arturo Luciani, Domenico Scarlatti. I: Note biograche, Rassegna musicale
11/12 (1938), 469.
The original text is contained in Luciani, Note I, 469, and Kirkpatrick, Scarlatti, 121, offers a translation.
16 The keyboard sonatas of Domenico Scarlatti
Ex. 1.3a K. 254 bars 1524
typical contrapuntal tag,
is taken as a pretext for the examination of various types
of counterpoint, mostly of a fairly bizarre sort. From bar 10 we hear in the left
hand an alla zoppa, or limping, gure, counterpointed against a straight-crotchet
right hand in a concertina-like pitch construction. The effect of this is indeed rather
lame, especially after the decisive opening and energetic continuation. From bar 17
the contrary motion between the parts is replaced by imitation, which goes badly
wrong, with the consecutive fourths at 19 and 23 having an obviously ugly effect
(see Ex. 1.3a). Even worse, the rst of each is an unresolved tritone. Slightly more
hidden are the parallel fths that follow on from these fourths in the same bars. The
true laws of writing counterpoint are not much in evidence here.
From bar 33 the previous methods of parallel and contrary motion between
the two parts are combined, but the result is much messier than this sounds. The
real relevance of this passage is more that it continues the ways of unsuccessfully
combining independent and notionally equal parts. The right hand especially here
has the avour of a voice in species counterpoint or a conventional ller motion
in a contrapuntal texture. Note too the staggered parallel fths at 335. Altogether
the passage sounds distended well beyond any functional basis. The right-hand part
moves down an octave before reversing its direction, as if to avoid a continuation
of the consecutives; meanwhile the left hand strides pompously down nearly three
octaves in an unchanging dotted rhythm. The literal repetition of the whole phrase
only emphasizes its uncertain import. The piece in fact seems to be going around
in circles.
One almost wonders whether the work has a specic target, whether
in fact it is a satire. Certainly the inconsequentiality of the contrapuntal textures
and the signs of mock ineptitude are hard to miss. At least one would think so;
This tag is virtually identical with that which opens K. 240, where it is, however, just one element in a very
heterogeneous sonata. Compare also the start of K. 463.
Note also the unexpected and awkwardly timed return of bars 6ff. at 25ff.; in addition, the cadential bar 32
recurs at 39 and 46, the passage from bar 10 is reworked from 29, and the left-hand line at this point recurs in
toto at 369 and 436.
Scarlatti the Interesting Historical Figure 17
Ex. 1.3b K. 254 bars 92101
in his recording of the complete sonatas, Scott Rosss version of the work is not
only soberly paced in the manner discussed before but nds a number of ways to
soften the harsh prole of the piece.
This is symptomatic of the embarrassment
that the composer often induces in the contemporary performer, who prefers to
retreat into the sort of good taste that may be rather more appropriate for various
contemporary keyboard repertories.
This softening is particularly unwelcome since the composer himself attempts
something of the sort shortly after the double bar. From bar 57 we hear a far more
acceptable form of imitative texture; even though the parallel fourths remain at
bars 58 and 60, they grate much less than those heard in the rst half.
At bars
612 we again hear earlier material that is contextually sounder and more directed;
the material from bar 10 is limited to two bars in duration and acts as a successful
transition. Another solution of a sort follows, when from bar 63 the opening tag is
reused four times in succession, as at the start of both halves of the piece. Here the
tag is transformed into a little galant episode; it is put into a homophonic setting and
becomes cadential rather than enunciatory. The change in texture is signicant, with
a striking move to three parts instead of the two associated with the would-be strict
style. The purpose of this transformation would seem to be to mock the pretensions
of the opening more directly than the intervening matter has already done.
This improvement in technique does not last, though, and the passage from bar
85 sounds even more confused than its rst-half equivalent. The right hand changes
direction more unpredictably, and the repetition of the phrase from bar 89 is now
For instance, he changes manual in the repetition of bars 339, to create an echo effect, and adds a number of
ornaments which to me suggest a civilizing inuence (Erato: 2292 45309 2, 1989). This complete recording
was made in 19845, and so nished in time for a tercentenary presentation on Radio France, in a series of more
than 200 broadcasts. Commercial release then took several more years.
This of course depends on the performance of the ornaments here if one realizes the appoggiatura and its
resolution in a minimcrotchet rhythm, then parallel fths will result! The very fact of the newnotation, however,
with the leeway in performance it allows compared to the original at bar 19, seems to signify some mollication.
18 The keyboard sonatas of Domenico Scarlatti
staggered to begin halfway through the bar. From bar 94, though, we have one of
the composers most striking inspirations. With any reasonable agreement among
the parts and hands obviously doomed to fail, here unanimity and coordination are
explicitly achieved in each hand successively (see Ex. 1.3b). Here nally there is
perfect imitation between the hands, but in a context that is clearly not contrapuntal
in any standard way. The change of texture and use of parallel sixths are enormously
striking in such a context, as is the change to stichomythic units after the prevailing
long-windedness of the syntax. The passage has a strong avour of elbowing out of
the way the previous nonsense. The repeated right-hand line from 98 also seems to
be part of the attempt to block the annoyances of previous material. In effect the
composer dramatically abandons the textural and syntactical premises of the piece.
In defence of the Ross recording, it must be said that such a work, like many others
by Scarlatti, is rather exhausting for the listener and performer to cope with. Alain de
Chambure has written of the slightly chaotic charm of the sonata,
which makes
it sound gentler than it really is. The intermittent ugliness and sprawl, even if to
parodistic ends, ask hard questions of what we are to prepared to accept in the name
of art music.
K. 193 in E at major also begins with an imitative point, but one that is rather
more problematic in execution (see Ex. 1.4a). The imitation in the second bar
immediately goes wrong, the left hand imitating at the seventh, without an initial
small note, which is then restored in bar 3 in both hands. The parallel tenths of
bar 3 also correct the very exposed parallel fourths of the previous bar, echoing
those we heard in K. 254. Bar 2 once again raises the issue of Scarlattis attitude to
counterpoint, and therefore, by implication, to the traditional musical values with
which it is associated. The composers tendency to abuse common practice in this
way exemplies what Giorgio Pestelli refers to as a quality of disdain in the sonatas.
Scarlatti often uses worldly trappings as a starting point for his structures here the
respectability of proceeding from an imitative point, in K. 277 a cantabile line of
the purest galant pedigree and then skews or discards them, often showing them
up by the passionate prole of later material. As well as a simple disdain for certain
conventions, the quality may also be dened as an unwillingness on the composers
part to be heard to be spelling out any creative intentions, and a reluctance to give full
elaboration to an affect (suggesting a strongly anti-Baroque orientation). It also seems
that the composer is not seeking approval through musical good behaviour. The
pride and delight in technique shown by Mozart, for example, are foreign to Scarlatti;
he is not so much a pragmatist as hostile to customary notions of craftsmanship. And
so artistically, as well as indeed historically, the composer seems to prefer not to
Catalogue analytique de loeuvre pour clavier de Domenico Scarlatti: guide de lint egrale enregistr ee par Scott Ross (Paris:
Editions Costallat, 1987), 99. He also writes, perhaps less acutely, that this uncomplicated little sonata appears
to be an experiment in the staggering of imitation voices.
See The Music of Domenico Scarlatti, in Domenico Scarlatti: Groe Jubil aen im Europ aischen Jahr der
Musik (Kulturzentrum Beato Pietro Berno Ascona: Ausstellung 24 August30 October 1985), second edn
(GermanEnglish) (Locarno: Pedrazzini Editions, 1985), 84.
Scarlatti the Interesting Historical Figure 19
Ex. 1.4a K. 193 bars 149
belong to the club. This can be seen too in the shaping of the rst ve-bar unit.
Given that Scarlatti does reuse its characteristic rhythm throughout the piece, can
this unit be described as a theme? It comprises just a scrambled opening and then
a cadence.
This question of terminology is again relevant to our immersion in nineteenth-
century models for musical conduct. We are used to understanding theme as being
cognate with idea. Of course, we would never expect the two to be identical, but
in practice we would expect an opening theme to have a good deal to do with
the creative idea of a work. In Scarlatti, on the other hand, we have a composer
who is almost uniquely offhand about his openings; only Haydn can compete in this
20 The keyboard sonatas of Domenico Scarlatti
Ex. 1.4a (cont.)
respect. (With Haydn, though, obstacles are generally set up as a creative challenge
to overcome. While this applies often enough to Scarlatti too, there can be another
sense that the obstacles are there to throw us off his trail.) The ideas behind the music
seem often to have nothing to do with any theme that we can recognize, yet our
intellectual habits tell us that any opening must be taken seriously and regarded as
some sort of denitive or purposive creative statement.
Scarlatti in fact provides his own commentary on the opening theme. At bar 6
he immediately moves away from the tonic, as if he wants to leave the mess behind.
Tellingly, the syntax becomes very square and solid, with prefabricated units moving
sequentially and by the circle of fths. The parallel sixths of bars 1012 and 1820
Scarlatti the Interesting Historical Figure 21
seem to represent an explicit correction of the parallel fourths of bar 2, this being
emphasized by the rhythmic identity of the respective units. This passage is succeeded
at bar 22 by an overt evocation of folk style. Barbara Zuber has nicely described the
subsequent material as a modal island;
diatonic progression is replaced by static
modal coloration, the prior duple organization is replaced by very distinctive three-
bar units. The harmony here should perhaps be understood less as V of B at minor
than as F Phrygian, with the left hand emphasizing the semitone of the descending
minor tetrachord BAGF. The form taken by this tetrachord, with raised third
and attened second, is, according to Jane Clark, typical of the Moorish version of
the Phrygian scale as commonly found in Andalusian folk music.
The right hands
alternation between raised and lowered forms of g
and a
is also a common property
of Andalusian chromaticism.
However, for all their extreme contrast, these three-
bar units also contract the pattern of the two previous eight-bar units: a scalic rise
leads to a fall followed by an appoggiatura ending.
As if thrown off course by such a rupture of musical style, the harmony in bars
345 retreats to VI of the tonic, E at major. These bars almost function as an
ironic echo of the modal scale activity. Compare for instance 345 with 234:
1. The right hand of bar 34 replicates the descending contour of 23 but takes its
rhythmic form from the preceding bar 22.
2. The right hand of bar 35 replicates the appoggiatura shape and rhythm found in
the right hand of bar 24.
3. In bar 34 the left hand contains the same repeated-note cell as 23 (and 22), but the
previous biting dissonance of a semitone, f
, is softened to a more standard
major seventh, Ba.
4. The bass motives in bars 35 and 24 are identical.
A fundamental difference, however, lies in the return to two-bar phrase units. Or
so we assume; but the sequential progression continued by bar 36 is cut dead by the
advent of a new phrase in 37, yielding another three-bar unit from 34 to 36! On
the other hand, the harmonic motion does continue to the expected F, yielding a
four-bar unit of BECF from 34. Technically, therefore, we have an overlap,
one that is given particular point through the play of stylistic properties to which it
itself contributes.
A more fully realized riposte to the exotic scale pattern ensues from bar 37. The
two-bar rise and fall patterns of 3742 sound like parodies of the modal passage,
here transformed into a lilting galant idiom. The chromatic tightness and clus-
tered harmonies are replaced by airy arpeggios and registrally isolated diatonic scale
Wilde Blumen am Zaun der Klassik: das spanische Idiom in Domenico Scarlattis Klaviermusik, in Domenico
Scarlatti (Musik-Konzepte 47), ed. Heinz-Klaus Metzger and Rainer Riehn (Munich: edition text +kritik, 1986),
Domenico Scarlatti and Spanish Folk Music: A Performers Re-appraisal, Early Music 4/1 (1976), 20.
See Zuber, Blumen, 28; Clark also mentions the ever-present chromatic hovering between the two versions

3 in Clark, Spanish, 20.
22 The keyboard sonatas of Domenico Scarlatti
progressions. Scarlatti thus seems to be working by a process of distortion, as each
new unit produces its commentary on the previous one. This process continues to
the end of the half, with the isolated tenor comment at 456 recalling the melodic
fragment of 234 in both pitch and rhythm.
What is most striking about this pattern is that here the composers disdain
seems to extend to the folk-like material as well; the Andalusian material cannot be
regarded as being any less mediated than the rest. Nevertheless, the second half of the
piece does concentrate on elements of the disruptive modal island. Zuber hears the
rst two phrases of the half (bars 5065) as the composers version of the melismatic
formulas of cante jondo (literally deep song), specically those that are heard before
the song proper begins. The vocal intoning of Ay is represented in bars 5053,
followed in 547 by an equivalent of the ornamental vocalizings known as salidas.
Is the odd rhythm at bars 5051 an attempt to capture the vocal inections of this
style? Several concrete instances of this feature from amenco song may suggest so.
In a ton a grande sung by Pepe de la Matrona, a ton a sung by Ramon Medrano and a
martinete sung by El Negro, contained in the recorded collection Magna Antologia del
cante amenco, one nds just this treatment of the initial Ay.
In the rst instance in
particular, with its marked crescendo to and accent on the end of the note, one hears
a marked correspondence to what seems to be suggested by Scarlattis notation.
Whether or not these phrases in K. 193 can have such specic folk models, they
are well integrated with earlier aspects of the sonata. They emphasize the neighbour-
note pitches of the modal island, the E and G that circle around F, with the G
here enharmonically treated as F. The recollection of the modal island as a unit
from bar 66 leads to a considerable change in its function. It is much more diatonic
in orientation, being clearly poised on V of G minor (with the F ( =G) being
placed in a functional context), and various changes of detail give the whole unit a
far less abandoned avour. Incredibly, the composer follows this with the exact three
bars that occurred after the original modal island: bars 724 are identical with 346.
Bar 72 sounds like a real harmonic non sequitur, but note that the new ornaments
found at bars 68 and 71 pre-echo those that will return from 73. The melodic
diction of the two passages is thus brought closer together, while this ornamental
link also helps to get us over the harmonic jolt.
This time, however, the passage
from bar 72 is not interrupted, as it was so disconcertingly at bar 37
, and is allowed
Note how unobtrusively the composer works in the basic cell of the opening. The neighbour-note basis of its
rst beat is heard both in its original shape, in the chain of gures in 43, and in inversion at the start of bars 45
and 47. The complete rhythm of the rst bar is present at bars 44, 46 and 48, now absorbed into the form of a
standard cadential closing gure.
Zuber, Blumen, 36, 38.
Magna Antologia del cante amenco (Hispavox: 7 99164 2, 1982), vol. 1 (7 99165 2), tracks 9, 14 and 19 respectively.
The obviously conjectural basis for such comparisons will be discussed in Chapter 3.
This harmonic juxtaposition is discussed by Joel Sheveloff, who notes the use of the pivot note (common tone)
to move from one chord to another. In this case it is the D that is barely heard in bar 72. He adds: It is normal
for Scarlatti to disguise the surface signicance of the common tone in this sort of situation; nineteenth-century
composers, on the other hand, tend to accentuate this detail. See Sheveloff, Keyboard, 3667. The composers
avoidance of best voice-leading behaviour, as thus elucidated, could be read as a perfect example of disdain.
Scarlatti the Interesting Historical Figure 23
Ex. 1.4b K. 193 bars 85101
to pursue its sequential course. This further emphasizes the corrective sense of the
second half, that it is an attempt to retell the story of the rst half in a more functional
The harmonic argument of the sonata, which has been tied up with the contrasts
of material, reaches a climax from bar 78. The attempt to project an unequivocal
dominant is clouded by the G from the modal island, and a vamp arrives from bar
86 to act as a musical melting-pot (see Ex. 1.4b). Vamp is a term coined by Sheveloff
to describe those apparently non-thematic, obsessively repetitive passages that occur
frequently in the sonatas.
The right-hand part makes continual reference to the
GG/EE axis around F, as if in an attempt to mediate between the modal and
tonal. The left hands role is unusually clear for a vamp; it features a big unfolding
between B and D in the bass, lled in by passing notes, in an attempt to establish
the dominant more securely. The vamp may also be conceived of as an effort to
overcome the sectionalized syntax of the work, with all its repeated units, either
sequential or at pitch. The passage does consist of course of endless repetitions of
the one cell, but precisely because of this we may also listen beyond the surface, to
one large phrase that will seemingly last for ever.
The right-hand line of the vamp is unusual in that, contrary to most similar passages
in Scarlatti, it is explicitly thematic, taking its cue fromthe opening cell. But, although
in sound and sense it clearly forms a climax to the other exotic suggestions found in
K. 193, the vamp still seems to issue from another world. There would seem to be
The vamp is christened as such in Sheveloff, Keyboard, 364.
24 The keyboard sonatas of Domenico Scarlatti
a basis in repetitive melismatic chant, which is what leads to the distinctly oriental
avour; but then, the location of an external source of inspiration is much more
comforting than ascribing such a passage to the mad genius of the composer alone.
To put this differently, the meaning of the passage is not exhausted by its possible
relationship to amenco song. We still have to ask what something so apparently
raw is doing in a nished art work. We must also remind ourselves that, if this does
come from the source suggested, then Domenico Scarlatti chose to listen.
Thus the vamp is integral yet separate to emphasize only its functionality and
compatibility on the large scale would be to swallow up what makes it so strange
along the way. Specically, this includes the sense of harmonic free fall, which
we can only grasp retrospectively from the standpoint of bar 100. We should note
also the clouding caused by the cluster of neighbour notes around the pivotal F.
That the grounds for the chromatic alterations in the right hand remain somewhat
obscure may be judged from several attempts to rationalize the passage. First of all
there are the corrections of Alessandro Longo, editor of the rst complete edition of
the Scarlatti sonatas in the early years of the twentieth century.
Among other things
he retains the G for several bars after bar 85 so as to avoid the abrupt resumption of
G in 86; he also cuts bars 9091 completely so as to shorten the endless reiteration.
These changes may be heard in the recording by Anne Queff elec, who applies a
dynamic arch shape to the vamp, fading away nearly to nothing by bar 99. This
treatment tells a familiar tale of nessing when I would argue for naked insistence.
Christian Zacharias substitutes E for E at bars 868 and 936, thus creating a
neatly consistent line of Es all the way through to bar 97. This attempts to clear up
the modal confusion that has been read as central to the argument of the piece.
From bar 100 the gesture towards greater continuity of syntax results in an al-
most uninterrupted stream of triplet semiquavers, like a release of energy after the
damming-up represented by the vamp. In this connection it is noticeable that the
rhythm of the opening bar of the piece is nowhere heard explicitly in the second
half, just as in K. 277 the most marked galant material disappeared for good be-
fore the second half had even begun. This is why Zubers (guarded) suggestion of
a seguidilla basis to the piece, with its rhythm being reminiscent of castanets,
is not ultimately of rst importance. That several other sonatas, such as K. 188 and
K. 204b, share both the repeated use of this same rhythm as well as exotic harmonic
coloration make a folk-dance basis for the material relatively likely. However, what-
ever the material origins of the opening of K. 193, it should be more than clear that
we cannot hear the whole as a dance form pure and simple.
The whole closing section of our sonata achieves its greater continuity by a radical
rewriting so as to maintain the momentum. The move towards harmonic clarication
Opere complete per clavicembalo di Domenico Scarlatti (Milan: Ricordi, 190610). K. 193 = L. 142.
Erato: 4509 96960 2, 1970 (Queff elec); EMI: 7 63940 2, 197985/1991 (Zacharias). Zacharias also alters the
Gs of bar 22 and so forth to Gs, although this might conceivably be a misreading.
Zuber, Blumen, 278. She also reports Alexandru Leahus belief that the similarly shaped material of K. 188
represents a malague na.
Scarlatti the Interesting Historical Figure 25
is made in earnest from bar 100, where the totally diatonic scurryings trump those
of the modal island. Note that all the high points of the right-hand runs occur on
, g
and e
, thus continuing the vamps business. In this sonata for once we may
claim that the composer does not in fact hold himself aloof from the various styles
and possibilities he introduces: in the end the work represents a decisive victory
for the diatonic and for the uent syntax it can generate.
In this conjuring with
eighteenth-century styles, the composer thus continues to elude any attempt to
schematize his artistic approach. This early confrontation with several sonatas should
have indicated some of the challenges involved in establishing a critical apparatus
adequate to Scarlattis stature and signicance. The following chapter reects in
more detail on the patterns of reception of this enigmatic gure.
This is meant from a rhetorical more than grammatical point of view, since in pure harmonic terms a victory
for the diatonic is the only possible outcome.
Writing about the sonatas, says Jane Clark, is a eld so full of pitfalls that anyone
willing to risk an opinion, however tentative, about the form, the chronology, the
Spanish inuence, the origins of the style or indeed anything else, is risking a great
The depth of uncertainty and, indeed, disagreement about what might in
normal circumstances be basic givens even about what the boundaries for en-
quiry are is surely unmatched among famous composers of such relatively recent
vintage. The wringing of hands has become more frequent with the progressive
institutionalization of musicology in the twentieth century and the perceived need
for accountable methodologies. Yet the uncertainties were felt before this, at least
in the negative sense that so little of substance was written about Scarlatti. It would
be wrong to suggest that Scarlatti had been neglected; the nineteenth century was
certainly familiar with Domenico, especially through the work of pianist-arrangers.
In 1898 Oskar Bie could write Scarlatti is especially remarkable to us in the present
day, in that he occupies the position of an early writer whose pieces still play a
part, though a small one, in modern public concerts.
While playing activity kept
the composer alive during this time, scholarly activity had to wait. The rst com-
plete edition, by Alessandro Longo, appeared in 190610.
The rst monograph on
Scarlatti, though, did not arrive until 1933. Perhaps not surprisingly, this honour fell
to a German scholar, Walter Gerstenberg. Books followed by Sacheverell Sitwell in
1935 and Cesare Valabrega in 1937.
It was not until after the Second World War, though, that the problems surround-
ing Scarlatti were fully confronted. Ralph Kirkpatricks 1953 volume marked a point
of arrival for its subject.
It was warmly received at the time and has continued to
attract acolytes up to the present day; indeed, most of the common currency about
Review of Domenico Scarlatti: Master of Music by Malcolm Boyd, The Musical Times 128/1730 (1987), 209.
A History of the Pianoforte and Pianoforte Players, trans. and rev. E. E. Kellett and E. W. Naylor (London: Dent,
1899), 69.
Opere complete per clavicembalo di Domenico Scarlatti (Milan: Ricordi, 190610).
Gerstenberg, Die Klavierkompositionen Domenico Scarlattis (Regensburg: Gustav Bosse, 1969; second reprint of rst
edn, 1933); Sitwell, A Background for Domenico Scarlatti (London: Faber, 1935); Valabrega, I l clavicembalista Domenico
Scarlatti: il suo secolo la sua opera (Modena: Guanda, 1937).
Kirkpatrick, Scarlatti.
Panorama 27
the composer still derives from Kirkpatricks thoughts and theories. Two subsequent,
fundamental texts, both taking issue with many of Kirkpatricks ideas, are unfortu-
nately not in general circulation. Joel Sheveloff s doctoral dissertation of 1970 rep-
resents the most important detailed work on the sources but was never published.
Giorgio Pestellis book of 1967, likewise based on a dissertation, remains the most
sustained aesthetic commentary on the Scarlatti sonatas.
No translation has ever
appeared; just as crucially, its great merits were obscured by controversy over its
nominal subject matter. Pestelli offered a replacement for Kirkpatricks chronology,
based roughly on the order of copying of works, by one based on stylistic analysis.
If this was speculative, its daring generality virtually placing its author in a no-win
situation, critics should perhaps have recalled that Kirkpatricks order was also specu-
lative. However, Kirkpatricks evidence was hard while Pestellis was soft, a reveal-
ing distinction in terms of the development of musicology outlined in Chapter 1.
If Pestellis approach was awed in principle, for example in its assumption of a linear
development of Scarlattis style or in its reliance on the Longo text, and certainly
debatable in its detailed realization, he nevertheless made a memorable attempt to
dene the artistic climate of this vast production of sonatas.
All of the above works equated Scarlatti with the Scarlatti of the keyboard sonatas,
leaving little room for the consideration of all his work in other genres and often
implying that much of this was not worth detailed consideration. This well-worn
opinion was nally contested by Malcolm Boyd, in his 1985 book that gave relatively
equal weight to all stages and products of the composers career.
By then the rst
complete edition of the sonatas in the modern era had appeared, edited by Kenneth
the nal volume appeared in 1984. That this should have had to wait until
so relatively recently tells its own story. A second edition, edited by Emilia Fadini,
published its rst volume in 1978; it remains incomplete, with eight of the projected
ten volumes having now appeared.
The Gilbert edition was neatly completed just
in time for the tercentenary of the composers birth in 1985, which gave particular
impetus to Scarlatti studies, producing several volumes of conference papers and
stimulating some long-overdue Spanish interest in documentary issues. There also
appeared in this year Sheveloff s two-part article Tercentenary Frustrations, which
is the best concise introduction to the uncertainties that have hampered Scarlatti
While Scarlatti has arguably been lucky to attract so many ne minds to his cause,
not just in the landmark publications mentioned above but in many smaller-scale
operations, the wider picture is not so happy. Within the universal set of musico-
logical endeavour he has received scanty treatment for a composer of his stature. All
our potential pitfalls have no doubt warned off many specialists; in generalist terms
the principal factor has probably been his unclear historical and stylistic position.
Sheveloff, Keyboard.
Pestelli, Sonate.
Domenico Scarlatti: Master of Music (London: Weidenfeld and Nicolson, 1986).
Domenico Scarlatti: Sonates (Paris: Heugel, 197184), 11 vols.
Domenico Scarlatti: Sonate per clavicembalo (Milan: Ricordi, 1978).
Sheveloff, Frustrations I and II.
28 The keyboard sonatas of Domenico Scarlatti
Another obstacle to discussion is the lack of outward differentiation to Scarlattis
keyboard output. It is not only the lack of any rm chronology but also the elusively
standard appearance of the sonatas that makes any basic mental ordering difcult,
for professionals as much as amateurs. How can one keep track of such a production
when trying to draw comparisons between various sonatas? It would be like trying
to maintain discipline among a family of 555 children all demanding attention; and
so it is quite understandable that so much Scarlatti scholarship has been dedicated
to reducing the clamour in various ways, so that one can hear oneself think. Longo
ordered the sonatas into families of ve, calling them suites and thus aligning these
with a familiar Baroque principle of multi-movement organization. Before this, in
1864, Hans von B ulow had edited eighteen pieces in three groups of six, justifying
his conversion of the works by alluding to their terseness and brevity. He also
gave titles to all but two of the sonatas, which mostly referred, once again, to the
suite (Sarabande, Gigue, Capriccio, Courante and so forth). On this matter he
declared: Characteristic titles for the individual pieces were also called for, since the
generic title sonata . . . gives a faceless boring avour that could easily turn the public
off, whereas a harmless external change . . . may help sustain interest.
Another logistical deterrent is the existence of four separate numbering systems,
by Longo, Kirkpatrick, Pestelli and Fadini. This points to a fundamental aspect of
Scarlatti studies: the strange symbiosis that obtains between the state of knowledge
on Scarlatti and the efforts in dealing with it. Our piecemeal knowledge of cir-
cumstances and sources has been paralleled by scholarly activity which has likewise
been uncoordinated and partial. Above this lurks the fact that a collected edition
of Domenico Scarlatti has yet to be attempted, let alone completed. More notable
than the lack of such a monolith, though, is the absence, for example, of an edition
of the complete cantatas, especially given claims for their relevance to the keyboard
As pertinent here as all the specic problem areas is the fact that the com-
poser is uncomfortably situated culturally. Where after all would the natural home
for a collected edition be, or have been?
Equally, Scarlatti has not done very well out of the early music movement. For
all the advocacy of Wanda Landowska, the composer has not altogether been em-
braced by harpsichordists as fully as one might have expected. Paul Henry Lang
connects this with the purely musical humour that he believes makes its rst ap-
pearance in Scarlatti. He continues: These arrowshafts of wit, nicely calculated to
penetrate stuffed hides, were one of the reasons why the rst generation of modern
harpsichordists, well groomed, proper and enamoured of the bonbons of the res-
urrected French harpsichord repertory, were at rst puzzled and uncomfortable.
Preface to Achtzehn ausgew ahlte Klavierst ucke von Domenico Scarlatti, in Form von Suiten gruppiert (Leipzig: Peters,
1864), i.
For consideration of the relationship of the cantatas to the keyboard works see in particular Degrada, Lettere,
and Kate Eckersley, Some Late Chamber Cantatas of Domenico Scarlatti: A Question of Style, The Musical
Times 131/1773 (1990), 58591.
Scarlatti: 300 Years On, The Musical Times 126/1712 (1985), 588.
Panorama 29
One wonders in fact whether Langs wicked sociological assessment is yet obsolete.
A certain spiritual antiquarianism may still obtain; in such a context Scarlatti brings
an unwelcome ambience of rock and roll. Once again, this does not amount to a
claim for outright neglect;
it is more an attempt to determine why so many in
our various musical subcultures have chosen, no doubt often unconsciously, not to
engage with Scarlatti. Langs suggestion that, in the composers lifetime, neither
professional musicians nor experienced amateurs quite knew how to make peace
with this unusual music
might also be extended up to the present day.
Given all the circumstances outlined thus far, the possibility of any sort of
denitive and monumental study of Scarlatti seems remote.
Barring the bene-
cent intervention of a deus ex machina, the material will never be in place to allow
this to happen, even were such a study still felt to be desirable.
This lack of the appropriate material hard facts has often been mused on
by commentators, producing theories that bring almost the only colour to what
Sitwell called the blank canvas of Scarlattis life.
In fact any biographer is forced
to speculate. What we might call the modal verb tendency a liberal helping of
must, should and could have is indispensable for such activity.
When invoking this absence of information, writers have naturally favoured dark
imagery: Scarlatti is characterized as an obscure, shadowy gure. It is easy to take this
obscurity as a given without realizing how extraordinary it was in the circumstances.
Not only was the eighteenth century an age of (musical) gossip, but, more specically,
our interest does not lie in a journeyman musician working at a provincial court.
Scarlatti was a celebrated composer (and player), the son of an even more celebrated
composer, who throughout his life was associated with people of the highest rank.
Once more a strange symbiosis seems to be in operation, between the disdain
identied by Pestelli as a fundamental aspect of the composers artistic personality
and the disdain for the sensibilities of historians that seems to preside over the
biographical situation.
The dark imagery that dominates these assessments of the state of affairs is, of
course, itself a form of colouring applied to the blank canvas. Nowhere is this
clearer than in Gerstenbergs assertion of the aristocratic obscurity surrounding the
An instructive example of relative neglect may be found in The Harpsichord and its Repertoire: Proceedings of the
International Harpsichord Symposium, Utrecht 1990, ed. Pieter Dirksen (Utrecht: STIMU Foundation for Historical
Performance Practice, 1992). In these entire proceedings Domenico Scarlatti receives one passing mention, in
contrast with the plenteous references to such gures as DAnglebert, C. P. E. Bach, Chambonni` eres, the
Couperins and Froberger, while a whole section is devoted exclusively to J. S. Bach.
Lang, 300 Years, 589.
Peter Williams, Review of Domenico Scarlatti: Master of Music by Malcolm Boyd, Music and Letters 68/4 (1987),
Sitwell, Background, 166.
30 The keyboard sonatas of Domenico Scarlatti
composers life.
This suggestion of aristocratic reserve, a common enough strain
in the reception of the composer, puts a more positive spin on a situation that
has seemingly frustrated and enticed in equal measure. For Massimo Bontempelli,
Scarlatti has had the good fortune for almost all trace of his everyday life to have
disappeared, which he described as an enviable fate.
He was certainly correct in
his implication that this would help the poetry if not the prose of Scarlatti biography.
Gilbert Chase has poured historical cold water on all speculation by reminding us
of the disparity between the worldly appreciation of vocal and of instrumental music
at the time. It is certainly true that it is difcult for us to grasp the supreme position
of opera, in particular, in eighteenth-century musical life, when the instrumental
works of such gures as Bach and Haydn still bulk so large for us. Further, Chase
contends that this disparity may be seen by comparing the position at the Spanish
court of Farinelli the castrato who arrived in 1737, retired from public concert
life and became, amongst other things, a great operatic impresario with that of
Scarlatti. Scarlattis relative obscurity is indicated by the paucity of information that
has come down to us concerning his life in Madrid.
Such a at explanation would
seemto be borne out by the fact that Scarlattis name appears only two or three times,
always insignicantly, in chronicles of life at the Spanish court.
We must also bear
in mind the theories that Queen Isabel kept her stepson Fernando and his wife Mara
B arbara in the background as much as possible, with clear consequences for the role
of their employee Scarlatti at court.
If such factors might help us come to terms with the apparent lack of worldly
appreciation Scarlatti received in Spain, this would still not help us with the circum-
stances elsewhere. Scarlatti was hardly written or talked about in Italy and Portugal
either, when he was, it would appear, primarily a composer of vocal music. The
conspiracy of silence in fact extends well back. An early performance of Scar-
lattis opera Tolomeo et Alessandro in 1711, put on especially for members of the
Arcadian Academy (whose numbers included Alessandro Scarlatti) at the residence
of the exiled Polish queen Maria Casimira, prompted the chronicler of the Arcadian
nymphs and shepherds, Giovanni Crescimbeni, to extol the virtues of the produc-
tion. Although he mentioned that the music was very good indeed,
name remains unmentioned.
This is curious yet somehow typical.
From the evidence contained in an inventory of Farinellis instruments and scores,
drawn up in 1783 and recently published for the rst time,
it would now appear
Review of Domenico Scarlatti by Ralph Kirkpatrick, Die Musikforschung 7/3 (1954), 343.
Verga LAretino Scarlatti Verdi (Milan: Bompiani, 1941), 125 and 1256.
The Music of Spain (London: Dent, 1942), 109.
See Pestelli, Sonate, 181.
See for example Clark, notes to recording by Jane Clark (Janiculum: D204, 2000), [1]. This would only apply
to the period up to 1746, when Fernando ascended the throne.
Boyd, The Music very good indeed: Scarlattis Tolomeo et Alessandro Recovered, in Studies in Music History
Presented to H. C. Robbins Landon on his Seventieth Birthday, ed. Otto Biba and David Wyn Jones (London: Thames
and Hudson, 1996), 10.
See Kirkpatrick, Scarlatti, 52.
See the Appendix to Sandro Cappelletto, La voce perduta: vita di Farinelli evirato cantore (Turin: EDT, 1995),
Panorama 31
that Scarlatti may well have been in any case a more active composer of vocal music in
Spain than previously allowed. As well as the many solo cantatas he almost certainly
wrote in Madrid, there is the possibility that some of the unidentied serenades
mentioned in the inventory were also written during this period.
If so, one might
have expected these to have received some worldly appreciation.
Another, more reliable at explanation for the absence of source material, partic-
ularly musical scores, has often been sought in such disasters as the complete destruc-
tion of the Alba Library in the Spanish Civil War in 1936, the Lisbon earthquake
of 1755 and several res at the Escorial. Such possibilities may also account for the
absence of sonata autographs from three other very important eighteenth-century
Iberian keyboard composers Carlos Seixas, Sebasti an de Albero and Antonio Soler.
A means of uniting the dark imagery with the lack of information on Spanish
circumstances would be to invoke the Black Legend (leyenda negra). This term,
coined by Juli an Juderas at the beginning of the twentieth century, symbolizes
the image and historiographical treatment of Spain as an outsider within Europe,
certainly once its Golden Age was past. Judith Etzion suggests that Charles Burney,
for example, probably knew more about Spanish music than he chose to disclose
in his writings, and, more specically relevant to our case, that Farinelli probably
told him far more about the musical life of the Spanish court than is transmitted in
The Present State of Music (17713). This would reect the wider eighteenth-century
assumption that Spain was musically backward and peripheral.
The uncertainties reviewed thus far primarily concern absence of information.
Just as characteristic, though, are leads which only invite further detective work,
tantalizing fragments that raise more questions than they answer. To add a few more
ecks to the blank canvas, here are some additional questions and issues that have
been entertained by Scarlatti commentators.
1. The circumstances of the ofcial publication in London in 1739 of the Essercizi,
the only edition of sonatas published by the composer in his lifetime. Why was
there a rival, and much more successful, publication of the works by Thomas
Roseingrave, and why did Farinelli lead Burney to believe that the Essercizi had
been published in Venice?
2. Did Scarlatti play his own works as a young virtuoso?
3. How did the young Scarlatti receive his musical training?
See Degrada, Lettere, 31415. For now the Salve regina of 1756 and the Madrid Mass (possibly written in
Spain) are being left out of consideration.
Spanish Music as Perceived in Western Music Historiography: A Case of the Black Legend?, International Review
of the Aesthetics and Sociology of Music 29/2 (1998), 1045.
See Clark, His own worst enemy. Scarlatti: Some Unanswered Questions, Early Music 13/4 (1985), 543.
Compare Eva Badura-Skoda, Domenico Scarlatti und das Hammerklavier,

Osterreichische Musikzeitschrift 40/10
(1985), 525, suggesting that this must have been the case, and Clark, Enemy, 544, where the author stresses
that the reports of Scarlattis playing never suggest he was playing his own music.
See Sheveloff, (Giuseppe) Domenico Scarlatti, in The New Grove Dictionary of Music and Musicians, vol. 16, ed.
Stanley Sadie (London: Macmillan, 1980), reprint in The New Grove Italian Baroque Masters (London: Macmillan,
1984), 327, where, as an antidote to our modal verb tendency, Sheveloff states atly that this is unknown.
32 The keyboard sonatas of Domenico Scarlatti
4. What were the circumstances in which Scarlatti lived while in the service of the
courts of Portugal and Spain? And what were his exact working conditions and
duties at court?
5. Under what circumstances were the sonatas written; how many of them actually
originated as teaching pieces?
6. Why were so relatively few of the sonatas published in the composers lifetime
( just seventy-three, none in Italy or Spain)
and why have so relatively few
contemporary copies turned up? It has been suggested that Scarlattis situation
may have been similar to that of Jan Zelenka at the court in Dresden, whereby
any publication and copying of his works was forbidden; Mara B arbara had thus
claimed sole ownership. We should also note the later situation of the symphonist
Gaetano Brunetti (174498), who was forbidden from distributing his music
outside the royal court in Madrid.
7. Who was the scribe of the Parma and most of the Venice volumes? The initials
S or SA found at the end of several sonatas in the last two Parma volumes
seem to provide a clue. Was it Sebasti an de Albero, Antonio Soler, one Andres
Solano, or even, as Roberto Pagano fantasizes, the ghost of our composers father,
Alessandro Scarlatti?
8. Were the Scarlatti sonatas performed at the Spanish court?
If so, where, when,
by whom?
9. Did Domenico Scarlatti become Fatty Scarlatti? Since the reappearance of the
Velasco portrait of the composer at Alpiarca in Portugal it has mostly been assumed
that Scarlatti was constitutionally slim, but Jane Clark nds in the representation
a distinctly visible tendency towards corpulence. For her, this shows the danger
of taking anything at face value with Scarlatti.
Anumber of the issues outlined above concern the creative environment inhabited by
Scarlatti at the Spanish court. Many writers stress that it was an exclusive and isolated
Boyd, Master, 1589. All but the Essercizi would seem to have been unauthorized publications.
Macario Santiago Kastner, Repensando Domenico Scarlatti, Anuario musical 44 (1989), 151; David Wyn Jones,
Austrian Symphonies in the Royal Palace, Madrid, in Music in Spain during the Eighteenth Century, ed. Malcolm
Boyd and Juan Jos e Carreras (Cambridge: Cambridge University Press, 1998), 137.
Scarlatti Alessandro e Domenico: due vite in una (Milan: Arnoldo Mondadori, 1985), 45960.
Harvey Sachs reminds us that there is no general agreement among experts . . . whether or not the sonatas were
played publicly at court. Notes to recording by Ralph Kirkpatrick (Deutsche Grammophon: 439 438 2, 1971
[notes 1994]), 2.
See Boyd, Master, 165. The nal sentence of Scarlattis dedication in the Essercizi would certainly suggest, even
allowing for hyperbole, that Mara B arbara performed his sonatas on certain court occasions: the mastery of
singing, playing and composing with which she, to the astonishment and admiration of the most excellent
masters, delights princes and monarchs. See Boyd, Master, 140. Ralph Kirkpatrick notes the abundance of court
communiqu es reporting musical evenings in the apartments of Mara B arbara before she became queen and states:
At these evenings Domenico Scarlatti was undoubtedly present and active. It is difcult to dispute this, but for
all that it is quite remarkable that we have no records that are explicit on the matter. Kirkpatrick, Scarlatti, 87.
Clark, Boyd Review, 209. The portrait is reproduced on the cover of Boyd, Master.
Panorama 33
one that helped to determine the character of the sonatas. These two perceived
properties have led to highly determinist equations. The apparent isolation has been
used to explain Scarlattis originality,
but this is no more adequate an explanation
than it is for Haydn, with whose situation Scarlattis has sometimes been compared.
Admittedly, it was Haydn himself who offered the line that in his isolation he was
forced to become original, but it has been far too easy for traditional musicology to
take such a remark (born at least in part out of Haydns famously modest persona) at
face value, to ground the historically problematic category of originality in localized
circumstance. Isolation, after all, is not an absolute any more than originality is.
Other composers placed in similar circumstances would not have been able to react
in the alleged manner. At best we can say of both cases that an opportunity was
grasped because of certain creative proclivities.
The second equation suggests the production of exclusive music for exclusive
Howcan we square the notion that Scarlattis music was an upmarket
luxury item with the abundance of popular and ethnic elements in the sonatas? Most
of the commentators who stress the aristocratic nature of Scarlattis keyboard art are
also those who minimize the popular side, often for nationalist reasons that we will
contemplate later in this chapter.
In any case, it is debatable whether this environment was characterized by great
renement or equilibrium. Surprisingly little capital has been made in the literature
of the instability of the two Spanish monarchs Scarlatti served under; perhaps this
was one hypothetical step too far for most commentators. The fragile mental state
of Felipe V, for instance, reached a crisis in 1728, not long before Scarlattis arrival in
Spain. The King would bite his arms and hands and spent the night screaming and
shouting; he believed he had been turned into a frog; he was afraid of being poisoned
by a shirt and would only put one on that had been worn by the Queen; he ate vast
quantities and would then spend entire days in bed in the middle of his excretions.
Although far less disturbed than his father, Fernando VI was also prone to depression
and notorious for his sexual appetite; he then behaved in extraordinary fashion after
the death of his consort.
If we wish to pursue such connections between creativity
and locality, surely Scarlatti would have been at least as affected by such an atmosphere
as by the apparently exclusive, elite environment evoked above he had, after all,
known little else throughout his career. Might the compulsive, repetitive, unstable be-
haviour of the vamp sections not owe something to such royal example? In fact, only
For example in Philip G. Downs, Classical Music: The Era of Haydn, Mozart, and Beethoven (New York: Norton,
1992), 49.
See for example Frederick Hammond, Domenico Scarlatti, in Eighteenth-Century Keyboard Music, ed. Robert
L. Marshall (New York: Schirmer, 1994), 178, and Anne Bond, A Guide to the Harpsichord (Portland: Amadeus,
1997), 180.
See Kastner, Introduction to Carlos Seixas: 80 Sonatas para instrumentos de tecla (Lisbon: Fundac ao Calouste
Gulbenkian, 1965), xxxiii, and Degrada, Lettere, 315.
See W. N. Hargreaves-Mawdsley, Eighteenth-Century Spain 17001788: A Political, Diplomatic and Institutional
History (London: Macmillan, 1979), 64. Further gruesome details may be found in John Lynch, A History of
Spain: Bourbon Spain 17001808 (Oxford: Blackwell, 1989), 6772.
See Kirkpatrick, Scarlatti, 131.
34 The keyboard sonatas of Domenico Scarlatti
Kirkpatrick has addressed such connections, but came to the opposite conclusion
that the sonatas functioned as an antidote to melancholy and madness.
Occasionally more particular environmental linkages have been sought: for in-
stance, that the sonatas echo the different attractions of the four royal palaces around
which the Spanish court moved on an annual basis the Pardo, Buen Retiro,
Aranjuez and the Escorial.
If this suggests a certain biographical desperation (quite
understandable of course in our circumstances) and a pictorialist reception of the
music that issues from the Kirkpatrick tradition, it is hardly to be dismissed in princi-
ple. It has been noted, for example, that Felipe V, the French grandson of Louis XIV,
tried to soften the rugged Castilian landscape he found himself in with the adorn-
ments of Italian and French art and architecture.
This is arguably reected in the
landscape of the sonatas, in their topical play of high and low, in the contrast be-
tween international and local musical images. This is not to suggest a direct causal
connection from one set of physical circumstances to another set of musical ones,
since again we must emphasize the element of choice. Scarlatti could have remained
as unaware as most at court apparently were of the cultural incongruities of the living
environment; but he seems at some level to have chosen to reect or accommodate
these in his work.
One other matter involves us again in contemplating an absence the fact that
Scarlatti took no part in the opera craze
which began after the arrival of Farinelli
in 1737. (Nor, curiously, did he play any part in the grand festivities at Aranjuez
masterminded by Farinelli.) This may be interpreted as a straightforward matter it
was not within the terms of Scarlattis job or seen as a further puzzle. The younger
Scarlatti had after all written a good number of operas and consequently had had
plenty of contact with the operatic world, if in the mostly sheltered form of private
commissions and performances.
Contemplating this puzzle brings us within range of another set of speculations
concerning Domenicos real-life personality. The consensus of opinion would offer
Kirkpatrick, Scarlatti, 120; see also 91. Kirkpatricks melancholy includes not just the circumstances at court
which the sonatas had to ward off but the entire baggage of Spanish gloom. In Domenico Scarlatti, written
and narrated by David Thompson, devised and directed by Ann Turner (BBC television documentary: broadcast
20 April 1985), we are told that the composers music was an antidote to Mara B arbaras disappointed life.
See Hammond, Scarlatti, 161.
Barry Ife and Roy Truby, Introduction to Early Spanish Keyboard Music: An Anthology, Volume III: The Eighteenth
Century (Oxford: Oxford University Press, 1986), 4.
This is Ifes term. Echoing the comments of Jane Clark, he believes that human malevolence on the part of
Queen Isabel may explain the composers non-participation, pointing out that, while Farinellis was a crown
appointment, Scarlatti was the personal servant of Mara B arbara and Fernando. Domenico Scarlatti (Sevenoaks:
Novello, 1985), 16. Since Fernando was not Isabel Farneses child and she was a notorious schemer on behalf of
her own children, to have allowed Scarlatti to participate, according to this line of thought, would have been to
lend unwanted prestige to the Prince and Princess.
Scarlatti only wrote two operas for a public theatre Ambleto of 1715 and Berenice regina dEgitto of 1718. See
Degrada, Lettere, 272.
Panorama 35
that he was not cut out for the theatrical world, perhaps even that he actively resisted
recruitment to the cause. This line of thought takes its cue from the words of John
Mainwaring, Handels biographer, who wrote that Scarlatti had the sweetest temper,
and the genteelest behaviour;
it is another way of making positive sense of the
absence of information we are faced with, suggesting an obscurity determined by
shyness. One version of this by Lang reveals the larger contradiction implied by
this portrait: Perhaps . . . there was something in the whirlwind lifestyle of Italy that
he found uncongenial; Domenico seems to have been a rather private person who
avoided publicity.
If this were the case, there would be an enormous contrast
between the alleged retiring nature and the artistic products whirlwind lifestyle
would describe a lot of the sonatas perfectly!
Indeed, were we to speculate on Scarlattis character from the evidence of the
music, we might imagine it to have been unstable or even schizophrenic. Some have
in fact hinted at such a possibility.
The danger with all such snapshots is, naturally,
one of circularity, as one moves too effortlessly from work to life and back. Yet it
would not quite be fair to ascribe the collective efforts to sketch the real Domenico
Scarlatti simply to a certain Romantic ideology. The music, after all, projects itself
so strongly and characteristically as positively to demand active curiosity about its
creative source. This is not always the case with composers of whose circumstances
and characters we are relatively ignorant.
The methodological problems inherent in such sketches are multiplied when
attempting a biography of the composer. Information is so thin that a biography
cannot really work. Ralph Kirkpatrick did an astonishing job, though, of making
us forget that there was little tale to tell, if at the expense of what have been chided
as creative excesses.
The most inuential of these was the overinterpretation of
Scarlattis relationship with his father. Subsequently the patriarchal bogeyman has
stalked many accounts of Domenicos life.
In fact, the looming gure of the father
is as tedious a historical leitmotiv as the rise of the middle classes. He is central to
biographical studies of Mozart, Kafka and Beethoven, to name but a very few. For
such a device to become a convincing argument, one has to prove that the father
was more than usually inuential. And dont all sons rebel yet also perpetuate certain
attitudes and modes of behaviour?
Roberto Pagano continues this line in his romance biography of 1985, Scarlatti:
due vite in una (two lives in one). In the name of his avowedly fantastic thesis that
Alessandro and Domenico merge into a single ideal character he formulates such
Cited in Kirkpatrick, Scarlatti, 33.
Lang, 300 Years, 585.
See Hermann Keller, Domenico Scarlatti, ein Meister des Klaviers (Leipzig: Peters, 1957), 86, and Clark, Enemy,
Sheveloff, Frustrations I, 399.
See for instance Hammond, review of Scarlatti Alessandro e Domenico: due vite in una by Roberto Pagano, Music
and Letters 69/4 (1988), 520, and Roman Vlad, Bach, H andel e Scarlatti nella storia della musica, in Metamorfosi
nella musica del novecento: Bach, H andel, Scarlatti (Quaderni Musica/Realt` a 13), proceedings of conference in Cagliari
on 1214 December 1985, organized by the Associazione Spaziomusica with Musica/Realt` a, ed. Antonio Trudu
(Milan: Edizioni Unicopli, 1987), 15.
36 The keyboard sonatas of Domenico Scarlatti
statements as only after the death of his father could Domenico begin to become
properly himself and Domenico obey[ed] an obscure need not to lose completely
the state of uncertainty and unease to which his relationship with his father had
habituated him.
The main thesis in fact provides a rather thin rationale for a
narrative in which Domenico plays a statistically minor part, as Pagano has the grace
to acknowledge at one point.
The same is evident in the 1985 BBC television
biography of Scarlatti, in which the personage of the composer disappears more and
more as the lm progresses and the facts become fewer and fewer. In the latter
part the facts of the royal lives (of Fernando and Mara B arbara) are used to create
a phantom biography for the composer.
In both these stories Scarlatti leads of
necessity a vicarious life through others. Both in fact point to the same difculty
that Scarlatti is incapable of emerging as a fully formed and independent historical
personage through non-musical data alone.
If the literature has had difculties creating an independent logic to the sequence of
biographical events, the same has been true when trying to make sense of the vast
sequence of individual sonatas. One of the commonest strategies for overcoming the
lack of outward differentiation highlighted before has been to see the sonatas as an
all-embracing panorama. This subsumes the claims of the individual pieces under
the banner of a meta-work. Each sonata becomes a miniature, a spot of colour
contributing to the complete canvas.
Sometimes the resulting panorama is casually
construed, as in this typical formulation from Stephen Plaistow: There are dances
and estas and processions here, serenades and laments, and evocations of everything
from the rudest folk music to courtly entertainments and churchly polyphony; and
as the kaleidoscope turns you marvel at the composer who could embrace such
diversity and shape it and put it all on to the keyboard.
This nice list of musical
styles and avours represents the more innocent side of the panorama tradition. After
all, isnt this just a function of such a large quantity of works in one genre, an honest
response to sheer weight of numbers? With comparable cases, though, such as Haydn
symphonies, Bach cantatas or Schubert songs, the problems of comprehension have
not led to what we often nd in the case of Scarlatti the suggestion of a more or
less deliberately coordinated whole. This implies a controlling world view behind
the entire production of sonatas.
Pagano, Vite, 462, 409, 461.
Besides, the attention given to monarchs and ministers has distracted me from the events of Domenicos life.
Pagano, Vite, 407.
Thompson, Scarlatti.
John Gillespie writes of a multitude of exquisite miniatures, Cecil Gray of a delicate, miniaturist, epigrammatic
style. Gillespie, Five Centuries of Keyboard Music: An Historical Survey of Music for Harpsichord and Piano (New
York: Dover, 1965), 69; Gray, History, 140.
Review of recording by Mikhail Pletnev (Virgin: 5 45123 2, 1995), Gramophone 73 (1996), 72.
Panorama 37
This is what Giorgio Pestelli complained of at the outset of his 1967 book: that the
sonatas had been treated as an undifferentiated block, like a single continuous poem
in more than ve hundred verses.
In fact, quite specic poetic analogies have been
made, with the sonnets of Petrarch and Belli; these too are held to accumulate into a
larger whole.
Bontempelli, who was one to offer a comparison with Petrarch, also
saw Scarlatti as a representative of pure music. Hence he found it enigmatic that
when we think of the [555] sonatas in their totality, what remains in our memory
is not a musical particular, but a panorama, a spell, of a nature that one would
today call metaphysical.
These analogies all have the virtue of responding to a
crucial aspect of Scarlattian art: the democratic openness, the sense that any and all
sounds may be incorporated in the name of music. But they are also transparently a
mechanismfor avoiding detailed contact with the sensuous particularity of the music,
the tendency described as endemic in the opening chapter. While in principle this
approach appears to celebrate diversity, to use a current phrase, and to emphasize
the comic variety of the surface, in reality it abstracts us from it.
Sometimes this approach is couched in more historically plausible terms: that the
sonatas summation of a world of musical possibilities embodies the encyclopedic
spirit of the Enlightenment.
Yet this also brings prescriptive associations that do
not ring true. Perhaps a more useful working concept when trying to move be-
yond notions of an even-handed, programmatic diversity is that provided by Piero
Santi, who allies Bontempellis metaphysical spell with the magic realism that is
the quintessence of twentieth-century Italian art.
Magic realism captures per-
fectly the alchemy of Scarlattis pluralistic appropriations. This formula also helps
us approach the synaesthetic genius of the sonatas, which the panorama tradition
illuminates in its frequent turning of sound into visual image.
A sideshoot of the panorama tradition is the procedure of evoking the world of
the sonatas by means of parataxis, of expressively loose syntax. One of the earliest
examples in the literature, with its obligatory collocation of characteristic features,
come from Oskar Bie:
It is a spectacle of reworks. Deep bass-tones are suddenly introduced; high thirds y off;
thirds and sixths are darted in; close arpeggios swell into monstrous bundles as they are lled in
with all possible passing-notes; octaves are vigorously introduced; the hands steer in contrary
motion, to one another, away from one another; they are tied into chains of chords; they
release themselves alternately from the same chords, the same groups, the same tones; unison
passages in the meanwhile run up and down; chromatic tone-ladders dart through, then
slowly moving phrases or still-standing isolated treble notes are seen confusedly dotted over
the changing bass as it runs up and down, in a kind of upper pedal point; harsh sevenths one
after another; repeated notes, syncopated effects, parallel runs of semiquavers with leaping
Pestelli, Sonate, 2.
See Luciani, Domenico Scarlatti creatore del sinfonismo, Musica doggi 8/2 (1926), 43, and Hammond, Scarlatti,
Bontempelli, Verga, 128.
See Ife, Scarlatti, 21.
Domenico Scarlatti fra i due nazionalismi, in Metamorfosi nella musica del novecento, 53.
38 The keyboard sonatas of Domenico Scarlatti
side-notes, such as we know so well in Bach; sudden interchanges from major to minor,
a device of which the Neapolitan operas are so fond; bold characterisation by means of
sudden pauses; startling modulations by means of chromatic passages; embellishments rarely
introduced; a delicate arrangement of tones from the severest fugues to the most unrestrained
bourr ees, pastorales, or fanfares such is the world of Scarlattis clavier-music.
On a purely syntactical level, too, this passage correlates with the panorama tradition.
Note how everything is contained within the one superabundant sentence, just as
all the sonatas are held within a single picture.
This superlative straining prose that Scarlatti attracts, while underpinned struc-
turally by the guiding critical image of the panorama, also has a technical counterpart
within the music. It seems to be a recreation of or response to the frequently fever-
ish, supercharged syntax of the sonatas themselves. Bies sentence, with its wealth
of strongly physical metaphors of movement, has the additional virtue of respond-
ing in kind to another vital feature of Scarlattian style its pronounced sense of
The alternative to the poem described by Pestelli, examination of individual sonatas,
has proved much less attractive. There has been an extraordinary if understandable
reluctance to engage with individual works. By training and inclination most histori-
cal musicologists have avoided such activity anyway. Analysts are in the same position
as historians unsure of the rules of the game, they have collectively kept well clear.
A simple example of the sort of analytical issue that would act as a deterrent is phrase
duration. If we nd phrase units of irregular length, such as the rst ve bars of
K. 193 (see Ex. 1.4a), should we assume that this is a marked deviation or incidental?
An adequate answer cannot be found purely by contemplating the individual work
alone, since we are reliant for our working assumptions on the historical concept
of style. In this case, the problem hinges on the duality of Baroque and Classical,
and the very different syntactical ideals we associate with the two style-periods. My
reading of K. 193 presumed that the irregular opening was indeed supposed to stand
out, but such an assumption must be more provisional than it would be were we to
analyse a piece by, say, Clementi.
What almost all the few existing analytical readings have in common is that they
are not integral.
Nevertheless, such contributions are at least refreshing in their
Bie, Pianoforte, 889.
For examples see Eytan Agmon, Equal Division of the Octave in a Scarlatti Sonata, In Theory Only 11/5
(1990), 18; Peter Barcaba, Domenico Scarlatti oder die Geburtsstunde der klassischen Sonate,

Musikzeitschrift 45/78 (1990), 3869; Downs, Classical, 52; Carl Schachter, Rhythmand Linear Analysis: Aspects
of Meter, The Music Forum 6 (1987), 459; Heinrich Schenker, Domenico Scarlatti: Keyboard Sonata in D
minor [K. 9] and Domenico Scarlatti: Keyboard Sonata in G major [K. 13], from Das Meisterwerk in der Musik,
vol. 1 (1925), trans. Ian Bent, Music Analysis 5/23 (1986), 15185; Janet Schmalfeldt, Cadential Processes: The
Evaded Cadence and the One More Time Technique, Journal of Musicological Research 12/12 (1992), 710;
Panorama 39
novelty value. Carl Schachters discussion of K. 78, for instance, is a nice reminder of
Scarlattis art at a level almost unknown in the general literature. His reference to the
fantastic motivic references that enliven the foreground of this tiny masterpiece
may seem too characteristic of analytical rhetoric but still carries some force if
we want to take seriously Sheveloffs claim that Scarlattis style is composed of an
abundance of tiny, special details.
Janet Schmalfeldts study of the use of evaded and
elided cadences in K. 492 is refreshing for another reason: the relevance of a crucial
aspect of Scarlattis technique is placed straightforwardly in an eighteenth-century
context, with a clear implication that the composer is post-Baroque.
Heinrich Schenkers analyses are born from his conviction that, on the evidence
of his keyboard works alone, Domenico Scarlatti is Italys greatest musician.
this was a radical stance for 1925 (just as it would be now), his use of two Scarlatti
sonatas to demonstrate his principles of voice leading and tonal coherence would
have been seen as eccentric. While his demonstration of a tempestuous unfolding
of purely musical sonorities
again makes (and, even more then, would have made)
a bracing change from the normal critical preoccupations, his choice of K. 9 and
K. 13 is a little disappointing. These very contained and controlled numbers from
the Essercizi were obviously appropriate to Schenkers demonstrations of unity and
logic. One longs to know how he would have coped with the clusters of K. 119,
the eternal reiterations of K. 317, the stylistic ruptures of K. 402. We know he
had access to Czernys edition of two hundred sonatas, and he certainly knew the
arrangements of Tausig and B ulow, since he goes out of his way to comment on
their gross barbarities.
It may that he was another Scarlattian who wished to avoid
the prevalent image of the composer as sprightly buffoon.
The only sustained reading of a Scarlatti sonata is of K. 296 by Peter B ottinger,
and it is quite a model for future emulation. It begins in unexceptionable fashion,
then becomes more and more fantastic, as normal discursive syntax breaks down,
to be replaced by fragments, quotations, unusual arrangements of music and text
on the page, burblings as if out of Beckett, and an obsession with the mechanics
of the keyboard and the hand movements needed to stir it into life. This is clearly
designed as an analogue to B ottingers view of the sonata (and Scarlattis style), in
Sheveloff, Keyboard, 41529; and Sheveloff, Uncertainties in Domenico Scarlattis Musical Language, in
Domenico Scarlatti e il suo tempo (Chigiana 40), proceedings of conference in Siena on 24 September 1985,
sponsored by the Accademia Musicale Chigiana Musicologia and the Universit` a degli Studi in conjunction with
the Societ` a Italiana di Napoli (Florence: Olschki, 1990), 14550.
Schachter, Rhythm, 48.
Sheveloff, Keyboard, 258.
Joseph Kermans term, used in Musicology (London: Fontana, 1985), 53.
Schenker, Meisterwerk, 153. For an account of Schenkers general treatment of Scarlatti see Ian Bent, Heinrich
Schenker, Chopin and Domenico Scarlatti, Music Analysis 5/23 (1986), 13149, especially 13940.
Schenker, Meisterwerk, 154.
Schenker, Meisterwerk, 176. Compare the reaction of Sebastiano Luciani, writing a year later than Schenker:
B ulow did not hold back from contaminating and weighing down . . . the airy grace of Scarlattis compositions.
Luciani, Sinfonismo, 43.
Kirkpatrick, Scarlatti, 281.
40 The keyboard sonatas of Domenico Scarlatti
which everything sounds multilevelled and unreal.
If the author goes too far in
pursuit of this ambiguity, he clearly intends to go too far. One can glimpse here a
distant cousin of the ecstatic prose of the panorama tradition.
B ottinger makes the composer almost too unutterably strange for words, too
sensational, but many of his formulations are welcome the description of Scarlattis
unreiner Satz,
the concept of irritation as a formal principle
and represent a
rare attempt to square up to the ambivalent and enigmatic side of the composers art.
If the essay is unhistorical in some senses, in another it has the true historical spirit
of trying to recapture what was new about a given phenomenon. In addition, it is
prepared to take risks and may stand as a polemical corrective to those who imply
that all the difculties around Scarlattis keyboard output are factual, practical and
A very different type of meaning is assigned by two other global rationales for the
sonatas improvisation and pedagogy. These might seem to be mutually exclu-
sive categories, one suggesting the sonatas issue straight from the composers n-
gers, the other that they were carefully written to aid the technical development of
Mara B arbara. Nevertheless, they both fall under the category dened by Pestelli as
While neither seems an unreasonable angle of approach to the
sonatas, both have been overplayed, and not just in Scarlattis case. Their covert pur-
pose, I believe, is to explain an embarrassingly large output froma later point of view
that of the work-concept that became fully established in the nineteenth century
and which is effectively as dominant today as ever. They are a way of justifying the
apparent fact that composers did not give such individual attention to their works, by
appealing to historical circumstance; but they skirt all questions of artistic creativity.
Improvisation is one of the commonest elements of the Scarlatti litany. It is a
problematic rationale because it implies that, whatever other supreme merits the
uvre possesses, considered thought is not one of them. This becomes explicit in
Boyds comparison of the sonatas and cantatas. First we read: Much of the keyboard
music of the period, and Scarlattis perhaps more than most, sprang directly from
the composers ngers, as it were, in the act of improvising. On the other hand,
though, the writing of vocal music was a considered activity, subject to the demands
F. 244: 4 Ann aherungen an eine Sonate, in Musik-Konzepte 47 (1986), 80.
B ottinger, Ann aherungen, 75 (Die Kunst des unreinen Satzes); the concept is amplied from 75 to 92, and I
return to it especially in Chapter 5 of this study. The phrase itself, playing on Schenkers Der freie Satz, means
unclean or impure composition.
B ottinger, Ann aherungen, 101.
I translate Kathleen Dales characterizing phrase tecnico-manualistico from her review of Le sonate di Domenico
Scarlatti: proposta di un ordinamento cronologico by Giorgio Pestelli, Music and Letters 49/2 (1968), 184, rather than
Pestellis commonly used tecnico-pianistic[o] (see Pestelli, Sonate, 3 and 5, for example), since the latter may
sound too narrow in its application. The only logical English word that can cover all the necessary ground,
keyboardistic, is too ugly ever to have caught on.
Panorama 41
of the text and the rules of good composition .
This sense of looser creativity
inherent in keyboard music has also led to the suggestion that the sonatas may have
been dictated improvisations, attractive because it seems to offer an explanation for
the absence of autographs.
This is not to deny the particular physical immediacy of so much keyboard music
of the eighteenth century, and certainly not Scarlattis, nor the sense of rhetorical
freedom in this repertoire, compared to, say, a string quartet or a cantata, but these
properties need to be reformulated. Reference to improvisation can become a tool
of evasion unless the terms of its employment are carefully thought through. It is ne
if it can be understood in the applied Schenkerian sense that all tonal composition
(at least at the highest creative levels) partook of improvisation. This was possible
because of the relatively secure nature of tonal rhetoric in the eighteenth century, all
its syntactical, harmonic and melodic manoeuvres what Rose Rosengard Subotnik
calls the supreme condence of a style in which. . . tonality was so secure.
because of such condence, the distinction between creating and playing about
was far from hard. And so to single out keyboard music for improvisatory attributes
misconceives the nature of creativity altogether at the time. Indeed, Charles Rosen
comments: The forms and textures of the early eighteenth century altogether are
closer to improvisation than those of any other time in Western music before jazz.
Within those terms of reference we may then allow that the physical engagement
entailed in keyboard composition may have made improvisation an even more vital
force. Without those terms, though, we have an approach that simply denies Scarlatti
his extraordinary compositional virtuosity.
The other technical-manual rationale pedagogy has been a millstone round the
neck of all eighteenth-century keyboard music.
The perception of the keyboard
sonata, for example, is that it is a small form small not just in the obvious
physical senses but also in aesthetic import and an amateurs form, predominantly
female, domestic, didactic. Such associations very often seem to circumscribe the
scope of scholarly treatment, which is modest, careful, clean: in other words, all the
undeconstructed feminine virtues. In a wider context, the frequent and logistically
Domenico Scarlattis Cantate da camera and their Connexions with Rome, in H andel e gli Scarlatti a Roma,
proceedings of conference in Rome on 1214 June 1985, ed. Nino Pirotta and Agostino Ziino (Florence:
Olschki, 1987), 2589.
An extreme version of this claim may be found in Chambure, Catalogue, 910.
Deconstructive Variations: Music and Reason in Western Society (Minneapolis: University of Minnesota Press, 1996),
239n. The Schenker disciple, Felix Salzer, expressed something similar when he wrote that music had reached
that unconscious stage of musical expression so vital to the development of an artistic language. Structural Hearing:
Tonal Coherence in Music (New York: Dover, 1962), 6.
Bach and Handel, in Keyboard Music, ed. Denis Matthews (Harmondsworth: Penguin, 1972), 74n.
Some of the material that follows in this paragraph has been drawn from my review of books by Bernard Harrison
and John Irving, No Small Achievement, Times Literary Supplement 4949 (1998), 20.
42 The keyboard sonatas of Domenico Scarlatti
understandable recourse to keyboard music, especially that of the eighteenth century,
in any number of teaching contexts in the present day has reinforced the didactic
image. The Scarlatti sonatas have not suffered from such taints as badly as many
keyboard repertories; the sheer difculty of so many of them has seen to that.
Nevertheless, one wonders whether an implicit feminine gendering of most of the
repertory does not play a part in its relative obloquy in the current climate the
dolls house of domestic connement next to the mans world of public genres like
opera and symphony.
Even at a higher level of technical prociency, after all, many
eighteenth-century keyboard composers were associated with distinguished female
protagonists: Scarlatti with Mara B arbara, Mozart with Barbara Ployer, Haydn with
Therese Jansen and Rebecca Schroeter.
The force of such associations may be seen in the book by Hermann Keller, who
describes the sonatas as a Hohe Schule des Klavierspiels, a complete course in key-
board playing.
He devotes a long section to enumerating all the technical features
in which the sonatas were intended to develop prociency scales, arpeggios, oc-
taves, leaps, repeated notes and so forth. In the course of this pedagogical exposition
he notes Scarlattis tendency to use repeated-note chords that evoke the guitar, and
comments: They give the sonatas in which they appear a marked masculine charac-
ter in contrast to the keyboard music of the minor masters of the later eighteenth
century destined more for the use of ladies. Elsewhere he describes C. P. E. Bach as
a feminine composer and Scarlatti as a masculine one, noting too that Bach never
steps outside his bourgeois North German atmosphere.
Kellers anxiety on this
score is instructive; Scarlatti had to be rescued from the female domestic associations
of his genre.
The pedagogical rationale for the sonatas turns up frequently elsewhere: already in
1839 Carl Czerny had asserted the great utility of the sonatas for pianistic study.
That this category again slights the place of artistic creativity is apparent in the much
more recent estimation by Howard Ferguson that, though he may never aim for the
heights reached so effortlessly by Bach, [Scarlatti] extended the technical possibilities
of his chosen medium in a way unmatched by any other composer.
The clear
implication is that, in their concentration on athletic training and development,
For an entertaining consideration of such issues see Richard Leppert, Music and Image: Domesticity, Ideology and
Socio-cultural Formation in Eighteenth-Century England (Cambridge: Cambridge University Press, 1993), especially
Chapter 3, Music, Sexism and Female Domesticity. If my claim about the causes of the unexciting current
image of most eighteenth-century keyboard music is reasonable, then it would appear that the assumptions
explored in this chapter are still with us. One should note too that the eighteenth-century keyboard sonata often
functioned as a safe laboratory for the era of historical (and analytical) positivism.
Keller, Meister, 39.
Keller, Meister, 44 and 83.
Czerny cited in B ulow, Klavierst ucke, i; see also Longo, Preface to Opere complete per clavicembalo di Domenico
Scarlatti, [i]. Note too how the work of Klaus Heimes on Scarlattis near-contemporaries Seixas and Soler gives
central importance to this category through a huge chapter on tutorial aspects in Soler and an extensive
citation of technical passages in Seixas. Antonio Solers Keyboard Sonatas (M. Mus. treatise, University of
South Africa, 1965), 55100; and Carlos Seixass Keyboard Sonatas: The Question of Domenico Scarlattis
Inuence, Bracara Augusta 28 (1974), 45367.
Early Keyboard Music, in Keyboard Music, ed. Denis Matthews (Harmondsworth: Penguin, 1972), 40.
Panorama 43
the sonatas lack any inner content. The assumption that there is a necessary gulf
between the two areas, that one either composes proper music or satises pedagogical
demands, is creatively and historically unrealistic. In any case, we should bear in mind
that the systematization of technique, the isolation of digital features to be practised
independently, did not truly arrive until the nineteenth century. It was this view that
then made of so much eighteenth-century keyboard music a useful stepping stone,
both technically and musically, to the later repertory.
Two other fundamental areas of investigation have been held up as the salvation
for Scarlatti studies chronology and organology. The reliance on a well-established
chronology for almost any form of scholarly musical study has already been explored.
The particular terms of reference for any discussion of this matter have been set by
Kirkpatrick; one of the main reasons he was able to tell such a good story in his 1953
book was that he was so condent of his chronology. All the standard parts of the mas-
ter narrative
can thus take their place, in the conspicuous stylistic development . . .
from the ashy and relatively youthful sonatas of [V 1749] and a few already copied
out in [V1742] through the poetic richness of the middle period of 1752 and 1753. . .
to the most complete and digested maturity imaginable in the late sonatas from 1754
to 1757;
subsequently we read that in the late sonatas everything is at once thin-
ner and richer.
What rendered Kirkpatricks wholly traditional narrative rather
incredible, if not absurd, was that he believed the dates of copying almost coincided
with those of composition. Thus, as he conceded himself, the development of a
was compressed into a remarkably short period.
Malcolm Boyd has made a useful distinction between the two separate strands
of Kirkpatricks chronological claims. He believes there is a good deal of stylistic
evidence to support Kirkpatricks general theory of a direct relationship between
the order of composition and the order of copying into the two main sources; on
While the cure-all of improvisation has never been disputed in the literature, a number of writers have distanced
themselves from the pedagogical view. Roy Howat, for example, believes that the character of the Essercizi
has nothing to do with the dryness of purely didactic exercises, while Massimo Bogianckino states that the
intentional dealing with any one technical problem is not to be found in Domenico Scarlattis sonatas. Howat,
Domenico Scarlatti: Les XXX Essercizi, notes to recording by Scott Ross (Stil: 0809 and 1409 S 76, 1977), [4];
Bogianckino, The Harpsichord Music of Domenico Scarlatti, trans. John Tickner (Rome: De Santis, 1967), 116n.
For a full discussion of the standard evolutionary master narrative see James Webster, Haydns Farewell Symphony
and the Idea of Classical Style: Through-Composition and Cyclic Integration in his Instrumental Music (Cambridge:
Cambridge University Press, 1991), 33547. The power of the traditional narrative is also evident in the BBC
biography, in which at the appropriate stage of the programme we are informed of a late surge . . . a creative
outpouring of old age; Thompson, Scarlatti.
Kirkpatrick, Scarlatti, 145. I have substituted here in square brackets Sheveloffs designations for V XIV and XV,
since the numbering of these last two volumes does not make clear that they antedate those numbered I to XIII.
Kirkpatrick, Scarlatti, 173. This is in itself a standard gambit, what Janet M. Levy calls the concentrated late style
in Covert and Casual Values in Recent Writings about Music, Journal of Musicology 5/1 (1987), 11.
Kirkpatrick, Scarlatti, 145.
44 The keyboard sonatas of Domenico Scarlatti
the other hand, he nds it hard to credit the special theory. . . that the sonatas
were copied into the Venice and Parma sets more or less at the time that Scarlatti
completed them.
This incredulity seems to have been shared by most other writers.
The general theory has been widely accepted; or, it might be more accurate to say,
it is often tacitly applied as a working tool without any direct acknowledgement of
its shaky basis. If one rejects the intrinsic musical status of the pairs, for instance
seeing them as acts of compilation rather than composition then chronology is
immediately destroyed in any specic, if not altogether in a broader, sense.
That some broader sense remains is apparent in the existence of like-minded
groups of works through the Venice and Parma collections. Roughly speaking,
this is most apparent in the sonatas now numbered in the K. 100s, 300s and 500s
and much less so elsewhere. If one accepts the existence, if intermittent, of fairly
homogeneous groupings, then are they the product of retrospective planning or a
reection of the composers various creative periods?
Among those who believe that the groupings reect a real chronological succession
are Kenneth Gilbert, who tells us that the three successive colours used for his edition
correspond to the three creative periods proposed by Kirkpatrick, youth, middle age
and maturity.
The standard developmental narrative is thus coloured in in the most
literal way, as the colours on the covers change from a ery red to a ourishing green
to a rich gold. On the other hand, it has been suggested that that the compilers of
the volumes were creating a sort of anthology, bringing together compositions with
common linguistic characteristics.
Such decision-making, though, would have brought on a headache; how similar
did sonatas have to be, for example, in order to qualify for such adjacency? While
sonatas undoubtedly were brought together to make pairs on the basis of key, the
notion that they were also brought together on the much wider and less quantiable
basis of style and language, in bulk, seems highly unlikely. The case of the sonatas
in Parma VIII and IX (roughly equivalent to Venice VI and VII), as mostly found
in Volume 7 of the Gilbert edition, seems to conrm this. The majority of these
sonatas are so distinctive texturally, topically and even, it would appear, aesthetically,
compared with the rest of Scarlattis output, that it is difcult to believe that they were
not written in a delimited period, prompted by external considerations on which
we can only speculate.
The idea that they were written on and off throughout the
Boyd, Master, 16061.
Gilbert, P eriple scarlattien, in Musiques Signes Images Liber amicorum Franc ois Lesure, ed. Jo el-Marie Fauquet
(Geneva: Minkoff, 1988), 132.
Pestelli, Sonate, 222.
Sheveloff suggests that some of these works may be for clavichord; he seems to believe, however, that there are
only about ten of these pieces, whereas there are surely many more in this distinctive stylistictextural group.
See Sheveloff, Frustrations II, 99101. It is also worth noting that almost no sonatas from the K. 300s appear
in the Lisbon Libro di tocate volume recently published by Doderer, nor in the Vienna II volumes unearthed
by Eva Badura-Skoda in 1971. Roberto Pagano notes an indisputable falling-off in quality in Venice V to
VII (K. 266355) and conjectures that these sonatas may have been intended for the instruction of a new
pupil Fernando. Domenico Scarlatti, in Dizionario Enciclopedico della musica e dei musicisti, Le biograe, vol. 6,
ed. Alberto Basso (Turin: UTET, 1988), 635.
Panorama 45
composers career, closing off most of the avenues freely chosen by Scarlatti in the
surrounding works, then brought together later, seems counterintuitive.
Uniting the concerns of chronology and pedagogy is Emilia Fadini, who offers the
hypothesis that the Venice volumes of 17527 were ordered so as to provide a gradu-
ated keyboard course: the didactic aspect of the production cannot be minimized.
She essentially offers a new telling of an old story with a series of technical crescendi,
traced several times over until the nal synthesis of the last volumes. Her grand plan
certainly has a feel-good factor in the way it emphasizes the coherence of the Venice
collections and skirts any nasty thoughts about chronology. The argument that most
of the sonatas are etudes dex ecution transcendante or, on a lower level, quasi-didactic
lessons transparently acts as yet another attempt to avoid any awkward contempla-
tion of the aesthetic character of the sonatas, never mind the source situation. Much
to be preferred is Kathleen Dales optimism in the matter: because no chronology
is known and hence we cannot follow his development as a composer, playing all
the Scarlatti sonatas is like journeying in a land where it is always spring.
No issue in Scarlatti studies has raised more strong feelings than that of organology,
at least since the time of Sheveloffs provocative theory that the fortepiano may have
been the instrument of choice for a large number of sonatas.
(He also suggests the
suitability of clavichord and organ for a relatively small number of them.) Before
then there seemed no doubt that this was harpsichord music. Many still believe that
the harpsichord was central to the sonorous and technical conception of the whole
output; others simply exclude any reference to the fortepiano.
For this camp the
only question worth debating has been just what sort of harpsichord might best
project these sonatas.
Backing up the Shevelofffortepiano axis has been David Sutherland, who has
reinterpreted existing evidence to suggest that Scarlatti was the pianos rst great
He notes that Scarlatti must have tried Cristoforis new instrument on
trips to Florence in 1702 (with his father and family) and 1705 (with the singer
Grimaldi), at the palace of Prince Ferdinando de Medici, who had supported
Cristoforis work. This does indeed seem so likely that our familiar modal verb
hardly seems necessary. He also observes that the diffusion of Cristoforis pianos . . .
is largely congruent with the geography of Scarlattis career, suggesting that Scarlatti
Hypoth` ese ` a propos de lordre des sonates dans les manuscrits v enitiens, in Domenico Scarlatti: 13 Recherches,
Hours with Domenico Scarlatti, Music and Letters 22/2 (1941), 115. Note that this was written in 1941, before
Kirkpatricks chronology destroyed such enviable possibilities of innocence.
First suggested in Sheveloff, Keyboard, 31941 and 357, then presented more denitively in Sheveloff,
Frustrations II, 90101. He notes that only seventy-three sonatas lie beyond the range of the Queens
For example Alberto Basso, notes to recording by Christophe Rousset (Decca: 458 165 2, 1998), and Gilbert,
P eriple.
Sheveloff, Frustrations II, 90.
Sutherland, Piano, 252.
46 The keyboard sonatas of Domenico Scarlatti
himself was the agent of that diffusion.
This is particularly striking when we recall
that the rst published keyboard works specically designated for the fortepiano, the
twelve sonatas by Lodovico Giustini of 1732, were dedicated to the Infante Don
Antonio of Portugal, brother of King Jo ao V. We should also note the title given
to the extraordinary works of Scarlattis younger colleague Albero in the Madrid
manuscript dedicated to Fernando VI Obras, para clavicordio, o piano forte
which must have been written between 1746 and 1756.
Other scholars have made
a case for the viability of the fortepiano, either through primary research (Pascual,
Pollens, Tagliavini, Badura-Skoda)
or for stylistic reasons (Pagano, for instance,
who believes the young Scarlatti must have realized that the Cristofori instrument
would give him a better way of realizing on the keyboard certain vocal aspects of
his inspiration
). A number of recent writers have mentioned the likely relevance
of the piano to at least a good number of the sonatas as a matter of course.
Just how much does all this intensive research matter? A large industry has grown
up around the attribution of specic works of the eighteenth-century keyboard
repertoire often within the uvre of a single composer to specic keyboard
instruments, perhaps most notably in the case of Haydn.
Yet the most important
lesson to observe from what seems to us now like a muddle of different instru-
ments, makes, ranges and special devices must be that they coexisted for most of
Sutherland, Piano, 250.
Linton E. Powell, The Keyboard Music of Sebastian de Albero: An Astonishing Literature from the Orbit of
Scarlatti, Early Keyboard Journal 5 (19867), 10, 12. It seems most unlikely that the phrase o piano forte was
added later, a point agreed by Powell, Antonio Baciero and Genoveva G alvez; see Powell, Albero, 14 (n.17).
Note too that clavicordio refers here to the harpsichord rather than the clavichord.
Beryl Kenyon de Pascual, Francisco P erez Mirabals Harpsichords and the Early Spanish Piano, Early Music
15/4 (1987), 507, 512; Stewart Pollens, The Pianos of Bartolomeo Cristofori, Journal of the American Musical
Instrument Society 10 (1984), 656; Pollens, The Early Portuguese Piano, Early Music 13/1 (1985), 19; Luigi
Ferdinando Tagliavini, Giovanni Ferrini and his Harpsichord a penne e a martelletti , Early Music 19/3
(1991), 399; Badura-Skoda, Hammerklavier. Badura-Skoda claims that two of the pianos at court went up to
, an assertion that I have not been able to corroborate; Badura-Skoda, Hammerklavier, 528. However, Beryl
Kenyon de Pascual, in discussing the piano in the Seville Museum with a ve-octave range (G
to g
), notes: If
we accept . . . that not all Domenico Scarlattis sonatas were written for the queen, perhaps we should consider
the possibility that some of the works for a 61-note instrument (G-g) were also played on, and perhaps even
composed for, a piano; Pascual, Mirabal, 512. Cristina Bordas notes several other references to the Spanish
piano from before 1750 in Musical Instruments: Tradition and Innovation, in BoydCarreras, Spain, 185n.
Pagano, Vite, 173.
See Bengt Johnsson, Preface to Domenico Scarlatti: Ausgew ahlte Klaviersonaten, vol. 1 (Munich: Henle, 1985), vi,
and Rafael Puyana, Inuencias ib ericas y aspectos por investigar en la obra para clave de Domenico Scarlatti,
in Espa na en la m usica de Occidente, vol. 2, proceedings of conference in Salamanca on 29 October5 November
1985, ed. Emilio Casares Rodicio, Ismael Fern andez de la Cuesta and Jos e L opez-Calo (Madrid: Instituto
Nacional de las Artes Esc enicas y de la M usica, 1987), 56. Note also the arguments in favour of Mara B arbaras
likely early ownership of pianos in Michael Cole, The Pianoforte in the Classical Era (Oxford: Clarendon Press,
1998), 1517.
For recent examples see: A. Peter Brown, Joseph Haydns Keyboard Music: Sources and Style (Bloomington: Indiana
University Press, 1986), especially 16071, including a list of preferred instruments for particular sonatas; L aszl o
Somfai, The Keyboard Sonatas of Joseph Haydn: Instruments and Performance Practice, Genres and Styles, trans. the
author in collaboration with Charlotte Greenspan (Chicago: University of Chicago Press, 1995), Part I; and
Bernard Harrison, Haydns Keyboard Music: Studies in Performance Practice (Oxford: Clarendon Press, 1997), 132.
Panorama 47
the eighteenth century. We must also bear in mind that the various instruments
were not necessarily as distinct in sonority as we might imagine today, that they
were in many cases tolerably similar.
The very title of Alberos Madrid works
suggests a relaxed attitude to what was deemed appropriate for a particular keyboard
instrument. Yet this in turn seems unsatisfactory. If we concentrate once more on
the case of Scarlatti, no one could deny for all the differences of organological
opinion the extreme sensitivity to sound exhibited by the composer. There is no
aspect of his style more marked by originality, both in conception and execution.
Given this, what did Scarlatti actually hear when he created his soundscapes? Surely
his point of departure was the colours and possibilities of particular instruments.
The implications of this organological indifference have not really been followed
up in the literature. Even if one prefers the notion of discrete groups of sonatas for
different instruments, it is difcult to imagine any keyboard composer, including
Scarlatti, schizophrenically conceiving rst one sonata or group of sonatas for one
instrument, then a second for another, especially when his larger style remains seem-
ingly immune to such proposed shifts.
And there is a larger question of composing
principle: it is not within the gift of the composer to control the precise sound qual-
ities of a performance. Even if we accepted that all the sonatas were conceived on
and meant for the harpsichord, we would then have to ask which particular harpsi-
chord in which royal palace was the authentic source for the technical and sonorous
properties of an individual piece. Perhaps one might claim the sonatas of Scarlatti
are keyboardistic in the rst instance, that gesture could be as important to their
conception and realization as is sonority.
Yet one feels that the battle will continue, particularly on the part of the
Kirkpatrickharpsichord axis. Frederick Hammond, for example, has recently writ-
ten: Scarlatti might have conceived a fewmonochromatic sonatas for the early piano,
but there is no reason to suppose that a composer already acknowledged in his youth
as a master of the visually and aurally splendid harpsichord should have taken any
more interest in the fortepiano than Artur Rubenstein took in the clavichord.
The rst part of this sentiment rests on Kirkpatricks speculation that a few of the
pieces in Venice I and II with inert bass lines might represent experiments in writ-
ing for the early piano;
but why would the composer respond to a touch-sensitive
instrument with monotonous and thin textures? The most likely answer, that the
piano could do with dynamics what the harpsichord had to do with texture, does
not seem adequate.
Also highly sceptical about the possibility of the piano is John Henry van der
Meer, who has recently constructed a new chronology for the sonatas based on
See Edward Ripin, Haydn and the Keyboard Instruments of his Time, in Haydn Studies, ed. Jens Peter Larsen,
Howard Serwer and James Webster (New York: Norton 1981), 305.
With the seeming exception of the group of sonatas concentrated in the early K. 300s referred to earlier.
Hammond, Scarlatti, 167. Another instance of such an anti-piano reaction, misquoting David Sutherland
along the way, may be found in PaganoBoyd, Grove, 403.
Kirkpatrick, Scarlatti, 184.
48 The keyboard sonatas of Domenico Scarlatti
the development essentially the progressive outward expansion of harpsichord
range. He believes with Kirkpatrick that the pianoforte was used at the Spanish
court only for accompanying and cites as evidence the lack of dynamic nuances in
the manuscripts;
one might counter with the relative lack of any clear indications
for changing manuals.
Not all the harpsichords we have information about were
one-manual instruments and there may well have been others we do not know of.
The inventory of the Queens instruments drawn up in 1758 has been used as the
basis for most discussion of Scarlattis keyboard instruments, but its value rests on two
assumptions: that the sonatas were conceived only in terms of her instruments and
that the collection remained unchanged over a long period.
In fact, as Sutherland
points out, Kirkpatrick chose to ignore certain evidence he had himself quoted
which argued against his assertion of the pianos accompanying role (reinforced by
an appeal to Farinellis fondness for the pianoforte).
This was a reference by
Burney to a harpsichord with a transposing keyboard, which was probably the third
instrument in the Queens inventory and which could only have been used for
The recent publication of the Farinelli inventory conrms this
claim that the accompanying instrument was a harpsichord.
Van der Meers new chronology suggests that the composition of the sonatas was
spread over most of Scarlattis career: thus 13 per cent of the works (including almost
all the Essercizi) were probably written in Italy and 24 per cent in Portugal and Spain
up to c. 1740. These remarkable claims rest on shaky methodological foundations.
The author states that when a sonata with one range is paired or arranged in a group
of three with compositions with a larger compass, it has been taken for granted that
the work in question belongs to the group with the larger compass.
This is a fatal
aw; van der Meer does not so much as acknowledge the very many writings that
point up the clear weaknesses in Kirkpatricks pair theory.
It is also surely dangerous
to base a chronology purely on range. Might the Essercizi, for example, have been
The Keyboard Instruments at the Disposal of Domenico Scarlatti, The Galpin Society Journal 50 (1997), 153. The
only exceptions to this are K. 70 and K. 88, two sonatas widely believed to be accompanied works (Sheveloffs
term is melo-bass sonatas). Van der Meer makes the rather extraordinary suggestion that the dynamics, rather
than giving instructions to the string player(s), imply that the accompaniment to the violin would have been
performed on the piano.
Some possible instances are discussed in Sheveloff, Keyboard, 34251.
See Hammond, Scarlatti, 168, and Sheveloff, Keyboard, 327.
Sutherland, Piano, 251 and Kirkpatrick, Scarlatti, 184.
See Sutherland, Piano, 251.
See Cappelletto, Farinelli, 210, and also Boyd, Scarlatti and the Fortepiano in Spain, Early Music 24/1 (1996)
(Correspondence, with reply by David Sutherland), 189.
Van der Meer, Keyboard, 140.
Almost uncannily appropriate to the present case are Sheveloffs words from his 1970 dissertation: It is dangerous
to make . . . assumptions about range based on the evidence of pairing; in fact, one of the most inadvisable of
procedures would be the formation of a tripod based on chronology-organology-pairing[,] using the evidence
of one to justify the other. This sort of circular logic creates a series of links, any one of which, if effectively
broken by the introduction of new evidence or the more efcient and logical use of old evidence, causes all
three basic elements of the tripod to fall. Sheveloff, Keyboard, 337.
Panorama 49
deliberately restricted in this respect because of the organological imponderables of
a foreign market?
Aside from this, it is not clear why the sonatas use of range should always auto-
matically expand outwards, even assuming the progressive expansion posited by the
author. It may be true more often than not that, as van der Meer claims, Scarlatti
does try to use the highest available note in individual sonatas, but for the substantial
number where this is not the case, such an ordering principle is misleading. And
what about revisions to the sonatas? An example that has recently come to light serves
to illustrate the slippery nature of such considerations. In the copy of K. 474 found
in the Lisbon Libro di tocate, published in 1991,
there are several points where a
lower-octave doubling is given with the high e
, presumably in the manner of an
ossia. It would seem as if d
was the highest note available on the instrument for
which the copy of the sonata was made. Or perhaps this was an example of caution-
ary editing, with the copyist or composer unsure about the range of the instruments
at court in Lisbon. Or perhaps the sonata originally existed in the narrower-range
version and this was a suggested expansion based on knowledge of the instruments
at Lisbon. In any case, this very tangle of possibilities gives some sense of how pro-
visional conclusions based on compass can be. Deepening the mystery in this case
is that the surrounding sonatas in the Lisbon collection are registrally much more
It was suggested previously that harpsichordists have by no means all shown the
great interest in Scarlatti that one might have expected. Keller, writing in 1957,
felt that some peculiarities of Scarlattis writing seemed atly to contradict harpsi-
chord style, especially the use of octaves (see K. 487 for an example of the sort of
texture to which he was referring). The fact that many sonatas sounded better on
the piano than on the harpsichord, and vice versa, had not helped: conscientious
pianists shy away from playing [Scarlatti] on their instrument, thinking it is really
harpsichord music; harpsichordists are uncomfortable, feeling that this is no longer a
clear-cut harpsichord style, and so dont play him. If only both sides would play him
at all . . . !
Albert Einsteins maxim that the secret of creativity is knowing how to hide your
sources would seemto apply particularly well to Domenico Scarlatti. Two corollaries
of this have already been stressed: the composers relative lack of historical situatedness
and the consequent claims for an absolute originality. In a sense, Scarlatti is only one
Sheveloff believes that the avoidance of notes above c
in the Essercizi may represent cautionary editing;
Sheveloff, Keyboard, 327.
The circumstances of this publication are explained more fully below on pp. 6970.
Keller, Meister, 378.
50 The keyboard sonatas of Domenico Scarlatti
of the most distinguished victims of a musicological malaise about mid-eighteenth-
century music, which is treated from another angle in James Websters study of
Haydns Farewell Symphony. Whereas Haydn has suffered from inadequate critical
apparatus with respect to the rst half of his output, all Scarlattis keyboard works
may be said to fall within the age of uncertainty, born of what Webster calls the
notion of a general inadequacy in mid-century music.
The difculties of style
classication that beset Scarlatti and others, though, are not so much a symptom of a
general historiographical problem as one particular at least in its intensity to the
eighteenth century. Of course, as soon as musical periods or eras are invoked, there
will be grey areas that affect all sorts of composers and genres, involving dualities
such as mainstream vs. peripheral or central vs. transitional; but it is not a question
of whether composer x receives a good deal or suffers from distortion. After all, the
position of those who are securely based within a period is just as constructed, just as
conditional, as the position of those who are a bad t. More to the point is whether
the prevailing system of thought allows for ease of treatment.
This is why the problem is so acute for the perception of eighteenth-century
music. The looming edice of Classicism, so tightly dened and entrenched in its
stylistic and aesthetic values, has made it very difcult to deal with a vast quantity
of surrounding music without a bad conscience. The traditional consensus has
been that (Viennese) Classicism is fully operative only from about 1780. It is also
generally felt that what we call the Baroque has begun to unravel by about 1720,
if not earlier; only the activity of J. S. Bach, until 1750, has distorted this in the
popular imagination. This yields a period of uncertainty and transition of some sixty
years, comprising most of the eighteenth century; absurdly, this is longer than the
ensuing Classical style itself! There is a corresponding difculty on the other side
of the edice, although not one that is chronologically so xed. Composers such as
Hummel, Dussek and Clementi have also fared badly, tainted with similar epithets
inadequate, impoverished, illogical, extravagant, manneristic as their pre-Classical
soul mates. Schubert, whose instrumental music might also belong here, has proved
somewhat less problematic, perhaps because his songs have offered writers a get-out
However, we cannot overcome such uncertainty by starting with a clean slate,
free of any periodization; it would be unrealistic, perhaps even dishonest, to claim
that we can dispense entirely with such ingrained terms of reference. If it is the
associations of the two terms as much as anything else that have caused such recent
disquiet the extravagance of one, the ordered, exemplary, and indeed geographically
specic nature of the other Baroque has just about become the equivalent of a
dead metaphor, its original associations now invisible to us. Classical, on the other
hand, seems unlikely to atten out in the same way; it has too charged a history.
Nevertheless, the persistence of the two terms testies to the sense that there is indeed
a fundamental artistic and cultural change at issue, but this needs to be treated in a
Webster, Haydns Farewell Symphony, 340.
Panorama 51
more nuanced way and cannot be thought of as having a clear point of arrival. A
starting point is the distinction suggested by George Hauer, that an aristocratic
courtly attitude to art produces the Baroque, while a democraticbourgeois attitude
produces Classicism.
In this problematic quest for historical identity I have already suggested that Scar-
latti can be understood as being as much a willing accomplice as a helpless victim,
given his high level of self-awareness.
The famed originality has a part to play in
this equation. It is often painted as a relatively innocent, inherent quality anything
but self-conscious but, while it may have a spontaneous side, it is also calculated.
If the historical recipe for the pre-Classical transitional period offers confusion, un-
certainty and plurality, for Scarlatti it offers opportunity. It is as if, sitting on our
shoulders, Scarlatti revels in his historical status.
This is not a luxury that critics can share. In their attempts at stylistic classica-
tion, they have chosen to emphasize different ingredients: the past (Baroque, but
sometimes also Renaissance polyphony), the uncertain present (galant/Rococo/
pre-Classical/post-Baroque/mid-century style), the near future (Classical), the far
future (modernism) or none of the above (originality). The ultimate in uncertain
stylistic placement is, of course, absence. The silent discrimination practised by many
generalist works has already been noted. Many Italian writers of more recent vintage
have emphasized the Baroque orientation of the composers work, and more gener-
ally his indebtedness to native traditions. One of the central strands of Pestellis book
details Scarlattis war against the modern galant style, one which is nally openly
declared in the many of the alleged late works; Scarlattis weapon of choice is the
Italian toccata as practised by his father.
On the other hand, many other recent
writers have claimed the composer for Classicism almost as a matter of course.
The two nal categories, originality and modernism, have rarely been invoked
in recent times. The sense that we can make relatively direct contact with the past,
that we can engage in unmediated dialogue with earlier gures, has fallen from
favour: contextualization is all. Equally, notions that composers exist outside time
(originality) or for the future (modernism) are an embarrassment, even though on
one level their very treatment in a present-day context is premised on just such
attributes. The balance of historiographical consciousness has shifted: we are now
almost painfully aware of our partiality as interpreters of the past, condent only that
Cited in Pestelli, Sonate, 23.
Several other writers have pinpointed this quality. For Bogianckino, Scarlatti is extremely conscious of his own
style, for Boyd he is one of the most style-conscious of all composers, while Degrada writes of sua sempre
vigile ricerca espressiva (his ever-vigilant search for expression). Bogianckino, Harpsichord, 43; Boyd, Master,
116; Degrada, Lettere, 309.
See Pestelli, Sonate, 25963.
These include Lang, Rosen (for whom the sonatas provide the rst signicant examples of [the] new dramatic
style), Alexander Silbiger (Scarlattis spirited buffo style) and Daniel K. L. Chua (who cites a passage of reiterated
gures from K. 521 to illustrate how cadential forces generate the energy of the Classical language). Lang,
300 Years, 5878; Rosen, Classical, 43; Silbiger, Scarlatti Borrowings in Handels Grand Concertos, The
Musical Times 125/1692 (1984), 93; Chua, The Galitzin Quartets of Beethoven: Opp. 127, 132, 130 (Princeton:
Princeton University Press, 1995), 166.
52 The keyboard sonatas of Domenico Scarlatti
our deliberations and (even more) those of past generations will reect the present.
Scarlatti intrudes with particular urgency on this newer state of affairs, since all his
circumstances seem to demand bolder explanations.
These were provided en masse in times that were less deferential to matters of
historical method and generally asserted an absolute independence from any geo-
graphical or temporal location. All of these seem to have taken a lead from Burney,
who had written in the 1770s that Scarlatti was truly inimitable . . . the only original
Genius, who had no Issue; and who formed no School.
Such declarations have
made of Scarlatti a force of nature, not a product of culture. It would be insufcient
simply to align them with older historical ways; in a similar manner to the panoramic
prose exemplied earlier, these declarations take their cue from the transcendental
physicality that so many have found in the sonatas.
The other way of removing Scarlatti from the clutter of contemporary association
was to assert his modernism. In one of its incarnations this is not a historically
problematic claim. Scarlatti can readily be situated in the context of the quarrel of the
ancients and moderns. For Burney again, Scarlatti was the rst composer to embody
the modern spirit, the rst who dared give way to fancy in his compositions.
Theodor Adornos denition of what modern meant for Bachs time involves similar
claims: it meant to throw off the burden of the res severa for the sake of gaudium. . . ,
in the name of communication, of consideration for the presumptive listener who,
with the decline of the old theological order, had also lost the belief that the formal
vocabulary associated with that order was binding. Wilfred Mellers dubbing of
Scarlatti as an eighteenth-century modernist is meant in the same spirit. The com-
poser, he tells us, wanted to do his own thing.
This use of colloquialism cleverly
reminds us of the perennial nature of this process. Claims for greater human relevance,
after all, accompany every artistic change, which is thus by denition modernist; this
even applies to new conservative strategies.
The other kind of modernism associated with the composer claims him as a
prophet or kindred spirit of the twentieth century, when the term also comes to
describe an artistic movement or period. Thus Max Seiffert in 1899 saw in Scarlatti
a prophet on the threshold of the modern epoch,
while Edward Dent in 1935
took him to be a sort of primer to modernism:
One result of that musical revolution which began with Debussy and is still in the process
of discovering the music of the future is that we have learned to appreciate and enjoy much
of the music which theorists and historians of the last century condemned as barbarous or
even licentious Mussorgsky, Berlioz, Gesualdo, Prince of Verona, P erotin and the early
Cited in Kate Eckersley, notes to recording (Loves Thrall: Late Cantatas, vol. 3) by Musica Fiammante
(Unicorn-Kanchana: DKP(CD)9124, 1992), 5.
Burney cited in Eckersley, Thrall Notes, 4; Adorno, Bach Defended from his Devotees, in Prisms, trans. Samuel
and Shierry Weber (Cambridge, Mass.: MIT Press, 1981), 141; Mellers, The Masks of Orpheus (Manchester:
Manchester University Press, 1987), 84.
Geschichte der Klaviermusik, third revised and expanded edn of C. F. Weitzmann, Geschichte des Klavierspiels and
der Klavierliteratur, I: Die

Altere Geschichte bis um 1750 (Leipzig: Breitkopf und H artel, 1899), 426.
Panorama 53
medieval composers . . . We need today, more than our grandparents did, width of musical
receptivity. We are faced everywhere with new types of music which are at rst difcult
to understand and enjoy. The study of such a man as Domenico Scarlatti will help us to
adapt our minds to new outlooks and to look forward to the future with sympathy and
If this associates Scarlatti with a humanistic breadth and tolerance, with openness
to the new, there was a very different modernist interpretation, the moderno mec-
canismo cited by Longo in the preface to his edition. This pregured Scarlattis
adoption by the Italian Futurist movement, which found in him just the image of
unyielding speed, elemental rhythmand il movimento aggressivo to serve their anti-
Romantic ends.
As noted at the outset, this image has persisted more than many
critics would like. Gino Roncaglia in fact subsequently dubbed Scarlatti a futurist,
although as much with respect to his supposed anticipations of the nineteenth as the
twentieth century. On the bicentenary of the composers death he wrote, Domenico
Scarlatti . . . is more alive than ever in the sensibilities and tastes of modern life.
Again it would be easy to emphasize only the historical moment of such a remark;
but it is worth more contemplation, since of all eighteenth-century masters Scarlatti
surely awakes the least nostalgic sentiment. (At least, if he is played with the vigour
which seems to be his due. The history of culinary interpretation of the sonatas
will be addressed in Chapter 6.) From this point of view at least, it is no accident
that Scarlatti also acted as a catalyst for neo-Classicism. Other modernist attributions
do not just invoke the spirit of the music; they suggest that its very materials and
techniques are comparable to those of the twentieth century.
It is worthy of note that similar sentiments whether they are simply determined
by this imagery or not have come from outside the musical world. In his historical
novel Baltasar and Blimunda, set in the reign of Jo ao V, the Portuguese writer Jose
Saramago includes the personage of Domenico Scarlatti. In tandem with the main
ofcial thread of the novel, the building by the King of a monastery at Mafra, the
characters of the title are involved with one Padre Bartolemeu in the construction
of a ying machine, known as the passarola, to which Scarlatti lends his enthusiastic
support. If Padre Bartolemeus Passarola were ever to y, I should dearly love to travel
in it and play my harpsichord up in the sky, urges our composer. Subsequently we
come across the description: Meanwhile the musician tranquilly composed his music
as if he were surrounded by the vast silence in outer space where he hoped to play
one day.
Thus are the tendencies to futurism and moderno meccanismo united.
Another recent manifestation of such modernity comes from the choreographer
Domenico Scarlatti: 16851935, The Monthly Musical Record 65/770 (1935), 177.
See Pestelli, Sonate, 323.
Domenico Scarlatti nel secondo centenario della sua morte, in Immagini esotiche della musica italiana, Accademia
Musicale Chigiana (Siena: Ticci, 1957), 67 and 69.
See Alain de Chambure, Les formes des sonates, in Domenico Scarlatti: 13 Recherches, 53, and John Trend,
Manuel de Falla and Spanish Music (New York: Alfred A. Knopf, 1929), 149.
Saramago, Baltasar and Blimunda, trans. Giovanni Pontiero (London: Jonathan Cape, 1988), 161.
54 The keyboard sonatas of Domenico Scarlatti
Siobhan Davies: Ive just started working with two pieces of music by Scarlatti . . .
and I nd this well of idiosyncratic, imaginative verve just racing through the music.
Scarlatti begins to seem remarkably contemporary. You feel the way he has extem-
porised and gone beyond the familiar.
There are many more writers, of course, who have dug for Scarlattis roots in the
past, in search of inuences on and sources for his style. Claims have been made on
behalf of such composers as Pergolesi, Corelli, Frescobaldi, Greco, Durante, Vivaldi,
Alessandro Scarlatti and Marcello and such other ingredients as the Neapolitan opera
sinfonia, the Italian operatic aria, the rened aristocratic sensibility of the Arcadians
and the polyphony of the sixteenth-century ecclesiastical masters.
Perhaps the
most intriguing suggestions do not involve direct instrumental precedents. If the
Italian operatic world is a reasonably well acknowledged part of any equation, less
commonly considered have been certain specic aspects of the Neapolitan scene.
One might recall Burneys story of the Neapolitan violinists who amazed Corelli
with their easy, brilliant sight-reading of passages he found very difcult;
note also
this description of seventeenth-century Neapolitan singing style fromJ.-J. Bouchard:
Neapolitan music is striking above all for its lively and bizarre movement. The manner of
singing . . . is brilliant and rather hard: in truth, not so much gay as odd and scatty, pleasing
only by virtue of its quick, dizzy and bizarre movement . . . ; it is highly extravagant in its
disregard for continuity and uniformity, running, then stopping suddenly, leaping from low
to high and high to low, projecting the full voice with great effort then suddenly containing
it again; and it is in precisely these alternations of high and low, of piano and forte, that one
recognizes Neapolitan singing.
These traits might remind us of many aspects of Scarlattis virtuosity and melodic
Siobhan Davies, A Week in the Arts, The Daily Telegraph, 20 May 1995, A5.
For these attributions see: Richard L. Crocker, A History of Musical Style (New York: McGraw-Hill, 1966), 349,
Degrada, Lettere, 31011 (Pergolesi); David Fuller, The Dotted Style in Bach, Handel, and Scarlatti, in
Bach, Handel, Scarlatti: Tercentenary Essays, ed. Peter Williams (Cambridge: Cambridge University Press, 1985),
117 (Corelli); Ife, Scarlatti, 9 (Frescobaldi); Friedrich Lippmann, Sulle composizioni per cembalo di Gaetano
Greco, in La musica a Napoli durante il Seicento, proceedings of conference held in Naples on 1114 April 1985,
ed. Domenico Antonio DAlessandro and Agostino Ziino (Rome: Edizioni Torre dOrfeo, 1987), 293 (Greco);
Degrada cited in Pagano, Vite, 183, and Pagano, Piena utilizzazione delle dieci dita: una singolare applicazione
della parabola dei talenti, in Domenico Scarlatti e il suo tempo, 857 (Durante); Michael Talbot, Modal Shifts in
the Sonatas of Domenico Scarlatti, in Domenico Scarlatti e il suo tempo, 334 (Vivaldi); Pestelli, Sonate, 6786,
and Pestelli, Bach, Handel, D. Scarlatti and the Toccata of the Late Baroque, in Tercentenary Essays, 27791
(Alessandro Scarlatti); William S. Newman, The Keyboard Sonatas of Benedetto Marcello, Acta Musicologica
29/1 (1957), 38 (Marcello); Rita Benton, Form in the Sonatas of Domenico Scarlatti, The Music Review 13/4
(1952), 270 (the opera sinfonia); Dent, A New Edition of Domenico Scarlatti, The Monthly Musical Record
36/430 (1906), 221 (Italian operatic arias); Degrada, Scarlatti[,] Domenico Giuseppe, in Enciclopedia della
Musica, vol. 5, ed. Claudio Sartori (Milan: Rizzoli Ricordi, 1972), 358 (the Arcadians); Kirkpatrick, Scarlatti,
115 (sixteenth-century polyphony).
Cited in Pagano, Dita, 845.
Cited in Pestelli, Sonate, 44.
Panorama 55
Another ingredient of this sort, as found in the list above, is the seemingly surpris-
ing one of sixteenth-century vocal polyphony. When mixed with the example of
certain Italian Baroque masters, this has often been said to provide the secure techni-
cal basis from which the sonatas could take ight. Kirkpatrick, for instance, held that
the example of Gasparini, Corelli and Pasquini gave [Scarlatti] the same power to
tame the luxuriance of his fancy; furthermore, the Spanish inuence was assimilated
and distilled with all the rigor that Scarlatti had learned from his sixteenth-century
ecclesiastical masters.
Whether many subsequent similar conclusions were inde-
pendently reached or simply form part of a characteristic Scarlattian litany is not
clear, but it is difcult to see the logical connection they all make.
Why did
other well-schooled composers, of Scarlattis or another generation, not attempt the
same originality or experiment? Such judgements remove the crucial element of
choice from the stylistic equation. The same is true with Bouchards description of
Neapolitan song; if this bears on Scarlattis style, it will be because the composer had
an ear open for it. Leonard B. Meyers words on the nature of inuence will prove
germane here. They are crucial to situating all aspects of Scarlattis style, above all
the vexed matter of Iberian inuence.
The nature of inuence, like that of creativity, has been misunderstood because emphasis
on the source of inuence has been so strong that the act of compositional choice has been
virtually ignored. And when the importance of the prior source is thus stressed, there is a
powerful tendency unwittingly to transformthat source into a cause, as though the composers
choice were somehow an effect, a necessary consequence of the mere existence of the prior
source . . . Other composers were in all probability exposed to the same piece of music or
external conditions without being inuenced, or they might have been affected in quite
different ways.
Turning the telescope in the other direction, to try to determine the extent of
Scarlattis inuence on subsequent generations, is just as vexed a procedure.
Leaving aside the various possible interrelationships with the other most promi-
nent members of the Iberian Keyboard School, Seixas, Albero and Soler, and the
English cult of Scarlatti,
as well as the different inuence provided by the com-
posers modernism, we have great difculty in making strong connections. This
difculty once more serves the cause of Scarlattis originality and the cause of the
Kirkpatrick, Scarlatti, 42 and 115.
See for example Ife (the relative conservatism of his musical training in Italy is often the key to his originality)
or Andreani (the solidity of an acquired older technique, that of Palestrina and the motet style, can allow (and
explain) all the anomalies of the composers writing). Ife, Scarlatti, 19; Andreani, Sacr ee, 98.
Style and Music: Theory, History and Ideology (Philadelphia: University of Pennsylvania Press, 1989), 143 and 144.
Malcolm Boyd devotes a whole chapter to this area; see Boyd, Master, 20523.
This is examined in Richard Newton, The English Cult of Domenico Scarlatti, Music and Letters 20/2 (1939),
13856. Linton E. Powell looks to possible inuence on a subsequent generation of Spanish keyboard composers
in The Sonatas of Manuel Blasco de Nebra and Joaqun Montero, The Music Review 41/3 (1980), 197206.
56 The keyboard sonatas of Domenico Scarlatti
isolationists among the critical community, who maintain that the composer did
not inuence the wider course of music history.
Uppermost in the sights of the stylistic assimilationists,
on the other hand,
has been the edice of Viennese Classicism. This interest follows naturally from
the ideological drive of eighteenth-century music history as characterized earlier.
For a long time this was thought to be primarily a question of spiritual orienta-
tion, since there seemed no evidence of any widespread promulgation of Scarlattis
works in Vienna. In 1971, however, the discovery by Eva Badura-Skoda of twelve
collections of Scarlatti sonatas in manuscript, in the archives of the Gesellschaft
der Musikfreunde, led to some re-evaluation. There were of course close connec-
tions obtaining between the courts in Lisbon, Madrid and Vienna; Joseph II and
Mara B arbara were cousins.
Figures who may well have been involved in the
Viennese promulgation of Scarlattis keyboard music include lAugier, Metastasio
and Porpora.
Armer logistical link is Giuseppe Scarlatti, our composers nephew.
It is generally assumed that Giuseppe visited his uncle in Spain before 1755, possibly
as a result of the performance of his opera LImpostore in Barcelona in 1752, and thus
he may have been the agent of transmission for the sonatas to Vienna.
Before such revelations gave greater plausibility to any theories of inu-
ence, the spiritual orientation emphasized the possible Scarlattian inheritance of
Most of the proposed links hinged around a certain physicality and
taste for reiteration, most easily localized in the scherzo spirit that Scarlatti was
thought to hand to the later composer.
Many of these references in the older
literature may of course tell us more about the status of Beethoven at that time
the composer as touchstone for any form of music appreciation than the dynamics
of inuence. The composer who succeeded Beethoven as axial point of the musical
universe, Mozart, is omnipresent in Pestellis book. The author detects in Scarlatti
David Yearsley, in a study of hand-crossing in the 1730s, is able to construct a case for Scarlattis inuence on what
seems to have been a Europe-wide phenomenon, even though in a strict positivistic sense the documentation
does not support this. His arguments suggest that a certain conception of evidence is what has constrained
investigations of the composers historical impact rather than lack of evidence as such. The Awkward Idiom:
Hand-Crossing and the European Keyboard Scene around 1730, Early Music 30/2 (2002), 22435.
I borrow the concept and terminology of isolationism vs. assimilationism from Daniel M. Grimley,
Peripheralism, Acculturation and Image in Fin-de-Si` ecle Scandinavian Music (M.Phil. dissertation, University
of Cambridge, 1995).
Noted in Badura-Skoda, Die Clavier-Musik in Wien zwischen 1750 und 1770, Studien zur Musikwissenschaft
35 (1984), 74.
These are examined in Federico Celestini, Die Scarlatti-Rezeption bei Haydn und die Entfaltung der
Klaviertechnik in dessen fr uhen Klaviersonaten, Studien zur Musikwissenschaft 47 (1999), 967.
See Seunghyun Choi, Newly Found Eighteenth[-]Century Manuscripts of Domenico Scarlattis Sonatas
and their Relationship to Other Eighteenth[-] and Early Nineteenth[-]Century Sources (Ph.D. dissertation,
University of Wisconsin, 1974), 10812.
See Philip Radcliffe, The Scarlattis: (ii): Domenico Scarlatti (16851757), in The Heritage of Music, ed. H. J. Foss
(London: Humphrey Milford/Oxford University Press, 1934), 29; Luciani, Sinfonismo, 44; Dent, Scarlatti,
176; Henry Cope Colles, Sonata, in Groves Dictionary of Music and Musicians, fth edn, vol. 7, ed. Eric Blom
(London: Macmillan, 1954), 896.
See for example B ulow, Klavierst ucke, ii; Malipiero, Scarlatti, 480; Villanis, Italia, 170.
Panorama 57
the gleam of future Mozartian spirituality
and is able to point to a number of
linguistic similarities. Indeed, there are moments where the diction of the sonatas is
uncannily Mozartian which is just the traditionally unhistorical way of pointing to
Mozarts Italian operatic and galant heritage. For all that, the surfeit of comparisons
with Mozart does not get us very far, since there is a more rewarding comparison to
draw with Haydn, whose keyboard works are sometimes held to show a Scarlattian
Quite often, though, the attribute Scarlattian amounts to no more
than a avour of rapidity and agility. Any links between Haydn and Scarlatti, as
already proposed in the rst chapter, are more a question of creative mentality than
coincidences of material or texture.
An element hinted at in the stylistic classications reviewed above, and a major fac-
tor in Scarlatti reception, is nationalism. This operates at the two levels suggested
in Chapter 1. First there is the overarching characterization of Latinate art in op-
position to the Austro-German mainstream, one largely subscribed to by Latin and
non-Latin writers alike. The attributes evoked are highly essentialized and must be
so to full the cultural dynamic of the comparison. The mainstream represents the
universal set, within which the other culture must establish its particular niche.
Thus while Austro-German music may or may not demonstrate qualities such as
elegance, logic or precision, these qualities are inherent in all Latin art. The criti-
cal activity of those members of the other culture may take an isolationist stance,
emphasizing even more the attributes found in the subset, or it may attempt assimi-
lation, minimizing the differences, as found in some of the connections drawn with
Beethoven or Mozart. Even in the latter case, though, the category of Latinate art
is still epistemologically active; it forms the starting point for all activity, whether
positive or negative. Below are the qualities consistently attributed in the literature
to Scarlatti the Latin composer.
Instant Latinate Essentials Generator
1. elegance and grace
2. rationality and logic
3. Mediterranean, Classical
4. detachment, dryness, precision
Pestelli, Sonate, 187.
Eric Blom claimed a strong inuence of Scarlattis spare keyboard style on that of Haydn and L aszl o Somfai
believes that such works as Haydns Sonatas Nos. 42 in G and 50 in D show a Scarlattian inuence. Blom,
Review of Il clavicembalista Domenico Scarlatti: il suo secolo la sua opera by Cesare Valabrega, Music and Letters
18/4 (1937), 422; Somfai, The Keyboard Sonatas of Joseph Haydn, 253. H. C. Robbins Landon suggests that the
young Haydn may have known some of the Scarlatti sonatas; see Haydn: Chronicle and Works, I: The Early Years
17321765 (London: Thames and Hudson, 1980), 84. However, the most detailed investigation of this topic
is contained in Celestini, Haydn.
58 The keyboard sonatas of Domenico Scarlatti
5. joy and happiness
6. clarity, limpidity, transparency, lucidity
7. brightness and brilliance
8. lightness.
These constructions have often been used as a stick with which to beat the music of
the mainstream. Such attacks, however, have been made from a position of weakness:
in most of the institutional contexts of Western art music, the Austro-German, having
built up a degree of immunity, retains its aura of being law-giving and universal.
Indeed, the attitudes of its adherents often show what Alan Sineld has called the
difculty of a dominant culture in realizing the relativity of its own perceptions.
Such an assumption of universality controls the following discussion by Charles
Rosen of the place of national styles in the Baroque: In the great German masters
Bach and Handel, the contrasts are of little importance, the styles fused. They pick
and choose where they please; it is perhaps one of their advantages over Rameau and
Domenico Scarlatti.
This asserts that Scarlattis style is less varied and less exible
than that of the German masters, a conclusion that is difcult to accept. It points
to another common corollary of the basic cultural dynamic, that Latin music is an
acquired taste, that it will only satisfy in certain temperamental circumstances. We
can see this, admittedly self-consciously introduced, in Eric Bloms review of Cesare
Valabregas book on Scarlatti:
The author is not often betrayed by a Latin hankering after ne phrases into such false
metaphors as lopulent[o] giardino scarlattiano. The cool and [sprightly] wit of Domenico
Scarlatti, generally heartless and material but always exquisite and cunningly put together,
gives one nothing like the pleasures of a luxuriant garden, but rather if one must be
metaphorical like those of a perfect assortment of tasty and varied hors doeuvre accompanied
by the nest and driest of sherries.
It was precisely in the eighteenth century that our mainstream began to relocate
from Italy to Germany: by 1800, at least in terms of keyboard writing, German
composers had at last achieved a self-condence that enabled them to assert the
superiority of their music.
The terms of reference for this struggle, as outlined
I do not give specic references, since all attributions may be found almost anywhere with great ease. The
inuence of the Italian novelist Gabriele DAnnunzio, who brought Scarlatti into the cultural mainstream with
his 1913 story La Leda senza cigno, is discussed by a number of writers. Pestelli observes that his story had the
important effect of xing the Scarlattian image once and for all as implying health, joy, strength, brightness,
latinit` a, mediterraneit` a. See Pestelli, Sonate, 31.
Alan Sineld, The Migrations of Modernism: Remaking English Studies in the Cold War, New Formations 2
(1987), 116.
Rosen, Classical, 46.
Blom, Valabrega Review, 423. This was echoed by Kathleen Dale in 1948, when she suggested the sonatas were
so exquisitely precise and concentrated that to hear a long succession of them would be like sitting down to
a banquet consisting exclusively of hors doeuvres. Domenico Scarlatti: His Unique Contribution to Keyboard
Literature, Proceedings of the Royal Musical Association 74 (1948), 40.
Daniel E. Freeman, Johann Christian Bach and the Early Classical Italian Masters, in Eighteenth-Century
Keyboard Music, ed. Robert L. Marshall (New York: Schirmer, 1994), 230.
Panorama 59
by Dennis Libby, have been echoed up to the present day: In the confrontation
between German and Italian music, for Italians the very term musica tedesca was
one of reproach, signifying an inability to write for the voice, and a fondness for
excessive complexity. The partisans of German music saw the Italian variety as
insipid, shallow, imsy in construction and shoddy in workmanship. Libby points
out that the judgment of history has come down overwhelmingly on the German
side, adding what has now become an article of faith, that music history has still not
completely freed itself from attitudes prevailing in its formative days as a scholarly
discipline in nineteenth-century Germany.
These attitudes are apparent not just
in historical method and assumptions, of course, but also in the way we characterize
musics technical and expressive properties. In other words, a neutral model of
scholarly procedure has been determined by modes of enquiry which are coloured
by the attributes of a specic musical culture. So, for example, we look to harmony
where the Austro-German tradition has its greatest apparent sophistication as the
engine of tonal music at the expense of rhythm and syntax.
Scarlatti has needed to be rescued from the associations of superciality and im-
siness outlined by Libby the negative image of the Latinate agenda above. So
Schenker assimilated him into a sturdier tradition by claiming Italy was a part of
him, yet . . . he was no part of Italy.
We noted in the rst chapter the attempts to
downplay the fact that most of the sonatas are fast. The emphasis on the composers
roots in the world of Renaissance polyphony the perceived mainstreamof its time
seems to answer the same need.
The most common response to such lurking danger, though, has been to withdraw,
in isolationist fashion, into strains of ineffability, of racial mysticism. The classic
strategy of the Latinate other has been to cultivate a sense of inaccessibility. A good
example may be found in the performing tradition of French music, warning off
those who do not possess the most esoteric good taste from attempting their Faur e
or Ravel. For Scarlatti the ineffability is to be found, paradoxically, in clarity. This
Latin clarity is the master category around which all the other traditional associations
cluster. It is invoked, one must emphasize, not just by Latin writers ghting their
corner but also by outside apologists. Alfred Brendel, for example, believes that
in performance of the sonatas Scarlatti needs very clear contours, Mediterranean
Lang tells us that this spirited music offers the most welcome antidote
for everything that is heavy, dense and overloaded.
The wonderfully revealing
wording leaves us in no doubt of the national origins of the heaviness. At the same
Italy: Two Opera Centres, in Man and Music: The Classical Era, ed. Neal Zaslaw (London: Macmillan, 1989),
This is discussed in detail in Chapter 4, pp. 1457.
Schenker, Meisterwerk, 154.
Brendel, Music Sounded Out: Essays, Lectures, Interviews, Afterthoughts (London: Robson, 1990), 239.
Lang, 300 Years, 589. A corollary of this is the frequent assertion of a superior Latin taste. Witness this statement
by Macario Santiago Kastner in a discussion of ornamentation: What has always existed, and continues to exist,
is good taste and bad taste. French, Italians, Spanish and Portuguese lean instinctively towards go ut s ur, more
than do Anglo-Saxons and Germans. The Interpretation of Sixteenth- and Seventeenth-Century Iberian Keyboard
Music, trans. Bernard Brauchli (Stuyvesant: Pendragon, 1987), 40.
60 The keyboard sonatas of Domenico Scarlatti
time, of course, by accepting the basic terms of the debate, it reinforces the distinction
between what is solid and universal (the heavy main course of the Austro-German
mainstream) and what is an acquired taste (the antidote, the cleansing sorbet).
That this Latin clarity is not self-evident, but more of a defence mechanism, is
apparent from some of the unlikely contexts in which it is invoked. For example,
Verdi noted in 1864 of the Cats Fugue, K. 30, that with such a subject a German
would have created chaos, but an Italian made something as clear as the sun
surely an improbable verdict on the artistic end product.
Or Claude Rostand
tells us: Scarlattis rhythmic inventiveness is as inexhaustible as it is rened, yet it
never courts confusion but retains a naturalness and transparency that have never
been equalled.
How does this verdict, entirely typical, square with the zigzag of
Scarlattis actual syntax, the elisions, the vamps, the patterns that fail to complete
themselves? It is no accident that the vamp itself (briey dened in Chapter 1) had to
be conjured into existence by English-speaking scholars Kirkpatrick, who called
it the excursion, then Sheveloff, who gave it the name used here. Before that,
when even acknowledged as a vague quantity, it had no name and hence no real
existence. Instead, we have found Scarlatti being idealized as a counterweight to the
Austro-German mainstream.
Another example of Latin clarity at full power comes from John Trend:
The passion is there, but it is always expressed with concision and clarity; the music is dry
and sparkling, but it sparkles in the heat, not in the cold. Scarlattis music, indeed, glitters
like hot Spanish sunshine, illuminating impartially, but not unkindly, tragedy and comedy
alike. There can be tragedy leading to despair, as in the incomparable Sonata in B minor
[K. 87]; yet even the shadows are hard and clear, not only in outline, and the faintest approach
to sentimentality is interrupted by a dry cackle of laughter from across the way. Scarlatti is
the exact opposite of Schubert.
Even if the last sentence gives the game away, this is an imaginative realization of
the governing paradigm. It also points to an aspect of Scarlattis sonatas that can be
difcult for us to cope with, given our immersion in German musical manners. This
is a certain relentlessness, as identied by Cecil Gray (with the Latin sun now beating
down on Italy rather than Spain):
Indeed, his dazzling brilliance and grace seem at times almost excessive; one comes to long
for a sombre, shadowy passage as one longs for a cloud to come and veil, if only for a brief
moment, the hard, white glare of Italian summer skies.
If based on a limited, or selective, reading of the sonatas, and if issuing from our
governing paradigm, this nevertheless identies a strand that has nothing to do with
Cited in PaganoBoyd, Grove, 406.
Notes to recording by Anne Queff elec (Erato: 4509 96960 2, 1970), 10.
Trend, Falla, 149.
Gray, History, 140. Compare Pierre Hantas much more recent comment that a certain jarring harshness
accompanies the gaiety that has so often been emphasized in Scarlattis work. Notes to recording by Pierre
Hanta (Astr ee Nave: E 8836, 1992/2001), 11.
Panorama 61
relentless tempi it is just as evident in Andante as Allegro movements. We have
encountered it already in the machinations of K. 254. Grays charge of excess also
represents a welcome delivery from Latin sweetness and light.
The second level of nationalism involves the treatment of our composer within
individual countries. It has already been suggested that Scarlatti lacks the weight of
any single culture industry behind him. This was already apparent to Dent in his
commemorative article of 1935:
What a scandal it would cause to all good German patriots if anyone suggested that Domenico
Scarlattis two hundred and ftieth anniversary should be celebrated this year on equal terms
with those of Handel and Bach! And it is a curious thing that the wish to celebrate Domenico
Scarlatti should be put forward in England, of all countries, and not (as far as I am aware) in
Italy, the country of his birth, or in Spain, the country of his adoption. We English people
have in fact had a particular affection for Domenico, which has manifested itself continuously
from his own times down to the present day.
This mischief-making dates of course from a time of more belligerent nationalism
throughout Europe,
yet it is valuable for its reminder of the link between national
consciousness and institutional support an equation which could not be so baldly
articulated in the present day of pan-European harmony. Aside from a rare objecti-
cation of the assumed German musical standpoint, we are reminded of an apparent
lack of organized interest from the two countries that should have the greatest stake
in Scarlatti. A similar observation was passed by Max Seiffert in 1899, before the
advent of Longos edition: The duty to present to the musical world a complete
critical edition of Scarlattis epoch-making works should have been incumbent upon
Italy; but she has yet to remember this. The honour was left to foreign countries, . . .
although not through complete editions.
Seiffert was referring to the German
editions by Czerny and B ulow, so scoring a nationalistic point for the more univer-
sal culture. Indeed, Dents statement is not without its own element of nationalist
preening; it also ts into the wider patterns of English adoption of outside com-
posers and English love of eccentrics.
The two main players in this story, though, have still exhibited a form of Latin soli-
darity at this second level of nationalism. By and large, they have been happy to agree
that Scarlatti is Italian. Thus the Italian tradition tends to reclaim him for the coun-
try of origin; the Spanish has been difdent, and even defensive, about the adopted
Dent, Scarlatti, 176.
See Santi, Nazionalismi for an account of the Italian context for this what Santi calls the second nationalism.
Seiffert, Klaviermusik, 420.
This is the point Richard Newton seems to miss when he writes that Scarlattis special excellences are of so
un-English a character that we could hardly have been surprised if they had been but coldly appreciated here.
Newton, Cult, 138. I am not suggesting that any perceived eccentricity was sufcient cause for the cult in
its own right.
62 The keyboard sonatas of Domenico Scarlatti
Spaniard in its midst. This has been fuelled by the too easy assumption, following
Kirkpatrick, that Scarlattis music represents the essence of the Spanish musical soul.
Just as nationalist rhetoric may have gone underground while its supporting oper-
ations remain in place as has been implied in the study of style classication so
it can be argued that these traditions have retained much of their force up to the
The manner in which Scarlatti was nally embraced in twentieth-century Italy has
been the subject of two compelling discussions by Piero Santi and Giorgio Pestelli,
both of which concentrate specically on the nationalist (which also means Fascist)
The composers reclamation also coincided, of course, with a modernist
disparagement of Romantic style, hence the particular signicance of our master
category of clarity. Not only was Scarlatti reclaimed for Italy, but so was supremacy
in instrumental music altogether, which had been usurped by Austro-German parti-
sans: this particular strain of initially anti-Germanic Mediterraneanism
was soon
muted by political events. In order to make Scarlatti specically Italian (again), writ-
ers had to differentiate within the elements of the Latinate paradigm. This meant not
only distancing the composer from any Spanish elements (Portugal was hardly men-
tioned), but, less obviously, distinguishing the Italian artistic spirit from the French.
Valabrega, for example, compared the terse, virile quality of Scarlattis energetic
musical laughter with the French harpsichord art of Couperin and Rameau, which
was dened by its sentimental and sensual languor and its adorable preciosity.
Furthermore, he likened the frenzy of embellishments in Couperin to a coral in-
vasion, so unlike the clean vitality of Scarlattis musical lines.
The healthy body
of Scarlattian art was also unaffected by any Spanish clothing: It doesnt matter if
[he] spent a number of years at court in Spain and Portugal; his creative spirit, even if
breathing the ckle vapours of the Spanish guitar, remains essentially Italian and free
from any deep ethnic inuence. For the author there were three Italian founders of
instrumental style and technique Corelli on the violin, Scarlatti on the harpsichord
and Clementi on the piano. Scarlattis pioneering approach was particularly to be
aligned with that of Clementi, another great Italian. . . of the next era
one irresistibly of the wind-up number on Frank Zappas album Tinsel Town
Rebellion, containing the repeated acclamation Lets hear it, folks, for another great
Healthy innocence was also the key for Gino Roncaglia, for whom Scarlatti was
one of the greatest interpreters of the elegance, urbanity, grace and serene spirituality
of the rst half of the eighteenth century in our Italy. His music conjured up clear
skies, sweet waters, the pure joy given by meadows in ower and the interior joy
Santi, Nazionalismi. Pestelli, Sonate, Introduction/II, Il mito di Domenico Scarlatti nella cultura italiana del
900, 2556.
Pestelli, Sonate, 39.
Valabrega, Clavicembalista, 978 and 99.
Valabrega, Clavicembalista, 88 and 8990.
Frank Zappa, Tinsel Town Rebellion (Ryko: RCD 10532, 1981/1995), at the end of Peaches III.
Panorama 63
born of harmony of spirit with the natural surroundings.
Remarkably similar
imagery is found in the account by Luigi Villanis, from three decades earlier, in
The comic seems to be the prerogative of the Italians, just as wit is characteristic of the French
and depth of philosophical thought distinguishes the German races. So in the laughter of
Neapolitan opera buffa there sparkles some of that joyful sun that plays on the waters; it gives
us a breath of that fragrance given off by gardens in ower; it lls us for a moment with that
child-like gaiety that is found in the games of young boys, half-naked, running on the beach.
When the French muse laughs, there is an undercurrent of malice; with Germans a gentle
melancholy of spirit is revealed . . . With us music laughs happily and then calms down;
Couperin laughs and dances in a thousand mincing affectations, while C. P. E. Bach simply
The amusing anticipation here of one of the central images of Thomas Manns Death
in Venice is not entirely incidental; the journey of Gustav von Aschenbach to Venice
is, after all, in search of precisely the restorative qualities lauded by Villanis and so
enshrined in European cultural lore.
This exaltation of clear healthy Italian simplicity, together with a tendency to
ignore alien elements, is also found in a musical tribute, Alfredo Casellas Scarlat-
tiana of 1926. This divertimento on music by Domenico Scarlatti for piano and
small orchestra works in references to many sonatas in a collage-like structure.
Gianfranco Vinays description of the fourth of the ve movements (Pastorale) as a
celebration of the Italian character of Scarlattis art, of the deep ties between certain
Scarlattian melodic inections and Italian popular song,
misses the point that
the whole piece does this cultural work. It is Italianized and picturesque, its syntax
tting in neatly with the literary tradition of the panorama. There is scarcely a hint
of any sonata that might have been thought of as overtly Spanish in avour. The
only real candidate, K. 450, which Clark has recently classied as a tango gitano,
would presumably have been taken by Casella to be Italianate. This is a kind of ethnic
This appropriation of Scarlatti has left its traces, if, as suggested earlier, in more
covert form. The continuing status of the Longo edition is one indicator. Roman
Vlad, writing in 1985, admits to having caused a scandal by saying that the Longo
edition still infests Italian conservatories.
In the Siena conference of the same
Roncaglia, Centenario, 64.
Villanis, Italia, 170.
Malcolm Boyd has given a list of references to sonatas in Boyd, Master, 2334. To this list, which the author
acknowledges is incomplete, I would suggest the following additions, in order of the ve movements: I the
countersubject of K. 41 (which Gianfranco Vinay misidenties as a distortion of K. 257; Le sonate di Domenico
Scarlatti nellelaborazione creativa dei compositori italiani del Novecento, in Domenico Scarlatti e il suo tempo,
128), K. 64; II K. 259, 162; III K. 450, 64; IV K. 446, plus something akin to the vamp of K. 439; V
K. 96. Some of these derivations are also spotted in Vinay, Novecento.
Vinay, Novecento, 136.
La port ee de linuence andalouse chez Scarlatti, in Domenico Scarlatti: 13 Recherches, 667.
Vlad, Storia, 22.
64 The keyboard sonatas of Domenico Scarlatti
year, the sonatas were still referred to by Italian scholars according to Longo numbers,
and music examples were drawn from the Longo edition. Pestelli, having preferred
Longo to Kirkpatrick numbers for his 1967 book, had by 1989, in a review of two
new volumes, jumped straight to the Fadini numbering, Kirkpatricks system being
given rst in parentheses, then not at all. If there were simple logistical reasons for
such preferences, the cultural status of Longo having precluded its disappearance
from general circulation, a more secure piece of evidence might be the fact that no
Italian translation of Kirkpatricks fundamental book appeared until 1984.
There are, however, more ne-grained instances of appropriation. The emphasis
by Italian scholars on Scarlattis strong roots in the past, if not simply reecting a more
intimate knowledge of the repertory, may (as already proposed) reect this tendency.
After all, Casella in Scarlattiana chose to incorporate several of the plainly archaic trio
sonatas (K. 81, 89, 90), so remote fromany notions of Scarlattian style, in preference
to works that might have disturbed his stylistic picture.
Pestellis assertion of the
general conservatism of [Scarlattis] work can be understood in this way too.
Nevertheless, Pestellis 1967 book could hardly be accused of ignoring the Iberian
avours or failing to deal with their implications, since they feature fully in the
discussion; but at a structural level his classication of the sonatas does just that. Many
of the most apparently Spanish or gesturally extreme works are made coeval with
the Essercizi, for instance, placed in the categories of toccata and study, effectively
deecting attention from their national allegiance or the creative temperament
on display. The structuring he adopts favours the sonatas displaying clear Italianate
roots or moderation and polish in their approach. How, for instance, can K. 120
simply be buried amongst the studies, or K. 99 and 114 among the toccatas?
Such a reception history, perhaps with an element of protesting too much about
the purely Italian, does not contradict the underlying assertion of the composers
statelessness. The scale of the operation in Italy has after all not been that great
as we have seen in other contexts, the non-activity is more signicant than the
activity. Indeed, in 1971 Kirkpatrick wrote, perhaps mainly as a polemic against
the continuing use of Longo: It is [in Italy] that the conception of Scarlatti as no
better than any of his mediocre contemporaries and that the inveterate scrambling of
chronology have retained an almost unshakable foothold.
In the case of Spanish
reception, however, the sense of absence is far more pronounced. Very little indeed
has been published on Scarlatti until the relatively recent past. There are, it must
be said, some simple logistical rationales for this state of affairs. It is only from the
1980s onward that musicology has become institutionalized in Spain meaning the
There can be no doubt that Casella was aware of the generic status of these works he published two of them
in an arrangement for violin and keyboard in 1941, discussing their status in his Preface. See Rodolfo Bonucci,
Le sonate per violino e cembalo di Domenico Scarlatti, Studi musicali 11/2 (1982), 249. Bonucci believes the
themes taken from the sonatas for Scarlattiana were chosen by Casella for their intrinsic fascination; Bonucci,
Violino, 258.
Pestelli, Sonate, 54.
Scarlatti Revisited in Parma and Venice, Notes: The Quarterly Journal of the Music Library Association 28/1
(1971), 7.
Panorama 65
establishment of separate music departments within universities with their teaching
staffs, students, training programmes and the sense of professional identity that arises
from that. Prior to this Spanish music researchers tended to receive an ecclesiastical
training. Without a musicological industry, the mass production of evidence so
typical of other countries had barely begun, and therefore many resources such as
libraries and archives have remained untapped. As Juan Jos e Carreras relates, it is
not only the heritage contained in these archives which remains unknown, but the
very existence of the archives themselves. This is a situation which is particularly
serious in the case of private or semi-private archives, many of which are in danger of
getting irretrievably lost. There is a long way to go, therefore, before the catalogues
of the Spanish musical archives can be said to be complete. These difculties of
training and resources have led to what the author characterizes as the problem of
individual, isolated and uncoordinated research in Spain.
If this sounds like the
summation of Scarlatti research given earlier, largely for different historical reasons,
then if one combines the two sets of circumstances, it would appear that Scarlatti
has been doubly affected!
For Scarlatti, however, the recent institutionalization of musicology in Spain seems
to have borne some fruit, in the form of documentary and manuscript discoveries
which will be described further on. Boyd has framed the prior lack of material very
It is a strange fact, quite as remarkable as the complete disappearance of the autographs of
Scarlattis keyboard music, that when Kirkpatrick (1953) and Sheveloff (1970) compiled their
exhaustive lists of Scarlatti sources neither writer was able to cite a single manuscript copy of
a sonata in any Spanish library or archive. This is a situation no less singular than would have
been the complete disappearance from England of all Handels oratorios, or the loss of all
trace in Germany of Bachs church cantatas, and it cannot be explained simply as the result
of negligence on the part of librarians, archivists and scholars.
That neglect may play a part in the situation, though, is implicit in Boyds wording,
and this is where nationalistic concerns blend into institutional rationales. At the
beginnings of Spanish musicology in the mid-nineteenth century, the focus of most
scholars was on the so-called Golden Age of Spanish music, the sixteenth and
seventeenth centuries. The eighteenth century in Spain had witnessed great foreign,
and above all Italian, inuence. (There was a comparable Italianization of Portuguese
musical life. This was made possible by the end of the war with Spain in 1713 and
the discovery of gold in Brazil, allowing Jo ao V to buy in great numbers of Italian
) Hence scholarly activity focused on those periods and repertories that
had suffered less contamination by alien inuence and were thus regarded as having
Musicology in Spain (19801989), Acta Musicologica 62/23 (1990), 266 and 287.
Boyd, Master, 153.
See Manuel Carlos de Brito, Scarlatti e la musica alla corte di Giovanni V di Portogallo, in Domenico Scarlatti
e il suo tempo, 69 and 72.
66 The keyboard sonatas of Domenico Scarlatti
been more intrinsically Spanish.
In this currency, Scarlatti, an Italian working in
eighteenth-century Spain, was not going to fetch a high price.
This was not just retrospective resentment, however; it may have been an active
force at the time, particularly in the eld of opera. Italian opera in Madrid under
the Bourbons seems to have been resented by both middle-class audiences and
Spanish singers, composers and players.
The focus of this enormously success-
ful venture, Farinelli, even witnessed the circulation of a pamphlet against him in
Recent revisionist views suggest, however, that the Italian invasion may
have been somewhat overplayed by historians. Carreras believes that the whole
process of Italianization was by no means a struggle between Italians and Spaniards,
but a process undertaken by the Spanish composers themselves.
One indigenous
form, however, did appear as a minor challenge to the dominance of opera. The
tonadilla esc enica, normally performed between the acts of a play, appeared on the
Madrid stage in the middle of the century, lasting until about 1800. It was the Span-
ish equivalent of the intermezzo, largely comic and unpretentious. One of the most
common character types was the majo, who would have fun at the expense of Italian
fops and French dandies; he was a theatrical representation of Spanish resentment
towards foreign cultural invaders.
For the variety of reasons mentioned so far, it has been very rare for Spanish
musicologists to work on non-Spanish themes. This in itself is telling in the context
of Scarlatti research, as is the blunt assessment of Xo an M. Carreira that xenophobia
and patriotism, even short-sighted parochialism, still inform not a little Spanish and
Portuguese musicology.
We need to recall in connection with this the defensive
attitude towards the alleged Spanishness of the Scarlatti sonatas; many Spanish musi-
cologists appear to feel that the issue has been prejudged, without their having been
consulted, as it were. Also at stake is the wider assertion of Scarlattis inuence on
Iberian keyboard music, which has also been treated with scepticism. One of the
strategies in response has been to retreat into notions of an ineffable Spanishness,
one that is inaccessible to outsiders the same cultural dynamic that has shaped the
performance tradition of French music. This response is more specically culturally
See for example

Alvaro Jos e Torrente, The Sacred Villancico in Early Eighteenth-Century Spain: The Reper-
tory of Salamanca Cathedral (Ph.D. dissertation, University of Cambridge, 1997), ixx, and also Torrente, A
Critical Approach to the Musical Historiography of Eighteenth-Century Spanish Music (Cambridge: unpub-
lished, 1995), especially 118.
This view is represented in Mary Neal Hamilton, Music in Eighteenth[-]Century Spain (New York: Da Capo
Press, 1971; reprint of rst edn [Urbana, Illinois, 1937]), 100.
See Javier Herrero, Los orgenes del pensamiento reaccionario espa nol (Madrid: Editorial Cuadernos para el Di alogo,
1971), 63.
See Torrente, Villancico, 6n., and Carreras, FromLiteres to Nebra: Spanish Dramatic Music between Tradition
and Modernity, in BoydCarreras, Spain, 716. In the context of the villancico, Torrente also describes Dur on,
Literes and Torres as protagonists in the introduction of Italian operatic conventions; Italianate Sections in the
Villancicos of the Royal Chapel, 170040, in Boyd-Carreras, Spain, 79.
Craig H. Russell, Spain in the Enlightenment, in Man and Music: The Classical Era, ed. Neal Zaslaw (London:
Macmillan, 1989), 359.
Opera and Ballet in Public Theatres of the Iberian Peninsula, in BoydCarreras, Spain, 28.
Panorama 67
determined too, since it invests in the allure of a dark, mysterious Spain, the most
commonplace of outside images (the Black Legend). In this manner the members
of a marginalized culture collude in its essentialization.
The grandest example of these tendencies may be found in Macario Santiago
Kastners 1989 article, Repensando Scarlatti, a sustained exercise in scepticismabout
all the received wisdom on the composer. He refutes the image of technical novelty
and with it Scarlattis assumed leadership of a new keyboard school, pointing to the
example of K. 61, which shows many gurations deriving from the toccatas of [his
father] Alessandro.
Scarlattis possible inuence on the native Iberians Soler, Seixas
and Albero is regarded as insignicant. More important for our current purposes,
however, are Kastners intimations about the true essentials of Iberian musical feeling.
Thus he claims of Scarlatti: When the southern Italian appears to be moved or ery,
he does it in order to affect a pose, but this is not as convincing as Iberian depth or
tragic sentiment.
We are also told that the harmonic and intervallic turns and the
vernacular rhythms found in Scarlatti that are supposed to be so denitively Spanish
are also found in the works of Vicente Rodrguez (16901760), Seixas, Soler, Albero
and others, to such an extent that it seems more prudent to ignore folklore as a
particular explanation for Scarlattian style. Such musical colours, Kastner points out,
had in any case spread to Sicily, Naples (where Scarlatti, grewup, of course), Valencia,
Portugal and so forth. The real Spanish musical language is not simply an inorganic
mix with Arab, Sephardic and gypsy additions it has been judged as such by
musicologists . . . who have little familiarity with what is genuinely Iberian.
A more temperate expression of this cultural dynamic may be found in, for exam-
ple, the recent comparative study by

Agueda Pedrero-Encabo of Scarlattis Essercizi,
published in 1739, and the thirty sonatas of Rodrguez, which were written well
before the date that appears on the manuscript, 1744. The aim of the study is to
settle the question of whether Scarlatti inuenced the Spaniard. This is intended in
the factual sense of establishing prior claims to certain progressive features such
as formal shaping rather than simply for reasons of stylistic interest. Perhaps not
James Parakilas has read such withdrawal differently, in his account of nineteenth-century exotic constructions
of Spain: It seems to be a danger of exoticism that those who are its objects, when they conclude that they
cannot overcome their exotic relationship to the centers of power, begin to consider that they might be better
off with no relationship at all. How Spain Got a Soul, in The Exotic in Western Music, ed. Jonathan Bellman
(Boston: Northeastern University Press, 1998), 193. For all that the authors focus lies elsewhere, it seems
remarkable that Scarlatti receives not even a glancing mention as a possible model for nineteenth-century exotic
representations of Spain. This surely reveals some of the colonializing assumptions that are implicitly being
criticized, the historical and geographical marginality of Scarlatti placing him beyond consideration or even
conscious thought.
Kastner, Repensando, 152 and 151. No one would deny the older heritage of K. 61, but this is an anomalous
sonata in any case through its unique use of variation form; what about the hundreds of sonatas for which
Kastners statement would appear not to hold?
Kastner, Repensando, 137.
Kastner, Repensando, 154. A different sort of scepticism was evident in Roberto Gerhards 1954 BBC radio
talks, The Heritage of Spain. Gerhard did not deny the Spanishness of Scarlatti, but felt it was a avour, a
peculiar accent rather than consisting of direct incorporation of Spanish/amenco material. The scripts of the
talks can be found as Gerhard.11.18 (2.12) in the manuscripts room of the Cambridge University Library.
68 The keyboard sonatas of Domenico Scarlatti
surprisingly, the writer lays more weight on the differences of the respective com-
posers than on any similarities, some of which are very striking indeed. One of the
strongest distinctions between the composers, according to the author, is found in
Scarlattis harmonic usage, illustrated by the colourful melodic turns of K. 7 and
the rich chords of K. 6;
but not a hint of possible Iberian inspiration is given, a
topic that is studiously avoided in this article.
I do not disagree in principle with the scepticism found explicitly in Kastners
review and implicitly in Pedrero-Encabos account of inuence. Eternal vigilance
is after all an indispensable quality for all Scarlattian research in particular. When
it comes to the vexed question of Iberian inuence, though, there has been an
obvious and apparently logical way forward an ethnomusicological investigation.
Boyd, supported by Clark, has called for the services of an ethnomusicologist familiar
also with the art music of eighteenth-century Spain and Portugal.
Whether we
really need this providential gure is beside the point for now. If we accept this
as an urgent requirement, then there would be an obvious country of origin for
such an individual. But no one appears to have stepped forward. Once again in
the eld of Scarlatti reception, what has not happened is at least as signicant as
what has.
Of course there are other ethnic elements that need investigation: the role of
Portuguese and Neapolitan folk music has not been addressed either. When set
against the tone of many of the views expressed above, van der Meers description
of the composer as an Italian-Portuguese-Spanish genius represents a rare bit of
Food for future thought is provided by the comments of Burnett James
on the attitude to Scarlatti of Manuel de Falla, indisputably a Spanish composer. They
remind us that the Spanishness or otherwise of Scarlattis keyboard sonatas is not just
a question of national essences, but has a historical dimension too:
Ironic at rst sight is the way in which the leading composers of the [Spanish] renaissance
over a century later, notably Falla himself, who were in the habit of denouncing the Italian
inuence on Spanish music and its debilitating effects on the native product, themselves
looked to Scarlatti as mentor and exemplar.
It is probably no coincidence that, with the changing circumstances of musicological
activity in Spain, a number of Scarlattian discoveries have been made there in recent
times. Yet signicant new information has also emanated from England, Italy and
Portugal in this period, amounting to an extraordinary late (twentieth-century)
harvest of which Kirkpatrick would have approved. Whether benetting from any
Los 30 Essercizi de Domenico Scarlatti y las 30 Tocatas de Vicente Rodrguez: paralelismos y divergencias,
Revista de musicologa 20/1 (1997), 388.
Boyd, Master, 222; Clark, Boyd Review, 209.
Van der Meer, Keyboard, 157. Another diplomatic summation may be found in IfeTruby, Spanish, 6.
Manuel de Falla and the Spanish Musical Renaissance (London: Gollancz, 1979), 35.
Panorama 69
impetus provided by the tercentenary in 1985 or owing more to sheer chance, such
discoveries at least offer a few more ecks for our blank canvas, since they have
answered few questions and solved few mysteries. Even if they have only caused
one to pose the same questions again, one should bear in mind that what might be
crumbs with other composers make meals for the Scarlatti scholar.
Certainly one of the most important nds is the correspondence of Monsignore
Vicente Bicchi, papal nuncio in Lisbon from 1710 to 1728. Just when Scarlatti did
arrive in Portugal has long been a matter for speculation. We learn from Bicchi,
though, that Scarlatti entered Lisbon to begin his posts as Master of the Royal
Chapel and keyboard teacher to the Royal Family on 29 November 1719.
would appear to rule out Roberto Paganos attractive theory that Scarlatti resided in
Palermo from April 1720 to December 1722.
We also learn of the performance
of far more serenatas and cantatas than so far known, and astonishingly that the
composer made his court debut as a singer and appears to have sung on a number
of occasions. (Scarlattis vocal abilities have been conrmed by the still more recent
discovery that he sang and played the harpsichord for James III, the Old Pretender,
in June 1717 in Rome.
) We also nd out somewhat more about several major
breaks from the composers Lisbon routine during the 1720s. The last of these has
turned out to be longer than previously thought from January 1727 until probably
December 1729.
According to the nuncios letter, Scarlatti went to Rome to
recover his health, while we know that he married his rst wife in Rome in May
1728. Here is an example of more meaning less, since speculation may now begin
concerning the composers other activities during this period of almost three years.
It would now appear that, contrary to popular legend, Scarlatti was not present at
the exchange of princesses in January 1729, when Mara B arbara was married to
Prince Fernando of Spain.
This information is contained in the preface to the facsimile edition of the Libro di
tocate, briey mentioned before. This copy of sixty-one Scarlatti sonatas was acquired
by the Portuguese Institute of Cultural Heritage in 1982; it can, Gerhard Doderer
believes, be dated to the early 1750s. Many of the ramications of this copy will
be explored in connection with subsequent commentary on individual sonatas. The
hottest news, though, is the appearance of a new Sonata in A major, found only in
this source, and the presence in the collection of K. 145. The latter sonata, known
Gerhard Doderer, New Aspects Concerning the Stay of Domenico Scarlatti at the Court of King John V
(17191727), Preface to facsimile edn, Libro di tocate per cembalo: Domenico Scarlatti (Lisbon: Instituto Nacional
de Investigac ao Cientca, 1991), 910. See also Aurora Scotti, LAccademia degli Arcadi in Roma e i suoi
rapporti con la cultura portoghese nel primo ventennio del 1700, Bracara Augusta 27 (1973), 11530. See
Appendix, Archivio Segreto Vaticano Segretaria di stato, Nunziatura di Lisbona, vol. 75 (Portogallo). In
between entries, Scotti paraphrases Scarlatti arrives in Lisbon on 20 January 1720. Quite how one squares this
with Doderers information is mysterious, as is the fact that information contained in an article of 1973 should
have remained unknown for so long. It would have saved Pagano a lot of work.
Pagano, Vite, 35462.
See Edward Corp, Music at the Stuart Court at Urbino, 171718, Music and Letters 81/3 (2000), 35163.
Doderer, Libro, 11.
70 The keyboard sonatas of Domenico Scarlatti
only from a copy in the Fitzwilliam Museum, Cambridge, was placed by Sheveloff
in his doubtful category,
but its authenticity now seems conrmed. (The same
may also hold therefore for its companion, K. 146.) The Sonata in A major has been
accepted without reservation by Boyd and van der Meer.
Taking its authorship for
present purposes as read, its very existence raises doubts about the comprehensiveness
of the Parma and Venice sets. The general assumption has been that these copies
represented a sort of Gesamtausgabe, and that if any other sonatas were to turn up,
they would be very early ones that were not deemed worthy of inclusion even in the
earliest Venice manuscript of 1742. This would not appear to be the case with the A
major Sonata. Thus one of the commonly agreed near certainties may be crumbling.
Amidst the numerous Spanish copies of Scarlatti sonatas that have nally emerged
in the recent past (such as the copies of 189 works at Zaragoza
), a number of
new sonatas have been found attributed to Scarlatti. Some, such as the two sonatas
held at the Real Conservatorio Superior de M usica in Madrid, the four found in
Montserrat and the Sonata published by Rosario

seem quite unlikely.
Some of the other unknown pieces attributed to Scarlatti seem more promising
the three sonatas found in the cathedral of Valladolid, and especially the two from
the Biblioteca Nacional de Catalunya in Barcelona.
Perhaps the most unexpected of Scarlattis three countries for new discoveries
would be Italy, yet we have already given an account of the importance of the
Farinelli inventory, with its suggestions of a richer Spanish production of vocal
music than might have been supposed and the revelations of its list of keyboards.
In any case, the whole document suggests that any further searching for new music
might want to concentrate on Italy as well as the Iberian peninsula. Indeed, Pestelli
discovered in the late 1980s four (or six) Scarlatti sonatas in an unknown manuscript
at the University of Turin. This is a rare nd, since there are few copies of sonatas in
Italy dating from the rst half of the eighteenth century. Furthermore, the sonatas,
Sheveloff, Frustrations I, 41819. The sonata is found in the Fitzwilliam Museum at MU MUS 148 (formerly
32 F 13).
Boyd, notes to recording by Mayako Son e (Erato: 4509 94806 2, 1994), 6; van der Meer, Keyboard, 137.
Reported in Jos e V. Gonzalez Vall e, Fondos de m usica de tecla de Domenico Scarlatti conservados en el archivo
capitular de Zaragoza, Anuario musical 45 (1990), 10316. Note the fact that the copy of K. 206 carries the
date 1752. If this reects the date of copying rather than reproducing what was found on the source, then this
copy was possibly made before that which appears in Venice. K. 206 can be found as the rst sonata of P V,
dated 1752, and the rst sonata of V III, dated 1753.
The Madrid sonatas are published as Appendix III in Boyd, Master, 24052; Bengt Johnsson (ed.), Montserrat
Sonatas Nos. 14, in Domenico Scarlatti: Ausgew ahlte Klaviersonaten, vol. 1, 8190; Rosario

Alvarez, Una nueva
sonata atribuida a Domenico Scarlatti, Revista de musicologa 11/3 (1988), 88393.
Antonio Baciero (ed.), Valladolid Sonatas Nos. 13, in Nueva biblioteca espa nola de m usica de teclado, vol. 3
(Madrid: Union Musical Espa nola, 1978), 3750, and Mara A. Ester-Sala, Dos sonatas de Domenico Scarlatti:
un tema abierto, Revista de musicologa 12/2 (1989), 58995 (sonatas reproduced in facsimile on 5914).
In addition, under the category of Libri Differenti we nd between items 15 and 16, entitled Sonata (/Sonate)
per clavicembalo di Scarlati [sic], mention of a Spiegazione della Musica. What can this mean? The explanation
of (the) music seems to be of a piece with the sonata of item 15; is this Scarlattis explanation, of this particular
sonata, of music in general? It would certainly be nice to know. For further discussion of the inventory, see van
der Meer, Keyboard, 147. Once again here, more information brings more uncertainty.
Panorama 71
although copied in a different hand, form part of a manuscript containing toccatas
by Alessandro Scarlatti and Handel.
The Turin sonatas comprise, in three pairs: K. 76 and K. 71; K. 63 and a Minuet
in G major; K. 9 and a Minuet in D minor. The rst three of these sonatas might
well have emanated from an Italian environment, while the copy of K. 9 one of the
best-known Scarlatti sonatas in the nineteenth century, which earnt it the nickname
of Pastorale diverges markedly from the reading found in the Essercizi. Pestelli
notes that, without any elements to help us with the dating, this manuscript could
be later than the 1739 edition of K. 130, but it cannot derive from it because of the
divergences. However, he avers that the hypothesis that [these sonatas] returned to
Italy from Portuguese or Spanish sources, long after the departure of the composer
for these countries, seems among the least probable.
It is quite conceivable, of
course, that Scarlatti took them back to Italy himself, especially since we have to nd
something for him to have done during the now yawning gap of 17279. On the
other hand, the two new Minuets (the fact of whose pairing with established sonatas
speaks well for their status) also have strong Italian traits.
One should bear in mind
too Graham Ponts theory that K. 63 represents a written record of Scarlattis entry
in the famous (but, of course, unsubstantiated) keyboard contest with Handel.
If we compare the Turin version of K. 9 with the one published as part of the
Essercizi, one detail stands out above all the closing bar. This consists of a D
minor arpeggio falling from d
to D in even triplet quavers from the rst to the
fourth beat, whereas the Essercizi version features a held unison. Scarlatti virtually
never has this kind of arpeggiated close, so common in the works of contemporary
keyboard composers, which seems to exist primarily to ll in the bar. A good
example may be found in the nal bars of the Giustini movement given as Ex. 3.2a
in the following chapter. (Where it does exist, as in the nal ourishes of K. 115 or
K. 136, it is normally a means of dispelling the tension arising from prior cadential
reiteration, and is thus rhythmically integral.) It tends to have an unwinding effect,
both texturally and affectively, that the composer obviously went out of his way
to avoid. Instead we are more likely to nd unisons, which often have a relatively
taut and tense effect compared to the satisfaction provided by a full harmony, either
chordal or arpeggiated. Such a feature, both in its positive manifestation and in its
negation of a generic commonplace, exemplies Scarlattis critical distance from
even the most ingrained of habits, the least chosen parts of a piece.
The discovery of this version of K. 9 might indicate, as many have suggested, that
all the Essercizi were revised or polished-up versions of considerably earlier work as
Una nuova fonte manoscritta per Alessandro e Domenico Scarlatti, Rivista italiana di musicologia 25/1 (1990),
115 and 117.
Pestelli compares that in G major with K. 80 and allies the D minor Minuet with that chromatic caprice . . .
well known in the Neapolitan environment and cultivated by Scarlatti himself in the Essercizi, as in K. 3 and
K. 30. Pestelli, Fonte, 105.
See Handel versus Domenico Scarlatti: Music of an Historic Encounter, G ottinger H andel-Beitr age 4 (1991),
23247, especially 2434.
72 The keyboard sonatas of Domenico Scarlatti
the composers dedication to Jo ao V might also imply.
But we are straying onto
dangerous territory. Any judgement that the Turin reading is earlier and hence by
implication less mature, less Scarlattian is conditioned by unproblematic notions
of style, progress and chronology that have already been shown to have undermined
discourse about both our composer and the music of his century.
Also of rst importance has been the publication in 1985 of an edition by
Francesco Degrada of the comic intermezzo La Dirindina. The rst performance,
scheduled in Rome in 1715, was prohibited at the last moment. It would appear that
the libretto, by the notorious Girolamo Gigli, was found too offensive, for the public
good and the good of the singers it satirized so unkindly.
The scandal caused led
to considerable demand for copies of the libretto, which Gigli had printed outside
Rome so as to get around the prohibition order. Several aspects of the affair suggest
that Scarlattis association with such a libretto was not incidental. The use of the
word scarlatti (scarlet silks) in the libretto is, as Annabel McLauchlan has pointed
out, noteworthy since it indicates a collaboration between Gigli and Scarlatti at the
stage of the works construction, and thus implicates both men more seriously in
an organized satirical attack.
In addition, there is the unusual note found on the
nal page of the libretto: The excellent music of this farce is by Signor Domenico
Scarlatti, who will be pleased to oblige everyone. Evidently, Scarlatti too wished to
prot by the scandal and sell some copies of the score.
The work seems nally
to have been performed in Rome in 1729; we may now suggest that Scarlatti was
himself present on this occasion.
Malcolm Boyd observes of La Dirindina that it is surprising to observe Scarlatti,
whose whole life was spent in the service of monarchs, viceroys and princes, aligning
himself with one of most subversive writers of the time in a work explicitly designed
to call into question the values of an art form which, more than any other, served to
atter and support the established order.
Indeed it would appear surprising, but
then it is precisely such values that seem to emerge from the composers keyboard
These are Compositions born under your Majestys Auspices: Kirkpatrick, Scarlatti, 102. Sutherland, for ex-
ample, believes the dedication letter suggests that the works originated as teaching pieces in Lisbon; Sutherland,
Piano, 246. We have already noted van der Meers organological reasons for believing that they were undoubt-
edly composed at a considerably earlier date; van der Meer, Keyboard, 141. On the other hand, Kirkpatrick
wonders whether this phrase in the dedication means that Scarlatti, by virtue of continuing in Spain to teach
Mara B arbara, still considered himself under the Auspices of the King of Portugal and so the sonatas might
still have been written in Spain; Kirkpatrick, Scarlatti, 137. Heimes believes it extremely unlikely that the
Essercizi were written in Portugal before 1729, that Scarlatti would have selected ten-year-old pieces when he
wanted to put his best foot forward, so to speak, on the occasion of receiving his knighthood; Heimes, Seixas,
Pagano believes that the wrath of the censor was directed more at Gigli himself than at the matter and manner
of the libretto. The subject of the libretto was a usual satire for the time. Pagano, Vite, 336.
An Examination of Progressive Style in Domenico Scarlattis La Dirindina (M.Phil. dissertation, University
of Cambridge, 1996), 12.
See the commentary in Degrada, Preface to edition of La Dirindina (Milan: Ricordi, 1985), xxii.
Boyd, Master, 734.
Panorama 73
Among other discoveries, the most intriguing are more letters from Portugal, this
time from the secretary to Jo ao V, Alexandro de Gusm ao. In one of these he indicates
that the famous Catalan oboist Juan Baptista Pl` a had come to Lisbon in 1747 on
Scarlattis recommendation. The signicance of this becomes apparent in a letter of
the same year, in which Gusm ao writes that new and piquant Scarlatti sonatas had
arrived and that he had heard themplayed in his own house in a way that pleased Pl` a
even though the latter had heard them played by the composer himself.
This is
the nearest thing we now have to a conrmation that Scarlatti performed publicly at
court in Madrid (question nine of our earlier list of specic uncertainties), unless we
are rather perversely to conjecture that Pl` a was granted a private informal audience.
On a more personal note, Beryl Kenyon de Pascual uncovered in 1988 the details
of a dispute that arose in 1754 between Scarlatti and his daughter-in-law Mara del
Pilar concerning her dowry following the death at the age of eighteen of his son
Alexandro (who had married secretly at the age of seventeen).
Also casting a
pall over his nal years is the memorial to the composers will, published by Teresa
Fern andez Talaya in 1998. The goods inventoried in the two additions show that
Scarlatti had enjoyed a very comfortable position; among them we nd reference to
a clavicordio, valued at 3,000 reales. The other signicant musical news contained
is that there were two keyboard instruments in Scarlattis house that belonged to the
Queen; these were returned along with various scores on the composers death.
Another precious piece of evidence, but one that has been in circulation since
1739, is the preface that Scarlatti provided for the publication of his Essercizi. Here
is Kirkpatricks translation:
Whether you be Dilettante or Professor, in these Compositions do not expect any profound
Learning, but rather an ingenious Jesting with Art, to accommodate you to the Mastery
of the Harpsichord. Neither Considerations of Interest, nor Visions of Ambition, but only
Obedience moved me to publish them. Perhaps they will be agreeable to you; then all the
more gladly will I obey other Commands to please you in an easier and more varied Style.
Show yourself then more human than critical, and thereby increase your own Delight. To
designate to you the Position of the Hands, be advised that by D is indicated the Right, and
by M the Left: Fare well.
The primary importance of this paragraph has been taken to lie in the unique
declaration of his art apparently given here by Scarlatti. We have already seen, though,
that the composers letter to the Duke of Huescar in 1752 has been read as another
such artistic document. The methodological problems apparent in the interpretation
of the Huescar letter need to be addressed in conjunction with our preface; this
See Brito, Portogallo, 789.
See Domenico Scarlatti and his Son Alexandros Inheritance, Music and Letters 69/1 (1988), 239.
Memoria con los ultimas voluntades de Domenico Scarlatti, m usico de c amara de la reina Mara B arbara de
Braganza, Revista de musicologa 21/1 (1998), 162.
Kirkpatrick, Scarlatti, 1023.
74 The keyboard sonatas of Domenico Scarlatti
may explain the ungenerous insertion of apparently in the sentence above. What is
needed before we judge the contents is a little light deconstruction. Both documents
ought, for a start, to be situated in an epistolary practice of the time. Before we take
literally Scarlattis complaints in the Huescar letter, we need to ask questions such
as who the composer was addressing and what both parties stood to gain from the
transaction. The letter accompanied scores supervised by the composer from the
parts for two hymns written by the composer Pierre du Hotz. These were rst
performed in Brussels in 1569 in honour of two ancestors of the current Duke of
Such a background to the letter might suggest that, in honouring the
past through scorning the present, Scarlatti was honouring also the ancestors and
hence the current Duke himself; they linked him with better, more illustrious
times. The composers remarks may simply have been a way of complimenting
the good taste of the Duke, either because of the particular circumstances of the
commission or because the Duke was known to be partial to the older polyphonic
ways. Perhaps too Scarlatti had a particular gain in mind. After all, immediately
after the condemnation of the moderns for their contrapuntal ignorance, Scarlatti
requests a visit from the addressee: I cannot go out of my house. Your Excellency
is great, strong and magnanimous, and full of health; why not come therefore to
console me with your presence: Perhaps because I am unworthy?
These are some
possibilities to consider before the composers remark can be taken as a statement of
artistic faith.
Indeed, hanging on every last artistic word is a tradition that dies hard. Having a
large quantity of material to draw on only increases the difculty of disentanglement.
One only need consider the cases of many voluble twentieth-century composers,
such as Schoenberg or Messiaen, who are so consistently taken at their word, even at
a sophisticated critical level. It is fatally easy to allow composers pronouncements to
dictate the terms for the reception of their music. This is not to deny the relevance
of such commentaries, merely to suggest that composers have an obvious stake in
how their music is understood. So they create to an extent personal mythologies,
leading us toward certain preferred angles on their output and away from others.
Scarlattis mythology is of course very different from the norm, since it rests on such
negative (or absent) foundations. Nevertheless, we can still ask the same question
of the pronouncement in the Huescar letter, in addition to those already posed. If
we take it objectively, as a genuine expression of artistic taste, as sincerely meant,
then given the consistent slighting of the old ways in the sonatas we would
have to conclude that Scarlatti was a hypocrite.
These concerns are especially relevant as we return to the preface to the Essercizi,
since its tone is so remote from that of the strictures contained in the Huescar
See Kirkpatrick, Scarlatti, 120.
Quoted in Kirkpatrick, Scarlatti, 121. We might also note the characterization by John Lynch of the Duke as
a malicious man . . . who, it was said, would betray his own mother to further his ambitions; Lynch, Spain,
184. This might help distance us further from any notion that Scarlattis letter simply represents an amicable and
sincere transaction.
Panorama 75
letter. Not only that, but it almost seems to set up an anti-mythology, inducing the
reader not to take the works too seriously, since the composer himself disclaims all
seriousness (profondo Intendimento). It almost seems as if Scarlatti is colluding in
the subsequent image of himself as light, supercial, the class clown, as if he does
not want to play the game of being a great composer. Note in this respect the
denial of Visions of Ambition. There is, of course, a historical way of rescuing the
preface, by referring it to the epistolary tradition of the modest disclaimer (such as
we nd in Mozarts dedication to Haydn of the six string quartets of 17825). It was
customary for the composer to downplay the quality of his efforts in this manner.
It was, however, equally customary for the composer to stress the seriousness of his
labours, as Mozart does, rather than to imply that the music has been shaken out of
his sleeve, as Scarlatti seems to.
One also needs to consider the conjunction of the preface with the dedication to
Jo ao V that precedes it, a matter that has rarely been considered. This does indeed
contain the standard obsequious gestures, but how can one square these gestures,
the magnitude of the dedicatee and the honour of the event that seems to have
occasioned the publication (the conferring of a knighthood on Scarlatti by the
King) with the trivial tone of what follows? This is not to imply that any offence
would have been taken by the monarch, who would surely have known what to
expect from his former employee, but to suggest that there is a certain breach of
decorum inherent in the conjunction of the two passages. It would also seem to
be a strange way to respond to a knighthood to write light music, and further
to imply (as is indeed the case) that the works are difcult to execute and that the
works are not very varied in style. Even the nal words, Vivi felice (live happily),
although once more a common enough formula, arrive abruptly, with a distinct lack
of ceremony. The implication, if we want, like Mellers, to appropriate a modern tag
on the composers behalf, is: dont worry, be happy.
It is in these terms rst of all that we must grapple with the preface as a public
staging of the gure of the composer and not simply as the outlining of an
artistic creed. As a rare gift for the Scarlatti scholar, the preface has commanded
many imaginative readings, even though most have looked simply for such a creed
or for evidence of the composers real-life personality. Pagano, for instance, thinks it
consonant with the qualities lauded by Mainwaring, of charming modesty, without
acknowledging the historical roots of such self-deprecation.
Sebastiano Luciani
rightly characterizes the preface as a delicious display and suggests it is repre-
sentative of Scarlattis mordant character;
presumably he is referring to musical
character, since the only grounds for lending the real-life Scarlatti such attributes
lie in a realist reading of the preface itself. Several writers do in fact dwell on the
He does, however, note that the preface is terribly remote from the previous adulatory Baroque delirium and
wonders, playing on the title of an article by Kirkpatrick, Who wrote the [sic] Scarlattis dedication?. Pagano,
Vite, 413. The article alluded to is Who Wrote the Scarlatti Sonatas?: A Study in Reverse Scholarship, Notes:
The Quarterly Journal of the Music Library Association 29/2 (1973), 42631.
Luciani, Note I, 469.
76 The keyboard sonatas of Domenico Scarlatti
sociological implications of the delicious display. Zuber notes its astonishingly free
tone, associating this with the fact of its publication in progressive London rather
than in Spain.
Many aesthetic readings of the preface concentrate on the two key phrases,
profondo Intendimento and Scherzo ingegnoso. For F. E. Kirby, the conjunc-
tion of the ingenious jesting with the following aim of mastery of the harpsichord
shows something characteristic of the new galant taste the emphasis on enter-
tainment and diversion coupled with a didactic aim. For Gretchen Wheelock, in
putting both expert and amateur on notice to expect challenges to traditions of solo
keyboard composition, Scarlatti acknowledges that his ingenious jesting intends
serious didactic ends. Here the didacticism seems to lie not in Kirbys technical
programme but in the very ingeniousness of the jesting. In other words, behind the
ingeniousness lies learning, but it is not the kind of learning a composer traditionally
displays. In a twist on this line of thought, Zuber reads profondo Intendimento as a
reference to the strict or learned style: Scarlatti is, in fact, opposing the rationality of
musical hearing with an outmoded strict style. She believes the whole document
has been underestimated, as merely the programme of a galant virtuoso. What
is at stake, we might claim, is a new kind of artistic intelligence. The readings by
Wheelock and Zuber spell a modernist refutation of traditional techniques and aes-
thetic attitudes, just the refutation that Burney championed in Scarlatti. This would
seem to be endorsed by the subsequent phrase Show yourself then more human
than critical, which could be understood as an appeal to contemporary relevance.
Loek Hautus in fact invokes Burneys claim that the composer knowingly broke
the rules, from a position of strength, as it were; he observes the modication of
Scherzo, the playful element so striking in the composers music, by ingegnoso,
which makes it clear that naive cheerfulness is not implied. In short, Scarlatti is
a reexive composer.
This reading supports the earlier assertion of Scarlattis
No such quality is implied by those writers who take the composer at his word,
those for whom profundity must be a demonstrable intent both musically and
verbally. This reects the kind of cultural conditioning that has already been discussed
in several contexts. His art has its limits, writes Klaus Wolters, and [Scarlatti] is
modest and honest enough to mention these in the preface. For Keller, the preface
conrms Scarlattis lack of spiritual depth and universal signicance when compared
to Bach. For Philip Downs, Scarlattis warning not to expect profound art was not
a facetious warning, for his readers did not want the profundity of a J. S. Bach.
Even Boyd, admittedly in something of an aside, writes One may argue about the
extent to which Scarlattis intentions went beyond a mere ingenious jesting with
Zuber, Blumen, 19.
Kirby, A Short History of Keyboard Music (New York: Free Press, 1966), 1656; Wheelock, Haydns Ingenious
Jesting with Art: Contexts of Musical Wit and Humor (New York: Schirmer, 1992), 18; Zuber, Blumen, 1819;
Hautus, Insistenz und doppelter Boden in den Sonaten Domenico Scarlattis, Musiktheorie 2/2 (1987), 137.
Panorama 77
art in the sonatas.
The presence of mere speaks eloquently for the force of the
dominant cultural model. Are we to take it that Scarlatti is not serious (enough)?
With the name of Bach acting as the natural touchstone, whether implicit or
explicit, for these discussions of Scarlattis seriousness, we might also consider the
claim of Robert Marshall that the Goldberg Variations were inuenced by the Essercizi.
He conjectures that, in placing madcap and strict-canon variations side by side, Bach
might have been responding to the preface, with its duality of profound learning
and ingenious jesting.
If this were indeed the case, it would represent a char-
acteristically systematic response to and misunderstanding of the terms of the
preface. Scarlattis denial of profound learning should surely be taken in the spirit
which Rosen nds in the ingenious jesting of Haydns popular style, which can
ingenuously afford to disdain the outward appearance of high art.
A similar mock
ingenuousness can be found after all in the very title given to the collection. This
might be another customary way of expressing humility, but there is certainly an
ironic gap between this claimed modesty and the arrogant uency, if one will, of the
technicalmusical contents. In this respect at least Scarlatti seems happy to throw us
off his trail.
Wolters, Domenico Scarlatti, in Handbuch der Klavierliteratur I: Klaviermusik zu zwei H anden (Zurich: Atlantis,
1967), 155; Keller, Meister, 29; Downs, Classical, 53; Boyd, Master, 190.
Bach the Progressive: Observations on his Later Works, The Musical Quarterly 62/3 (1976), 3489. Roman
Vlad also thinks it highly likely that Bach composed the Goldberg Variations in full knowledge of the Essercizi;
Vlad, Storia, 14. Sheveloff expresses grave reservations about such a proposal in Sheveloff, Frustrations II,
Rosen, Classical, 163.
If Scarlatti had a genius for leaving few traces in life, he showed the same talent
in his work. We have already reviewed the difculties of classifying his style in
the large, issuing both from the broader historiographical problems associated with
eighteenth-century music and from the composers own anomalous position within
such a system. On a more intimate scale too the details of Scarlattis language are
difcult to x. In particular, his relationship to such notions as topic and genre is
ambiguous and elusive. The exact source or stylistic location of what we are hearing
at any one moment is often quite unclear. On the other hand, the sonatas seem
unprecedentedly open to a range of inuences hence the panorama tradition
and unusually direct in their presentation of them. This is particularly true of all the
popular material, which rarely offers occasion for pastoral nostalgia or a culinary
exoticism. Thus we are faced with the paradox of a music that is turned outwards
yet resists classication. If Scarlattis range is democratically wide,
there also ap-
pears to be a certain reserve in the avoidance of explicit allegiances of topic and
The critical difculty has been to hold these two conicting elements in some sort
of equilibrium. The panorama tradition rushes to embrace the diversity of material
in the sonatas, but, by making such varied manifestations of style a global attribute,
it skirts the question of how they are to be identied and how they operate in
particular instances. The alternative, approaching the paradox from the other side,
is to deny programmatic or picturesque intent. Not surprisingly, this has been less
in evidence in more recent writing, since the formalist line that meaning is found
beneath rather than on the surface has fallen from grace. Thus Kirkpatrick, having
done so much to esh out a panorama, especially a Spanish one, stated nevertheless
that it does not nd expression merely in loosely knit impressionistic programmusic,
but is assimilated and distilled with all the rigor that Scarlatti had learned from his
sixteenth-century ecclesiastical masters, and is given forth again in a pure musical
language that extends far beyond the domain of mere harpsichord virtuosity.
was of course an attempt to give creative respectability to a gure who so often was
Mellers, Orpheus, 86.
Kirkpatrick, Scarlatti, 115.
Heteroglossia 79
(and still often is) seen as the provider of light relief; more importantly for present
purposes, it tries to rescue the composer from any implications of an indulgent
Many other writers were anxious to distance Scarlatti fromthe sullying associations
of the programmatic.
The desirability of an abstract view is evident in Degradas
assertion that, whatever the [external] suggestions from which the imagination of
Scarlatti takes its cue, each element quickly loses its ties to an empirical reality,
becoming puried in the nervous ow of the music and being reduced, without
the least descriptive ambition, to the abstraction of a formal game.
Even if such
pronouncements seem quite obvious in their historical moment, they can certainly
not be rejected completely: if the external suggestions were as fundamental as the
panoramists imply, then they would surely be more transparent in their presentation.
What the abstractionists play down, though, is the very fact of the mixed style itself,
and in particular the historical force of such impurity. After all, if Scarlatti was intent
on purity, he was certainly at liberty to ignore the outside voices that seem to press
in on his musical world this is what all composers to a greater or lesser extent had
always done.
This very fact of a mixed style, not just globally but more often than not at
the level of the individual sonata, allies Scarlatti unambiguously with modernist
tendencies. An essential difference between our binary pair of Baroque and Classical
lies in the sense of musical argument that arises in the latter through a pronounced
variety of material. Because this variety issues from a single organizing gure, the
composer, there is an inherent sense of critical perspective on or distance from the
material. There can be no feeling of absolute authority to the discourse when so many
different voices present themselves; instead, language assumes a relativist signicance.
Replacing the gure of the Baroque is the Classical topic, a term which by
denition refers to a larger musical world, one of which it forms just a constituent, a
possibility. It is axiomatic to this study that Scarlattis sharp variety of musical materials
encourages us to view him in a Classical light; yet such a classication seems difcult
when the topical operations so often appear to be covert. How democratic can this
variety be when its manifestations are not readily accessible and comprehensible by
either musical amateur or professional?
The elusiveness of such open music may be illustrated by several cautionary tales
deriving from commentary on particular sonatas. At this stage our concern will be
with sonatas that present a relatively unied surface, where topics, to use Leonard
Ratners distinction, act as types rather than styles.
In other words, the topics
ll the frame of a section or movement (as in the labelling of a piece according to
the dance type of a minuet) rather than simply being one element among several or
many (as when minuet style forms just part of a more varied whole).
See for example Willi Apel, Masters of the Keyboard: A Brief Survey of Pianoforte Music (Cambridge, Mass.: Harvard
University Press, 1947), 164 (the sonatas are entirely free from programmatic connotation), or Valabrega,
Clavicembalista, 111 (his pronounced aversion to any programmatic design).
Degrada, Enciclopedia, 358.
Classic Music: Expression, Form, and Style (New York: Schirmer, 1980), 9.
80 The keyboard sonatas of Domenico Scarlatti
Ex. 3.1 K. 238 bars 15
Commentary on the Sonata in F minor, K. 238 (Ex. 3.1 presents the opening),
has uncovered a nice variety of attributions. David Fuller suggests it is reminiscent
of Corelli preludes and allemandes. Pestelli also hears the sonata in terms of older
models, comparing it to K. 8 and K. 92, both essays in a dotted style, and so assigns
it to the Portuguese period of the 1720s. In the light of Kirkpatricks remark, My
Portuguese friends tell me that [K. 238] resembles a folksong from the Estremadura,
Pestellis chronological suggestion is fortuitous! (Rafael Puyana tells us it was Kastner
who made this suggestion, and that the melody derives from a popular ballad that is
still sung today.) To back up his classication, Kirkpatrick further suggests a scoring
for outdoor wind band and a possible processional context. Gianfranco Vinay notes
that Casella used bars 26ff. in the Sinfonia of Scarlattiana; the reference is found
in the Grave introduction that calls up a glorious past, suggesting that Casella too
heard this passage, if not the whole sonata, as antique. Boyd counters the folk-song
classication with: But [Kirkpatrick] did not quote the folk-song, and the style of
the sonata as a whole seems to derive more from French court music than from
what we would normally recognize as Spanish folk style. (Boyd confuses Spain with
Portugal, perhaps assuming an Iberian musical solidarity that we may pass over for
the present.) For Clark, any French aspect is surely a matter of the look of the
notes on paper; like many others, [this sonata seems] to be lled with that intense
loneliness so typical of so much Spanish folk music. Subsequently she has stated
that the sonata uses a well-known tune from Segovia, sung to the romance Camina
la virgen pura.
Thus, just in terms of national style or identity, K. 238 has been
found to be Italian, Portuguese, French and Spanish.
If this is indeed a Portuguese or Segovian folk tune, then it shows the dangers of
a too easy categorization based on apparently familiar surface phenomena. On the
Fuller, Dotted, 117; Pestelli, Sonate, 161; Kirkpatrick, Scarlatti, 167, 201 and 294; Puyana, Inuencias, 52;
Vinay, Novecento, 128; Boyd, Master, 180; Clark, Boyd Review, 209; Clark, Clark Notes, [5].
Heteroglossia 81
other hand, in even the most folk-like of Scarlatti sonatas, there will inevitably be
interference from other musical styles or from other types of syntax we are not after
all dealing with transcription. Even given the most genuine attempt to render what
is heard, this can only take place against the linguistic constraints of the time. The
sequences at bars 11
or 2630
, for example, surely offer a Baroque style and
syntax; Casella chose wisely for use in his archaic movement. On the other hand, a
passage like 8
to 11
seems very near to a possible folk model, especially with the
isolated melodic impulses in the right hand. These raise questions not just of critical
interpretation but of performance practice. If one reads the piece as French dotted
style, then these melodic units can be heard and played as straightforward ourishes
within the style. If, on the other hand, they are felt to be vocal exclamations, then
a different execution may be in order, less clipped and more expansive.
Another case where differences of aural opinion testify to the composers powers
of suggestion suggestion rather than statement is K. 435 in D major. This has
been heard as implying Italian, French and Spanish musical imagery: castanets jostle
with mandolins and echoes of the French clavecinistes.
The material at bars 45 also
nds a counterpart in an untitled piece in D major (47v) by Santiago de Murcia
from his Passacalles y obras (1732), the most extensive collection of Spanish guitar
music of the time. This reminds us that Scarlatti may have responded to the guitar
playing found at court rather than just that heard in popular contexts, as so much
of the literature implies. The use of the gure by de Murcia may suggest a French
source, given the French background to the popularity of the guitar at court.
Such variety of stylistic and topical characterization does not have to be seen as
in any way problematic. Sonatas such as the two above are an open invitation to
the ear. They solicit the imagination of the listener. The very lack of specicity
of association must be understood as essential to the works, and indeed to most of
Scarlattis sonata output; we are not, in other words, faced with eighteenth-century
naive pictorialism. (Not that there is anything wrong with that; we only need call
to mind what wondrous ends it serves in The Creation and The Seasons.) Even where
aural evocations in the sonatas become quite explicit, they are rarely sustained. In spite
of Kirkpatricks reservations, impressionism, as dened for Scarlatti by Donald Jay
Grout, covers these qualities quite nicely: Because his sonatas absorb and transgure
so many of the sounds and sights of the world, and because he treats texture and
harmony freely with a view to sonorous effect, Scarlattis music may be termed
impressionistic; but it has none of the vagueness of outline that we are apt to
associate with that word.
The strength of such a term lies in making clear that the
Compare the interpretations in: Kirkpatrick, Scarlatti, 204; Sacheverell Sitwell, Appendix: Notes on Three
Hundred and More Sonatas by Domenico Scarlatti, in Southern Baroque Revisited (London: Weidenfeld and
Nicolson, 1967), 290; Chambure, Catalogue, 147; Pestelli, Sonate, 24950; Vinay, Novecento, 123.
See Neil D. Pennington, The Spanish Baroque Guitar, with a Transcription of De Murcias Passacalles y obras (Ann
Arbor: UMI Research Press, 1981). The corresponding passage is found from bar 25 in the untitled Murcia piece.
Pennington reminds us that when Felipe V arrived from France in 1700, he brought with him about twenty
members of the French court, who were used to hearing the guitar played in court entertainments.
Grout, History, 456.
82 The keyboard sonatas of Domenico Scarlatti
evocative material, however suggestive of particularities, is a means to an end, that
the composer is not interested in static depiction.
The often widely diverging readings or better, hearings of individual sonatas
are thus very much in this dynamic spirit. Even though they may play some part
in particular instances, historical distance or ignorance cannot account for such
conicting reactions. Just as the works themselves incorporate different voices, this
generosity is extended to the granting of interpretative room for different listeners.
Another tempting apparent anachronism that can help us capture this quality is
Mikhail Bakhtins heteroglossia. The following denition by John Docker will
have the most force if we understand language to include the musical language
which is our concern here:
[Heteroglossia is] the operation of multi-voiced discursive forces at work in whole culture
systems. For Bakhtin heteroglossia is clearly evident in the workings of language, where the
ction of a unitary national language is always trying to contain the stratication, diversity and
randomness produced in the daily clash of professional, class, generational, and period utter-
ances. Existence itself is heteroglossia, a force eld created in the general ceaseless Manichaean
struggle between centripetal forces, which strive to keep things together, unied, the same;
and centrifugal forces, which strive to keep things various, separate, apart, different.
Such a governing concept is relevant not just to the original cultural sense of Scarlattis
clashing utterances which I claim is conceived as such by the composer but to
what we make of them. It was suggested earlier that the elusiveness of their denition
seems to contradict the democratic accessibility that the variety itself promises to
deliver. Now we may understand, however, that it is precisely the elusiveness that
delivers the democracy. If the framing of topics were to be too neat and clear,
then the sense of heteroglossia would fall away. As things stand, we are offered not
just a successive, but a simultaneous variety of the musical surface, tempting us to
x our impressions in specic terms but allowing for few right or wrong answers.
The exteriority is what counts, not the absolute value of particular references we
think we can identify. Such a process can only unfold because Scarlatti presents a
studied elusiveness it is no accident of spontaneous or improvised Latin invention.
Gino Roncaglia grasped this beautifully in 1957 when he wrote that nothing is
programmatic, but everything is intensely evocative.
To an extent this reects the limits and difculties of topical identication alto-
gether (and this also applies to the classication of gures). There is always the danger
of nominalism in topical approaches; labelling a topic as such does not exhaust the
signicance of the relevant material, since its associations may be a matter of relative
indifference to an argument. Further, topic theory does not readily allow for the
relative neutrality of some material. This is clearest in the case of singing style,
which often seems more like a given of most later eighteenth-century language than
a marked and discrete type of invention. On a larger scale one might question the
Postmodernism and Popular Culture: A Cultural History (Cambridge: Cambridge University Press, 1994), 171.
Roncaglia, Centenario, 65.
Heteroglossia 83
premise that changes of material evince a basic dramatic or theatrical orientation.
How surprising can variety be in the mixed style? Contrasts of material may be-
come as much self-evident as dramatic. If there is also a certain inbuilt interpretative
promiscuity to topical thinking, though, which will rely on the assessment of the
individual context for its explanatory power, this is as much a strength as a weakness.
In the particular case of Scarlatti, however, I have just argued that such issues take
on a harder edge. Further, in his case this all takes place over and above the reader
[or listener] authority that is a basic assumption today the emphasis on the power
of a listener to construct a framework of understanding rather than deferring to the
authority of the composer. Even if we allow and celebrate the variety of responses
according to cultural knowledge and circumstances, there is nevertheless a remark-
ably low level of intersubjective agreement about the likely identity and provenance
of so much of Scarlattis material.
Such concerns seem especially urgent when it comes to classifying dance types
amongst the sonatas. So much, after all, is at stake when trying to x a national
identity for the composer, as manifested in the claims for prevailing ethnic colour.
It is indeed easy to become mesmerized by a concern for dance derivations, and
once more this is due to the seeming directness of presentation. Even if we assume
for the moment that some or many of the individual sonatas are based on particular
dances, we need to stand back in order to grasp the larger point, one that is not easy
to see because it involves a typical Scarlattian absence. This is that the sonatas rarely
identify the dance forms on which they might be based.
The composer, we should
remind ourselves, was free to provide titles and topical designations. The very fact
that he does not label very frequently when he often could speaks volumes. The
eighteenth-century tendency was after all to provide such designations wherever
possible, bearing in mind the pictorial and programmatic tradition. Only in the
case of some minuets and pastorales does Scarlatti align his invention with particular
forms. There is only one exceptional case amongst all the sonatas K. 255, which
contains the words oytabado and tortorilla in the course of the rst half. As
we might somehow expect, these little bits of evidence have proved completely
mysterious; it has been suggested that they refer to organ stops or to bird calls, but
they might also refer to dance types.
If this is indeed the case, it stands as a salutary
exception to the composers silence on such matters. A possible rejoinder to this
interpretation, that there was no need for the composer to label music that was
conceived primarily for private royal consumption, does not seem adequate to the
scale of the silence.
When we move beyond the assumption that particular sonatas must summon up
particular dance forms, we may nd that dance per se becomes the governing topic.
Take the case of K. 496 in E major; in 3/4 time, this has been identied by Pestelli as
Compare Basso, Rousset Notes, 6.
Luigi Tagliavini has wondered whether oytabado is a corruption of the Portuguese word oitavado, which
was a popular dance there in the eighteenth century. See the discussions in Boyd, Master, 178, and Sheveloff,
Frustrations II, 989.
84 The keyboard sonatas of Domenico Scarlatti
a minuet. It seems difcult to agree with this; the basic rhythmic cells, the rst-beat
triplet and the repeated-note crotchet gure, are surely rather too insistent in manner
and gesture for the upmarket dance form. Equally, although this represents a modern
style, it is not courtly-galant. We might compare it rather with the Sonata in A at
major, K. 127, which, while in cut time, has a quite similar atmosphere. Both works
represent that distinctive Scarlattian category of what we might call fresh-air music,
turned outwards but lacking formal ties to any one topic or genre. Surely K. 496
embodies dance as a basic impulse rather than any particular dance form. Suggestions
of a minuet are therefore not excluded, but they cannot be denitive either.
We may certainly presume that a sonata like K. 305 in G major has a dance
basis, but it is so remote from a Baroque stylized form that one should really make
comparisons with a work such as Coplands El sal on M exico, which aims to capture an
essence through the free working of fragments rather than reproduce one single form
or type. Many of these fragments in K. 305 can in fact be heard in other sonatas:
in K. 311 (compare bars 824 with 268 of the present work
), K. 284 (compare
its opening material, with drone pedals, with bars 57 here), K. 413 (compare bars
910 with 1213 of K. 305), or K. 372 (see bars 379, which are very similar to bars
57 here).
The opening unit of K. 305 is almost impossible to scan. Performers are generally
chronically underaware of the implied cross-rhythms of many of the dance-like
sonatas, unless they are clearly indicated by the notation. To give just one possible
version of the opening unit, it could be heard and played as a frankly jazzy succession
of (counting from the initial left-hand G) 5/8 3/8 2/8 5/8. Indeed, irregular
rhythmic handling generally counts for everything in these dance-like movements.
The composing against the bar line suggests that the energy of the dance cannot be
contained in conventional notation. Most of the rst eleven bars are written against
the bar line, from 12 we are back on the downbeat, then 1921 are very ambiguous
in this respect. The phrase elision halfway through bar 33 places the subsequent
music against the downbeat once more.
The second half gives the impression of accelerating. After the newinitial material,
we simply hear permutations of what was heard in the latter part of the rst half,
from bar 22, as the music seems to gallop to a close. In other words, the thematic
treatment is mimetic of the way a dance, once warmed up and having left behind
its preliminary skirmishes, develops an unstoppable momentum.
Scarlatti thus responds to the structural dynamics of the dance and in a way leaves
the world of avowed craftsmanship there is little feeling of the high-art social
context within which the work by denition is situated. Instead there is the sort
of uncanny directness that encourages us to identify such sonatas as the real thing.
To hear such a sonata as some might, as a rened reection from above of real-life
material, is to underplay precisely what is most radical here, the immediacy of tone
and technique and the sensation of rude energy.
Noted in Chambure, Catalogue, 113.
Heteroglossia 85
Of course, such irregularity as we nd in K. 305 is not to be thought of as
inherently realistic. This would be to reinscribe what Lawrence Kramer calls the
sentimentalization of wildness,
the myth of the music of the people being unin-
hibited and free, as opposed to an art music constrained by syntactical and expressive
convention. In reality folk music is often more ordered and regular than art mu-
sic. If the spirit of the dance governs K. 305, and this becomes even more striking
in highly impetuous works such as K. 262, it is idealistically irregular, expressing
the blur of activity, the frenzy, the exhilaration of bodily movement. Once more
evocation counts for more than any programmatic delity. So Scarlatti manages to
give an impression of unprecedented commitment to popular dance forms without
necessarily being highly naturalistic.
This contradictory combination of immediacy and distance tends to be replaced
by simple distance in many other topical and generic contexts. Once more it is a
question of notable absences. Many writers have implied the relevance of generic
categories such as concerto, toccata and suite for the sonatas. Yet two of these are
hardly to be felt at all. Only the toccata seems to have a real generic identity for
Scarlatti, and even then it is not often presented in pure form, being mixed up
with other types of material. There are certainly works that recall or depict the
concerto many of the Essercizi and sonatas such as K. 70 and K. 428 but these are
relatively few and relatively indirect in their references to the genre. The frank-
ness apparent in many of the sonatas of Marcello and Seixas, to name two near
contemporaries, provides a notable foil to this. The straightforward suggestions of
the solotutti divisions of the concerto and the continual presence of overt string
gurations, in such works as the third movement of Marcellos Sonata No. 7 or
Seixass Sonata No. 5 (1980), bring home how subdued such manifestations are in
The composers relationship to the suite category, however, provides the most
telling absence. Gerstenberg noted in 1933 that Scarlatti made little apparent use of
suite movements as models, except for the (fashionable) minuet.
Indeed, few are
the movements that will submit to such generic dance classications; as we have
seen, there is another type of dance altogether that Scarlatti prefers to cultivate. The
actions of B ulow and Longo in creating suites out of the sonatas were thus not just
determined by the problematic brevity and independence of so many individual
pieces they were also an attempt to provide the sort of generic security that most
of the sonatas conspicuously deny. Indeed, the whole notion of genre is held at
a distance. Even in the case of the works labelled as minuets, the most signicant
element is a wider statistical one: there are not many of them. Jo ao Vs appreciation
of French culture and ways, for instance, seems to be reected in the work of Seixas:
very many Seixas sonatas contain short minuets that follow larger movements in a
prevalent two-movement structure. The minuet was of course the aristocratic French
Classical Music and Postmodern Knowledge (Berkeley: University of California Press, 1995), 125.
Gerstenberg, Klavierkompositionen, 85.
86 The keyboard sonatas of Domenico Scarlatti
dance form par excellence. It was also one of the genres most cultivated in Spanish
keyboard music, but not by our composer.
This apparent indifference to certain external allures, what Henry Colles nicely
described as the glamour of conscious association,
is fundamental when con-
sidering Scarlattis relationship to genres. One can see the same attributes outside
the realm of the keyboard sonatas. Magda Marx-Weber nds it striking that, in his
Stabat mater, Scarlatti makes sparing use of the standard chromatic formulas that
occur in most church works with a serious text such formulas as the pathotype
fugue theme that falls by a diminished seventh and the chromatic fourth. She also
notes that the word-painting traditionally associated with words such as agelli and
tremebat is almost entirely absent.
Equally, the ground-bass structure found in
the rst aria of the early cantata Bella rosa adorata is the only known example in all
of the composers music.
In these instances too, Scarlatti seems to prefer not to
In at least one instance, though, such militant creative disdain leads to the opposite
result. There is one topical signal about which Scarlatti is normally absolutely explicit:
the fanfare or horn call. In this case, the individuality is found precisely in giving the
topic such a gloriously full and open expression. Most contemporary keyboard music
did not of course even attempt such effects; but where it did, as with the French
pictorial school, the result is generally restrained, quite unlike the boldness of the
Scarlattian versions. This use of fanfare forms part of a wider predilection for rudely
vigorous open sonorities, including unusual octave doublings and parallel fths, that
would normally have been considered out of scale for a keyboard instrument. The
treatment of the horn call by two other composers provides a telling comparison (see
Ex. 3.2a and b). Towards the end of the gigue nale from his Sonata No. 3, Giustini
introduces an unmistakable reference to a horn call. Note, however, the frequent
insertion of an E in between the C and G. This provides a gentrication of the gure,
the third softening the bare fth of the underlying model,
which was clearly too
rude to stand by itself. Almost exactly the same process is evident in the nale of
Galuppis Sonata No. 6,
in the rst few bars, but here the horn fth is avoided
altogether. It is difcult to imagine any Scarlatti sonata being so coy about this topic.
In most cases, though, the topics found in the sonatas are not self-evident in
manner or presentation. They tend to be skewed in various ways. In the Sonata
in C major, K. 398, the topical basis is the pastorale. The indicators of this topic
This comparison between Seixas and Scarlatti is explored in Kastner, Repensando, 1512.
Colles, Sonata, 895. Notably, Colles also remarks that the sonatas do not appeal to the familiarity of established
dance rhythms.
Domenico Scarlattis Stabat mater, Kirchenmusikalisches Jahrbuch 71 (1988), 19.
Pointed out in Boyd, Cantate, 253.
The classic horn-call gure is made up of two parts: while the upper line descends by a third, the lower descends
triadically, creating intervals between the two parts of a third, fth and sixth respectively. A reversed, ascending
form is also common.
The numbering is taken from Baldassare Galuppi: Sei sonate, ed. Iris Caruana (Padua: Zanibon, 1968),
No. 5052.
Heteroglossia 87
Ex. 3.2a Giustini: Sonata No. 3/iv bars 5677
Ex. 3.2b Galuppi: Sonata No. 6 bars 18
remained very stable over a long period of time: use of drones, parallel melodic
intervals, relaxed repetitions, setting of the music in simple keys such as C, F and
G major that could plausibly be tackled by rustic musicians. A subset of the drone
involves a transformation of the static bass note into a rhythmic pedal, almost always
oscillating between two notes an octave apart. Very frequently this converts into a
crotchetquaver unit in the compound time signatures (such as 6/8 and 12/8) most
favoured for the pastorale. This can be seen in the extract from the Pastorale for
organ by Domenico Zipoli, published in 1716, given in Ex. 3.3a.
The oscillating
octave pattern that opens K. 398 (see Ex. 3.3b) undoubtedly refers to this common
bass gure, but the composer presents it in disembodied form. It covers the full
range of the keyboard, using all available Cs; the gure is reinvented to become a
play of rhythm and sonority. This demonstrates well the composers independence or
critical distance from found material; what should be subordinate becomes central,
what should be restricted in compass becomes wide-ranging. The effect is so gently
playful that one scarcely notices the disruptive wit that underpins it.
The Sonata in F major, K. 379, carries a dance title. The most striking feature
of this Minuet are the demisemiquaver gures marked con dedo solo, meaning
glissando. In the rst half these gures only appear once the dominant, C major,
has been reached, allowing for their simple execution on white notes only. The
second-half equivalents, although not marked as such, should also presumably receive
a glissando treatment. For this to happen in the context of the tonic, however, B
would need to be used rather than the B demanded by the notation. Such an odd
A subset within this subset involves the lling-in of the jumping octave by notes approximately halfway between.
See Der ruhende Pan, the interlude for strings alone from Telemanns Overture for Four Horns, Oboes, Bassoon
and String Orchestra, F11 (1725). A very similar bass pattern is found in the slow movement of Beethovens
Pastoral Sonata in D major, Op. 28, testimony to the remarkable durability of such topical signals.
88 The keyboard sonatas of Domenico Scarlatti
Ex. 3.3a Zipoli: Pastorale bars 4756
Ex. 3.3b K. 398 bars 19
bitonal effect can, however, only just be glimpsed given the speed and register of the
right hands gures. More disconcerting than this, though, is that these nger solos
appear in a work named Minuet, which is hardly the most appropriate home for
them. This is perhaps acknowledged in the title carried by the M unster and Vienna
readings of the sonata, Minu e stravagante. The feature is not simply introduced as
a novelty; it is a natural extension of the earlier rapid scalic shapes in both direc-
tions. Such thematic respectability cannot disguise, however, the obvious infelicity
of this freakish effect appearing in the context of a sociable and fashionable dance
As well as the sort of outright disembodiment found in the examples above,
Scarlatti also deects topics in an indirect manner. K. 18 furnishes an example. It is
built from the busy Fortspinnung found in so many of the Essercizi, but the treatment
does not entirely match. The sonatas repetitive syntax removes the archaic character
from the governing style of the material.
This subtle conict of means and manner
is most apparent in the reiterations of bars 413, and especially from halfway through
This is also discussed in Pedrero-Encabo, Rodrguez, 382.
Heteroglossia 89
Ex. 3.4 K. 263 bars 134
bar 42, a moment when the semiquaver patterns suddenly achieve an extraordinary
poetic stillness.
The Sonata in E minor, K. 263, begins with material of older vintage (see
Ex. 3.4). Like K. 402, in the same key, it presents antique modal polyphony. The con-
trasting lines in thirds in high and low registers found from bars 6 to 11 simulate
antiphonal exchanges.
Compare Scarlattis own Miserere in E minor, which features
The anonymous writer of notes to a recording of K. 263 suggests that it recalls old music almost ironically in
its polyphonic gravity. Notes to recording by Gustav Leonhardt (Harmonia Mundi: BAC 3068, 1970), [1].
90 The keyboard sonatas of Domenico Scarlatti
Ex. 3.4 (cont.)
just such exchanges of parallel thirds.
These are succeeded by imitation and further
antiphony, retained throughout the rst half except for cadential points.
This is a sonata that works by transformation, so that although there is a tonal
return the dramatic progress is from A to B. There are no harsh edges to the piece,
and the decorum of the opening style is never overtly undermined.
It is not so
much that Scarlatti suggests a stylisticaesthetic gap between past and present, but
rather he is playing with a sense of time. The opening has the quality of a memory,
strengthened by its failure to reappear. Through the course of the sonata a musical
present tense of the sort entertained in the discussion of K. 277 becomes more
insistent, especially in the second half. Such playing with past and present may indeed
have been inspired by the schizophrenic professional and geographical circumstances
of the composers career. A sonata such as K. 513, as we shall see in due course,
presents this more overtly.
In spite of the fact that the opening does not return, its presence is felt everywhere
in the rst half. All the octave scales, rising except for the elaborated extended format
Quoted in Marx-Weber, Domenico Scarlattis Miserere-Vertonungen f ur die Cappella Giulia in Rom, in Alte
Musik als asthetische Gegenwart: Bach, H andel, Sch utz, proceedings of IMS congress, Stuttgart, 1985, ed. Dietrich
Berke and Dorothee Hanemann (Kassel: B arenreiter, 1987), 136.
Kirkpatrick includes the sonata as example of a type in which a free succession of ideas brings about gradual
changes of mood. Kirkpatrick, Scarlatti, 278.
Heteroglossia 91
bars 245 and 312, are reections of the opening, with its rising octave followed by a
gap-lling stepwise descent. With the rst such derivative, in bar 12, the initial rising
octave g
is lled in by step; it then contains the descending steps approximately
to the equivalent point in bar 2 where the ear is diverted by the imitative reply of
the right hand. On a larger scale the soprano from 13
to 16
elaborates a simple
stepwise descending octave; note also at bars 18 and 19 the rising octaves then
stepwise descents of the stretto pattern. Later versions of the scale are pointed by the
prominence given to E in various contexts: the suspended e
in the tenor at 20
initiates the falling linear intervallic pattern; the way the right hand curls back up
to e
at 25
before its conclusive descent; the very prominent e
reached in the left
hand by the jump of a third at 26
; the corresponding right-hand shape, imitating
the left across the phrase structure, at 278. In almost every case, the E falls to the
D as in the model.
A number of other archaizing features maintain the suggestion of the antique:
the chromatic imitation from 16, which could be from a ricercare; the subsequent
linear intervallic pattern and sequence from bar 20
; and, very noticeably, the parallel
fourths at 26
and 33
. Because of their position within the structure, and the secure
establishment by this point of a stylistic context for their archaism, these fourths do
not share the anomalous avour of those heard in the second bar of K. 193 (Ex. 1.4a).
In addition, frombar 20 to the end of the rst half all the material is composed against
the bar line, the bar line needing to be displaced to the third beat of the bar. This,
quite different in character from the metrical slipperiness we noted in K. 305, is
suggestive in its own right of earlier practices. One could imagine this piece in a
stylistic context where the bar lines were editorial. This also issues from the rst
material; Kirkpatrick cites K. 263 as a conspicuous example of the undesirability of
the bar line, although restricting his remarks to the opening.
Through all these means the opening is kept alive while its features are absorbed
into somewhat more modern idioms. On a large scale, even the lack of harmonic
adventure in the rst half (which continues to be the case later in spite of the more
active harmonies) ts with the decorum of the opening style; after the chromatic
passage that follows the rst structural cadence in III there are no chromatic notes
whatever and no attempt to inect or shift fromGmajor. The most current-sounding
material forms the coda from bar 34
, more open in sonority and expression than
anything previously, but even this takes its cue from the opening the right-hand
shape at 5
that formed a cadence to the opening phrase is here expanded to
articulate cadentially the whole rst half.
The open fth that starts the second half is a familiar sonority at this point of
Scarlattis structures. It often seems to act as a pivot to a new harmonic world,
clearing the air by invoking an elemental interval. (Compare K. 490, discussed in
Chapter 5.) This casts the closing material from the rst half in a new light. The
second half-bar unit of bar 41 is now a repetition rather than being a third higher,
Kirkpatrick, Scarlatti, 298.
92 The keyboard sonatas of Domenico Scarlatti
the total effect now being musing and introspective, less ordered, and the left hand
follows suit at 45. The sonority grows richer on the turn to A minor. In fact, after the
rst bar, the whole of the second half is in the minor mode. The rst-half material is
thoroughly reordered, and expression becomes more urgent as past associations turn
into present experience, disturbing the previous equilibrium. From bar 47 there is
no imitation; the right hand continues its line, giving a lyrical sweep to the ascending
sequence as opposed to the ordered turn-taking of the rst half. The gure in the
second half of each bar is composed of steps rather than the previous falling thirds
heard in 28 and 29; the painfully dissonant appoggiaturas on the third crotchet of
each bar make this narrower range very audible.
Further intensications follow. The linear intervallic pattern from bar 53 is much
higher than before. The chromatic gure from 58 is greatly intensied through its
presentation in a stretto form. From bar 64 the cadential phrases that were separated
by six bars in the rst half (25
and 32
) are now juxtaposed, again in
ascending sequence. In bar 68 we hear a richer and higher version of 15, with our
parallel fourths now placed in a clearly diatonic rather than archaic context.
The register continues to be higher in the transposed closing material from bar 69.
The penultimate bar carries the emphasis on seconds to a logical climax, as the har-
monic texture is invaded by crushes. If this is remote from the language of the
opening, so is the marked sense of a personal lyrical voice above them. The hint of
exotic-Spanish avour here, which has been tasted briey at several other points in
the second half (especially in the scales at bar 62), acts as an index to the change in
orientation of the material since the outset. The nal bar may be an archaic reference
(the ending in minor that omits the third as a propriety), but the E octaves in each
hand can also be heard as a verticalized reference to the octaves of the initial entries in
bars 13. In both respects this nal bar constitutes a somewhat grim gesture towards
the decorum of the opening topic. K. 263 is dramatically conceived, yet there is no
rupture of style of the sort we will observe in many sonatas to come. The stretto
from bar 58 is emblematic of this quality; it is at once a climax of learned style and
the passage of most intense lyricism in the sonata.
Such inherent creative polyvalence means that few sonatas seem to display an
absolute delity to their putative topics. Some possible examples are K. 446, a past-
orale, and K. 365, a rare example of apparently unbroken Baroque decorum. A work
like K. 198 in E minor sustains a two-part invention texture almost throughout, but,
rather like K. 263, it nishes in a very different place to that where it started,
becoming more and more racy and shading into the territory of the dance. A
number of the Essercizi appear not to share such topical wavering. K. 4 in G minor,
for example, is impelled along at an even rate, never really strays from its opening
material, is not premised on surprise.
The splitting of the texture into distinct
Peter Williamss comment that it looks as if it is meant to be played dolce afrms this latter sense. The Chromatic
Fourth during Four Centuries of Music (Oxford: Clarendon Press, 1997), 106.
Several writers suggest that K. 4 is an allemande: Pedrero-Encabo, Rodrguez, 375; Pestelli, Sonate, 138; Seiffert,
Klaviermusik, 422.
Heteroglossia 93
voices near cadence points to provide a richer sense of closure the voices often
chase each other towards the nal chord is retained by the composer as a device
long after most of the elements of this language seem to be abandoned.
A more intriguing test case for topical delity is provided by the Sonata in B
minor, K. 87. Sheveloff claims that this work, like K. 8, 52, 69, 92 and 147, seems
to be arranged from some sort of large homogeneous ensemble work, like a string
fantasia or concerto grosso.
Yet the freedom of part-writing and informality of
texture we nd in these works are surely only possible precisely because all the lines
are conceived for one instrument. The intimacy of tone and technique also rather
argue against this attribution. A most telling piece of evidence is that, in his 1746
concerto arrangements for string orchestra of many of the Essercizi, Charles Avison
does not arrange K. 8! There is in any case a sonata that ts the bill better than any
of those listed by Sheveloff: K. 86, which suggests a Corellian trio sonata, although
even there the counterpoint is too wide-ranging and free for this to be a reality.
What all these works do share, though, is a certain ambiguity of creative stance.
How style-conscious is Scarlatti here; is he inside the style or detached from its
techniques? Is K. 87 an attempt at a genuine stile antico or a nostalgic glance?
The rst aspect to consider is the very undramatic harmonic movement; this sonata
barely leaves its tonic. The end of the rst half is more on than in the dominant,
featuring an imperfect cadence, IV of B minor, at 31
32. The e
in the soprano
on the last quaver of bar 33 provides only the weakest of tonicizations of V. In any
case, there is the plainest of moves back to a root-position B minor at the start of the
second half. We should note too that V was not a normal destination for the rst
half of a minor-key work. Compare the end of the rst half of K. 60, which is also
very much on the dominant of G minor rather than in it; K. 67, another likely early
work, shares this feature. The tonic acts as constant magnet in K. 87, in particular
in the thematic form of bar 1. There is no articulated opposition of keys in other
words the harmonic language is not really diatonic, and this reinforces the sense of
the archaic topic.
On the other hand, what models are there for this free counterpoint? The texture
is congested, and this is not claried by the small amount of imitation. Here, as in
K. 52 and K. 69 in particular, Scarlatti seems happy to write contrapuntally with-
out an explicit formal basis. Such textures are hardly unknown elsewhere in the
eighteenth century, but the degree of informality seems unique to Scarlatti. If we
compare K. 87 with a movement such as the Allemande from Bachs Partita No. 4
in D major, we nd that, for all its freedom, the texture there is much more hierar-
chically conceived; and Handels free contrapuntal textures are neater and less dense
than what we nd in the present work.
We must acknowledge, however, that this
Sheveloff, Frustrations I, 416.
A number of Handels Courantes approximate to this sort of keyboard texture. Compare, for example, the
Courante from the Suite in C minor, HWV 445, which has a good deal in common with K. 69 (however, some
of its initial material is used as a Sarabande in the fragmentary Suite in C minor for two keyboards, HWV 446!),
or the Courante from the Suite in G minor for Princess Louisa, HWV 452 (c. 1739).
94 The keyboard sonatas of Domenico Scarlatti
may be more our problemthan Scarlattis. Hans Keller once asserted that there was no
adequate language for discussing textures that are freely polyphonic. Commenting
on this, Philip Weller has suggested that most of our terminology for dealing with
polyphony is derived from the teaching of strict counterpoint. He believes we need
to acquire a exible vocabulary and mode of discourse capable of dealing with
freely unfolding polyphonic textures.
Although the existence of this conceptual gap may mute any claims for uniqueness,
there does seem to be something special about Scarlattis free polyphony. In its
unsystematic texture it is, I would suggest, reminiscent of the composers dislike
of formal neatness in other contexts and his aversion to formality altogether, to
overt structural, topical and generic control. We might look also to his fugues,
which subvert the genre,
and the imitative openings that are quickly abandoned
or undermined. These imitations are sometimes taken to arise from sheer force of
habit, but, in that they suggest a relatively strict contrapuntal conduct that is almost
always denied, they may also embody disdain.
K. 87 is particularly close in spirit and substance to K. 69; compare the respective
nal bars or the constant use of a rhythm in conjunction with a stepwise descent.
Both seemintense and tender in mood yet there is also some sense of distance framing
the music. Of course this is in a way inevitable and prompts some renement of our
central point of enquiry. By denition there will be a gap in the perception of the
piece, since the style it embodies is not compatible with the modern musical dialects
of Scarlattis time. This gap was exploited as such by composers in the sacred genres
which were the usual home of the stile antico, so as to suggest the historical and moral
authority of a past style. Its very inaccessibility to a modern sensibility (both then
and now) is what guarantees its effect. The crux of the matter, therefore, is whether
we can locate anything within the sonata itself that suggests this distance.
There appear to be no breaks of decorum in K. 87: the inexorable quaver pulse,
with scarcely a trace of normal periodicity, seems to increase the external gap and
weaken any internal one. The music seems to renew itself without the overt cre-
ative intervention so favoured elsewhere by the composer. The descending dotted-
rhythmic bass and the constant return to a soprano b
, as agents of this renewal,
underpin the piece. They act like a refrain or disembodied subject. However, the
musical character does surely change in the second half the parts become less in-
dependent, and sequence and the circle of fths are employed, as the music achieves
greater direction (compare bars 63ff. with their equivalent at 27ff.). Is there a hint
of irony in the very deliberate sequences of 48ff. and 57ff.? They are certainly more
modern in style than anything we heard before. Is there some suggestion that the
remote beauty of the rst half must be compromised by the action of the second, a
sense of regret? More patterning is certainly required in the second half to ground
the music syntactically and affectively; perhaps the antique counterpoint cannot be
Frames and Images: Locating Music in Cultural Histories of the Middle Ages, Journal of the American Musicological
Society 50/1 (1997), 33n.
Sheveloff states that each [fugue] is in some way ighty, overcomposed or grotesque; Sheveloff, Grove, 343.
Heteroglossia 95
plausibly sustained in an age when diatonic functionality must take rst place. Do
these changes simply represent technical necessities, though, or are they calculated
to create an aesthetic distance?
The ambiguous creative traces in this sonata are reected in its comparatively vo-
luminous reception, which tends to fall into two categories: the authenticist, which
hears K. 87 as a straight exercise in recreation of an old style, and the anachronistic,
which hears it as the height of emotional poetry.
For Christian Zacharias, in the
former camp, K. 87 is an embodiment of the Spanish past, a Vittoria madrigal re-
born, austere yet unfettered by the conventions of counterpoint. A different sort
of Spanish colouring is detected by Donna Edwards, who says that bars 279 are
characteristic of the siguiriya gitana. This is not completely implausible in its own
right, especially since the rhythmic and syntactical character of the material is very
different from what surrounds it. Puyana believes K. 87 is Portuguese in the character
of its melancholy expression, that it reveals that state of mind known as saudade an
untranslatable mixture of bitterness, grief, anxiety and nostalgia.
However, even the most Romantically or ethnically inclined accounts of K. 87
would not presumably deny the older lineage of its basic material. From these points
of view, though, the language employed would simply be an old means to a new end.
For the authentic interpreter, any added value would already be inherent in the
very use of the language outside its effective time period. Such issues can of course
arise with any use of older styles. What makes them more pressing in the current
case is the feeling that Scarlatti is so keenly aware of what it implies to cultivate older
means, especially when, on the keyboard, there is already a gap between material
and medium. K. 87 seems to be more than a display of science would one be
wrong to suggest that it is more affecting than the real thing, like Richard Strauss
being Mozartian? The many recorded performances seem to share this historicist
relish. Only Zacharias does not favour the prevalent remoteness and self-regarding
but are these a product of history or are they encouraged by a similar
creative stance on the part of Scarlatti?
The historiographical malaise that affects mid-eighteenth-century music means that
the galant style is both difcult to dene and difcult to defend. Collectively we are
not quite sure what it is, but we know we dont like it. The common image of galant
style involves mannered melodic manoeuvres, thin textures, an articial simplicity,
For some of the varying verdicts see Anonymous, Notes to recording by Vladimir Horowitz (RCA: RL 14260,
1982), [1]; Pagano, Dizionario, 635; Pestelli, Sonate, 523; Roncaglia, Il melodioso settecento italiano (Milan: Hoepli,
1935), 261; Valabrega, Clavicembalista, 286; Vinay, Novecento, 123.
Zacharias, notes to recording by Christian Zacharias (EMI: 7 63940 2, 1991 [notes 1985]), 8; Edwards, Iberian
Elements in the Sonatas of Domenico Scarlatti (DMA dissertation, North Texas State University, 1980), 2930;
Puyana, Inuencias, 52.
EMI: 7 63940 2, 197985/1991.
96 The keyboard sonatas of Domenico Scarlatti
a dull moderation of expression, an aristocratic ambience and impoverished technical
means. It has effectively been regarded as a sort of dumbing-down, but not seemingly
in the name of a bracing populism. It therefore involves the unappealing combination
of being intellectually low and socially high.
The basic historical moment of the style is, however, well enough understood:
it is a reaction against the technical and cultural features of Baroque art. What has
not been well dened is the connection of the galant with other anti- or post-
Baroque styles. Crucially, we tend to separate galant from the world of comic opera.
This might seem reasonable enough, given our association of galant with moderate
and buffa with quick speeds, the racy naturalness of opera buffa against the rened
naturalness of galant, the Italianate roots of one against the French roots of the other.
Yet the two styles must be seen as two sides of the same coin, as the public and private
faces of the same tendency. Both were premised on a desire for greater accessibility
and informality, and both achieved this by denying the authority of the church or
strict or high style. While the appreciation of comic opera in these terms has not
been impeded, it has proved difcult to grasp the modernity of the galant. Of course
new simplicity will always tend to impress less than new complexity, but the new
linguistic means of the galant have been stigmatized as mere fashion, as a parade
of trite formulas. On the other hand, opera buffa is cherished in spite of, or even
precisely for, its highly formulaic aspects.
Perhaps the difference in image can be summed up in one word: Mozart. While
the example of Mozarts comic operas gives a retrospective blessing to all that went
before under that rubric, the future issue of the galant has never been so clear. Yet
Mozarts instrumental works, for instance, inherit a galant instrumental style just as
surely as his operas relate to an earlier tradition, only we prefer not to phrase it in
these terms. The bad press that the galant has had obscures the simple reality that
it did win out, not just by weight of examples, but at the highest artistic level. Its
simplicity of surface means and moderation of manner, in the name of more direct
communication with the listener, seem unpalatable to us today as the basis for a
paradigm shift.
The crux of the negative reception of the galant style is the resulting abandonment
of artice and complexity in general, and the abandonment of counterpoint in
particular. Signicantly, one reads frequently about the thin textures of the galant,
while it is unknown to nd disparagement of the thick textures of the older style.
Equally, while the galant is short-winded and features too many cadences, one
does not nd the older style described as long-winded. Further, the galant is dened
by its mostly melodic clich es, while Baroque contrapuntal tags do not suffer from
this ignominy. I have written elsewhere that our superstitious awe of counterpoint
gives it a moral authority [that] seems to place it above . . . critical scrutiny.
authority does indeed seem to be relished quite uncritically by a high proportion
of the musical community. In a sense, this inconsistency of response is determined
Chopins Counterpoint: The Largo from the Cello Sonata, Opus 65, The Musical Quarterly 83/1 (1999), 122
and 117.
Heteroglossia 97
by the very different aesthetic premises of the old high style and the newer galant
one. For Carl Dahlhaus the period of the galant saw the beginnings of true aesthetic
reection, in contrast to the socially exclusive absolutismof the seventeenth century,
where aesthetic judgment was never really an issue.
In inviting a personal response,
indeed an individual view of what music should mean or be, the galant was opening
itself up to rejection by the powers of aesthetic judgement that it was the rst to
allow! The fact that we take the accent of much galant music to be as courtly or
high as the music it replaced is not of the rst importance; what it speaks of is
quite different. This is why there is no necessary credibility gap between Hauers
democraticbourgeois orientation and an often gracious style of delivery.
A useful recent reminder of the foundations to the galants bad press has been
provided by Laurence Dreyfus in his study of Bach. The author states: The Enlight-
enment in the rst half of the eighteenth century resulted in a kind of catastrophe
for serious musical artice, through its naive worship of nature, facile hedonism,
uncritically afrmative tone, appeal to public taste, privileging of word over music,
emphasis on clearly distinguishable genres, [and] rejection of music as metaphysics.
By catastrophe for musical artice we are obviously to understand above all the
decline of counterpoint. Artice in this context seems to be value-free, thus also
reinforcing the absolutist terms outlined in the previous paragraph. On a different
cultural level, uncritically afrmative brings home another current difculty with
the galant aesthetic, what Voltaire dened as its seeking to please.
Dreyfuss phrase
logically implies there must also be an uncritically negative way of seeing things, but,
with our elevation of the tragic and the broken (in modern and postmodern thought
respectively), one should not hold ones breath for it ever to be acknowledged.
Subsequently Dreyfus claims that the progressive musical thought of the day, for
all its elegance and charm, had signalled a regression in technique.
This implic-
itly narrow denition of what constitutes good musical technique, based again on
uncritical elevation of the stricter styles, has dogged not just the galant but all post-
Baroque idioms (in which we should also include the styles of sensibility and Sturm
und Drang). The problematic technical image of these idioms largely explains why
we have the absurd situation referred to earlier of a sixty-year interregnum between
High Baroque and High Classical styles. After all, the High Classical is dened to a
great extent by its recovery of serious technical means, above all counterpoint.
Dreyfuss rather grudging acknowledgement of elegance and charm is typical of
See David A. Sheldon, The Concept Galant in the [Eighteenth] Century, Journal of Musicological Research 9/23
(1989), 9091.
Cited in Daniel Heartz, Galant, rev. Bruce Alan Brown, in The New Grove Dictionary of Music and Musicians,
second edn, vol. 9, 430.
Laurence Dreyfus, Bach and the Patterns of Invention (Cambridge, Mass.: Harvard University Press, 1996), 2434.
Note also the verdict of Daniel E. Freeman: From the standpoint of the modern critic, many composers of the
mid-eighteenth century had much better luck relying on tried and true techniques held over from the Baroque
rather than experimenting with new styles. Freeman, J. C. Bach, 256.
See the formulation by Julian Rushton that the complexity of the Classical style is partly the result of its historical
consciousness, its assimilation of those styles against which the galant was in revolt. Classical Music: A Concise
History from Gluck to Beethoven (London: Thames and Hudson, 1986), 29.
98 The keyboard sonatas of Domenico Scarlatti
our wider failure to enter imaginatively into the claims of the new simplicity. The
pleasant melody and sociable tone of the galant must have been stirring in their
provision of a more immediately human scale to music. We need to try to hear it in
the same fresh light, odd though the comparison may seem, as Debussys monodies,
which also gain their effect partly through the polemical overturning of a weighty
technical apparatus.
In view of the bad press accorded to the galant style it is not surprising that
attempts have been made to distance Scarlatti from its associations. Paul Henry Lang
asserts that while the rest of Europe took readily to the aristocratic style galant, to
Scarlatti this style evidently appeared frozen on the surface and hollow within, a
series of habits and prescribed customs and clich es. Degrada notes approvingly how
the late cantatas contain a density and severity of structure that is far removed from
galant blandishments. A large proportion of Pestellis study of the sonatas is devoted
to disentangling Scarlatti from the style, of which we nd traces and hints which
are only short-lived. If Scarlatti was touched. . . by the galant but not attracted to
it, this was due to the impatient sensibility that never let the keyboard rest.
detailing this war against the galant it is notable that Pestelli hangs on every scrap
of counterpoint found in the sonatas.
For all his protestations, at the very end of the study Pestelli calibrates the sig-
nicance of Scarlattis connection to the galant very nely when he characterizes
it as a love-hate relationship (attrazionerepulsione).
The anti-pedantic orientation
of the style nds an obvious counterpart in Scarlatti, especially in the form of the
freedom of dissonance treatment that was at the centre of its technical identity (and
of many disputes between old and new schools of thought).
Even if Scarlattis
treatment of dissonance goes well beyond what would have been acceptable to the
disciples of the galant, there is still a shared assumption. The same goes for the
prevalent two-part textures found in the sonatas, as well as the moderation of slower
tempo markings that characterize the galant approach
tempo markings slower
than Andante barely exist in the Scarlatti sonatas. It is in any case inconceivable that
Scarlattis music could exist entirely outside the galant, especially when dened in-
clusively to conjoin with the world of opera buffa. The highly articulated syntax and
associated cadential formulations, for example, were inescapable for any composer
who wished to speak in a modern voice. If on the other hand Scarlatti can hardly
be thought to embody all the attributes of the galant spirit, this is no different from
his reserved relationship to all other musical types and styles. The most important
cautionary note is sounded by David Sheldon, who reminds us that most musical
applications of the term galant were made by German writers, and that to ignore
this would run the risk of projecting German values onto all of Europe, and actually
I am thinking of such monodic openings as those to Pr elude ` a lapr` es-midi dun faune and the preludes Bruy` eres
and La lle aux cheveux de lin.
Lang, 300 Years, 587; Degrada, Lettere, 300; Pestelli, Sonate, 232 and 86.
Pestelli, Sonate, 271.
See Heartz, Galant, 431, and also Sheldon, Galant, 97100.
This is discussed in Freeman, J. C. Bach, 239.
Heteroglossia 99
continue the tradition of Germanic bias in historiography. This is valuable in its
implication that the theoretical disputes over the galant may not have carried quite
the same edge for the LatinCatholic Scarlatti.
The love-hate relationship may be seen in two sonatas paired in both V and P,
K. 308 and 309 in C major. K. 308 shows like K. 277 (Ex. 1.2) the galant evocation
of the individual voice. Kirkpatrick suggests it might have been inspired by Farinelli:
one wonders whether Farinelli in his later years was singing with similar purity and
restraint. Ann Livermore writes of a vocal sense of line . . . developed with simplicity
and restraint against a sparse accompaniment.
Historically such suggestions are on
rm ground, given the association of the galant with the operatic world that we
are liable to overlook. To judge from the writings of Quantz a German, be it
said the sonorous ideal of galant music was Italian bel canto, which reached its
height in the rst third of the century, when the greatest castratos, such as Farinelli
and Carestini, were in their prime.
If the example of Farinelli was inuential in
Scarlattis particular case, whether by his presence in Madrid from 1737 or by earlier
repute, then this would have extended beyond the melodic delivery as such to the
constitution of the whole style.
Any ascription of restraint to K. 308, however, risks confusing texture with affect.
Of course there is a certain purity and simplicity to the writing, but to leave it at
that suggests a lack of sympathy with the new sensations offered by the galant. We
tend only to hear what sound to us like thin textures and short-winded melodic
lines, yet, given the preponderance of sigh gures throughout the sonata, one could
speak of a stylized eroticism. Note in particular the deepest sigh, found when the
tenor unexpectedly answers the upper voice at bar 30
; and the frequent grace
notes seem to signal a sort of amorous irtation. To get the full effect of this idiom in
historical context, one must set it beside a more established type of slow movement
K. 69 or K. 86, for example. The nakedness of the texture in K. 308 is shocking by
comparison it is the space between and around the strands of the texture that is so
expressive, indeed seductive. The lack of fullness in note values and texture can thus
be construed positively, not merely as a symptomof technical undernourishment. It is
just such attributes that help to create the galant notion of voice. Language metaphors
dominated eighteenth-century discourse on music, and their force increased along
with the increasing cultivation of shorter syntactical units that could be equated with
speech rhythms. Hence the common metaphor of music as conversation and, more
broadly, the sense of a voice that was exible and attentive to changing circumstances,
that seemed to engage directly with the listener.
This texture promotes an atmosphere where the slightest inection registers, in
which the sighing appoggiaturas can achieve their full sensual effect. Note in partic-
ular the magical conduct of a circle of fths in bars 1115. The unprepared sevenths
Sheldon, Galant, 103. Compare Bogianckinos assertion that the Latin-Catholic world found it relatively easy
to leave behind the Baroque. Bogianckino, Harpsichord, 20.
Kirkpatrick, Scarlatti, 169; Livermore, A Short History of Spanish Music (London: Duckworth, 1972), 11617.
See Heartz, Galant, 431.
100 The keyboard sonatas of Domenico Scarlatti
in bars 12 and 14 illustrate that freedom of dissonance treatment, the force of which
is hard for us to recapture today. The galant was a domestic as well as a courtly
language and therefore likely to be associated with the feminine; all the sighs found
in K. 308 further this sense. One thinks of an eighteenth-century literary equiv-
alent, the epistolary novel, which seems undramatic in its structure and premises
yet can convey great intensity within its world. Indeed, Goethe dened the style
of the related sentimental novel as being typically feminine, full of full stops and
short phrases.
Leaving aside any arguments on essentializing of the feminine, one
wonders whether the galant, like the eighteenth-century keyboard sonata which
often embodies the style, has been downgraded for just this reason. A certain covert
sexism seems to operate in both cases; and this is intensied by a perception that
galant sensibility was conned to a comfortable social world. These generate the
unattractive combination of a style that is intellectually low but socially high.
If one accepts that the feminine sighs of K. 308 should convey some intensity, it is
up to the performer to make this happen. This is particularly true by denition of the
galant, which is a style of personal inection. It is all too easy for the contemporary
performer not to hear beyond Langs prescribed custom. The tone should not be
breezy or innocent or decorative; all the appoggiaturas invite some heaviness of
execution, con amore rather than simply pleasing.
If the companion work, K. 309 (Ex. 3.5a) is not galant in the more specialist sense,
it does exemplify the galant in our inclusive sense (equivalent to the unfortunate
terms pre-Classical or mid-century style), as being the modern vernacular.
Its most striking feature is undoubtedly the long-note melody rst heard from
bar 10. When this enters, interrupting the start of a parallel phrase from bar 8, it
seems to come from nowhere, with the new right-hand note values and left-hand
repeated notes. This sense of incongruous interruption is encouraged by the return
to the opening gure at 1415. The predominant conjunct movement up to bar 9
is replaced by grotesquely sprawling wide intervals the voice leading is as poor as
could be imagined. As with K. 254 (Ex. 1.4), but even more so, this simply must
be a parody of some sort. Is this a joke on the galant? The literature of the time
teemed with complaints on the part of the ancients about the galants inability or
unwillingness to observe the proprieties or rules of composition; we have here
an extreme instance of a lack of learning.
The second version of the long-note
melody, like the rst outlining a diminished seventh, is even more awkward, in
Cited in Pestelli, The Age of Mozart and Beethoven, trans. Eric Cross (Cambridge: Cambridge University Press,
1984), 11.
Note the working denition by Ann Jessie van Sant that greater degrees of delicacy of sensibility often to a
point of fragility are characteristic of women and upper classes, in Eighteenth-Century Sensibility and the Novel:
The Senses in Social Context (Cambridge: Cambridge University Press, 1993), 1. Note that I am making the
assumption throughout this argument that, for musical purposes anyway, galant and sensibility are closely related
We might compare this with Charles Rosens citation of a passage from Sammartinis Symphony No. 6, which
has a rather similar sprawling transitional top line, described by Rosen as unbelievably ugly. Sonata Forms (New
York: Norton, 1980), 14041.
Heteroglossia 101
Ex. 3.5a K. 309 bars 151
the hiccups of its bass accompaniment and its ve-bar duration. There is then some
attempt to repair the damage by gap-lling, the rising leaps being answered by falling
steps at 21 and 23.
The improvement continues with bars 22 and 24 forming
together with bar 20 a larger-scale falling progression, from d
to c
to b
. However,
there is something rather clumsy about the cadential approach of bars 25 and 26,
making one realize that there is another level to the apparent parody an inability
to handle modulation as well as voice leading.
This was also present in bar 14, in the lling of the previous c
gap by the stepwise descent from a
to d
but rather disguised by the thematic role of the bar as a return to earlier material, not the sort of continuation
found at bar 21.
102 The keyboard sonatas of Domenico Scarlatti
Ex. 3.5a (cont.)
Bars 34 to 37 refer again to the problem passage. It is turned into a minor-mode
enclave, with repeated semibreve Gs replacing the gawky leaps. The repeated upper
appoggiaturas and harmonic colour even offer a hint, if no more than that, of Spanish
colouring (compare the similar melodic gures heard at the start of second half of
K. 490). The harmonic movement of the surrounding material is very straightfor-
wardly diatonic. That there might be something rather pointed about this simplicity
is suggested most strongly by the repeated left-hand Gs from bar 28, which cling
to the safety of the dominant after the laboured effort required to reach it. Like the
right-hand line to follow in bars 347, they also offer an emphatic correction of
the pitch contours of the initial sequence of four semibreve values. Thus an entirely
Heteroglossia 103
Ex. 3.5b K. 309 bars 5777
typical bass afrmation of the new key leaps into the creative foreground. The clos-
ing idea also seems pointedly simple; bars 436 could easily be imagined as the
peroration of a comic operatic number, demonstrating again the stylistic adjacency
of buffa and galant.
The second half not surprisingly makes further efforts to put right the problem
passage (see Ex. 3.5b). The rst and second tries, from bars 61 and 67, are less awful
than the rst-half versions because there is a better balance between rise and fall
and the intervals described are narrower. The third version plays even safer, with its
repeated notes perhaps taking their cue from bars 347, but, with its very plainly
exposed tritone caused by the leap up to b
then tamely back to the safety of the
repeated f
, it is actually the ugliest of all. Once more the bass accompaniment
is unsettled in its precise rhythmic form, and the feeling persists that it is quite
incongruous anyway as a companion to the semibreve values. Another irritant is
that, as in the rst half, the passage seems undecided about whether it should last
for four or ve bars.
The second subject from bar 82 is quite drastically rewritten it is only four
instead of six bars long, and the bass is more shapely with its stepwise movement
104 The keyboard sonatas of Domenico Scarlatti
Ex. 3.5c K. 309 bars 8392
than the previous repeated notes. Indeed, the very marked fall from a semibreve
G to a semibreve F, producing a highly directional 4/2 harmony in bar 83, offers
another type of correction to the prevalent leaping about of the semibreve rhythms.
Finally at bars 869 the problem passage is put right (Ex. 3.5c) and its stylistic origins,
totally obscure to this point, are made clear. From galant ineptitude we arrive at
a solution that is like a typical contrapuntal tag (compare several famous Mozart
examples, such as those found in the nales of the Jupiter Symphony and the G
major Quartet, K. 387). We should also note that the bass at bars 869 seems to nd
a settled accompanying rhythm and that it complements the intervallic trend of the
tag in exemplary fashion. From this perspective the previous passages suggest the
composing of someone with a little learning who has a notion of using a clever old
tag, but cant remember how it goes. This sonata parodistically embodies the very
criticisms that were made of the modern manner at the time.
There are many other works that seem to parody aspects of the galant manner,
especially its tendency to produce chopped up music. Deliberately poor continuity
of thought is displayed in the initial parts of sonatas like K. 106, K. 524 and K. 170
(the tempo designation of which, Andante moderato e cantabile, already tells us
what style to expect). On the other hand, just as many sonatas are eager to test the
genuine charms of the style, even if, as in K. 277 or K. 384, these are ultimately mixed
with other, incompatible ingredients. Only in one section of his sonata output does
Scarlatti produce a series of apparently straight galant essays. These, the works centred
around V VI and VII (K. 296355), are what I would dub the modest sonatas;
the chronological implications of their production have already been considered in
Chapter 2. Certainly many of them t oddly in the wider context of the whole
uvre. Their demeanour is introverted, the composers customary nervous energy
and use of sharp contrast being largely absent. They feature no registral extremes,
no marked popular colours, no overt virtuosity.
There is, however, a fundamental contradiction in the relationship of style to
technique in these works that has not been pointed out. Scarlatti treats a galant
idiom treble-dominated, with high and continuous bass lines, an emphasis on
Heteroglossia 105
graceful symmetry, and a pervasive modesty of tone in a rigorous manner, very often
monothematically, as if he is trying to force the idiom into the genre of an invention.
The composer becomes obsessed with pattern-making, so that the personal freedom
of inection that should be at the core of the galant is straitjacketed. The music wears
a xed smile, as it were, and begins to suggest a mode of toy music. In other words,
the galant idiom is forced to march to an uncongenial syntax. It is almost as if the
technique of the vamp has been transferred to the work as a whole, with the hypnotic
fascination of undifferentiated movement; it is noteworthy, though, that the modest
sonatas never employ vamps as such.
One can sense an equivalent to the concept of Classical tone in such works; there
is no way of knowing the extent to which the composer is standing aloof, and even
to ask, as Rosen says, is to miss the point.
Such works seem to bespeak a kind
of boredom, but it is as if the theory that through boredom comes fascination
being put to the test; the fascination comes from the sense that the composer may
be treading a ne line between giving the listener enough to go on and not enough
to go on. Thus such works can both repel and fascinate. The Sonata in A major,
K. 286, provides such an instance. The idea rst heard at bars 8
10 can easily
fascinate; it has an odd avour, with its staggered parallel intervals of fths and
octaves. Unlike most star turns in the sonatas, though, it is not transformed in any
way, nor does it interact with other material; it seems simply to be put through
its paces. Whether or not we choose to become engrossed, one should note that
the composer is being characteristically extreme in his gestures, as we nd with
comparable works such as K. 274, 291, 334 and 342. The tenor of the basic material
is accessible, the treatment rather forbidding in its ascetic minimalism. A sonata like
K. 286 is an entrancing object, one that is perfectly formed and spins indifferently
around before our eyes and ears.
Perhaps the most extreme work of this character is the Sonata in A major, K. 322.
Pestelli comments: Even when using the more casual locutions of international
language, Scarlatti loads his works with suggestions of popular song; see the simple
extended melody that emerges in the codas of [K. 322]. One needs to have heard the
pianist Arturo Benedetti Michelangeli play this passage in full voice . . .; the performer
seems to react polemically to the clich e of a rened and slightly anaemic Scarlatti. In
his hands the passage emanates good health and outdoor singing.
While certainly
agreeing with such an approach in principle, one wonders if it applies to K. 322,
which does seem pallid, not so much because of the nature of the melody, but
because of the thinness of the total texture. The melody is accompanied throughout
only by bass minims.
It would surely be difcult to hear this merely as popular
Rosen, Classical, 317.
Discussed by Diane Arbus in Diane Arbus (New York: Aperture, 1972), 13.
Pestelli, Sonate, 195.
Georges Beck calls these boring implacable minims . . . without variety or vigour. R everies ` a propos de
Scarlatti, in Musiques Signes Images Liber amicorum Franc ois Lesure, ed. Jo el-Marie Fauquet (Geneva: Minkoff,
1988), 15.
106 The keyboard sonatas of Domenico Scarlatti
Pestelli further suggests that K. 322 is composed at the absolute limits of economy
imaginable, achieving a sort of virtuosity in saying everything with a minimum of
means . . . This is an inexpressive work par excellence, without the least tension it
would have been incomprehensible to the masters of the galant.
This is hard to
square with his comments above, but seems more attuned to the spirit of the work.
The means are certainly minimal: this is not simply a dull work, but pointedly,
exotically dull. This quality inheres not in the character of the material as such but
in the implacability of its treatment. By far the most dramatic moment is the simple
diminished seventh arpeggio in bar 63. Really the sonata offers a virtuoso proof
of our boredomfascination symbiosis. There is something akin to what we nd in
Shostakovich or Mahler a mixture of being drawn to and repelled by the banal
but because of the terms of eighteenth-century musical language, dening such a
process is more elusive. After all, it was precisely the galant (remembering its broader
sense) that aspired to the naturalness and simplicity that were seen as the supreme
merits of folk music, which led to a narrowing of the gap between popular and
high-art idioms. Thus K. 322, while patently galant in manner, could also be heard,
in its apparent unselfconsciousness, as a form of stylized or idealized popular song.
The work cannot, however, be heard as a parody, because it lacks any foil within
K. 322 also illustrates the composers concerns with space and register that will be
explored in Chapter 6 this is all keenly felt as narrow and conned. The diminished
seventh of bar 63 is the one expansive gesture, but it is immediately gap-lled.
sonata presents a completely stratied texture there are holes above and below as
well as in between the two lines, and still the whole sounds narrow, because there
is absolutely no depth of eld to the sound.
The extreme, seemingly mechanical
continuity of texture and of syntax remind one again of the phenomenon of the
vamp. This means that, pace Pestelli, the sonata does express a certain sort of tension,
like that of someone who needs to run for a train but is forced to walk.
In bar 65 of his rendering of the sonata, Zacharias plays two minim As in the
bass instead of the correct semibreve.
While this change may represent the sort
of tidying that almost no performer of Scarlattis sonatas can resist, it might also be
that he has quite understandably become mesmerized by the established pattern
of the bass line. The semibreve A in 65 produces a brief loss of momentum that
seems to be occasioned by the mild shock of bar 63. It is unfortunate that the
performer does not observe this semibreve value, since in the terms of K. 322 it is a
Pestelli, Sonate, 239.
This may owe something to the diminished-seventh chord outlined in the treble at 48
49, part of a singleton
phrase that causes an unexpected blip to the repetitive symmetry. On the other hand, this unit is not rhythmically
anomalous as is that heard in bar 63.
Beck, perhaps misunderstanding the world of the modest sonatas, believes K. 322 is a typical example of a
sonata that needs textural lling-in: Shouldnt one breathe into [the bass minims] the life they are lacking by
adding some notes? One could drive a coach with ve horses through the gap between treble and bass. Beck,
R everies, 15.
EMI: 7 63940 2, 197985/1991.
Heteroglossia 107
momentous happening. Apart from this bar, the bass plays absolutely nothing apart
from minims.
It should be clear in the light of a number of earlier discussions that I believe the issue
of Iberian inuence has been largely misconceived. It has been regarded principally
as a question of essence, at the expense of certain historical considerations. We have
seen, for example, that Falla looked to Scarlatti as the classic Spanish composer,
and there is no doubt that Scarlatti had an inuence on later Spanish art music,
whether in dening an approach to the incorporation of popular elements or whether
in suggesting a certain compositional ethos. If we accept that this inuence was
practical as well as spiritual, then the authentically Spanish becomes unknowable.
If we suppose for a moment that nothing about Scarlattis sonatas is intrinsically
or extrinsically Spanish, then the mistaken application of certain features of his
sonatas in the name of Spanish music would logically lead to the exclusion of such
works also from any ethnic canon. This would hardly be a tenable position. The
fact that Scarlattis Spanish idiom may be no more truly representative of Spanish
popular music, or various subsets within that, than horn calls are of German folk
music is not of fundamental importance. Spanishness is what we or a composer
construct as being Spanish; it is in the rst instance a question of tradition, of
cultural determination, rather than one of essence. Furthermore, our impressions of
Spanishness derive in the rst instance from its embodiment in art music, even for
those who have direct experience of, say, amenco guitar and vocal performances.
This is because a natural ltering occurs when we listen to folk music, or when a
composer listens, then attempts to incorporate its elements. An assumption of creative
selective hearing normally operates in the transmission of folk music in an art-music
context that which cannot be captured within certain bounds of coherence and
decorum is omitted. Leonard Meyer tells us that a composers representation of
such sounds is itself always partly dependent upon prevalent cultural traditions for
hearing and conceptualizing the phenomenon in question.
This will vary over
time and according to the properties of the language within which a composer
works, but it also interacts with those ltered features found in previous art music.
In this sense folk elements cannot really be heard at all until they are brought into a
high-cultural context and thus given a basis for comparison.
Style and Music: Theory, History and Ideology (Philadelphia: University of Pennsylvania Press, 1989), 126. This and
the other matters of principle discussed here help render less urgent the logical objection to the whole enterprise
of identifying Iberian strains: that we are in no position to assess the form taken by folk idioms well over two
centuries ago and should not extrapolate back on the basis of knowledge of later examples. See for example
Frederick Hammonds remark that until we know more about eighteenth-century Iberian folk music, detailed
documentation of its inuence on Scarlatti is impossible. Hammond, Scarlatti, 178. We might also note at this
point one of those Scarlattian absences the fact that Joel Sheveloff studiously avoids all questions of Iberian
material and inuence.
108 The keyboard sonatas of Domenico Scarlatti
The notion of being able to recover the essential features of a folk style or a national
character, removing later accretions to reveal an authentic original, is a common
cultural trope. For clear historical reasons, though, it has had particular force in
a Spanish context. The image of Spain, and in particular Andalucia, as Europes
oriental other, as a place where one could see the Middle East without leaving the
is well established. So well established as a musical construction, in fact, that
a composer like Debussy could, in works like La Puerta del Vino and Ib eria, simulate
it with almost no direct experience of the country or its indigenous folk music.
It is hardly surprising that such easy appropriation has led to the defensive and
sceptical attitude characterized in Chapter 2. On a broader scale, Xo an M. Carreira
has noted that a conviction that the task of the musicologist should be to retrieve
what is essential/national and to identify and dene what is articial/foreign is
a constant feature of standard reference works by Spanish and Portuguese musical
The attempt to recover an uncontaminated form of amenco, one which is not
on general access and has not been corrupted by cultural or actual tourism, is rooted
in the same dynamic. This process was initiated with the organization of a cante
jondo festival in Granada in 1922, by Federico Garca Lorca and Falla among others,
the goal of which was to attempt such a recovery after the nineteenth-century
commercial debasement of the style. Its participants, and later representatives, have
been described by Timothy Mitchell as avant-garde primitivists who wanted to
shun history, to escape urban society, to ee the pollution of modernity.
denial of history, of its sullying associations, avoids the central point that authenticity
is not essential to the experience of such music in the sphere of high art. By denition,
it does different cultural work in this context.
Against the relativismwhich has been offered above, one might argue that, without
some attempt to isolate the truly Spanish elements in Scarlattis style, however fraught
that operation might be, we cannot properly judge his style. We will be in danger
of attributing to the composers powers of invention, to his originality, what is
in fact a more or less direct rendering of popular material. A particularly striking
harmonic progression, for instance, a strong use of dissonance, an unusual texture or
type of phrase structure might simply be indebted to a folk model, for all the ltering
involved. How can we possibly grasp the nature of the composers creativity unless
we can identify such sources with reasonable condence, assess the relative delity
of the rendering, note the purposes served by transformations of material? This,
however, misses the point that the very incorporation of these elements, certainly
Timothy Mitchell, Flamenco Deep Song (New Haven: Yale University Press, 1994), 112. See also Etzion, Legend.
Debussy attended a bullght in San Sebastan and was able to hear amenco singers and guitarists at the Exposition
Universelle of 188990 in Paris. See Chase, Spain, 299.
Opera and Ballet in Public Theatres of the Iberian Peninsula, in BoydCarreras, Spain, 17.
Mitchell, Flamenco, 169. For a more traditional account of the circumstances of the 1922 Concurso de cante
jondo see Marion Papenbrok, History of Flamenco, in Flamenco: Gypsy Dance and Music from Andalusia,
ed. Claus Schreiner, trans. Mollie Comerford Peters (Portland: Amadeus, 1990), 457.
Heteroglossia 109
given their apparently vivid manifestation in Scarlatti, is already a form of originality.
As has been stressed before, inuence is only what the imagination of the artist
chooses to make of it. It is a question of more or less conscious creative choice.
Other composers may have heard, but did not listen; at least, they did not let such
elements into their artistic world. This could, of course, relate to circumstances of
employment as well as temperament.
Remaining unaddressed, though, is the question of identication. Just what is this
material that is incorporated by our composer? We have already replied that the an-
swers lie in the future, as it were, in those features that were reected in later Spanish
music, whether issuing from that national environment or simulated elsewhere. But
if we indulge a natural curiosity about origins, we must wonder from whom Scarlatti
derived his Spanish features. When it comes to the incorporation of exotic elements,
he does appear to stand at the beginning of the line. Two younger contemporaries
of Scarlatti, Seixas in Lisbon and especially Albero in Madrid, appear to explore
similar areas, but it would be difcult on the basis of known circumstances to allow
them prior claim to this honour (and if this is so, then in what circumstances and
environments did Scarlatti acquire his familiarity with the style?). The implications
of this literal originality are, as already explored in Chapter 2, uncongenial both to
historiographical and nationalistic thought.
If this exoticism really is without precedent, this is less important in an absolute
sense than in the way it is contextualized within the art work. The exotic sounds
so novel in Scarlatti because it is placed in contexts that exaggerate its difference, or
in contexts that suggest the impossibility of its artistic presence.
In other words, it
forms part of the composers pointedly mixed style. The exotic will assume a harder
edge when it is an unexpected visitor than when it presents itself from the start. In-
deed, incorporation would generally be an impossibility under these circumstances.
Only a few Scarlatti sonatas present themselves in this way. K. 450 in G minor, the
sonata identied by Jane Clark as a tango gitano, is a rare example. Here is a sonata
that really acts as if it were in its entirety a functional amenco dance. The Spanish
element lls the screen. Consequently, there is no sense of argument in the work.
Commentators have often written as if many sonatas were simply to be explained as
this or that dance, but in fact any use of amenco topics seems to be almost entirely
as styles rather than types. Here, on the other hand, there is no overt sense of critical
distance we are simply presented with the whole object. Without the sharpening
provided by the presence of conicting material, the effect of the work is, in fact,
relatively unremarkable. K. 532 in Aminor also assumes a relatively functional aspect,
but we will see in the following chapter that it contains plenty of added value.
The intermingling of terms like Spanish, folk and amenco in the recent discussion
raises the familiar problem of determining the ethnic origin of popular elements in
In this connection I dissent strongly from Richard Taruskins suggestion that the Scarlatti sonatas represent a
typically eighteenth-century use of stereotyped local colour which is essentially comic. The weakness of both
notions should be apparent from the arguments presented not just in this section but throughout my study. See
Taruskin, Nationalism, in The New Grove Dictionary of Music and Musicians, second edn, vol. 17, 692.
110 The keyboard sonatas of Domenico Scarlatti
the sonatas. The ambiguities of classication can be conceptualized as a series of
binary pairs held within ever-widening circles: Andalusian folk music vs. amenco,
Andalusian vs. Spanish, Spanish vs. Portuguese, Iberian vs. Italian. It has already
been suggested, of course, that a precise sourcing of popular elements is not always
possible or even desirable. The use of exotic in recent paragraphs was calculated
to bridge such taxonomical gaps, to focus attention on what counts in an art-music
context. It should also be read as implying something different from popular, but
without any binary opposition: exotic represents the hard edge of the popular. The
most sustained exotic colours, however, are undoubtedly associated with amenco.
The musical and cultural problems inherent in the denition of amenco are
legendary. The enormously complicated schemes for classifying its various vocal
and dance forms are less relevant for current purposes than its comparative cultural
interpretation. For one, amenco cannot be straightforwardly regarded as folk music.
It is not rural, it is urban. It is not timeless, but arose in the relatively recent past (by
general consent it had been clearly established by the start of the nineteenth century).
Its image is not healthy and merry; rather, it tends to connote fatalism, histrionically
expressed, and has strong associations with alcohol, prostitution and that despised
group, the gypsies. Perhaps most importantly, the authenticity of amenco cannot be
equated with anonymity, since its music is largely generated by specic individuals,
whose works carry their name when used by subsequent singers. However, this
no longer seems such a crucial distinction; the presence of specialized practitioners
in all sorts of folk traditions around the world is now fairly widely understood. It
is also very difcult to extricate amenco from the more traditional folklore of the
region. It is generally agreed that the safest distinction is made less on the basis of
material than on that of style of performance. Flamenco is more introverted, tense
and highly ornamented than traditional popular forms. This style is often associated
with the term cante jondo (deep song).
Not only is there considerable ambiguity about the boundary between amenco
and Andalusian folk music, but there is also a tendency to conate Andalusian and
Spanish folk music. This is not just the product of outside ignorance, though; it has
a historical dimension. In the nineteenth and twentieth centuries, as Mitchell has
suggested, promoting Andalusian culture was the best means for promoting central-
ism and defusing incipient Catalan or Basque nationalisms.
When we reach our
wider circles of classication, the same overlappings occur. For instance, Jane Clark
has suggested that the sonatas K. 49092 make up a triptych of forms associated with
the music of Holy Week in Seville: thus K. 490 represents a saeta, K. 491 a seguidilla
sevillana and K. 492 a bulera. This has been disputed by Rafael Puyana; while ac-
cepting that K. 490 recalls the saeta, he believes K. 491 is a Majorcan bolero, and
K. 492 a Portuguese fandango. Further, he believes that many other sonatas belong
to the same family of Portuguese fandangos.
Finally, there are the same grey areas
between the Iberian and the Italian (or Neapolitan). Surprisingly few writers have
Mitchell, Flamenco, 156.
Clark, Andalouse, 635; Puyana, Inuencias, 53 and 52.
Heteroglossia 111
suggested that, rather than being a problem of classication, such ambiguities may
derive from a calculated stylistic crossover (leaving aside for now the question of any
open invitation to the ear). Puyana does make such suggestions. For example, he
notes that Scarlatti often cultivates the rhythm of the Italian gigue and complements
it with Hispanic accentuation, as in K. 525. In other sonatas the Neapolitan alter-
nates with the Iberian, thus amalgamating [Scarlattis] two fundamental sources of
inspiration; K. 429, with its barcarolle rhythm, offers such an alternation.
From this grand confusion we may reasonably assume that identication on the
basis of supposed dance rhythms always, of course, to the extent that such identi-
cation is regarded as necessary is the least reliable of indicators. In material terms,
one could propose that an order of melodic, then harmonic, then rhythmic features
corresponds to relative ease of identication. There is, in other words, less difculty
in disentangling say, amenco, from the Italian when we consider melodic style than
when focusing on rhythmic conformations. Of course, such an ordering is highly
provisional, but I believe it forms an index to relative levels of exoticism, which, it has
been argued, play a cardinal role in the larger stylistic framework. Thus, in the modal
islands of K. 193, for example, the melodic shaping sounds highly exotic, the har-
monic basis somewhat less so, and the underlying rhythm rather less again. The
other determining factor is the implied performance style, and the atmosphere that
this engenders. These considerations suggest, once again, that amenco should stand
somewhat apart when we ponder Scarlattis incorporation of elements from below.
Such a distinction matters because of the socio-political implications of the com-
posers use of amenco elements. We must rst acknowledge, though, that what the
composer incorporated could not have been dened as such at the time. Flamenco
music only assumed any sort of ofcial public identity once the edict of Charles III
in 1782, which sought to end the persecution of gypsies, allowed gypsy music to
emerge from its isolation. Within Scarlattis time at court in Madrid, for example,
Fernando VI decided to have some nine thousand gypsies rounded up and sent to
work in his munitions factories of C adiz, as a sort of nal solution to the gitano
problem. Nevertheless, it is clear that amenco must have developed from a source,
and that elements in the sonatas represent such source material. If so, then what was
a court composer doing bringing such disreputable elements into his music? This
question has been entertained by only a handful of writers. Barbara Zuber offers a
strongly political reading of these circumstances. She reminds us that before Scarlatti
received his knighthood in 1738, he had to attest to his purity of blood that he had
no Jewish or Moorish ancestors (the other two persecuted minorities of the time).
She believes that Scarlatti like other artists such as Cervantes in effect sided with
the gypsies, and that possibly [his] advocacy for the music of Spains lowest social
classes . . . was also a political and social index for his circumstances in Spain, of which
we know so little. Increasingly through the nineteenth century, especially with the
Puyana, Inuencias, 53. See also Clark, Andalouse, 63; for Clark, though, it is more a question of an Italian
sensibility which modies or controls the Spanish elements rather than Italian features being included as such.
112 The keyboard sonatas of Domenico Scarlatti
establishment of the so-called caf es cantantes from the 1840s, amenco became a more
respectable and commercial proposition. Zuber reminds us that Scarlatti put such
material into his sonatas at a time when it was less opportune to perform this strange
music to the Madrid court aristocracy.
Such interpretations have been ventured
explicitly by no other writer.
On the other hand, Mitchell has persuasively documented the fact that upper-
class interest in under-class expressive styles goes back a very long way in Spain,
especially in southern Spain.
This may be correlated with the social phenomenon
whereby upper classes may cultivate a certain roughness of manner to distinguish
their behaviour from that of the aspirational middle classes, ever on the rise. In a
Spanish context, this meant amenquera an aristocratic adoption of gypsy manners
and even dress, to distance themselves from the enlightened Franco-Italian ways of
their middle-class inferiors. Again, though, it is difcult to assess the applicability of
this largely later behaviour to the composers environment. So often in the sonatas
one wonders what Mara B arbara would have made of a particularly vulgar or
irrational passage. After all, there is surely a big difference between the idealized
folk styles that were acceptable enough for court consumption and the electric
intensity more typical of Scarlatti. Kirkpatrick, as we have seen, suggested that such
elements functioned primarily as a distraction, quite the opposite of the gritty
realism we might prefer to hear in them. Only Pagano has suggested that they
may have been understood and enjoyed as such: the insertions of low-life material
seem to have been born from a sort of courtly connivance between master and
Nor must we forget the Queens absolute passion for the dance, as attested
to by the English ambassador of the time, Benjamin Keene; perhaps this passion
extended beyond the execution of the normal courtly forms. Of course, we must
not overlook the possibility that the royal couple, and presumably the court in the
event of those sonata performances we have no record of, could not distinguish
between particular references to amenco-type material and more general popular
inections. Such political and environmental speculation should not in any case
overshadow the broader cultural moment of Scarlattis amenco manner, radical
beyond any doubt.
So what features may be proposed as indicators of a amenco style or manner? The
most salient, we have already suggested, may be melodic. The style is melismatic,
featuring ornate embellishment, incessant repetitions of a single note decorated by
appoggiaturas above and below, a limited melodic range and portamento effects. The
Sonata in C major, K. 548, features from bar 22 a modal island with such melodic
characteristics (see Ex. 3.6). Most notable are the harsh dissonances of bars 3033.
See Zuber, Blumen, 814. Note too the comments of Linton Powell when assessing Scarlattis apparent use
of guitar effects in his sonatas: Curiously enough, native Spanish composers of the eighteenth century did not
show an overwhelming predilection for emulating the guitar in their keyboard works. Perhaps they considered
such gypsy music vulgar. A History of Spanish Piano Music (Bloomington: Indiana University Press, 1980),
See Mitchell, Flamenco, 99.
Pagano, Vite, 447.
Heteroglossia 113
Ex. 3.6 K. 548 bars 1943
For all the apparent renement of notation, what the ear accepts is the insistent
repetition of a melodic cluster that always sounds dissonant against the changing
harmonies. The following texture, featuring purely diatonic sixths in the right hand
and bass octaves, with a clean gap between the hands, forms an effective antidote to
this exotic display. The strange melodic cluster is an outcrop of the previous material,
specically the ourish heard every two bars from bar 22
The cluster is briey heard again at bar 40, followed by a reintroduction of the
syncopations from 22, in a passage that seems like a parody of the exotic. (Note
114 The keyboard sonatas of Domenico Scarlatti
the rough voice leading at bar 43, which is even more apparent at 48.) It may not
be that, but it does lighten the mood by being less static. Such apparent jauntiness
should not necessarily be thought of as antithetical to amenco, which involves
more than just pain and harshness; the sadness of cante jondo is a ritual aspect of its
expression, not unlike what one nds in the blues. It might be preferable to think of
Scarlatti as moving between more or less stylized forms of the idiom; this is in any case
inevitable, given the very act of composition and its high-artistic context. Stylization
is also tied up with the question of how the composer hears his source material.
Klaus Heimes, reviewing such melodic writing in Scarlattis disciple, Soler, suggests
that the conventional notation of such passages often represents but a courtly
purication of a vocal gliding through vacillating intervals.
Such purication,
though, is more an inevitability than the implied concession to royal taste. While the
writing at bars 22ff. might exemplify this process, the clusters at 3033 do not seem
to be very ltered at all. The notational suggestions of various forms of appoggiaturas
and neighbour notes are not very convincing in other words, this is just the sort
of material one would expect a composer not to incorporate, because it cannot be
heard within the constraints of the language and notation of the time. Yet Scarlatti
allows this irrationality into the nished artistic product.
The Sonata in F major, K. 107, is also notable for an apparent attempt to portray
amenco vocal effects (see Ex. 3.7a). The right-hand guration found at bars 1723,
with its outlining of a scale through repeated and implicitly slurred pairs of notes, is
found very frequently in the sonatas. However, is the current example, rather than
necessarily being heard as toccata-like, Scarlattis approximation to vocal portamento?
The repetitiveness of the cadential units and their extravagant ourishes at bars 2530
do suggest amenco melismata, even though the harmonies are diatonic. The related
melodic material from bar 33 is more clearly ethnic, but different only in degree
rather than kind. Also worthy of note is the effect of bars 3943, which do more
than display exotic scale forms; the clashes between the hands produce a compo-
site sound picture that may be suggestive of quarter-tones, of something beyond the
diatonic system and its notation. Such teeth-grinding dissonance is at least equalled
by bars 11213 in the second half. K. 55 is another work which takes great delight in
the displaying of exotic-sounding scale forms, which are surrounded by exuberantly
physical, entirely diatonic material. Again, the narrow clashes of the total texture
seem to reproduce the melismatic microtonal inections of amenco song.
climax of the exoticism in K. 55 comes at bars 8895 (see Ex. 3.7b); it requires a
real act of will not to hear such a passage as Spanish.
To return to K. 107, there seems in fact to be a amenco takeover of the sonata,
symbolized by the very unusual minor ending to a work that begins unexceptionably
in major. So often, when considering the harmonic indicators of amenco style,
Heimes, Soler, 172. The author gives as an example bars 4852 of Solers Sonata No. 19 in C minor.
For another example of the isolated display of such scales, see K. 232, especially bars 278 and 678, although
the effect here is much more quizzical. For an example by Albero, see Sonata No. 19 in B minor, bars 1821.
Heteroglossia 115
Ex. 3.7a K. 107 bars 1743
Ex. 3.7b K. 55 bars 8596
116 The keyboard sonatas of Domenico Scarlatti
Ex. 3.8 K. 313 bars 6576
minor goes with the ethnic and major with the normal musical world. Exoticisms
ourish in the minor, while the major is more brilliant and accessible. This is nothing
very special in terms of tonal rhetoric, except that for Scarlatti the minor allows access
to all those oriental scalic avours. In Ex. 3.8, for instance, from K. 313, a turn to
the minor prompts a marked Spanish coloration in bars 712, where we nd one of
the closest approximations to that now ingrained marker of Spanishness, the rapid
turn gure. Here one could even say that the rubato is notated. However, this gure
is as much thematic as realistic; it reects the second subject of the sonata, heard
from bar 42.
Ex. 3.8 also illustrates the harmonic feature traditionally taken as axiomatic to
Scarlattis representations of the Spanish: the Phrygian progression or cadence. This
involves an emphatic leading towards the dominant by the note a minor second
above, which may be present in the bass or a higher voice; here it is found in the alto
in bar 71. Just as common is the hovering around the dominant by both

4 and

6. This can be seen in the activity of C and E around D in Ex. 3.7b. Boyd is more
sceptical: The frequent Phrygian progressions . . . are often said to derive from
the modes and cadences of Spanish folk-song, but they also occur prominently in
a cappella church music and as cadences in slow movements of Italian concertos and
sonatas. He also points out that the oscillation between the two chords is found
often enough in Scarlattis earlier vocal music.
This cautionary note overlooks the
conrming role that may be played by other factors, such as the stylistic clothing
and wider syntactical context of the progression; Exx. 3.7b and 3.8 seem to leave
little doubt about their ethnic roots. Nevertheless, one must temper ones certitude
when encountering examples such as Ex. 3.9 below, from an aria in Leonardo Leos
opera Amor vuol sofferenza:
Boyd, Master, 18081.
Heteroglossia 117
Ex. 3.9 Leo: Amor vuol sofferenza Tu si no forfantiello bars 79
Ex. 3.10 K. 218 bars 7784
This is a minor enclave that postpones the nal cadence of the opening ritornello.
The bass line hovers around the dominant, with dynamic and accentual weight
falling on

4 and

6; there is also a repetitive syntax that is clearly at odds with the

galant style of the surrounding melodic writing (both the dotted rhythms and the
cadential sextuplet are strong markers of the style). This suggests that such a feature
may be as Neapolitan as it is Spanish. That the Naples of Scarlattis boyhood was
under Spanish rule, however, suggests a partial explanation for such an ambiguity.
One should also be careful not to place too much weight on modality in general
when assessing popular simulations, since the modal functions as such an all-purpose
folk indicator. That said, in practice the role played by other parameters can remove
some of the uncertainty of attribution.
This is undoubtedly the case with the Sonata in A minor, K. 218. In bars 7982
(see Ex. 3.10) the composer strips away the melodic formulae that have domi-
nated the piece. We are left with the



5) bass that in some form or other

has been present for much of the time and an accompanying upper part in voice
exchange with it. What remains is the engine of Spanish harmony as Scarlatti con-
ceives it in this sonata, IV or IV
alternating with V in the Phrygian progression.
118 The keyboard sonatas of Domenico Scarlatti
Ex. 3.11a K. 182 bars 7485
Ex. 3.11b K. 188 bars 12430
This is a moment of unusual creative frankness which lays bare the imagined essence
of a Spanish sound. Indeed, it is like a form of Klang-meditation,
rather com-
parable to those extraordinary moments in the Fandango by Soler in which the
melody drops out and we are left to contemplate the ritualistic bass line alone. The
sense of this being distinct from the preceding music is enhanced by the rhythmic
opposition between these bars and the right-hand hemiola in the previous four
(bars 758).
Another harmonic feature found frequently in the sonatas seems to evoke the
world of amenco: the emphatic ninth above the dominant bass, often texturally
reinforced. Two similar realizations of this feature from K. 182 and K. 188 are given
in Ex. 3.11. (See bars 80 and 130 respectively.)
This ninth may even be related to the Phrygian cadence, as a verticalized form of
the semitone progression. Like the exotic scales, it has a fraught quality that seems
to place it outside the orbit of more open or relaxed folk idioms. Such expressive
denition in the composers use of modality, the pronounced sense of estrangement
from more customary musical languages, is what seems to have inspired a refresh-
ingly critical assessment from J. Barrie Jones. Writing of the Granados arrangement
of twenty-six of the sonatas, the author states: The occasional modal avours of
I borrow this term from James Hepokoski, Sibelius: Symphony No. 5, Cambridge Music Handbooks (Cambridge:
Cambridge University Press, 1993).
Something remarkably similar occurs in the Fugue No. 1 in D minor by Albero, at bars 204ff .
Heteroglossia 119
Scarlattis music were to some extent inspired by Spanish folk-music, and that, no
doubt, is one of the explanations of that curious and sometimes unsatisfactory stylis-
tic mixture that is so characteristic of Scarlattis music.
Although one might jib
at the cautious line on the extent of modal activity, it is nice to nd a direct attack
on Scarlattis mixed style. This at least acknowledges just how incompatible the
different styles are in principle and the extent of the risks Scarlatti runs.
When considering those rhythmic factors that may capture amenco style, we
may set to one side the identication of dance rhythms that has already been shown
to be fraught in its own right. Instead, we may concentrate on several more abstract
matters. Over-repetitiveness is one recurring feature of amenco representation, for
example when the repetition of a normally simple cadential unit turns into the
opposite of what it normally connotes, stability. We have already seen this in K. 107
(see bars 2430 of Ex. 3.7a). In the Sonata in G major, K. 105, we nd in bars 649
three consecutive versions of the two-bar unit previously heard just once at bars
523 to clinch a phrase.
Here repetition is made exotic and therefore stylistically
unstable. This may be related in principle to the vamp, which does the same on a
much larger and more disruptive scale. The repetitions found in bars 3944 of K. 502
go further than this, though. Here we are treated to six consecutive bars of the same
The sense of irrationality is magnied in the second half of K. 502. From bar 94,
with the changes of time signature in conjunction with the crude sequential patterns
and the agglomeration of different melodic rhythms, one senses perhaps more than
anywhere else in the Scarlatti sonatas a straining towards something that cannot be
expressed in the notation, that is quite beyond the comprehension of the world of
high art. Nowhere else does the music break down quite so openly and vividly. To
hear this just as a particularly lively translation of folk idiom is to miss the main point.
Recalling our principle of creative selective hearing, we would expect such music
never to have made it onto the page. The problem could not be much more acute
than that faced by a composer trying to assimilate amenco idioms, which are not
entirely European in origin and expression, and in the eighteenth century. What
Scarlatti is presumably trying to capture here above all is the metrical complexity of
amenco rhythms.
Another rhythmicsyntactical factor is more abstract still. It was suggested in the
early account of K. 277 that the inuence of amenco, and to an extent all folk music,
Enrique Granados: A Few Reections on a Seventieth Anniversary, The Music Review 47/1 (1986), 22.
Malcolm Boyd discusses copying matters with respect to the P and Madrid versions of K. 105 in a review of
the recording by Scott Ross (Erato: ECD 75400, 1989), Early Music 17/2 (1989), 272, and Scarlatti Sonatas
in Some Recently Discovered Spanish Sources, in Domenico Scarlatti e il suo tempo, 667. Both scribes seem to
have copied from a source in which repeat signs and great curves were used as a shorthand, leading to some
confusion in the nal product. On great curves, see Chapter 4, pp. 1735.
Clark says that K. 502 is a peteneras; Clark, Spanish, 20. The closing material of each half, from bars 60 and 119
respectively, is strongly echoed in several other Iberian sonatas. Compare it with that found in the same structural
position in the Sonata No. 117 in D minor by Soler, and also with the closing gures found in Alberos Sonata
No. 8 in F major, at bars 40 and 456. These similarities are so pronounced that they suggest less that the two
younger composers might have been inspired by Scarlattis piece than a shared external model.
120 The keyboard sonatas of Domenico Scarlatti
may have operated on a level beyond the appropriation of various idiomatic features
that it encouraged a sense of the contingency of musical style altogether. Equally,
it was suggested that the composers sense of temporality may have been affected.
Such considerations are plain in the case of vamps, but they may intrude in quite
different contexts. The Sonata in A major, K. 404, plays with time through a rather
cubist assemblage of sequences the sort of intoxicating monotony that Scarlatti
may have cultivated under the impact of amenco. The descending scales with upper
pedal that recur again and again (from bars 36, 52, 75, 83, 117, 133, 156 and 164)
are clearly too thin or slow-moving in context to sustain the listeners attention in a
normal manner. Instead, we may nd ourselves listening to the passing of time and
becoming lost in the mechanics of the pattern. The texture too is absorbing in its
dryness. The material may be Arcadian, but the treatment goes way beyond that.
Everything seems to happen in slow motion.
The half-imitative texture heard at the beginning of the second half at last speeds
up the rate of events, but this is then countered by a slowing of momentum. Bars
113 comprise a magical moment when time seems to stop here we have
small sequential repetitions instead of the very long-winded ones that have been
the norm thus far. We hear several frozen gestures, rst of all a sort of idiomatic
musette, then a Spanish turn of phrase, both chiselled out by rests, and then, at bars
11415, a clear reference to a standard sequential syntax. Then the music returns for
good to the previous inscrutable manner. K. 404 could almost be a Satie piece about
boredom, alleviated only by these heart-stopping moments early in the second half.
If we are to connect the temporal sense of this sonata with anything, it might
be better regarded as Spanish rather than specically amenco in character. Indeed,
in some respects it seems opposed both to amenco intensity and to the nervous
character of most of Scarlattis syntax. As one hears it in K. 404, or other works like
K. 296 and K. 544, this is a passive attitude to time. Time is not used efciently or
functionally. Linton Powell has commented on this sense in the works of Rodrguez:
[Rodrguez] tends to carry on gurations and sequences much too long and to wander
harmonically with no clear sense of tonal goal. Anyone who has examined Spanish keyboard
music of the seventeenth and early eighteenth centuries will nd these faults long-
windedness and harmonic meandering. They appear to be native Spanish traits, endemic
to the music. But . . . perhaps they are deliberate esthetic aims. Could centuries of intimate
exposure to an alien Near Eastern culture have left a lingering fondness among the Spanish
people for the static, the contemplative, the immobile, the goal-less, in contrast to Westerners
continual haste to be in motion from one preplanned point to another through the most
efcient means of transport? At any rate, we have not seen the last of this characteristic in
Spanish keyboard music.
Pletnev is surely right to adopt a deliberate Andante tempo in his recording of K. 404. Virgin: 5 45123 2, 1995.
Powell, Spanish, 10. Note also the comments of John Trend on a similar quality in Granados: Yet his sense of
form or, as some critics hastily conclude, the absence of it was also new; he rambled on, making his points
by repetition (like a Spanish poet) and saying the same thing in a number of delightful and decorative ways.
Trend, Falla, 33.
Heteroglossia 121
If this seems to collude too easily with the essentializing of the land of ma nana,
one simply has to have played through some of the fugues of Albero and Soler, the
tientos of Jos e Elas, and even more the sonatas of Rodrguez. To this Westerner
at least, the gigantic sequences one nds may be exotically enticing, but they can
equally be infuriating and upsetting, so implacably do they continue on their way.
Contemplation of this temporal property in conjunction with K. 404 convinces one
of the force of Puyanas denition of an intrinsically Spanish form of expression
that comes from an old tradition in which the passions and temperament are con-
trolled, leading to an intense expressive austerity. He believes that many sonatas
without the slightest folk colour show. . . that the composer had also acquired this
The uncertainties of classication reected upon in this section have an executive
counterpart. This is the question of the relative degree of stylization appropriate
to the performance of perceived popular, and especially amenco, material. With
exceptions like the criticism by the Italian Claudio Bolzan of a recording by Alexis
Weissenberg, where the Spanish dance rhythms are too marked, so transforming
some sonatas into real Iberian dances,
this area has hardly been touched, in per-
formance as well as writing. Often, of course, no particular intervention is required
for such a avour to emerge.
In many cases, though, particularly in the rendering
of cante jondo elements, there is more room to manoeuvre. In Wanda Landowskas
performance of K. 107, the melodic style of which was discussed above, she slows
down markedly for the most exotic melismatic elements (from bars 33 and 107),
to convey what she calls the sensous and provoking nonchalance of the sonata.
Mikhail Pletnev, at bars 25
ff. and 45
ff. of his recording of K. 24, likewise exagger-
ates the exotic by a marked slowing of tempo, as well as the application of Spanish
avour a sort of mannered, histrionic tenderness. In particular he leans on the
alto ninth found at 30
and 50
for a real groan of misery. Emilia Fadini claims that
many sonatas begin in the manner of a guitar introduction, discursively, without
regular metre, even if the notation suggests otherwise; the notated tempo applies
only to the (amenco) song or dance that follows.
Such bold claims are realized
in her performances of works like K. 99 and K. 184. Jane Clark is another recent
performer who sometimes takes a radically direct route. In her version of K. 225, for
example, she replaces the simple crotchet accompaniment of the left-hand chords
Puyana, Inuencias, 54.

Agueda Pedrero-Encabo similarly evokes an eminently Spanish compositional tradition
of an austere, expansive and reiterative character. For her, however, this is an essence inherited by Rodrguez
from Cabanilles, and denitely not encountered in his contemporaries Scarlatti and Seixas. La sonata para teclado:
su conguraci on en Espa na (Valladolid: Secretariado de Publicaciones e Intercambio Cientco, University of
Valladolid, 1997), 248.
Reviewof recordings by Vladimir Horowitz (CBS: MP 39762) and Alexis Weissenberg (Deutsche Grammophon:
415 511 1), Nuova rivista musicale italiana 22/1 (1988), 101.
Try, for example, Virginia Blacks driving, exuberant performance of K. 187. United: 88005, 1993.
Landowska on Music, collected, ed. and trans. Denise Restout, with Robert Hawkins (Secker and Warburg:
London, 1965), 249.
Notes to recording by Emilia Fadini (Stradivarius: 33500, 1999), 1819. Even within this scheme she differentiates
between the relatively rigid instrumental and free vocal elements that follow an introduction.
122 The keyboard sonatas of Domenico Scarlatti
with the rhythm of the seguidillas, upon which dance form she believes the sonata is
Should the player put on such an accent? Should a performance in such circum-
stances be stylized, assimilated to a perception of the prevailing style of the composer
or his era, or should it be realistic? Naturally, this realism is itself a highly stylized
construct, leaning heavily on the inherited notion of Spanishness dened at the
outset of the whole discussion.
The most common reaction to this matter of performance practice would be to err
on the side of caution. But such histrionic exaggeration as we nd in Landowska and
Pletnev is arguably very much in style. Wouldnt a straight and sober performance
represent a lesser degree of taste? The same issue arises with Andreas Staiers version
of the Sonata in Amajor, K. 114. He gives an exceptionally ery performance, which
spills over into hysteria when, in the passage beginning at bar 34 and especially from
bar 144, he speeds the music up almost beyond belief.
Here once again some
will feel that Staier fails to keep the suggestions of amenco at a distance, that this
is too much of a good thing. At this point the well-worn notions of eighteenth-
century moderation and distance will come into play. These are all too evident in
the moderate character of many Scarlatti performances. One might also recall the
topical reserve that, it has been argued, denes the composers wider approach to
style, but this is an inherent property that hardly requires executive demonstration.
Indeed, such reserve would be positively misleading if translated into performance
it may deny styles their absolute claims but it does not deny them their vitality or
right to speak. The splendidly unreserved Staier in fact makes the vitality of K. 114
frightening rather than in any way picturesque. Perhaps he was heeding the advice
of Kirkpatrick, given when considering the Romantic inheritance that still denes
so many of our attitudes to music, post-authenticity mood notwithstanding:
The type casting of eighteenth-century music that was common in the last century was by no
means eliminated by twentieth-century restorers and enthusiasts. Rather they forced it into
an even tighter costume, into a kind of strait jacket created by the newer notion of a profound
and impassable gulf between eighteenth-century and romantic music. Consequent on the
rise of a sense of style, rose a conception of Stilechtheit that was often quite unsupported by
the historical researches with which it pretended to justify itself. Eighteenth-century music
was forced to be pure and abstract; humanity was permitted it only in the most limited
form. . . There is no nobler mission for a harpsichordist or for a player of Scarlatti than to
frighten such people to death!
Explained in Clark, Clark Notes, [5].
The performances reviewed in this section derive from the following recordings: EMI: 7 64934 2, 1949/1993
(Landowska); Virgin: 5 45123 2, 1995 (Pletnev); Stradivarius: 33500, 1999 (Fadini); Janiculum: JAN D204, 2000
(Clark); Teldec: 0630 12601 2, 1996 (Staier).
Kirkpatrick, Scarlatti, 280. Perhaps reecting such a tradition, Pestelli makes the strange comment that the
interpretation Scarlatti gives of folklore is free from the slightest vulgarity; Pestelli, Sonate, 193. Surely it is
precisely the sensation of rude vulgarity that is so novel in the composers incorporation of folk elements. In
any case, folk music itself never comes across as vulgar in an aesthetic sense vulgarity requires an aiming high
to be noticeable.
Heteroglossia 123
The topical plurality and ambiguity that characterize Scarlattis mixed style have
been read by Giorgio Pestelli as indicators of the composers theatrical vocation.
The sonatas overow with the animated life of the stage, they offer us a musi-
cal spectacle. In eshing this out, the author reminds us of the theatricality with
which eighteenth-century life was often conducted, its fondness for disguises and
This is an attractive metaphor for the sense of musical process found
in the sonatas. Similar imagery is found elsewhere in the literature: Sacheverell Sitwell
found the sonatas inhabited, alive with gures.
Clearly related to the panorama
tradition, such conceits have the advantage of stressing more clearly the agency of
the different types of musical material, that they are not simply held within a sort of
tableau. The effect of the mixture may, in other words, be dramatic.
An example of such inhabited music is K. 96 in D major (see Ex. 4.12), one of
the best known of the sonatas. In its wide range of imagery, it seems to aim for
a carnivalesque inclusion of the whole (musical) world. The second half enriches
this sense of generosity by containing a good deal of new material or old material
radically transformed compare the repeated-note mutandi i deti passages, for in-
stance, found from bars 33 and 145 respectively. The equivocation over mode at
the end of each half also strengthens the sense that we are in a world of bound-
less possibility, one that is both democratic and comic. Everything and everybody
have their part to play; all the worlds a stage. The overcoming of the minor in-
terpolations in each half could even be seen as symbolic of this comic viewpoint.
Even if some of the materials such as the fanfares might seem to be indebted to
French models, the sonata as a whole is very far indeed from the rather formal pro-
grammatic approach of the French keyboard composers. K. 96 is unthinkable in the
French tradition, or indeed any other tradition at all, given its directness, its worldly
A sonata such as K. 96 has an indubitably panoramic aspect which has then been
extrapolated, rather too easily, to the entire output of sonatas. In the majority of
cases the effect of such a mixture of material may be more disputatious or uncertain;
it may even, as J. Barrie Jones found, be unsatisfactory. This is where the theatrical
metaphor loses its force, unless it can be broadened to take account of the harder-
edged opposition of different topics and styles encountered in many works. At
this level it may seem less a case of conicting characters placed on one stage as
characters that inhabit different stages altogether. The outcome of such conict can
Pestelli, Sonate, 1956.
Sitwell, Background, 131, 135.
For an account of this type of patterning, deriving from the minor echo-repeat so familiar from the world of
the Italian concerto, see Talbot, Shifts, 314.
Several musicians have heard K. 96 within just one topical frame. B ulow calls it Gigue in his Achtzehn ausgew ahlte
Klavierst ucke (Leipzig: Peters, 1864), where it forms No. 6 of Suite No. 3, and Alfredo Casella arranged it for
small orchestra as the nal movement of his Toccata, Bourr ee et Gigue (Paris: Maurice Senart, 1933). Puyana
counts this as one of the many Portuguese fandangos amidst the sonatas; Puyana, Inuencias, 52.
124 The keyboard sonatas of Domenico Scarlatti
Ex. 3.12 K. 402 bars 1102
be difcult to gauge: does Scarlatti fuse opposites or narrate the impossibility of
their convergence?.
The Sonata in E minor, K. 402, opens in strict or learned style (see Ex. 3.12). The
crucial elements of this style were, according to the theorist Heinrich Koch in 1802:
This is how Kevin Korsyn encapsulates a comparable issue in J. W. N. Sullivan and the Heiliger Dankgesang:
Questions of Meaning in Late Beethoven, Beethoven Forum 2, ed. Christopher Reynolds (London: University
of Nebraska Press, 1993), 172.
Heteroglossia 125
Ex. 3.12 (cont.)
a serious conduct of the melody, using frequent stepwise progressions which do not
allow ornamentation and breaking-up of the melody into small fragments; frequent
use of bound dissonances (suspensions); and strict adherence to the main subject.
All of these elements obtain here, with suspensions being especially prominent. More
Cited in Ratner, Classic Music, 23.
126 The keyboard sonatas of Domenico Scarlatti
Ex. 3.12 (cont.)
than that, though, this is a topic old-fashioned even in the rst half of the eighteenth
century it is in the mould of the sixteenth-century vocal polyphony, with the same
antiphonal suggestions, that we found in K. 263. Note that the left-hand writing of
bars 7
mirrors that found in the right hand at bars 1
; less exactly, the left
hand at 5
follows the right hands 3
. This is a perfectly formed and highly
unied texture.
From bar 9 there is an immediate shift from the opening idiom. The music
continues to move in precise two-bar units, but the effect is very different. To
begin with, the Palestrina style cannot have this repeated-block syntax. In place
Heteroglossia 127
Ex. 3.12 (cont.)
of the long descending phrases we hear a breaking-up of the melody into small
fragments in the repeated melodic unit, while the left hand jumps between the thirds
ACE heard at bars 111315. For all the marked difference in stylistic premises,
the right hand uses two of the earlier basic shapes scalic descent in plain crotchets at
, then the neighbour-note motion towards a cadence (compare bars 1011
with 45
). However, these shapes are now treated insistently, unlike their previous
calm distribution. The texture becomes much more homophonic, with narrower
doublings (parallel thirds against the previous sixths), and the tessitura is drastically
compressed as all parts remain within the span of an octave. In addition, the very
presence of trills is a strong signier of change: remember that ornamentation should
not occur in the strict style. With its abrupt reharmonizations of the right-hand line,
the passage from bar 9 is also explicitly diatonic in its harmonic versatility after the
modal world evoked at the start. At bars 14
and 16
in the left hand we nd a
rhythmic hint that the opening has not entirely been subjugated. Bar 16 in fact
moves back towards genuine part-writing.
Then at bar 17 the opening tries to reassert itself. This is immediately apparent
in the reappearance of b
, the rst note of the piece and the rst suspension. It will
continue to act as an important reference point, both registrally and as a concise
reminder of the opening stylistic world. The composure of the opening is not
128 The keyboard sonatas of Domenico Scarlatti
regained, however. The imitations come sporadically; ascending lines cross against
falling ones (see the voice exchanges at 1920 and in bar 21). Note also the presence
of parallel thirds at 17 and 19
and the plain outlining of a tritone at 1921.
From bar 22 the music has clearly returned to the melodic stasis found from bar 9.
Signicantly, the density of trills increases (hinting at an oriental melodic style, as
might the


6 bass). The strict topic survives only in the tenor interjections of the
original 76 suspension cell, now heard very much as a remnant. This rst section
has an ABAB expressive-material structure that will also hold for the entire rst half.
A rest with a pause follows, the rst of many in K. 402.
If the strict topic was undermined within the rst section, then bars 26ff. blow
it away. We move to the most up-to-date style, the galant. For Koch in 1802, the
dening elements of the free, or unbound style were: many elaborations of the
melody, with more obvious breaks and pauses in it and more changes in rhythmic
elements; a less interwoven harmony; the fact that the remaining voices accompany.
The harmonic non sequitur emphasizes the stylistic leap: we move from a bare fth
FC, which could be heard either modally or as a dominant of B minor, to
D major. While this sounds abrupt, from a more abstract technical viewpoint it is
actually smooth: the omission of any A at 25 avoids a clash with the As at 26, and the
heard in the soprano can be heard retrospectively as the leading note of D. There
are other points of economy too: bar 26 begins with a falling triad just like bar 1,
while at 29 in the right hand we hear a reworking of the CBBC succession of
bars 245!
The differences are of course more to the point. The chain of falling steps in
bar 1 is replaced in 26 by a chain of falling leaps (in other words, an arpeggio); the
bass also leaps about, quite gratuitously, especially at 3031; we hear a homophonic
texture; minor-modal is replaced by the sociable major; the harmonic rhythm is
much slower, with all harmonies in root position until bar 34; there are very wide
gaps between the hands; and the chromatic appoggiatura at 27 is a real marker of the
galant. This appoggiatura is an echo, across the chasm, of the one we heard in bar 25,
but with the dissonance now approached by leap. The same happens with the dis-
sonant d
of bar 29. These unprepared dissonances display the modern style which
caused such theoretical anguish to the upholders of the old ways. In addition, the
asymmetry of detail within a symmetrical framework is very modern, a technique
found constantly in the later galant language of Mozart, for example.
Note how
the sequential repetition in bars 289 is not exact, with the melody of bar 29 being
a free decoration of that in bar 27. The two rising arpeggios in the bass at 30 and
Cited in Ratner, Classic Music, 23.
In his recording of K. 402, Andr as Schiff employs a heavy legato from bar 26, which seems odd stylistically.
This free and mixed style needs mixed articulation. In his second-time performance of the rst section he
adds ornaments at bars 6
and 20
; the inappropriateness of such additions will already be plain from the earlier
discussion of the strict style. This reects not, of course, a wrong performance but the difculties of stylistic
apprehension posed so often by this music. Decca: 421 422 2, 1989.
Heteroglossia 129
32 then balance the two falling ones heard before in the treble, another form of free
More striking is the extravagant leap at the end of each rising bass arpeggio from
down to D, as if to emphasize the freedom from bound style, the difference be-
tween modern instrumental and old vocal ways. Nothing could be more antithetical
to the language of the opening than this detail. Try singing that, the modern style
seems to demand. The new triplet gure at 31, with its chic decorative air, represents
one of those pronounced changes in rhythmic elements noted by Koch. It is per-
haps the obviously inorganic nature of such an element that has caused the negative
characterization of galant language as being full of articial formulas, without the
realization that such looseness was delivered in the name of freedom. The protracted
formulaic cadence at bars 367 then widens the stylistic gap still further. The wittiest
of all the oppositions, however, is half buried in this formula: the melodic gure
from bar 35
, with the same initial long-note syncopation, transforms the stepwise
descent from B to E heard at the start.
The subsequent pause is once more broken by completely new material and a
disorientating jump of a third. This D to B move is at once more shocking than the
previous jump and less so, because the new material itself enters less demonstratively
than did the D major arpeggios in bar 26. The B (A) will in turn move back to F,
thus rmly ensconcing the use of thirds-relations. This relationship was set up by the
shift from A to C to E in the bass at bars 1115; but more important than the connec-
tion of intervallic shape which is now writ large is the principle of harmonic exi-
bility that underpins this diatonic behaviour. It now contradicts the opening style on
the largest possible scale. We also nd ourselves a tritone away from the tonic.
Again, there is some voice-leading continuity across the void: the closure of the
second section on a unison D provides a smooth pivot to what follows. Like the
opening, this section begins on the second beat, while the parallel thirds provide a
textural reference to strategic points in the rst section. The answering unit more
explicitly revives earlier material compare the right-hand line at 40
with bars
or the whole of bar 41 with bar 8, to give the most obvious parallels. Altogether
this material seems to mediate between previous extremes. The sequential repetition
of the rst four bars up a step recalls the procedure heard in bars 269 of the second
section, while the suggestion of antiphony between the units revives the opening
texture. Yet the very alternation of style between phrase units in questionanswer
fashion is only possible in the modern manner.
A mini-vamp follows from bar 46 as a melting pot for the disparities presented
thus far. A suspension gure occurs four times from bar 46, vaguely echoing the
suspensions that characterized the opening learned style. Now, however, they are
restruck and move (incorrectly) upwards.
The exact counterpart of the gure
This feature is noted by Hermann Keller, who suggests (not in a schoolmasterly tone) that such voice-leading
misbehaviour hurts the ear; see Keller, Meister, 71.
130 The keyboard sonatas of Domenico Scarlatti
at 46
, however, is found at the very end of the rst section, in the BBC
succession across the bar at 245. The hint of the exotic found there is now more
openly realized, with the insistent repeated chords and the abandoned atmosphere
of the whole. From bar 55 there are clear echoes of the end to the rst section,
culminating in bar 58, with its pause, matching bar 25. There is, however, no
remnant syncopation in the tenor now. The music almost seems to have turned full
circle; we are back where we were before the rst rupture. Might this imply that
all the intervening material was a big interpolation, or, more extraordinarily in
view of the destruction of all precepts of good continuation, of stylistic and affective
integrity, that we have just witnessed that it was all redundant? On other hand, the
fact that we nd ourselves back in bar 25, so to speak, might suggest that D major
is about to recur.
In addition, the pauses have by now conditioned us to expect an ensuing surprise,
so it is doubly surprising when the same harmony is resumed after the gap. It is a
double bluff, one which also continues to hold back a viable alternative key area,
whether III or V. This section fromthe end of bar 58 again appears to have mediating
force, but now leans more openly on material from the second section, with the bass
arpeggios and melodic repeated notes (compare bar 30). Bars 624 contain multiple
echoes of the multiple material we have been confronted with so far:
1. The melodic peak on a syncopated two-beat b
in bar 62, followed by a descending
scale, recalls the rst section, bars 12 and 1719;
2. The syncopated rhythm with neighbour note in bar 62 alone may be compared
with bars 4, 41 and especially 78
3. The right hand in bar 63 reintroduces the previously anomalous triplet rhythm
of bar 31, now put in a directional rather than decorative context;
4. The contour of 624 as a melodic whole resembles bars 357, especially with the
initial second-beat syncopation on B and the following elaborate ornamentation;
5. The immediate cancellation of the leading note in the ABA line of bar 62
replicates at the dominant the DED of bars 89.
From bar 69 there is another descent from b
, eventually moving down a whole
octave. The outline of the falling triad from bar 1 can be recognized in bars 69 and
70. The left hand brings back the vamping middle-register crotchets (with more
textural thirds) from the previous section, emphasizing the


6 progression, FG.
There is also a consistent use of harmonic interruption, at bars 64, 69 and 72, when
each time the expectation of reaching a root-position dominant becomes stronger.
Such teasing harmonic detours are of a piece with the stylistic interruptions of
the discourse. The root position is nally granted at bar 75, which brings a more
conclusive assemblage of elements, seemingly in the name of nding a middle style.
We hear another descent fromb
down an octave; the triplets are nowintegrated into
the surface rhythm instead of representing sporadic outbursts; the thirds in the left
hand achieve direction. Above all we have harmonic security; until we reach bar 75,
Heteroglossia 131
the next best thing was found in the second section. This described a complete
rounded harmonic movement of D major early in an E minor work! This was
an illusory harmonic security. Given such harmonic and stylistic uncertainties, the
unison texture that articulates B minor from bar 77 makes a very decisive impression.
We then receive a rude surprise over the double bar into the second half B
to C is the largest-scale interrupted progression of the piece. Immediately at bars
823 the opening gambit from bars 12 is harmonized IIVV and thus brought
within the realm of contemporary style.
The bass sonority and note values recall
those of the modern second section, while our thirds intrude again at 83
. In
a stunning display of topical transformation, the opening unit is brought back ve
times successively from the start of the half, each time differently treated, as if to
purge it thoroughly of its original strict associations. The passage as a whole is of
course anything but strict, being keyed around a modern versatility, with several
changes of mood.
After the galant reworking of the opening at bars 823, bars 845 present a more
contrapuntal version. An exact transposition of bars 12 occurs in the left hand,
which also of course answers the right hand of the previous two bars. The upper voice
of 845 moves in contrary motion, as at bars 19 and 21, before disappearing into thin
air at the start of 86, a charming way of undercutting the return to counterpoint. The
third version is like a textural halfway house, with its chorale-style setting. At bar 89
of the fourth working, the expected dissonance wrought by a suspension is replaced
by a triple chordal dissonance. The fth version proceeds from the same basis but
reharmonizes the augmented second, expands in duration and reveals more clearly
than the fourth version a debt to the rst-half vamp rhythm in the left hand. The
insistent syncopated rhythms clearly derive from the same area. Thus the rst dozen
bars of the second half compress all the stylistic and textural possibilities presented so
disconcertingly in the rst half. At the same time, the consistent use of one piece of
material the opening two-bar unit as a pivot for the invention reveals a certain
debt to the precepts of the strict style. In bar 94 the voice-exchange pattern heard
most recently in bar 85 is nally put in a more stable harmonic context. From bar
95 a sixth form of the gambit, the same as heard at bars 88 and 90 (except that the
C is replaced by C), leads to a third harmonization, now much more consonant as
a simple dominant seventh. Signicantly, the effect of the suspension that we would
expect on the downbeat of bar 96 has now completely worn off.
The ending of this section in G major means that the return of the second section
in G (down a fth from its rst-half form) plays a different role. Instead of being
a harmonic shock, it gives us more of what we have just reached. Its harmonic
meaning also changes in that it has a straightforward harmonic relationship to the
starting key of the second half. The IV, C majorG major relationship is what we
might have expected to hear in the rst half. There is a fairly extensive rewriting of
bars 1014, which now have a more transitional character (note especially the exact
A similar transformation of a strict-style opening can be heard at the same point of K. 240.
132 The keyboard sonatas of Domenico Scarlatti
sequence created in 1034). Bars 1001 and 1023 in the right hand now integrate
the syncopated rhythm of the original learned cell quite explicitly. The dissonance is
prepared and resolved in more respectable fashion too, through a cambiata formation.
The section even begins on the same note, g
, as the start of the half. The arpeggios
themselves are no longer such a surprise after all the versions of the descending-
triadic Kopfmotiv in the previous section. Altogether this passage now forms a more
integrated part of the argument.
Subtle changes made in the version of the following section, from bar 113, also
suggest greater continuity. The left-hand material of bars 1313 comes straight from
bars 224, not from 557, as it ought. Thus the suspension gure in the tenor is
reintroduced. Crucially, there is no pause marked at bar 134. Even if so much had not
changed in the mean time, the device would anyway have exhausted its potential by
this stage. It also disappears because, with the changed form of bars 1313, Scarlatti
is in effect taking us directly from the equivalent of bar 25 to 58
, so cutting out our
big rst-half interpolation. The most signicant changes, though, are found in the
bass of bars 145 and 148, with their echo of the sustained surprise C that began the
second half. Thus even the constant interrupted progressions themselves are now
less jarring, since reference to C has been made a way of integrating the harmonic
action of the second half.
K. 402 as a whole drives towards greater coherence of its very disparate elements.
To speak of a comic variety of the surface seems inadequate to the scale of the
contrasts or better, ruptures presented to the listener in the rst half. The very
act of composition itself seems to be under scrutiny, with the sense that the pauses
represent a creative abandonment of the prior material, that the sonata begins several
times over in a new key and in a new style. After all, if such incompatible styles are
to be housed within a single work, one might expect a structure that contrived to
de-emphasize the awkwardness. Instead, the silences (which the performer might be
advised to make long and outside the basic pulse) and the harmonic shifts advertise
the fact. On the other hand, the very lack of smooth (re)transitions in the rst part
of this work may show a particular sophistication of technique, born from an under-
standing of the potential and relative compatibility of different materials. From this
point of view, the awkward silences and harmonic jumps represent correct syntax.
Much broad symmetry is then needed in the second half to act as a counterweight
to the disruptive force of the rst, but with many important adjustments at a micro
level reecting the changed signicance or weight of materials. At the end we ar-
guably have, as suggested earlier, a middle style it is certainly not especially modern.
Here and at rst hearing, from bar 75, this sounds like the recollection of a Baroque
concerto grosso idiom, in the manner of Corelli or Vivaldi: is this a middle way?
The structure and material of the opening sections might almost be conceived
as a reply to critics, ctional or actual. They could certainly be allied with the
quarrel of the ancients and moderns. The beginning might convey the message to
the ancients, So this is how you want me to write music? The composer then shows
how it does not and cannot work in the present day. We could even place this sonata
Heteroglossia 133
in a specically Spanish context of theoretical controversies, above all the dissonance
war unwittingly started by Francisco Valls in 1715 (to which we shall return). Of
course, the very intense working of all the basic material of the sonata, as explored
above, itself reveals learning, in the name of nding some common ground. The
contrasts turn out not to be as abandoned as they rst appear.
A number of sonatas raise such contrasts onto a more explicit structural plane.
In his chronological classication Pestelli brings together a group of six sonatas that
consist of a dialogo tra musica antica e moderna. One of these, the Sonata in E
major, K. 162, alternates Andante and Allegro sections. The Andantes are in an idyllic
pastoral vein. They offer a very polished and idealized naturalness Arcadian, in
other words. The Allegros, on the other hand, have a bustle about them and some
suggestions of string guration that prompt rmer comparisons with the world of
Vivaldi and Corelli.
For all the Italianate pedigree of the materials in this sonata,
the formal nature of their juxtaposition again suggests concerns apparent elsewhere in
the Spanish musical environment of the time. We nd a similar plan, for example,
in Alberos Sonata No. 22 in F minor. Here, an evocation of antico style in the
Adagio sections is followed by an exhilarating romp of modern guration in the
Vivo sections. The unusual formal plan, particularly in the way the rst B section
of the ABAB alternation straddles the double bar, is shared by K. 162.
The contest of ancient and modern is found on a larger scale in the six works
by Albero entitled Recercata, fuga y sonata. Powell has suggested that the titles imply
sixteenth-, seventeenth- and eighteenth-century inuences respectively,
and these
three-movement works do seem to represent three historically progressive styles:
ancient preludizing (recalling not just the genre of the title but also the French
unmeasured prelude), the contrapuntal tradition (issuing from the past if still alive in
the present) and the popular/galant world of the current time. Further suggesting a
conscious eclecticism are the Obras de organo entre el Antiguo y Moderno estilo by Elas
of 1749, for which Albero himself wrote the preface. This also obtains in the case
of the twelve piezas and toccatas found in the Montserrat collection entitled Obras
del Maestro Jos e Elas, and several of the piezas are quite explicit about their stylistic
allegiance: the indications en forma de aria and en forma de concierto are found
in the tenth and eleventh respectively.
It is very characteristic that, while Albero
and Elas make plain the nature of their stylistic project, Scarlatti does not spell out
such a plan. Although the contest of styles is built into the basic structure of the
work, K. 162 offers no title to help the player or listener. As ever, it contents itself
with the anonymity of Sonata.
That the composer was not inspired by external trappings, whether taking the
form of titles or an explicit formal alternation of styles, can be seen in most of the
other alternating sonatas, such as K. 170, 176, 265 and 351. They tend to be curiously
Compare the guration that closes the rst half of K. 162 with that of the closing ritornello in K. 265, bars
Powell, Albero, 16.

Agueda Pedrero-Encabo, Some Unpublished Works of Jos e Elas, in BoydCarreras, Spain, 21415.
134 The keyboard sonatas of Domenico Scarlatti
nondescript. Of greatest interest is the possibility of operatic inuence on such forms,
given the composers own habits in his early operas. Changes of tempo, dynamics and
affect are strikingly frequent, for example in Ptolemys aria Tiranni miei pensieri,
from the newly recovered Tolomeo et Alessandro. Boyd has made a telling comparison
with Handels setting of an adaptation of the same libretto in 1728; in a number of
arias the unied Affekt of Handel can be set against the contrasting of particular
phrases in Scarlatti.
One might also compare our alternating sonatas with some
of Scarlattis orchestraloperatic overtures as in the sharp uctuations of Sinfonias
Nos. 9 and 14.
Thus the theatrical metaphor for Scarlattis opposing topics and
styles may have some literal roots.
In fact the composer tends to achieve stronger effects not by alternation of this
sort, but through interruption. The Sonata in D major, K. 236, contains a seemingly
inexplicable interruption in the rst fteen bars of its second half. There can be
no doubt of its older vintage, with the very clear large-scale imitations and linear
intervallic patterns suggesting perhaps a toccata idiom.
On the other hand, the
rest of the sonatas material is not exactly without toccata-like properties of its own.
These form part of a typical assemblage dominated by the racy dance rhythms of
bars 2030. Perhaps the greatest surprise afforded by the interrupting material is
simply its continuous semiquaver rhythmic values, whereas the rest of the material
comprises virtually continuous quavers, apart from the very occasional semiquaver
cell. Although it disappears as mysteriously as it arrived, the passage does leave its
mark; in bars 579 the raw popular dance material is given in melodic sequence, a
stylistically unlikely treatment for which the rst half provides no precedent.
More disconcerting still is the Sonata in B at major, K. 202. The return to rst-
half material from bar 110 in the second half, after an interruption, is very eeting;
and what we hear subsequently is really a coda using new material in a markedly
broader, more popular style than the music of the rst half. Strictly, the literal return
lasts for just one bar. The left hand does not wait its turn to provide an imitative
answer, as it did at the start of the rst half, but interrupts the right hand by moving
to the third pitch of the original shape, E, in a cross between imitation, stretto and
hocket. That effectively does for the opening material before we move on to the
populist coda.
The rst half of K. 202 is effectively a blend of toccata, galant and popular. In
the light of subsequent events, it may be regarded as a civilized version of the mixed
style, without hard edges. The middle, interrupting section is in Italian popular style,
whether one describes it as a siciliana, as do Sitwell and Chambure, or a pastorale, as
do Pestelli and Boyd.
In length and force of expression it quite outweighs the outer
Boyd, Tolomeo, 1819.
These works are discussed in Boyd, Master, 8083.
Pestelli describes it as a sudden aring-up of the toccata which breaks the unity of the discourse, a renewal of
the toccatismo of Alessandro Scarlatti. Pestelli, Sonate, 76.
For other examples of interruptions, see K. 282, 414 and 511, all in D major.
Max Seifferts remark that the structure of the whole is reminiscent of an Alessandro Scarlatti overture form
seems cold comfort. Seiffert, Klaviermusik, 422.
Sitwell, Baroque, 288; Chambure, Catalogue, 83; Pestelli, Sonate, 2023; Boyd, Master, 172.
Heteroglossia 135
sections. Indeed, as we have seen, it seems to blast away the material of the rst half.
It also shifts harmonic ground constantly and disconcertingly. This is very ambitious
for a folk style; compare the much more realistically modest harmonic activity of the
interrupting pastorale in K. 235. Also striking are the clusters and rough chordings
and the relentless drive intoxicating monotony of the rhythmic construction.
Such features, it is plain, do not have to connote amenco idiom, suggesting again
that the gap between Italian and Spanish folk languages is often not as wide as we
might imagine. Such considerations seem even less urgent than usual, though, if we
think through the implications of the whole structure.
The harmonic abstruseness, which almost turns a straightforward pastoral idiom
into a vamp, and the very calculated registral plan, which helps build the tension
towards a climax at bar 85, both lie outside customary conceptions of folk art. The
popular musical imagery thus has an articial, even fantastic character. In spite of
the fact that this section is so patently an artistic product, its interrupting presence
in the context of the whole marks a distinct step outside, or back from, the world of
high art. After all, this interruption is so lengthy that it effectively constitutes the
main material of the sonata,
giving the whole structure a centrifugal force. In
the confrontation it implies between what Peter B ottinger calls the closed sphere
of art and its acoustical environment,
the contingent nature of musical high
art is revealed: whatever its pretensions to comprehensiveness (hence the civilized
variety of the rst half), it remains a dialect of the few. The rest of the world may
not be listening.
This is the most radical implication of the rupture in K. 202, of its linguistic
incompatibilities. That Scarlattis sonatas are situated in a world that may not be
listening is brilliantly grasped by Jos e Saramago in Baltasar and Blimunda. When in
the novel Scarlatti took to visiting Baltasar and Blimunda on the estate of the Duque
de Aveiro, where they worked on the passarola:
He did not always play the harpsichord, but when he did he sometimes urged them not to
interrupt their labors, the forge roaring in the background, the hammer clanging on the
anvil, the water boiling in the vat, so that the harpsichord could scarcely be heard above the
terrible din in the coach house. Meanwhile, the musician tranquilly composed his music as
if he were surrounded by the vast silence in outer space where he hoped to play one day.
The last sentence here has already been cited for its implications of futurism, but what
precedes this offers the ideal expression, not so much of the composers aesthetics,
To give this some statistical support, Andreas Staier, in his recording of the work (Deutsche Harmonia Mundi:
05472 77274 2, 1992), takes 105 over two playings of the rst half and 220 over a single playing of the
second half. Of this, the pastorale section takes 200 and the coda just 20. With a repeated playing of the
second half, the interruption takes up over two thirds of the total performance time. On the other hand, I
believe that Staier takes the pastorale too slowly (Boyd comments on the tendency to play eighteenth-century
pastorales too deliberately in Master, 172). It may begin in charming fashion, but the brutal development of
texture and insistence of the governing rhythm seem to demand a livelier pace to have their full effect. Thus
the total length of the interrupting passage might lessen, but it would still be disproportionate.
B ottinger, Ann aherungen, 80.
Saramago, Baltasar and Blimunda, trans. Giovanni Pontiero (London: Jonathan Cape, 1988), 161.
136 The keyboard sonatas of Domenico Scarlatti
but of his philosophy. Art music, or at least Scarlattis art music, can have no prior
claims over the stuff of everyday life.
The governing irony that allows the artistic product that is K. 202 to exist at all
is the very manufactured, articial nature of the naturalistic pastorale section. It can
only realize this philosophy by in fact simulating the stuff of everyday life. It is this
ironic knowledge that allows Scarlatti to compose his music so tranquilly in the
din; he knows that his music, while surrounding itself with real life, stands ultimately
apart from it.
Another extraordinary counterpart to this, also lying outside the realm of the
normal critical literature, can be found in David Thompsons BBC television pro-
gramme of 1985. The challenge for this medium lies in nding appropriate visual
imagery to accompany the playing of eighteen sonatas over the course of the pro-
gramme, when, that is, the pictures do not simply showthe performance of the works
by Rafael Puyana. On the occasion that interests us here, the music of K. 240
a mixed-style sonata with a predominance of popular avours is set to picture-
postcard images of the canals of Venice, well stocked with gondolas. At the point
where the sonata swerves into an exotic passage (bar 43), the picture changes sud-
denly and most disconcertingly. We nd ourselves in a workshop watching the
activities of the gondola builders sanding, hammering, planing and cleaning. In
other words, we are viewing the labour that puts the gondolas in the postcards. The
correspondence to the stylistic sense of many Scarlatti sonatas should be clear. The
world of high-art music is analogous to the picture postcard, a controlled presenta-
tion of nished imagery, while sonatas like K. 202, and indeed K. 240, with their
rough edges and abrupt changes of perspective, allow us to glimpse the existence of
another, foreign world. This world may help create the material for (Scarlattis) art,
but we would not expect it to be directly acknowledged or glimpsed in the raw.
The Sonata in C major, K. 513 (Ex. 3.13), consists of an even clearer version of
the ABCshape that was implicit in K. 202. This work has been seized upon gratefully
by all writers on the sonatas, since for once, in the rst two sections, we can be quite
certain as to the topical references. The opening section (A) is marked Pastorale, thus
issuing from the same stylistic source as the interrupting B section in K. 202.
theme of the second section (B) is an Italian Christmas carol, Discendi dalle stelle.
This more clearly offers the pastoral vein as found in many Christmas concertos,
with drones and parallel thirds imitating the pifferari (players of pipes or fes). The
nal, very different, section (C) seems to present a toccata style with populist accents,
but the dance impulse is also certainly present.
The harmonic scheme of K. 513 is most unusual all the real action takes place
in A. The B section is entirely in G major, while the C section is entirely in C major
(although avoiding an articulated root-position I until bar 62). The odd harmonic
practice thus reinforces the stylistic dislocations.
The additional marking Moderato, however, gives it a more leisurely aspect than most of Scarlattis versions
of the topic (K. 446, for example, is marked Allegrissimo).
Heteroglossia 137
Ex. 3.13 K. 513 bars 116
The A and B sections represent two faces of the same pastoral idiom: B is artless
where A is artful. The A section seems to offer a nostalgic view, but the material
is worked and made more affective than the reality (a property suggested in our
earlier examination of K. 87). It is only naive in the rst two and a half bars. These
are followed by an exact repetition of the material down a tone in B at major,
which immediately undercuts the simplicity. The leaping octave gure in the bass,
heard early on in bars 3, 5 and 8, is the same marker of pastoral style we saw in
K. 398 (Ex. 3.3b). Whereas it was playfully disengaged from its proper function
138 The keyboard sonatas of Domenico Scarlatti
there, in K. 513, as in works such as K. 270 and K. 446, it carries its normal rustic
What follows, however, contains many sour notes. The sudden exposed dom-
inant seventh of bar 6 seems an intrusion, emphasized further by the parallel 6/3
movements onto it. Yet it is also a logical dissonance, fusing the C major tonality of
the opening with the following B. This phrase unit stops abruptly, followed by a
dramatic silence; its repetition then makes one line out of the top two voices, thus
exposing the tritone. (The subsequent parallel fths in the top two parts of 8
be a characteristic reference to rustic technique.) The reworked sequential repetition
of this three-bar unit, beginning on the nal quaver of bar 8, is more anguished,
with the succession of perfect and diminished fths heard in the right hand. Again,
this would appear to originate in the common technique of affectionate parody of
rustic players. If so, by sounding so harsh, it transcends this. The same could be said
of the howling f
at 10
, which might represent being out of tune. Our opening
idyll is now a distant memory.
More artice is apparent in bar 15, where we nd a wonderful overlap in the phrase
structure; 15
ought, like 14
, to represent the nal beat of a one-bar unit, but it also
functions as the downbeat of its own one-bar unit. This is conrmed by the parallel
one-bar unit beginning at 16
. At last here we reach the dominant, in conjunction
with a return to the initial texture and idiom: the purity of representation of the
opening is thus reasserted.
This has been a very convoluted mode of reaching the dominant; with the attain-
ment of the goal, it prolongs itself very sturdily, but by means of quite new material.
The A section has strayed from the authentic utterance promised by the generic
title, through its artistic perspective on the pastoral material. B clears the air, gives
us the real thing; it creates a sudden sense of stylistic perspective. After the highly
strung core of A, it sprawls crudely and riotously. For all the greater realism of B,
the fade-out heard at the end, at bars 345, is certainly more arty than folksy. (It is
realized precisely in this sense in the fourth movement of Casellas Scarlattiana.) The
same three right-hand notes that effect the fade-out are reactivated on the return to
a repeated A section (with f
becoming f []
); this linkage technique is also plainly
artistic. It helps to create the striking effect on return to A, which now sounds
like an apparition. It becomes even more comprehensively framed than it already
implicitly was.
The start of C parodies the start of B; compare the pitch contours of the top
parts at bars 3638
and 17
. More generally, the parallel intervals seem to guy
those found in B. This all feels more like a coda than a second half. We had the
same sensation with the nal part of K. 202. Does this represent the modern or the
composers personal keyboard style; is it a distinct new stage in the argument or
more of a dismissive gesture? For Wilfred Mellers A and B are uproariously routed
by a whirlwind presto coda. He adds: Whats to come is still (very) unsure.
Mellers, Orpheus, 86.
Heteroglossia 139
K. 513 is certainly affectively open-ended. It would seem to present a narrative
Scarlatti throws a challenge to the listener to make sense of the story.
In recent times the conventional assumption that non-vocal music can tell some
sort of story has been subjected to intense scrutiny. Precisely in what sense can a
narrative voice be conceptualized in instrumental music and how can the distancing
from events essential to the act of narration possibly operate? The consensus that
only under special conditions can such musical narrativity exist has in turn been
queried, for example by Robert Hatten, who suggests that shifting the level of
discourse may not be enough to create literal narration, but it achieves one of the
characteristic aims (or consequences) of narrative literature that of putting a spin
on the presentation of events.
Such shifting is very clearly delineated in K. 513.
He also invokes Bakhtins concept of the polyphonic novel, in which characters
interact with the narrating voice to the extent that the narrator becomes a plurality
of centres of consciousness irreducible to a common denominator. Such interaction
of centrifugal stylistic forces, together with the overt signalling of the presence of a
narrator (the controlling composer) by means of arty devices, is also found in K.
513 and to a greater or lesser extent in all those Scarlatti sonatas that live by self-
conscious topical manipulation. The fade-out at the end of B, for example, clearly
creates a distancing effect.
Further, Hatten explicitly links the heteroglossia of
Bakhtin, the play of styles and language types in literature, with possible musical
equivalents: extreme contrasts in style or topic (especially those involving a change in
register), cueing of recitative as a topic, direct quotations, disruption of the temporal
norm can all enable the composer to present different perspectives in the music.
Three of these four possible conditions are met by the current sonata.
Mellerss interpretation of the story is that it might be said to [be] about the end
of the old world.
It certainly suggests some disintegration of a unitary experience
of the (musical) world. If this is an elaborate way of suggesting a post-Baroque
orientation that was hardly unique to Scarlatti, it is certain that Scarlatti pursued the
consequences and implications of a mixed style further than any other composer
of the time.
That this newly uncovered variety may be confusing as much as
liberating is apparent in the conundrums presented by K. 202 and K. 513.
Thus far our investigation of topical mixture has not touched on its most common
formin the sonatas outright topical opposition within a single integrated structure.
It is often difcult to determine the outcome of such oppositions. Of course the
mixed style as a whole is premised on a coexistence of its elements, but, as was made
On Narrativity in Music: Expressive Genres and Levels of Discourse in Beethoven, Indiana Theory Review 12
(1991), 76.
For Massimo Bogianckino, this morendo connotes a sorrowful fading out of the memory. Bogianckino,
Harpsichord, 110.
On Narrativity in Music, 95.
Mellers, Orpheus, 86.
Both Clark and Pestelli believe K. 513 to have been written early in the composers career. See Clark, Enemy,
545, and Pestelli, Sonate, 2034. Both writers unnecessarily assume that there must be a close temporal rela-
tionship between inspiration and composition, as if a composer of all people would not be able to retain or
remember material well beyond the time of rst acquaintance with it.
140 The keyboard sonatas of Domenico Scarlatti
clear during the study of the panorama tradition, it is inadequate simply to extend
such a principle of tolerance to the nature of the individual work. The outcomes
may suggest a fusion of elements (centripetal) or a separation of them (centrifugal);
the contest may produce a victor or at least a sense of progression fromone element to
another. In K. 256, for example, the dotted style that is prominent in the rst half has
to give way to the galant; in K. 434, the contrapuntal manner of the opening, while
never entirely abandoned, is overwhelmed in the second half by dramatic melodic
and textural developments. These works remind us that many of the styles and topics
juxtaposed by Scarlatti would normally be treated autonomously. This is certainly
the case with both the dotted style and imitative counterpoint, which we would
normally expect to exist without contradiction in entire sections or movements.
Such examples remind us not to be complacent about the achievement of topical
One particularly interesting phenomenon among the sonatas that seem cen-
tripetally inclined is illustrated by the Sonata in F minor, K. 386. The toccata is
surely the basic premise, but some of the syntax and inections suggest Spain and
the dance. Perhaps we need to apply the term fusion in its current popular musical
sense to understand the creative results fusion rather than the very frequent jux-
taposition. The second subject from bar 32 is clearly Spanish in its harmonic and
pitch contours but does not break the decorum of the toccata style. The left hands
falling thirds and the right hands d
succession t with earlier shapes. Does
this suggest that the exuberance of toccata and of amenco are the same thing, that
they represent the same human impulse? The physical and emotional exhibitionism
that they respectively represent mix very naturally here, in the name of extravagant
display. Both require many notes in their expression, the toccata by denition so,
but amenco does as well. As with other such sonatas, like K. 29, 48, 50 and 545,
there is here a dissolving rather than contrasting of topical categories: is this a way
of adding a passionate edge to the basic keyboard genre of the toccata? The genre
undergoes expressive renewal through this mixture, in best traditions of Verfremdung
Many other types and degrees of fusion are represented. The Sonata in G minor,
K. 476, offers a bracing mixture of Iberian dance and Baroque idioms. The two
often seem to go together, sharing a propulsive power that favours heavy and regular
accentuation. This is quite unlike the variety of weight within beats and bars and
phrases found in the modern style. K. 476 contains one of the most memorable
realizations of a common syntactical device in the sonatas: a three-part sequence
that involves the wholesale transposition, generally upwards, or reharmonization of
a phrase, often made dramatic by the use of silence around each of the units. In view
of the element of bluff that is frequently involved, as well as the sense that we are
K. 50 has the distinction of being found in a Portuguese copy in the Biblioteca Nacional in Lisbon, Ms. Mus.
338, entitled Sonatas para Cravo do Sr. Francisco Xavier Baptista, but without Scarlattis name being given. Does
this suggest it was a Portuguese work? See Kastner, Repensando, 149.
Heteroglossia 141
witnessing a performance by the composer, we shall be calling it the three-card
trick. An underlying coherence is provided by a circle of fths from bar 96. If this,
like the sequential organization, seems a standard linguistic feature of the time, the
manner of presentation suggests an Iberian inuence, which might be conrmed
by the stylistic basis of this sonata. It seems to be an example of the bien parado, that
moment in the dance when the participants freeze in their positions.
The Sonata in G major, K. 337, is another work assembling different styles that
share exhibitionist elements. First we hear a toccata which also has touches of violin-
ismo; then in bar 18 we have a perfect example of what Pagano terms the eruption
of another world.
This very amboyant amenco material (almost exactly par-
alleled at the start of the second half of K. 324) takes ones breath away. Here is an
example of a passage that surely does call for some slowing and exibility of tempo
it is difcult to assimilate the material with the rhythm and pacing of the rest. From
bar 23 we hear what is more obviously violin writing like a solo passage from a
concerto. The plunging arpeggios of bars 257 reect bars 59 and 1216. They are
certainly more idiomatic for the violin than the keyboard at this later stage, but the
more natural keyboard equivalent from bar 5 reinforces Pestellis argument that much
keyboard toccata guration was originally translated from violin technique.
From bar 34 we return to more folk-like material, but now with an Italian accent.
However, the closing cadential shape at 37 and then at 413 strongly resembles bars
19 and 21 of the amenco material. When it occurs at the end of the half, it is also a
typical Baroque bit of guration, another example of our Essercizi-type cadence. The
composer seems to be delighting in nding similar turns of phrase in incompatible
idioms styles are being united in a higher cause. What they have in common is
their public face. The very unusual full chords at the end of the half and the end of
the piece seem to renew the suggestions of an orchestral-concerto idiom.
The closing material from bar 34 is expanded in the second half, at the expense of
the string-crossing passage. It is heard rst in E minor, the mode seemingly at odds
with its populist character. Signicantly, towards the end of the passage it mutates into
something that derives clearly from the world of high art; after the simple popular
IV alternations, a 710 linear intervallic pattern sets in at bar 77. However, the
pattern is broken after a bar and a half; as in so many other sonatas, Scarlatti denies
the pattern its natural completion, which would require at least another bar and a
half. The popular character of this material is then strongly reafrmed by the rather
rustic decorations in the right hand once the material reaches the tonic. Although
the two styles are thus sharply differentiated, there is the suggestion that the two have
something in common. The high-art sequence emerges unprompted, as it were, in
a context of popular repetition. The common ground is a desire for and joy in
patterning and reiteration.
One nal case study presents the more abrasive side of topical opposition. K. 99
in C minor is a very clear case where the Spanish idiom does battle with a higher,
Pagano, Vite, 448 (queste irruzioni di altri mondi).
See Pestelli, Toccata, especially 279 and 281.
142 The keyboard sonatas of Domenico Scarlatti
international language what I have generally been calling the Baroque. What is
unusual in K. 99 is that the Spanish idiom unequivocally opens the work and also
frames it at the end of each half.
This opening material (a fandango?) contains a tension within itself, though, the
sort of harmonic tension apparent to one listening with tonal expectations. The
apparent tonic C minor is weakly articulated, and in fact G, which sounds like it
ought to be a dominant, seems to be the tonic. The combination of placement
within the bar and melodic contour stress the pivotal role of G. Note how the
melodic line, perhaps an attempt to reproduce cante jondo, moves from G up to D,
the latter emphasized by the preceding ornamentation. Bar 4 then has a stronger
double meaning: it represents a point of repose or resolution as it clears away the C
minor harmonies, but in the light of bar 5 it is also heard as dominant preparation.
The sense of C minor as I in the following bars is still equivocal, though. Note the


6 bass at bars 56, which will be more fully exploited from bar 31 to
bar 37. The AA false relations sound very exotically modal, and the nal arrival
on I in bar 8 is far from conclusive. The C is not supported by other members of
the triad; instead, the bass line rather fades away through downward octave coupling
and the pause clearly represents a question mark. Structurally this may be a cadence,
but rhetorically it is anything but.
It is quite logical that what follows is a sweeping C minor arpeggio an attempt
to assert tonal authority, and this is supported by a change of style that sets in rmly
from bar 13 with a descending Baroque sequence. At bar 26 we are still in C minor,
which makes sense in the harmonic context described above in other circumstances
it would be a remarkable disproportion. At bar 25 we have not so much an elision
as an interruption, with the sudden entrance of a new melodic style and repeated
chords in the left hand, and the material arguably acts as a transition in stylistic terms.
Nevertheless, bar 26 still sounds like a further, and more dramatic interruption, by
material that is passionate and histrionic, of classic Spanish formation. Note the exotic
effect of the appoggiatura at 27
and the accumulation of sound in the left hand by
means of clusters. Driving the point home, the contour of the right hand, especially
with this nal appoggiatura, resembles that at bars 56
. The rising third AB-C
at 27
then suggests the shape of 4
and 6
, so that the sense of a variant on the
earlier phrase unit is even more complete. The repetition of the phrase at 27
then represents a syntactical parallel to the earlier passage. However, the exibility
of syntax from bar 26 is worthy of remark; we basically hear three versions of the
unit, but only the middle one is complete. The rst lacks a beginning (although we
only hear this in retrospect, of course) and the third lacks an end. This is a common
technique in Scarlatti, and one we should not take for granted. An absolutely straight
series of repetitions of a phrase unit without some fudging of the edges is quite
The third unit is interrupted by bar 31; even though the voice leading from30 into
31 is passably smooth, there is another abrupt change of texture. Bars 31ff. could be
regarded almost as neutral ground in terms of style and keyboard writing, although
Heteroglossia 143
they still favour the Spanish. The bass line hovers around Din modal manner, picking
up on the


6 shape from the opening unit, bars 56, while the soprano varies
the up-and-down stepwise melodic motion of the previous section. On the other
hand, the total right-hand part carries a suggestion of cross-string writing, while
the left-hand leaps to the top of the texture revive the cross-hands writing of the
sequence from bar 13. The total texture is more aerated and stratied its more
formal conception also suggests the Baroque manner of before. Then, unusually,
bars 3
ff. return from 36
ff., as the Spanish material reasserts itself very directly. The
changes to the upper voice in bars 39 and 41, compared with the model, bring the
modal mixture fully into the melodic line itself.
The opening to the second half rewrites the opening to the rst half: the right-
hand material is essentially the same but with more ourishes, while the left hand
rather makes the point of the original by being anchored in G throughout. This
truer revelation of the openings harmonic nature is now of course possible in the
new harmonic context, following the cadence in G at the end of the rst half. From
bar 48 the most overtly Spanish material of the rst half (26ff.) is translated into, or
appropriated by, the international terms. It is treated in simple descending sequence,
thus taking on the syntactical character of the material played from bar 13 in the rst
half. The texture is again more aerated and stratied the clusters have gone, and
there is a comfortable gap between the hands. This now leads, more smoothly than
at the equivalent point in the rst half, to the neutral material at bar 52, but this
in turn has been clearly captured by the world of diatonic normality. The passage is
now in the major, III (the rst structural use of the major mode in the piece), the
stepwise movements of the original are replaced by VI successions in the bass and
triadic outlines in the right hand, and the upper-register material in the left hand
now occupies third as well as second beats. With this two-crotchet rhythm and the
outlining of a third from second to third beats, it recalls the left-hand upper-register
shapes at bars 13, 15, 17 and 19. The sequential shift upwards from bar 57 is also
telling harmonic progression replaces the inarticulate hovering of bars 317.
From bar 64 yet another abrupt change occurs, back to swooningly Spanish mate-
rial. This revives the music of bars 26ff., but now in plain quavers the broken-sixth
semiquavers have since been appropriated by the Baroque idiom at 4851. The more
neutral passage returns from bar 68 with its function as a melting pot claried, but
just when we might expect the return of the closing/opening material, at bar 75,
there is a dramatic intervention by the material from bars 13ff. This is now more
boldly shaped with its chain of falling thirds, but it leads to a pause and a rest in bar
80 that have a similar character to bar 8 a sense of impasse.
The closing material returns in the tonic, but one might say the sonata ends with
a sense of stalemate. There is neither strong harmonic resolution nor rhetorical reso-
lution. Harmonically the opening uncertainties return, and the lengthy preparation
of V (modal I) from bar 64 until bar 80 is met by a single tonic perfect cadence in the
last two bars. (The root-position tonics reached in bars 83 and 85 do not complete
their preceding V6/5 harmonies; they represent a backing-up to the beginning of
144 The keyboard sonatas of Domenico Scarlatti
the phrase.) The harmonic and rhetorical aspects are of course intimately connected,
since the two types of harmonic behaviour derive from the two different stylistic
worlds, which appear to be centrifugally incompatible. One should not imagine that
the Baroque idiom in some way holds back or interferes with the true expression
of the Spanish one; both are extrovert in their different ways, and in terms of gener-
ating momentum and incident they make a great team, but any closed structure in
a diatonic art-music context demands a satisfactory articulation of a primary tonal
area, and this does not happen here. One might say that the V 1749 indication to
move straight on to K. 100 (volti subito) represents a natural consequence of the
unresolved tension of K. 99.
If so, must it be this particular sonata? P does not link the two, and does not in fact
pair K. 99 at all. The V II version of the sonata precedes it by K. 139 in C minor.
Might Scarlatti have written a sonata that seems to demand a sequel, preferably in
a clear tonic major like that of K. 100, without prescribing or deciding which one
it must be? If not, we need to consider the composers sense of an ending. There
are certainly many other sonatas which do not conclude very conclusively (K. 277
from Chapter 1 was an example, and try sonatas like K. 416 or K. 132). There are so
many sonatas that do seem to end with a thorough sense of resolution, though, that
one cannot claim that such a structural dynamic is anachronistic when applied to
Scarlatti. We have seen how K. 193, for example, decisively embraces the diatonic.
What allows such a profusion of voices to enter the Scarlatti sonata? And allows
them to interact in such an extraordinary way? Leonard Meyer, in considering the
question of what makes composers (such as Scarlatti) innovators, seeks an inherent
artistic explanation:
Three interrelated personality traits seem to favor the use of innovative procedures and
relationships: (1) a distaste and disdain for whatever is highly predictable or is sanctied by
custom; (2) a complementary propensity to delight in conjoining seemingly disparate and
discrepant realms or in turning things topsy-turvy by, say, making old means serve new ends
(perhaps in order to mock custom); (3) an ability to tolerate ambiguity a necessary condition
for the actualization of either of the rst two tendencies. The ability to tolerate ambiguity is
important because it enables the artist to take time to invent and consider more alternatives,
and in doing so to nd more satisfactory ones than might otherwise have been chosen.
These three elements have all been amply demonstrated in our consideration of
Scarlattis creative personality thus far. The ability to tolerate ambiguity will be-
come even more apparent as we turn in the next chapter to an examination of the
composers syntactical style.
Meyer, Style and Music, 139.
What are we to make of a tonal language that appears to privilege rhythm over
harmony? In the keyboard sonatas of Scarlatti the exploration of rhythm or, more
broadly understood, the exploration of syntax would seemto take priority over har-
monic considerations as such. The identikit image of a Scarlatti sonata would involve
generous reiterations of short phrase units against a relatively lightweight harmonic
background, but a general impression of animation does not amount to the privi-
leging of rhythm one might claim for the composer. Rather, it is simply a part of a
larger campaign in which all elements of normative syntactical patterning are open to
investigation. Inevitably, these will turn around the matter of degrees of repetition.
Repetition at some level or other is of course an essential precondition for the
existence of music, for it to be recognized as constituting an artistic statement. In
Western art music we can account for it most comfortably when it fulls certain
roles or ts with certain models. For instance, it may be present in the name of a
larger symmetrical whole: thus an antecedent phrase is matched by a consequent to
make up the larger unit known as a period; an immediate repetition of a shorter unit
followed by an elaboration of the same constitutes a sentence; on a higher level larger
sections can be repeated to give us ABA form or rondo form. Such repetitions occur
in the name of structural comprehension, and all live by the basic duality of departure
and return. They lend hard edges to our listening experience; they guide us through
a process that is potentially less clearly focused and less immediately meaningful than
our encounters with other forms of artistic expression, where words and images
provide a more concrete starting point. On a lower level, repetitions may be used
both to create and to dispel tension; for instance, they may abound in a transition or
development section, promising a stability that will coincide with their disappear-
ance. On the other hand, repetition in codas aids a different type of articulation, but
one which is again the corollary of a primarily harmonic argument. In this instance
the repetitions imply the forced exclusion of alternative material different keys
or themes or textures and so strengthen a sense of closure. Such repetitions on
these lower levels generally exceed what we might call natural limits and so tend
This section is based on a paper given at the University of Surrey in October 1997 and subsequently at the
University of Cambridge.
146 The keyboard sonatas of Domenico Scarlatti
to draw attention to themselves. Nevertheless, this represents a well understood
rhetorical strategy the purpose of the insistence quickly becomes evident.
But what if repetition is unpredictable or seems out of all proportion, in other
words when its functional basis is unclear? The Scarlatti sonatas offer a wide range
of such non-functional moments. Seemingly excessive, unmotivated repetitions are
common, passages that test our tolerance levels and seem to rend large holes in the
musical fabric. Most frequently such repetitions are as direct and literal as can be; it is
worth noting that variation, in the sense of the immediate varied treatment of a short
musical unit, is largely foreign to Scarlatti.
Further, its large-scale manifestation,
variation form, is found just once among all the sonatas (K. 61). Variety of detail
tends to be found within rather than between units. Thus the smallest cells may be
subject to continuous changes of exact shape, but at the level of the phrase Scarlatti
is unlikely to provide the sort of varied repetition that was second nature to Mozart,
for instance. The exact repetitions we are faced with, at the level of the phrase unit,
may well occasion embarrassment on the part of a performer or writer. One strategy
for deecting this, the use of echo effects in performance, must be viewed with
suspicion, since it goes against the grain of Scarlattis style.
This style itself goes
against the grain of the level and type of repetition with which we feel comfortable:
insistence seems to count for more than minding ones musical manners.
These two characteristics or principles can hold good on a larger scale as well.
Repetitions are there when we dont expect them and absent when we do; they are
both lacking and excessive. One particular manifestation of the taste for excessive
repetition has even, as we have seen, earned its own label. In the vamps, one
cell, normally without any evident thematic relevance to the rest of the work, is
repeated ad nauseam against a changing and highly elusive harmonic background.
If this feature is quite well known, there are many other syntactical peculiarities
that are less widely acknowledged missing bars, whose absence tends to destroy
our sense of hypermetre, missing bass notes, whose absence tends to destroy our
sense of phrase, phrase elisions and overlaps, which may even occur between the
two halves of the entire piece, so undercutting the structural cadence at the end
of the rst half. In short, Scarlatti will do anything to undermine a normal sense
of patterning. Surprising irregularities and surprising regularities together suggest a
thorough questioning of syntactical models, yet all these features have not earned
Scarlatti the reputation for technical wizardry that a study of the works suggests he
deserves. He is allowed to be a technical wizard of another kind, but that is not what
is meant here. Scarlattis rhythmic and syntactical virtuosity have been undervalued
or not even acknowledged because our training leads us to value harmonic range
over a rhythmic one.
This at least is the conclusion one must draw from the evidence of the sources. The question of possible ex-
temporized variation and embellishment has been discussed at the level of the phrase by Boyd, Ross Review,
273, and at the level of repeated playings of entire halves by Sheveloff, Frustrations II, 1036. The addition of
individual ornaments, as it were spontaneously, was of course a possibility for any keyboard music of the time,
but this will not necessarily have the larger implications that are currently under discussion.
See Rosen, Classical, 623, as one example of many warnings against this practice.
Syntax 147
Both theoretically and compositionally, it would seem, harmony has been regarded
as the real motor of tonal music. A wide harmonic vocabulary is almost always to
be admired. Harmonic exploration is cognate with depth and mastery; rhythmic
exploration, including in its widest sense syntactical exploration, is more likely to be
regarded as an optional extra. It may be felt as quirky, offbeat, a special effect rather
than something that is intrinsically substantial or necessary. Thus a simple harmonic
vocabulary is more likely to draw comment than a simple syntactical vocabulary.
Simple harmonies may need to be rescued by some special appeal, leaning on the
text or notions of affecting simplicity, for instance, whereas four-square syntax may
well not even be perceived as a problem. In the classroom chorales are worked in
the name of good voice leading and of harmonic range; training in rhythmic and
syntactical skills, in order to acquire versatility in these areas, barely exists as such.
Ear tests concentrate overwhelmingly on ne differentiations of pitch rather than
To put this more abstractly, our cultural and theoretical training means that we are
better at dealing with progression than with proportion when it comes to the way
music moves. As if plugging the gap, Scarlattis most conspicuous efforts are directed
towards investigating proportions. If we are undersensitized to such matters, then it is
all too easy to assume an irrational basis for the consequent musical behaviour in the
Scarlatti sonatas. Notions of his geographical distance fromthe European mainstream
help too in simply making the composer a wild man of the Iberian peninsula. While
irrationality is a real presence in many of the syntactical oddities of the sonatas,
this presence is rationally conceived. Its effects are understood and calculated, even
if the results remain startling or unbalanced. Often we seem to witness a battle
between untutored physical impulse and the syntactical habits of art music, the
physical side invading and exposing the artice that surrounds it. This arises naturally
from the sort of topical manipulation examined in Chapter 3, although it is not
simply to be correlated with a perceived opposition between the popular and the
artistic, an opposition which we have seen is frequently compromised as well as
afrmed. Through this battle, as well as through all his other rhythmic and syntactical
peculiarities, Scarlatti makes us aware of the contingent nature of musical time.
A concise example of how such issues may be raised is found in the Sonata in F
major, K. 554. The opening idea (see Ex. 4.1a) consists of a chain of thirds from C
to C.
The latter part of this idea is expressed in rhythmic diminution, as if throwing
the idea away, and throw away is exactly what Scarlatti does with it. This arresting
opening sinks without trace. It must leave the listener with a sense of dissatisfaction
that something so characteristic should fail to return. That the chain of thirds could
have an indirect motivic inuence on later material is not to the point; it may have
an organic connection to subsequent events, but rhetorically there is no counterpart
at all. A very convenient point of comparison is what Handel does with the same
idea in the same key, in the nal movement of his Concerto, Op. 6 No. 2 (see
Ex. 4.2). This also falls a notional two octaves fromCto C, with a similar acceleration
towards the end. It constitutes a fugal subject whose many, inevitable, structural
148 The keyboard sonatas of Domenico Scarlatti
Ex. 4.1a K. 554 bars 15
Ex. 4.1b K. 554 bars 4657
returns form the exact syntactical opposite to Scarlattis neglect of his subject.
is pretty much an unwritten law of all Western composition one of those rules
of good continuation that the most characteristic feature, that which stands out
most clearly against a background of the familiar, should be reiterated, investigated
or developed. Handel takes his fresh invention and uses it to prove his craft, by
showing the capacity to integrate it into a musical argument. From this perspective,
Scarlattis procedure represents not so much a lack of craft as a deliberate refusal
to take up the expected challenge. Instead the challenge is of a different nature
it is to us as listeners, when faced, not here with unexpected repetition, but with
the unexpected absence of repetition. The failure of this opening to return simply
projects the unexpected absence onto a larger syntactical unit the entire piece.
This corresponds to a fugal theme type that Warren Kirkendale associates with the Rococo; it uses three descend-
ing thirds in succession as the repetend of a sequence. This might in turn suggest that Scarlattis unaccompanied
rst bar makes as if to evoke this theme type before throwing it away. Fugue and Fugato in Rococo and Classical
Chamber Music, revised and expanded second edn, trans. Margaret Bent and the author (Durham, NC: Duke
University Press, 1979), 98.
Syntax 149
Ex. 4.2 Handel: Concerto Op. 6 No. 2/iv bars 113
The second half of K. 554 also features something highly unusual and, leaving
aside the application of repeat marks, unrepeated, from bar 49 (see Ex. 4.1b). This
of course is some sort of episode rather than something that announces itself as
potentially thematic and form-determining, as we heard at the start of the piece.
What it has in common with that opening, though, is that it is an enticing pattern
that fails to nd any clear resonance elsewhere in the structure. It too stands as
an isolated sonorous object. After it has also disappeared without trace, the rest
150 The keyboard sonatas of Domenico Scarlatti
of the second half presents us with as much repetition as we could possibly want,
a literal transposition of the last twenty-seven bars of the rst half. And so these
two passages act as no more than irritants to the larger structure, which otherwise
proceeds as if nothing were amiss. Affectively, though, the balance is rather different.
The second passage in particular is enormously memorable in its sinuous sequential
movement. It is an example of what we might dub Scarlatti jazz, meaning that any
possible external inspiration seems to count for little; it seems rather to represent
the identifying personal manner of the player-composer. Inspiration instead seems
applicable in another sense the composer is visited by a single brilliant idea that can
only be properly captured at one moment in time. Against the plentiful repetitions
of the rest of the music, both immediate and rhyming between the halves, our two
unique passages give a sense of the here and now, of a sort of musical living for the
moment. It is as though they exist in real time as against the composed time of the
rest of the sonata.
Another concise example of a sonata where single events seemto inhabit a different
world is K. 525, also in F major. Writing in 1927, Gian Francesco Malipiero pointed
out the similarity of K. 525 to the Scherzo of Beethovens Seventh Symphony.
a comparison may easily be deconstructed as an attempt to add lustre to the Scarlatti
work, to lend it prestige by association; some other instances of this were noted in
Chapter 2. Nevertheless, even aside from the obvious kinship of material, there is a
remarkable kinship of spirit. The scherzo-like quality of K. 525 (perhaps attested to
by B ulows renaming of it as such in his arrangement
) reminds us that many of the
Scarlatti sonatas may be protably, if seemingly anachronistically, thought of in this
light. After all, the scherzo is one tonal genre where we do expect rhythmic handling
to occupy centre stage (in the case of Mendelssohn, for example, the frequent very
soft dynamics encourage us to concentrate on pure pulsation).
In one respect, however, this sonata does not t with our maverick syntactical
prole. Like many another work, K. 525 begins by means of imitation between the
hands, but whereas most of these sonatas abandon the imitation almost immedi-
ately, in another example of opening premises that are not carried through, K. 525
pursues the idea. The opening material governs the whole piece, very much in the
economical mode we associate with the later scherzo. Bars 9ff., for instance, are in
many more than the two or so notated voices we hear a piling up of entries in the
manner of a stretto. We are presented with a modern, racy contrapuntal texture. The
repetitive syntax that ensues throughout the sonata is not to be construed as in any
way exceptional in its own right; it is no syntactical aberration, but a logical conse-
quence of the textural mode adopted. However, the huge chords that occur shortly
after the stretto (bars 20, 22 and so forth; see Ex. 5.6a) provide a gesture that kills any
Malipiero, Scarlatti, 480.
It may be found as the nal, sixth piece in Suite No. 2 of Achtzehn ausgew ahlte Klavierst ucke, in Form von Suiten
gruppiert (Leipzig: Peters, 1864).
Syntax 151
Ex. 4.3 Platti: Sonata No. 3/iii bars 919
Baroque vestiges dead at a stroke. They are the antithesis of any and all part-writing.
So foreign are they to the contrapuntal style and the eet progress of the sonata that
they seem to occupy a separate temporal as well as textural dimension. Thus,
just like the two unrepeatable and seemingly incompatible passages in K. 554, these
chords come from another world. They suggest a collage-like conception of the
whole in the manner of Stravinsky. Crucial to this understanding is the invariance
of the chords; they are not worked, are not subject to a (temporal) progression that
would make good their anomalous status. In this sense, they do not participate in
the larger argument of the sonata; indeed, we could easily imagine a version of K.
525 that would be apparently unaffected by their absence. Ex. 4.3, from the nale
of the Sonata No. 3 in F major by Giovanni Benedetto Platti, published in 1742,
features a similar textural disruption.
This movement, entitled Gigue, is predominantly in two parts, and so the sudden
chords, with their arresting rhythm, disrupt both its textural and generic premises.
Platti, however, incorporates his shock into the larger argument and so assures the
coherence of the whole. The initial shock of the D
chords is somewhat assuaged
when they are immediately followed by G
chords, constituting exactly the sort of
progression that is lacking in the Scarlatti. The best touch, however, is found in the
nal bar of the half, after several bars that restore the customary two-part texture. The
nal C major chord clearly provides a textural counterpart to the earlier seven- and
eight-part chords, thus completing the progression. It also allows us to understand
the disruptive texture as a dramatic realization of the circle of fths, from D to G
to C, in the name of establishing the dominant. Not only that, but this nal full
chord would have been an expected gesture anyway. Countless movements from
the keyboard music of the time proceeded largely in two parts until such cadence
152 The keyboard sonatas of Domenico Scarlatti
Ex. 4.4 K. 27 bars 132
points, when it was common practice to ll in the harmony, either chordally or
by means of an arpeggio. (As we have already seen, Scarlatti goes out of his way
to avoid both possibilities.) Platti thus wittily justies the convention here through
the particular prior circumstances of the movement.
By comparison, the chords in
K. 525 are like inarticulate gestures, blobs of sound.
If the seemingly independent existence of the killer chords in K. 525 offers a
rather indirect example of a syntax that is both split-level and repetitive, there are
sonatas whose repetitive traits are more obvious to the listener. An example is K. 27
For another example, see Sonata No. 29 in C major by Rodrguez. The predominantly two-part texture, full of
familiar suggestions of string writing, is interrupted at bar 54 by huge eight-part repeated chords. These are then
assimilated by being treated in a characteristically generous sequence, with seven separate limbs, taking us back
to the departure point of G major in bar 68.
Syntax 153
Ex. 4.4 (cont.)
in B minor (Ex. 4.4 gives the rst half). Its stretch of apparently irrational repetition,
heard in the rst half from bar 11, is all the more exceptional in that it cannot be
rescued by any evocation of Latinate vitality. The repetition feels static rather than
The sonata in fact progresses by means of a dialogue between learned and toccata
styles; neither term is ideal, but they help to capture a clear opposition of syntactical
types. Of the passage from bar 11 Giorgio Pestelli writes:
Then there is something for which one can truly nd no source or reference: an insignicant
arpeggiated guration, instead of continuing on its way, begins to circle around itself like a
Catherine-wheel . . . Here the strophic logic of traditional musical discourse collapses, that
made up of antecedents and consequents, of attractions and repulsions always in motion.
This reiterative furore, for which time stops, so to speak, oscillates between a hedonistic taste
154 The keyboard sonatas of Domenico Scarlatti
that rejoices in its powers and a sensibility that is astonished by the possibilities of the world
of sounds.
The hedonism of which Pestelli speaks implies an inability or unwillingness to be
rational and measured in ones enjoyment, to know instinctively when enough is
enough. Here it must do business with the severity of a learned style. However, the
learned style of the rst three bars is not entirely blameless, with some clear denials
of voice-leading propriety a b
is missing from bars 2
and 3
. But it does better
than the toccata style from bars 4 to 6, which features the clearest of parallel octaves
between the outer and inner parts. Of course these could be understood as colouristic
doubling, and the B minor 5/3 chord of bar 4 is in fact succeeded by 6/3 chords in
the two subsequent bars, but the ear is so sensitized by the idiom of the rst three
bars that the parallels really do register as such. The following polyphonic texture at
bars 79 is more solid with its four parts, but again there are missing continuations
in individual voices. In the rst instance this is to avoid the consecutives that would
arise from their presence.
The toccata style responds by showing more exibility of
melodic movement; the g
traced by the upper line at 11
chimes with
the linear movement of the learned material, more specically with its falling thirds.
Compare, for instance, the bass line from 7 to 10, with its falling-third semiquaver
shapes and also the augmented version traced by the crotchets DCB, GFE
and FED.
On the next syntactical level up, though, there is no exibility at all, just a seem-
ingly endless repetition of the same bar. The hands swap roles twice, relieving the
monotony technically and visually, but not syntactically. Is this really music? is the
question that hovers over the passage.
Eventually something must give, and from
bar 17 the arpeggios form themselves into a linear intervallic pattern of 108, with
suspensions added to make a 1098 pattern (see Ex. 4.5a).
This swapping around of the roles of the hands in an extremely repetitive passage
is also found in the Sonata No. 1 in D minor by Rodrguez. The similarity of
conception is very striking. Frombar 49 of this piece a two-bar module of alternating
V5/3 and V6/4 harmonies is played twice in each disposition before the hands
exchange material, which consists, as in K. 27, of broken chords in a middle register
and widely leaping crotchets on either side. The ensuing four-bar units are played
four times in all, making sixteen bars altogether! This easily outdoes K. 27. Not
Pestelli, Sonate, 146.
Thus the implied tenor b at 8
would yield parallel fths with the alto. One bar later, the alto note is omitted
for the same reason to avoid a simultaneous DE in the tenor and AB in the alto.
Peter Williams compares the passage with the opening of Bachs Gigue from Partita No. 1 in B at major. In K.
27 this difference of articulation, depending on which hand does the leaping, seems to be a calculated effect . . .
Alas, once again we will never know for certain whether Scarlatti intended a distinction or, on the contrary, was
giving the player the task of producing the same effect by two quite different methods. Hints for Performance
in J. S. Bachs Clavier ubung Prints, Early Keyboard Journal 5 (19867), 323.
Syntax 155
Ex. 4.5a K. 27 bars 1721
10 10 98 10 98 10 98 8
Ex. 4.5b K. 27 bars 236
10 7 7 7 7 10 10 10 10 10
only that, but after a two-bar breather the same repetition is repeated up a fourth,
although this time it nally breaks into a harmonic progression from bar 81. This is
similar in effect to the linear pattern that takes over from bar 17 of K. 27. Although
in themselves much more extreme than what we nd in K. 27, the character of
these repetitions is far less certain. As much as anything, they revive the questions of
Spanish temporality discussed in Chapter 3.
This device that emerges in bar 17 helps to civilize the syntax of the mind-
less toccata.
The quasi-parallel octaves still obtain between the outer voices, but
these can now be more readily grasped as colouristic doubling. Bars 21 and 22
then form a sort of neutral link in the manner of bar 10. From bar 23 we hear
another linear intervallic pattern, a 107, that lies more in the province of the
learned style. Reductions of this pattern (Ex. 4.5b) and that of bars 1721 are given
In its rhythmic uidity, though, this pattern seems to take something from the
toccata passages. This suggests that the two styles are beginning to borrow, indeed
learn from one another. The rest of the half bears out this reading. Thus at bars
267 the rapid unfolded thirds of the semiquaver guration bear the imprint of the
toccata, but note the subtle imitation between the left hand of 26 and the right hand
of 27. There is also a rough inversion between the scalic quavers that pass from the
I dissent from Pestellis comment that bars 17ff. reveal a melody of facile sentimentality; Pestelli, Sonate, 146. He
overlooks the learned basis provided by the linear intervallic pattern, quite loaded in this context. A sentimental,
nostalgic impression may indeed be created, but this tells us more about how we hear such patternings today,
and our enjoyment in surrendering ourselves to their ancient lineage. See the discussion on reception of the
galant style, Chapter 3, pp. 968.
156 The keyboard sonatas of Domenico Scarlatti
right hand in bar 26 to the left hand in bar 27. The closing gesture from bar 29 is
of more uncertain import and has an enigmatic effect. Stylistically it lies within the
realm of the toccata, but its falling scale steps owe a debt to the learned material
from the start. These semiquaver scale steps seem to ll in the wide spaces of the
earlier toccata passages.
On a grander and more radical scale is the Sonata in G major, K. 260, where
once again passages of unreasonably extensive repetition alternate with more familiar
material. This work appears to invert the order of things: the normal passages (those
that the composers contemporaries would have recognized as proper music) do
not ultimately so much afrm the familiar diatonic world as represent a rather
pallid response to the vamps, which must be regarded as the real content of the
sonata. Found approximately in bars 2541, 6171, 10736 and 15578, these feature
obscure harmonic progressions, marked implacably by left-hand chords on each
downbeat, offset by oscillating quaver patterns in the right hand. All four passages that
followthe vamp sections are similar in material and seemuntouched by the foregoing
events. In another context they would be unexceptionable, but here, if they represent
reality to the vamps fantasy (since this can hardly be a viable way to go about the
craft of music), their reality the recognizable thematic patterns, the movement
by normal-length phrases, the rmly articulated tonality is dull, unsatisfactory,
perhaps even unreal. There cannot be much doubt that their plainness is deliberate;
they are effectively totally diatonic so that the contrast between what feels like
absolute freedom and Gebrauchsmusik is underlined. All four responding passages in
fact feature some chromaticism, but this is purely linear and never undermines tonal
Of course the vamps are totally dependent on the surrounding contextualization
provided by the normal sections, since, as we have seen with the composers use
of exotic elements, such music cannot exist without this regular framing but that
an independent existence can even theoretically be conceived for the vamps is the
radical possibility suggested by K. 260. Thus the contingency of musical norms
is suggested; they become disembodied through their relationship with the vamp
passages. Scarlatti goes further than any other composer of the common-practice era
in suggesting that diatonicism, and its syntactical clothing, does not encompass the
musical universe. We all must have wondered at some time whether this or that tonal
composer, while improvising at the keyboard or in the mind, played or imagined
combinations of notes and types of syntax that could not conceivably nd their way
into any nished artistic context. Only Scarlatti seems to have had the nerve to allow
such moments into his nal products.
This is not to say that we can advance improvisation as an explanation for these
moments, for the reasons detailed in Chapter 2. Nor can we rescue them by an
appeal to a form like the free fantasia. The fantasia was, after all, a distinct genre that
sanctioned all manner of freedoms within its frame, while Scarlatti impurely mixes
his fantasies with more standard material, in works that carry the title of sonata
Syntax 157
(how rich this bland title is turning out to be!).
It is characteristic, though, that
in K. 260 he seems disinclined to reassert the authority of the prevailing language,
hence the rather underwhelming response to the challenge posed by the vamps.
Kathleen Dale, writing in the 1940s, got this just right when she commented that
the visionary quality of these interpolations is emphasised by the prosaic character
of the surrounding paragraphs of scales and arpeggios.
Such questions may arise through the contemplation of any of the composers
vamps, but the difference in this sonata is that the vamp is not a single, if extended,
central event it recurs at regular intervals. The four separate sections belong together
as clearly as the diatonic sections do, and at each recurrence, the implication is that
the vamp, having been temporarily suppressed, has risen to the surface again as if
it insists on its rights to take a full formal part in the musical structure, as though the
structure is to be analogous to some kind of rondo form. In fact, the vamps assume
more prominence in the second half, as each one lasts about twice as long as its
rst-half equivalent. Thus their striving towards autonomy becomes more insistent.
Although the vamps seem remote from any eighteenth-century diction (even if
possibly taking their cue from Vivaldian concerto gurations
), they in fact contain
strong melodic impulses that never shape themselves into anything denitive. There
are plenty of rogue moments among the revolving right-hand patterns when the
rate of pitch change suddenly spurts ahead of what we might expect, particularly
in the second half. It is as though we are approaching an eloquent statement but
never achieve it. We can hear this best in the rst vamp of the second half, especially
between bars 115 and 126. Always becoming, never being, each vamp melts away,
and what is eventually delivered is mundane bustle.
In memory the piece exists not so much in its ofcial G major as in its timeless
moments. If Scarlatti wasnt a relatively peripheral gure, we could describe this as
a truly prophetic piece of the Ich f uhle Luft von anderem Planeten variety. It is
so exceptionally audacious that we dont have the historical or stylistic means to do
justice to it. Characteristically, Scarlatti doesnt explain the object is presented for
our contemplation, and nothing is signposted.
It is worth pointing out that K. 260 has not been much recorded. Indeed, players,
both in concert and on disc, have shied away from all the most excessively repetitive
sonatas, and especially those that contain vamp sections. It is not hard to divine the
reason for this avoidance. Excessive repetition is embarrassing for the performer
and possibly for the listener too. When it cannot be understood to fall within one of
the rhetorical categories outlined earlier, then it may seem antisocial, if not living on
I mention this genre by way of comparison because of its associations with the sort of harmonic freedom found
in K. 260. Historically, though, it does not have strong ties with Scarlattis cultural and working environments.
The toccata would be a more apt point of comparison, but since I believe Scarlatti uses this much more as a style
rather than as a type, the same reservations apply.
Dale, Contribution, 43.
See Sheveloff, Grove, 3389. We will return to this stylistic suggestion.
158 The keyboard sonatas of Domenico Scarlatti
the edge of sanity. After all, many forms of irrational conduct or mental illness involve
repetitive behaviour, arising from an inability to judge the line between enough and
too much. Or if we think of the reception of twentieth-century minimalism, many
hostile parties have accused it of an antisocial orientation, linking minimalism with
the hippy drug culture of 1960s California. The embarrassment for the player of
a Scarlatti vamp is one of having to act out such seemingly unbalanced, irrational
behaviour. The performer is uniquely exposed. This is a particular problemgiven the
traditional role played by eighteenth-century music in our culture as the embodiment
of civilized values; it offers an opportunity to advertise ones taste, ones sense of
style, as Kirkpatrick would have it, that has been taken up by many performers as
well as listeners. As vamps generally involve free guration and decontextualized
harmony, there is no style as such to immerse oneself in or to hide behind.
On the other hand, it is seemingly easier for performers to cope with Scarlattis
absent repetitions, and with the resultant lack of symmetry. The coping is often
achieved by means of various acts of subterfuge tidying up ornamentation, for
instance, so that parallel units automatically receive parallel embellishment, or by
adding bars at the ends of sections to make a phrase scan. Scarlattis habit of lopping
off a bar giving us one bar at the end of the rst half, for example, when two are
needed to balance the hypermetre of the whole phrase is disregarded by performers
almost without exception. An example may be seen at the end of the rst half of
K. 523 in Gmajor (Ex. 4.6). Bar 43 is preceded by three matching two-bar units from
bar 37 and should clearly be followed by another bar of the D octave to make up the
expected, indeed surely inevitable eight-bar phrase. The failure of the expected bar
44 to eventuate runs so strongly against the syntactical grain that it is hardly surprising
if most performers show themselves unable to cope, except by effectively rewriting
the close of the phrase. Indeed, in many cases they may not even be conscious of
ignoring the notation.
Mikhail Pletnev does exactly that in a performance that conveys a wonderful sense
of the registral play through the sonata, showing how much structural resonance and
colour may be invested in this parameter.
His deviations from any published text
may well trouble the Scarlatti acionado, but they form a useful index to the most
idiosyncratic aspects of the composers style in this piece. Everything that is most
individual here this most individual of performers smoothes out and regularizes. As
well as the addition of extra bars at the end of each half to make the numbers balance,
we nd the elimination of asymmetrical details that prevent the precise repetition of
small cells (such as the removal of the tenor d in bar 39 and the playing of the whole
bass line one octave higher), and the replacement of the open fth on the downbeat
of bar 21 Pletnev must consider this too raw a sound and so replaces the left hands
A with a C.
Virgin: 5 45123 2, 1995.
Exactly the same alteration is found in B ulows arrangement of K. 523, found as No. 6 of Suite No. 1 in
Achtzehn ausgew ahlte Klavierst ucke. This reminds us of Scarlattis relishing of such open sonorities, as detailed in
the discussion of horn calls in Chapter 3, pp. 867.
Syntax 159
We also hear notes added in the bass at bars 7, 9, 11 and 13. Missing bass notes are
one of the thorniest problems for the modern-day editor of Scarlatti sonatas. Bass
notes are frequently lacking precisely at important structural points, just when the
preceding harmonic activity most demands their presence and articulative power.
Their denial can create what Ralph Kirkpatrick called a sickening emptiness in
the bass which produces vertigo, and their absence often seems so incredible that
scribal error is generally assumed by editors.
The delicacy of the matter lies in the
probability that some of them may indeed represent scribal error but that all of them
together cannot they are too frequent an occurrence. However, as a species they
may be aligned with those missing bars at the ends of phrases; they also suggest a
determination to undermine precisely the most secure and automatic of syntactical
habits and assumptions. Kirkpatricks visceral reaction indicates the level at which
such denials affect us; intellectually we may just about be able to assent to them,
but the musical body rebels. Such details are, and should be, almost impossible to
live with. And so from bar 7 Pletnev spells out the linear intervallic pattern that
is only half articulated by Scarlatti, thus removing the teasing distortion of texture
and register. Ex. 4.7a shows the underlying pattern which Pletnev brings to the
More striking by far than these, though, is the addition of a companion phrase
unit at the beginning to match the singleton at 14: Pletnev replays these four bars
before proceeding further. He of course gives us what we have a right to expect the
sonata starts with a self-contained periodic phrase unit and with a sequential pattern
that seems to demand a response or continuation in kind. Everything would seem to
be set up for an immediate repetition. The mode (even the very key of G) and metre
(3/8) play a part in this too, implying a light style that would be structurally easy.
In fact, what we have is a version of what I call the opening stampede, quite
a common occurrence at the start of Scarlatti sonatas, which favours momentum
over clear articulation it is structurally breathless, we are given too much to take
in too quickly. The opening of K. 457 in A major furnishes another instance of this
stampede. We do not expect to nd such intensity and unpredictability of action
at the beginning of a sonata. There is no secure point of cadential or phraseal
articulation; instead, we are propelled forward in search of the stability that should
have formed the point of departure. The hectic patterns at bars 517 of K. 523 are
also very characteristic in this regard they twist out of any settled shape. K. 523 in
fact turns out to be a problem sonata, where all subsequent material represents some
sort of response to the initial challenge to our perception. In terms of shape, bars 513
are already an answer to the opening unit, given their basis in a stepwise descending
sequence. The phrase functions as a very indirect and expanded consequent to the
rst four bars.
In strict syntactical terms, though, these bars do not correct the impression of
lopsidedness. That process begins slightly later. The material from bar 21 is a clear
Cited in Sheveloff, Uncertainties, 159. This article offers an almost unique discussion of the feature, at 15965.
160 The keyboard sonatas of Domenico Scarlatti
Ex. 4.6 K. 523 bars 173
reference to the opening, with some simplication of the pattern but more impor-
tantly a new continuation the four bars 214 are balanced by the continuation
towards a cadence point, making eight bars in total. The whole is then repeated,
thus dealing with both original unsatisfactory aspects of the opening bars: the
short-windedness and the lack of phraseal balance. Even the closing material from
bar 37, with its melodic outline falling from

5 to

1 (see the stepwise fall from a
to d
at 379), reworks the contour of the start (the stepwise fall from d
to a displaced
Syntax 161
Ex. 4.6 (cont.)
in bar 5), and now there are three iterations of the unit, overlapping. Three is
certainly better than one.
That the opening is to be conceived as a problem becomes absolutely clear at the
start of the second half. This moves straight to the tonic minor and simply gives us
the opening four bars in that key (447). The initial harmonic sense is of course
different because of the opening D pedal. The minor key also works rhetorically
here, casting a shadow over the condent but wrong opening gesture. This explicit
tonic-minor version is given a new continuation, leading to a half-close at 50; the
original phrase has again been broadened.
There is immediately another recomposition from bar 50. The sequential con-
struction of the original right hand is now made more structurally sequential
in other words, into a linear intervallic pattern (76; see Ex. 4.7b). The original
compound melodic structure is now made explicit, with a clearly independent alto
line. And so we have a timely intervention by a more learned style; its associations
of sturdy technique and reliable patterning make it once more a good friend in a
162 The keyboard sonatas of Domenico Scarlatti
Ex. 4.7a K. 523 bars 715
10 (7 7 7 10) 10 10 7 10
Ex. 4.7b K. 523 bars 5054
Ex. 4.7c K. 523 bars 5764
4 6 4 6 4 6 4 6
Ex. 4.7d K. 523 bars 448
crisis. In fact, this passage is doubly learned, since, in addition, the bass is imitating the
right hand from the start of the half (compare bars 4448
of the right hand with the
left hand from50
to 54
). Both these elements of learning, the linear pattern and the
imitation, impose a rmer shape on the original unit. We should note especially that
the left-hand imitation of the earlier right-hand pattern means that we have two
phrases acting as question and answer, precisely the sort of relationship that was de-
nied at the start but which Pletnev decided to full. This second phrase too receives
a continuation, at bars 545, to lead to a half-cadence.
There follows yet another recomposition. With a phrase overlap, the right hand
from bar 56 traces the same line from d
to a
heard at the beginnings of both
Syntax 163
halves, while the alto becomes still more independent, forming its own 46 pattern
with the bass (see Ex. 4.7c). This contains all four original stepwise pairs, as found
too at bars 4448
: this is illustrated by Ex. 4.7d, which aligns the shared notes.
This then hooks into a repetition of bars 489 at 645, but note how the total
phrase has expanded. The phrase including 645 is at least two bars longer than that
containing 489; the exact length depends on whether one includes the overlap in
bar 56.
Thus we have a very comprehensive working-out of the original problem, sig-
nicantly involving learned devices coming to the rescue. Does their presence also
suggest that the very opening was based on serious patterning, but dressed in new
clothes and failing to cut a convincing gure? This could mark a syntactical plot
involving the collision between periodic and sequential impulses or the modern
manners of a galant style and the older ways of the learned. It requires a consider-
able effort on our parts to become alive to such possibilities of syntactical argument,
when, as outlined earlier, we most naturally read tonal music in terms of its harmonic
narrative. If we only have an ear for harmonic vocabulary, a sonata like K. 523 will
pass by all too easily. After all, it moves briskly enough to the dominant, which is
prolonged in totally diatonic manner, and then, remarkably, spends the entire second
half in the tonic (if mostly on its dominant), only changing mode halfway through.
Our training might suggest that there is nothing to detain us only a quirky open-
ing that could be ascribed to artistic mannerism. But it should be apparent that the
composer is well aware of the implications of his syntactical tricks, whether made
good, as here, or not.
What stimulus might Scarlatti have had for the cultivation of his peculiar syntactical
habits, aside from the workings of his own creative mind? K. 532 in A minor suggests
one answer. As proposed in the previous chapter, K. 532 is an unusual case in that,
like very few of the Scarlatti sonatas, it appears to be entirely Spanish, a dance scene,
presented as if it were a transcription. There is a sense of proud gesture in the ery
repeated units, which is perhaps easier to choreograph than to analyse in normal
terms. Repetition is always easier to evoke than to explicate.
While often it seems to be more the principle of irrational repetition, abstracted
from any localized source, that governs the vamps and comparable passages, K. 532
suggests that the principle may also be more locally grounded. It virtually begins
with a vamp, reharmonizing time and again the repeated melodic cell c
. This
is then expanded immensely from the start of the second half, starting with the
same notes as at the beginning (compare bars 63
66 with bars 4
7), in the most
common position for a vamp. While this may be a recreation of a frenzied ritual,
it also shows a fascination with a xed sonorous object. The repetition becomes in
fact more repetitive over the course of the passage.
To start with, Scarlatti replaces the endlessly repeated melodic cell with transposed
forms between each four-bar unit. Thus the reiterated CB becomes ED from bar
67 and then GA from bar 71. Unlike the rst-half model, though, the bass ostinato
164 The keyboard sonatas of Domenico Scarlatti
Ex. 4.8a K. 541 bars 1630
gure now remains constant, so that while the upper voices become less repetitive,
the bass becomes more so. A quasi-stretto speeds us towards an exact transposition
of the whole passage up a fourth (compare 63
with 83
). This leads, not
to more variance, but to a direct repetition of the start of the second larger phrase
(compare 83
ff. with 95
ff.), with the bass an octave lower. From here Scarlatti
reverts to the earlier principle of melodic insistence and harmonic change found in
the rst half. When from bar 107
we return for the third time to the identical phrase
(as at bars 83
and 95
, save for the change to minor), it is a powerful effect. After all
the animation, after all the repetitions, varied either in the upper voices or the bass
but never both at once, we win through to. . . more of the same. It is almost like a
victory for brute repetition over differentiated composition, the same principle we
saw in the treatment of the huge chords in K. 525, although on a broader level the
whole vamp-like passage obviously ts this bill.
A similar distinction also seems to inform the Sonata in F major, K. 541, another
work that strongly suggests the contingent nature of musical time. This sonata be-
comes dominated by material, rst heard frombar 19, that is less thematically distinc-
tive than anything else in the piece a routine left-hand guration and a right-hand
two-chord shape whose purpose is unclear (see Ex. 4.8a). Perhaps the right hand
punctuates the hectic repeated accompaniment, but it does not divert it from its
course. It suggests cadential closure note the sudden thick texture and the trills
but the left hand ignores the repeated cues. In effect we have an accompaniment
Syntax 165
Ex. 4.8b K. 541 bars 5772
to nothing that becomes the centre of attention. Ironically, the phrase from bars
to 27
is a perfect eight bars long after a characteristic opening stampede that
plays around with nuances of phrase rhythm in an idiom that clearly favours duple
sectional organization. Is Scarlatti saying from bar 19 Fill in your own melody?
as if the demands of rhythm and our sense of syntactical proportion, now satised by
the eight-bar unit, far outweigh the particular means by which these are realized.
This much might be suggested by the continuation from bar 35, after a minor-
mode repetition of our eight-bar unit. The left-hand gure remains in essence the
same except that it is no longer rooted to the spot, but now it clearly accompanies
the tuniest of tunes. Pestelli notes this tune as a fragment of an Italian Christmas
song, known as the Couperin pastorale.
If this is the case, it only strengthens the
sense of compositional gesture outlined above, that of lling in a melody so what
could be better than one which is pre-existing?
In the second half the purple patch is treated to a reductio ad absurdum and the
right-hand interjections become more obviously silly from bar 61, with the double
trills in the lower two parts of the three-part chords and the horrid voice leading
(see Ex. 4.8b). At the end of the rst unit, at bars 667, the left hand denies the V of
D minor implications that have been set up and goes its own way. It ceases, in other
words, to accompany. This is the surely inevitable outcome of the individualization
of an apparently subordinate line. The left hand reverses its direction and features an
awkward leap of the leading note down a major seventh. The right hand suddenly
Pestelli, Sonate, 2056. The same fragment can be found in K. 260, in fact compare bars 8891 of its rst half.
166 The keyboard sonatas of Domenico Scarlatti
nds life after this too and presents a new gure at bar 67. A logical pause follows
the left hand must lead on now that it has overtly assumed the initiative, but it is as
if the right hands dramatic shape has called the lefts bluff. Bar 68 represents the rst
point of rest in both parts.
The re-emergence of the left-hand guration from silence conrms the sense that
the gure simply marks time rather than representing truly composed material. The
failure of the left hand to do anything more than continue with its accompaniment
to nothing suggests that we are hearing meaningless sound against a background
of silence. The subsequent passages and their subsequent silences only strengthen
the impression. Scarlatti appears again to be playing with the boundaries between
composed time and brute, mechanical time.
During the third of these second-half passages the right hand returns to its rst-
half form, so dispensing with the chordal parallel fths, and from bar 86 the left
hands now expected change of direction is not allowed free rein. The right-hand
chords move in a pattern with melodic force, the left hand is forced to adapt, and
the spell appears to be broken. This is clinched by the cadential pattern at bars 889,
which picks up on the tune of the rst half compare bars 35
, for example.
Melodic and temporal coherence has been resumed. Now there occurs another bars
rest with a pause.
Once more, however, the left hand at bar 91 emerges with its pattern out of
nothing, so that the strange sequence of events in effect continues. The security
provided by the patterning of bars 889 now seems just as provisional as the non-
sense material. One barely notices that this is now a recapitulation of the rst-half
material. The Christmas tune, however, does not recur; instead, from bar 98, one
hears pairs of notes in the right hand that seem to compress the rising second of the
chordal motive, while the left hand asserts its authority by pushing up by step from
A to F. This is even more apparent from bar 101, where the right hand is clearly
accompanying, not melodic. Such changes of detail help make this sonata another
poor specimen of the balanced binary form in which Scarlatti is supposed exclusively
to deal. The piece is progressively drained of recognizable thematic content as what
should be an incidental detail overruns the structure. In the end composed time
seems to be an empty vessel, as rhythms and repetitions lose their phenomenological
Silence surrounds and inltrates the piece, and we are left with the impres-
sion of an empty chattering, as though Samuel Beckett had taken a hand in the
conception of this sonata. As we have observed Scarlatti shaking us free of various
syntactical dependencies and assumptions, offering a new perspective on the habits
that make up the art music of his time, we might not have suspected that he might
also call into question the largest syntactical unit of all the musical composition
Note the remarks by Jeff Pressing that systematic repetition of patterns can dull time perception, stretch or
even eliminate . . . the apparent time. His primary context for discussion is the music of (near) contemporary
composers, but he also notes the relevance of Scarlattis sonatas to the subject, mentioning K. 422 and K. 417.
Relations between Musical and Scientic Properties of Time, Contemporary Music Review 7/2 (1993), 109.
Syntax 167
We will now examine more closely some of the elements of Scarlattis syntactical
renewal. As already outlined, our prevalent assumptions about the relative weight of
different parameters in tonal music have led to a lack of awareness of rhythmic and
syntactical factors. Indeed, there is some lack of theoretical vocabulary for them, even
though they may often work more directly on listeners sensibilities than do harmonic
patterns. These factors do not of course operate independently of harmony: the two
are interdependent. Nevertheless, while there is a long tradition of considering
harmony more or less autonomously, abstracted from other musical parameters, the
same does not go for rhythm.
This should not be taken to imply that writers have failed to acknowledge Scarlattis
proclivities in this direction. Ralph Kirkpatrick described the composer as a past
master of phrase structure, noting Scarlattis employment of juxtaposition, contrac-
tion, extension and the insertion of irregular phrases, although, surprisingly, he did
not acknowledge the missing-bar phenomenon.
Signicantly, though, such re-
marks were subsumed under performance in the nal chapter of his book, while
consideration of Scarlattis harmony merited an earlier chapter to itself. Malcolm
Boyd counselled us to analyse not the statement and restatement of themes, but
rather the balance and imbalance of phrases, and the manipulation of motifs. He
adds that the phrase rhythm of the sonatas reects the composers position on the
stylistic border-line: while the music trades in short articulated phrase units, their
manipulation frequently results in a seamless continuity which has more in com-
mon with Baroque than with Classical methods.
As has been suggested elsewhere,
though, Scarlatti seems to make positive capital out of his transitional position, as
if he were colluding with the historical ction. This is not the same as a present-day
writer conveniently reading these features into the music and then connecting them
by means of the customary rhetorical identication with the composer. After all,
the same self-consciousness is evident in the play with various styles and linguistic
registers discussed in Chapter 3. Surely one of the reasons that the mixed style was
so attractive to Scarlatti was precisely that it allowed him to pursue his interest in
rhythmic and syntactical phenomenology different means of patterning, types of
reiteration and ways of constructing musical time.
It is Joel Sheveloff, though, who has provided the most considered commentary
on Scarlattis syntactical habits. Writing of the phrase structure of the Sonata in
D major, K. 140, he notes that its choice of a crooked, winding path may be of
a piece with other syntactical anomalies. He lists three examples: the beginning of
motives and phrases in the middle of a bar, stops in unusual places and relationships
of material between the two halves that are out of phase.
Elsewhere, he describes
how the uneven relationships between phrases produce a kinetic energy that helps
speed a piece on its way. The most frequent of techniques used to generate this
Kirkpatrick, Scarlatti, 311. See also the section Tempo and Rhythm, 292304.
Boyd, Master, 174.
Sheveloff, Uncertainties, 170.
168 The keyboard sonatas of Domenico Scarlatti
energy is phrase elision, which only Haydn cultivates as frequently and as interest-
ingly as Scarlatti.
While such elision produces energy, it also denies our instincts for completion and
for symmetry. It can therefore bear both a positive and a negative (anti-normative)
interpretation; it can be productive and subversive. While it is often understood
as a means of avoiding the over-sectional tendencies of the new periodic syntax,
leading to Boyds seamless continuity, less often remarked in this context are the
positive attributes of periodic organization itself, which is after all the basic modus
operandi of Scarlattis keyboard music. Yet, apart from anything else, it is this that
allows the very possibility of a mixed style ordering by discrete units of syntax
encourages the conception of discrete units of material. If the raw syntactical ele-
ments of the new style do court the danger of short-windedness, equally, those of
the Baroque may lead to shapelessness. (This danger would seem to be satirically
reected in two works already examined in Chapter 1, K. 39 and K. 254.) That
this is rarely, if ever, acknowledged reects the more respectable perceived technical
basis of the older style, as discussed earlier in connection with the reception of the
Many musicians, however, cannot see past the composers untidiness, often directly
or subliminally accounted for as being primitive or negligent. Robert Schumann was
unable to come to terms with this aspect of Scarlatti it is difcult sometimes to
follow him, so quickly does he tie and untie the threads
while many performers
of course do a good deal of housekeeping before presenting their sonatas to the
public. Especially revealing are the recompositions of Charles Avison in his Twelve
Concertos of 1744, based on the Essercizi and a number of other (presumably earlier)
sonatas. In the preface to the initial publication of a single concerto he wrote that
many delightful Passages [are] entirely disguised, either with capricious Divisions,
or an unnecessary Repetition in many Places. These are just what Avison tends to
remove. He also claimed to be taking off the Mask which concealed their natural
Beauty and Excellency,
thus providing inadvertently an apt image for Scarlattis
manipulation of syntactical norms.
Avisons arrangement of the Sonata in A major, K. 26, as the last movement of
Concerto No. 1 is a case in point. The original is full of discrepant details; nothing
quite matches or aligns neatly. At the equivalent of bars 1521 (see Ex. 4.9) he omits
a bar so as to yield a neater 3 2 construction. It is difcult, though, to say just
which bar is omitted it seems at rst to be 19 but is in fact probably 15 since the
passage is really recomposed. The harmonic sense is changed. At bar 15 we get the
root-position A minor denied by Scarlatti after the preceding dominant preparation,
and the following bars alternate between prolongations of I and V; compare Scarlattis
hovering on the dominant and consequently more uid, continuous syntax. In fact,
Sheveloff, Keyboard, 415 and 369.
Cited in Boyd, Master, 218.
Cited in Boyd, Master, 225.
Syntax 169
Ex. 4.9 K. 26 bars 1540
the sonata is all dominant preparations of various sorts until bar 43 (even the opening
tonic is not given proper cadential denition).
Bar 20
features an elision, with the upper-voice c
both completing the falling-
third motive and initiating a new downbeat-orientated module. Avison clearly can-
not cope with this, since he has removed a prior bar to make the syntax scan. Another
elision follows almost immediately at bar 22
. This both completes the melodic line
from the two previous bars and runs into a sequential repetition a step down of
bars 15ff. As in bar 20, it is the lower part which rst moves clearly to the next
unit. This time, though, the elision of the third two-bar unit of the phrase does
not happen (see bars 267). The simpler patterning may act as a corrective to the
rst whole phrase, but in context bar 27 seems unexpectedly bereft of new devel-
opments; it sounds unnaturally bare. When from the following bar (28) we hear
the same upper-voice falling third, if now a third higher, which then rises back to
170 The keyboard sonatas of Domenico Scarlatti
the initial note, the music seems to have caught up with where it should have been
two bars earlier (compare 1920 of the model). However, the inner part has already
abandoned its cross-string guration; bars 289 match 2021 in this respect, with
a rough inversion of contour. In other words, the inner part appears to be only a
bar behind. The bass octave gure goes with the sense of the treble in this game
of being out of phase it is two bars behind the model. Out of all this confu-
sion, Avison extracts material which makes the two phrases from 15ff. a matching
When the upper voice completes its rising third back to f
at bar 30
, the lower
parts have already moved on to a new texture. If a more straightforward patterning by
two-bar units seems to be re-established from this point, the strange clashes between
the hands mean that there is no chance to enjoy this. In other words, the sense of
material being out of phase continues. Conrming this sense is that while the upper
part seems to move to something new (in fact it is an intervallic distortion of the
rise and fall of 2021), the lower parts slightly rework the material that began the
two previous phrases. Compare these lower parts at bars 30
with bars 15
beginning on the second quaver of the bar, both feature a falling-third gure, doubled
in thirds, interspersed with a repeated-note lower strand.
This is answered by a rising third which the latter passage also doubles by thirds.
The difference in the latter passage is that the repeated notes now occur on, rather
than off, the beat. This creates a feeling of total syncopation, a way in which this
layer alone is out of phase with its earlier appearances. The threefold reiteration of
the lower parts from bar 30 also recalls the two previous phrases. This means that
30ff. constitute both a distinctly new section and a sequential continuation of the
earlier material. This is yet another layer of syntactical ambiguity, in the form of a
giant overlap of function.
A further complication is the role of bar 30
in the lower parts, thus far unac-
counted for. The parallelism with the two previous phrases encourages us to hear 32
as the last quaver of a six-quaver unit, but the fact that it matches the downbeat back
at 30
may encourage us to hear it rather as the rst beat of a six-quaver unit. Similar
ambiguities attend the top part. As in the lower voices, a six-quaver loop is set up,
but where does it truly start? On paper it seems to begin with the G on the second
beat of 30, but there is a grey area here caused by its continued stepwise movement
through from the D of 29. So perhaps we perceive a clearer beginning from the
subsequent D. It is not too surprising that Avison recasts the upper-voice line from
bar 30 and leaves out the accompaniment. What results is a resourceful rewriting
in the name of a much less remarkable half-cadential formulation. The confusion
of this whole passage from 15 is of course augmented by the left-over-right-hand
writing, especially from bar 30. Digital and syntactical strangeness are thus matched
in this topsy-turvy world.
Readers who have tried to follow all these twists and turns, or at least my account
of them, may well nd themselves in a state of nervous irritation. Yet this is exactly the
avour that tends to emerge fromthe sort of syntactical virtuosity on display. In many
Syntax 171
cases such material would simply have been unthinkable in an ensemble context
and this of course is one strong justication for many of Avisons alterations.
At the end of the half bars 656 are omitted another removal of unnecces-
sary Repetition. This makes for a neater, more controlled cadence. Yet it is also
unbalancing. The extra repetitions are both irrational and rational. In manner they
are overly insistent, but structurally they are needed to balance all the various in-
conclusive dominant hoverings that have gone before. In part, Scarlattis repetitions
signal a new importance for proportions in a musical argument, one based on a more
varied sense of harmonic and phrase rhythm. Avison has arguably not grasped this
sense of proportion. The fact that the same material is used from bars 43, 55 and 63
also makes clear that the same end is required a proper conclusive cadence in the
dominant (minor).
Revealing in a different direction is Handels treatment of the material he borrowed
fromthe Essercizi for his Twelve Concertos, Op. 6, of 1739. He consistently augments
Scarlattis material. Of course, Handels borrowing cannot be directly compared with
Avisons transcription, but it is noteworthy that both composers nd means of
making the original material more comfortable; one cuts while the other expands.
The nal movement of the Concerto in G major, Op. 6 No. 1, based on K. 2, is
the solitary exception. Elwood Derr suggests that this is probably the single instance
in Op. 6 where Handel reduces Scarlattis epigrammatic statements to still more
compressed terms.
Another form of reworking alluded to a number of times already is the addition of
extra bars at cadence points by performers. Missing bars are most commonly found
at the ends of the two halves of a sonata but may occur at any relatively important
point of cadential articulation. This phenomenon illustrates the composers constant
vigilance, his distance from the most ingrained of compositional habits. It may be
allied not just with the absence of important bass notes, as suggested earlier, but also
with the pronounced tendency to avoid fully textured closes, whether simultaneous
(chordal) or successive (arpeggiated). We noted in Chapter 2 the avoidance of a
closing arpeggio in the generally known version of K. 9 in D minor. Extreme
examples of denial of a closing chord, when the preceding dominant chord surely
demands such a resolution, may be found in the extracts from K. 208, 317 and 450
given in Ex. 4.10.
Scarlattis curtness at such junctures, whether achieved through textural or syntac-
tical denial, seems to react against the rhetorical relaxation that normally coincides
Nicholas Cook makes a comparable point about Geminianis more literal-minded and straightforward concerto
grosso version of a Corelli sonata: that it may be as much a function of genre as of personal disposition. At the
Borders of Musical Identity: Schenker, Corelli and the Graces, Music Analysis 18/2 (1999), 195.
Handels Use of Scarlattis Essercizi per Gravicembalo in his Opus 6, G ottinger H andel-Beitr age 3 (1987;
published 1989), 176.
172 The keyboard sonatas of Domenico Scarlatti
Ex. 4.10a K. 208 bars 1214
Ex. 4.10b K. 317 bars 11318
Ex. 4.10c K. 450 bars 4042
with the arrival at an important structural point. Such relaxation seems quite in-
evitable and natural; one only need think of the number of fugues that abandon
strict part-writing and a set number of voices in their nal bars. The interpretation
of such denial in Scarlatti can vary. If heard at the end of an entire sonata, it will tend
to suggest simple negation of the natural, that some tension built up towards the
cadential close remains unresolved; it denies us the full mental and bodily relaxation
we have been conditioned to expect. In a way, it may be seen as a specic embod-
iment of the taste for an open musical experience, as dened in the discussion of
the topically mixed sonatas in Chapter 3. Where else in the tonal repertoire of the
eighteenth century does one nd such an ambivalent attitude to closure? On the
other hand, when the syntactical side of such denial occurs at intermediate points
in the structure, it may serve the more positive ends of maintaining momentum.
Indeed, it may even be made good later.
One question that must arise when considering the missing-bar phenomenon,
a seemingly tiny detail with very big implications, is whether this is a considered
notation. Perhaps, if we bear in mind the unsatisfactory source situation, this reects
scribal laxity; or perhaps it reects an understood convention, with the performer
being expected to make up the missing bars as required. However, the sheer number
of missing bars or beats found in the sources overwhelms such commonsensical
objections. More specically, a number of sonatas are notationally explicit on this
Syntax 173
matter. In K. 149, for example, the rst-time bar at the end of the rst half makes
explicit that the performer should not wait for a whole bar to ll itself out before
continuing. The time signature is 4/4, and bar 16 follows a crotchet rst beat with
just a crotchet rest, marked with a pause. Gilbert inserts a 2/4 time signature in the
rst-time bar of his edition to guide the performer and changes the crotchet rest
to a quaver rest. The pause might admittedly be thought to allow for an effective
lling of the missing beats, but the second-time bar leaves no room for doubt, as
the second half continues immediately from the third beat of the bar. K. 199 offers
a more straightforward example. The nal rst-half bar of this 12/8 sonata consists
of just six quaver pulses; the third and fourth beats have gone missing. There is no
doubt that this phenomenon represents a highly individual effect, but we are not
trained to listen for individuality or to expect a personal stamp in such an area. Our
natural reaction is to deny it, from the point of view both of our body clocks and
of our theoretical training.
The Sonata in D minor, K. 120, represents an extreme and quite unequivocal
example of such abruptness at a cadence point (see Ex. 4.11, which gives the rst
half). The cadential reiterations from bar 22 build tremendous tension which more
than ever would seem to demand a spacious resolving gesture. Instead, we are given
a mere quavers worth of resolution on the downbeat of bar 27 before being whisked
back to the start of the sonata. The same operates in the continuation of the second
half from the second quaver of this bar. Tellingly, the second half provides a foil to
this: it ends, not with a quaver, but with a dotted semibreve marked with a pause!
Such contrasting treatment within a particular sonata again conrms that we are
dealing with a conscious technique.
A variant on the same principle is provided by what Sheveloff has dubbed great
curves, the large slurs found above and below staves most often in association with
repeat marks at the end of the rst half of a sonata. These slurs indicate that the
material contained within them is to be played rst time around and then omitted
on the second playing. Their effect is often to produce a large-scale structural elision
between the two halves. For Sheveloff they form a crucial part of Scarlattis radical
treatment of the midpoint of the binary form:
Most music in Scarlattis lifetime used a rst ending to provide a retransitional link from
the end of the rst half back to the opening material on the tonic; the second ending then
does away with this linking material, allowing the rst half to nish with the fullest, most
convincing stop in the piece, save for the parallel ending of the second half, and thus, of the
sonata. In Scarlattis usage of two endings, an opposite effect tends to prevail. In about 125
sonatas, he will allow the rst half to come to its fullest stop the rst time, and then use the
second ending to overlap the border between halves, so the musical fabric can ow seamlessly
[between] them, almost magically evaporating the usual brick wall between halves.
Sheveloffs phrase in a discussion of this feature in K. 125; he notes that it appears too often in Domenicos
keyboard works to be an accident. Sheveloff, Keyboard, 423.
Sheveloff, Uncertainties, 155. The great curve is also discussed in Sheveloff, Keyboard, 27988.
174 The keyboard sonatas of Domenico Scarlatti
Ex. 4.11 K. 120 bars 127
If the effect may be magical, it may also be plain disconcerting. In K. 535 the
second-time closing arpeggio rushes ahead into new harmonic territory before one
can adjust (see Ex. 6.13). Another example that disorientates both our harmonic and
syntactical senses is found at the mid-point of K. 253; here the falling B at major
arpeggio that ends the rst-time playing of the rst half is completed second time
around not by a b but by an a.
Even many recent performers who are clearly working from the best editions fail
to observe the indications of the great curves. Once more, we must acknowledge
how fundamentally our musical body clocks are being interfered with, making ex-
ecutive resistance almost inevitable. The composer is not simply scoring easy points
at the expense of conventional shapings and proportions; what is indicated in the
sources is often deeply upsetting to our musical instincts. The implication is that
even these are habitual as much as fundamental, that they are the product of cultural
Syntax 175
Ex. 4.11 (cont.)
training. They are so ingrained that our experience of them has become located
entirely in the body, instinctively felt rather than consciously measured. Scarlatti,
by interfering overtly with such natural phenomena of voice leading, timing and
texture, returns them to an intellectual level, in an extreme of relativistic thought.
It takes an iron will on the part of the performer to meet rather than evade such
Executive resistance is even plainer in performers approaches to the much more
frequent missing-bar phenomenon. One of the best examples of this may be found in
176 The keyboard sonatas of Domenico Scarlatti
Ex. 4.12 K. 96 bars 130
K. 96 (see Ex. 4.12). The stand-alone bar 25 presents a challenge to the performer
a mental one. In recordings sampled, Andreas Staier adds two bars and an aspiration
after 25 before proceeding to bar 26, Vladimir Horowitz adds three extra bars to make
a four-bar unit, Anne Queff elec adds almost ve to make six, Pletnev just over ve
(plus a tremolo and mock-heroic piano hustle) and Arturo Benedetti Michelangeli
just under six bars. Christian Zacharias manages the comparatively superhuman feat
of adding only one extra bar after 25.
Obviously the grand build-up of sonority
from bar 11 onward seems to demand time to resonate, or at least some clearing-
space, before any continuation, and the duple construction also seems to require an
even count of bars before the next phrase can proceed. (Some of these performances
must be following Longo, who adds a pause over 25, but one imagines that even
without Longos intervention most performers would quite naturally add one. In
his edition B ulow not only adds a pause but also the indication longa!) However, it
must be quite clear that the composer is not prepared to grant this and what should
be heard is an intrusion by another idea before we could possibly expect it. Yet there
is also a positive expressive point to this denial of the natural. The rushed syntax in
fact aids the impression given by K. 96 of a giddy panorama, as considered in the
previous chapter. The rst real breathing space does not arrive until bar 137, well
into the second half, and this is marked by a pause. What follows this is a return of
precisely the material that arrived too soon in the rst half, bars 26ff. This represents
Deutsche Harmonia Mundi: 05472 77274 2, 1992 (Staier); Sony: 53460, 1964/1993 (Horowitz); Erato: 4509
96960 2, 1970 (Queff elec); Virgin: 5 45123 2, 1995 (Pletnev); Grammofono 2000: 78675, 1943/1996 (Michelan-
geli); EMI: 7 63940 2, 197985/1991 (Zacharias).
Syntax 177
a clear correction of what was so unsettling before and proves the need for doing
exactly what was notated in the rst half. On the whole performers consistently
play fast and loose with the rhythmic and phrase-structural features of the sonatas
in a way that they wouldnt contemplate doing for, say, harmonic structure. One
might counter that, historically, these represent legitimate areas of freedom for the
performer timing and delivery whereas harmony is xed, beyond all questions
of intentionality. This would simply conrm the priorities suggested at the outset
of this chapter.
The missing-bar phenomenon forms part of a wider vigilance about cadences
altogether. As implied already, the sheer number of cadences in the sonatas can be
seen as inherently problematic. For Hermann Keller the too frequent and too sim-
ilar cadences were the weakest point of Scarlattis style, although this was a fault
shared by other composers of the epoch, one connected with the disappearance of
the basso continuo. Macario Santiago Kastner wrote that the constant repetition of
small units was a common stain on eighteenth-century keyboard music.
Music as
dance is nowhere to be seen in such judgements. Often of course Scarlatti does shade
these cadences differently, whether through the syntactical means already discussed
or through registral manipulation; often too his most brilliant invention accompanies
this regrettable stylistic weakness. First of all, though, we need to consider a wider
defence of this stylistic feature. It is difcult for us now to appreciate the vigour
of eighteenth-century tonal language from this point of view repeated cadential
formations were a new and exciting thing, they must have given a sense of freedom.
Our ears are more geared to nineteenth-century ideals, precisely when such con-
siderations led to a weakening of tonal logic. Charles Troy has noted in a study
of the intermezzo how the constant repetition of small units is sometimes carried
to absurd lengths. For example, Orcone in Alessandro Scarlattis comic scenes for
Il Tigrane (1715) is directed to repeat the same four-note motive during an aria as
many times as he wants, until he shows himself to be out of breath.
(Scarlatti may
owe something to such an approach in his vamps and elsewhere, but his passages
have no words and are thus less immediately comprehensible.) A high degree of syn-
tactical articulation, above all by means of cadences, is indissolubly associated with
the entry of pronounced popular, comic and dance elements into art music, all of
which were richly exploited by Domenico Scarlatti. They are also predominantly
associated with speed, whose problematic aspects were considered in Chapter 1.
These considerations offer a stylistic background to the cadential formations found
in Scarlatti. Although these often sound, or are made to sound, like one of his most
distinctive personal traits, they are one of the aspects of his style for which we can nd
the clearest precedents and echoes. In sonatas by composers such as Galuppi, Platti
and Paradies one nds very similar turns of phrase in closing cadential passages, and
Keller, Meister, 78; Kastner, Introduction to Carlos Seixas: 25 Sonatas para instrumentos de tecla (Lisbon: Fundac ao
Calouste Gulbenkian, 1980), xvii.
See Charles E. Troy, The Comic Intermezzo: A Study in the History of Eighteenth-Century Italian Opera (Ann Arbor:
UMI Research Press, 1979), 946.
178 The keyboard sonatas of Domenico Scarlatti
Ex. 4.13 Galuppi: Sonata No. 1/ii bars 5160
the same tendency towards sharp, jocular invention, suggesting a common Italian
comic-operatic heritage.
In particular one nds the repeated bass motion rising
mostly or entirely by step from I to V that is such a trademark at this point in Scarlatti
sonatas. Ex. 4.13 shows an example from the second movement of Galuppis Sonata
No. 1 in C major.
Such formulations remain a trademark of Italian operatic style well beyond Scar-
lattis and Galuppis time, of course, as does the relatively plain delivery of the perfect
Yet in spite of these shared cultural characteristics, Scarlattis cadences do
often sound highly distinctive. The composer appears to reinvent the cadence. One
of the means by which he manages this can be found in the Sonata in G major, K.
180 (see Ex. 6.4). At bars 30 and 32 there is a sudden blur of activity in the cadential
pattern, caused by the unexpected sounding in the upper voice of a D, its quick
cancellation by D and the uncertain place of the intervening E in the harmonic
scheme. Scarlatti is fond of putting in elements that make one look askance without
threatening the harmonic sense (which is usually overwhelmingly strong at such
nal cadential junctures). Similar examples of chromatic interference in cadential
approaches may be found in K. 242, K. 495 (in the second half), K. 184 (in the
form of a whole-tone scale) and K. 482 (note especially bar 90, with its underlying
parallel fths between the voices, made worse by the tritone heard on the fourth
beat). K. 224 also offers a dizzying turn of events at the end of each half, seen in bars
64 and 66 of Ex. 4.14. What do such rogue elements mean? Is Scarlatti suggesting
that any old notes will do given the impelling force of the basic progression, making
us aware of the articiality of harmonic habits? The cadence and the approaching
manoeuvres represent an area of denition usually taken for granted, of course, not
Note too Pestellis comment that Sammartini showed a liking for unpredictable ideas, reserved for the coda;
Pestelli, Mozart, 31. For acknowledgement of this trait in Scarlatti see Boyd, Master, 168, and Chambure,
Catalogue, 123.
The numbering is taken from Baldassare Galuppi: Dodici sonate, ed. Iris Caruana (Padua: Zanibon, 1974), No.
As noted in Peter Williams, The Harpsichord Acciaccatura: Theory and Practice in Harmony, 16501750, The
Musical Quarterly 54/4 (1968), 520.
Syntax 179
Ex. 4.14 K. 224 bars 638
thought to require detailed listening. Through this harmonic and the previously ex-
amined syntactical interference we are suddenly forced to perceive the object afresh,
as if for the rst time. Such a process can be understood not just as a manifestation of
disdain but as a form of renewal, through the agency of the concept of Verfremdung.
In his study of insistence in Scarlatti, Loek Hautus invokes Verfremdung, the
equivalent of a term originating with Russian formalist literary theory in the 1920s,
to help explain the composers use of repetition and dissonance. As he explains, over
time the means of art, through habit and automatism, become pale and schematic
and lose their effect; thus, although we know an object or image or syntactical
device is still present, we can no longer see or hear it clearly. Our perception of it
has been worn down by over-familiarity. Such artistic means can be revived through
the deformation of existing models. By making strange, by twisting something
out of its familiar contours or placement, our perception of it can be renewed. As
Hautus reminds us, the need to combat such wearing-out of perception helps to
explain the driving force of artistic innovation and the development of personal
style characteristics.
Thus historical changes in art the shift or drift from Baroque
to Classical, for instance and the particular ngerprints of an individual artist can
both be grasped through the agency of Verfremdung.
This term is most commonly associated with its adaptation by Brecht, and here
the artistic aims seem to bear more specic relevance to Scarlatti. By the application
of Verfremdungseffekte Brecht hoped to force an audience to attend to the implica-
tions of the material presented rather than being swept along by all the familiar
dramatic-narrative devices, with their culinary comforts; the audience was to be
made critically aware of the articiality of their artistic experience. Surely no com-
poser before the twentieth century is so preoccupied with intrusive devices that force
all manner of reevaluation from the listener, although Haydn would run Scarlatti
close in many respects.
To return to the more fundamental denition of the term, it should be clear that
Verfremdung does not in any way specically dene Scarlattis artistic attitude. The
term highlights a basic historical dynamic that helps us account for artistic change,
so that at most we can speak of greater or lesser degrees of Verfremdung in various
styles, genres, epochs and individual outputs. In generic terms, for instance, it is of
less relevance to sacred genres and the strict style, when continuity with the past
and passive contemplation are desirable ends. In terms of individual outputs, we can
Hautus, Insistenz, 142.
180 The keyboard sonatas of Domenico Scarlatti
certainly assert that Verfremdung is a constant presence in the structures of Scarlatti,
hence the category of originality that has been so frequently evoked.
Another type of cadential Verfremdung can be found in K. 120 (see Ex. 4.11). In
the rst half the nal cadential repetitions actually begin at bar 17
. The second
version from 18
is interrupted at bars 1921, then we hear ve more, every bar
ending the same way in all parts. The rst of these further ve repetitions emerges
during the course of bar 22; by halfway through the bar it has become clearly
recognizable, as if it is picking up from where the second playing was interrupted,
at 18
. The bass repetitions are essentially identical every time to get the real
avour of this insistence, the reader might be advised to play or sing through the
bass line alone from the onset of the passage. Such repetition arguably defamiliarizes
the cadence. Heard once or twice it is unexceptionable, but heard more often it
subverts the idea of cadence, which is now an object of contemplation in itself
even a fetish rather than a simply a mechanism or means of articulation. This is
particularly noticeable given the proportions of the structure (the cadential furore
begins not much beyond halfway through the rst half) and the Baroque manner of
the preceding material (note the sequence at bars 6
and the very metre, 12/8,
itself ) this is not a style that requires the frequent articulated cadential repetitions
that follow. These make us very aware of closure as a structural property, and, through
a repetition that has a delaying as much as a conrming effect, of the possibility that
closure might not eventuate. Thus we are again reminded of the articial nature of
musical time and its commonly agreed syntactical rules. Here we have an energy that
wont abate, an excessiveness that seems to refuse artistic control. The Verfremdung is
completed, as noted before, by the impossibly abrupt return to the beginning and
move onwards after a quavers worth of resolution. Over-preparation is succeeded
by under-articulation.
If the manipulation of cadence tends to upset comfortable expectations of ending,
the stampede technique upsets our equilibrium at the opposite end of a binary
sonata. As dened earlier in conjunction with K. 523 and K. 457, this occurs at
or near the beginning of the rst half of a sonata. Broadly speaking there are two
types of beginning to a Scarlatti sonata the difdent and the hyperactive. The rst
may be routine, conventional, low-key, often involving the use of imitation between
the hands that is then abandoned. This difdence is not necessarily a matter of
affective character but of structural function; if the opening material and texture are
abandoned, it raises the question of why the composer decided to begin with themin
the rst place, to place them in such a rhetorically and formally privileged position.
The hyperactive beginning, on the other hand, seems to present a celebration of
the tonic, the sheer excitement of being in motion. It is difcult for us to deal
with this except by evocation, since we are used to energy at this time being more
latent and channelled towards possible growth. K. 503 offers an example of this type
(although it also features initial imitation). The stampede can include both elements.
K. 268, for instance, suggests a certain creative difdence at the start, in that the
rst really chiselled invention is not heard until bar 15. On the other hand, this is
not simply a casual opening, and one could hear the rst section as expressing an
Syntax 181
energy level that takes a while to settle and channel itself. There is a blur of activity,
with one idea running into the next. After the initial formulaic gesture (compare
the opening of K. 339), new material occurs at bars 5, 7, 10 and 12. It is a sort of
montage technique.
That the composer is deliberately emphasizing animation at
the expense of shape is made clear in the contrasting syntax of the section from bar
15, where the phrase builds to a urry of movement after an arresting start using
unexpected dotted rhythms and syncopations. After the prior hectic activity, we hear
something distinct and memorable.
These opening urries normally take hold shortly after the beginning of a sonata.
They may create a blur of different patterns, as in K. 212 and K. 248, or they may
feature ritual repetition of a single gure. This is the case in K. 457, already consid-
ered, and sonatas like K. 194, 195, 375 and 447. Such ritual repetition invariably has
a pronounced popular character; these passages produce the sensation that we have
been caught up in something like a dance, without prior warning. In the case of the
Sonata in A major, K. 221, we are thrown off balance from the outset. The opening
presents a sort of grand preludizing with material that is hard to dene, but seems
to be a cross between a fanfare and a dance step.
It is a rhetorically memorable
version of a process by which momentum is gradually achieved by changing more
and more elements of a static repeated phrase, as in K. 457. This fascinating opening
gesture, not surprisingly, fails to return, suggesting the sort of musical living for the
moment outlined in the earlier discussion of K. 554 (Ex. 4.1).
One other syntactical feature that should be reviewed here is the three-card trick,
introduced earlier in conjunction with K. 476.
Other examples of this upward
transposition of an entire phrase may be found in K. 215, 261, 264, 268, 434, 449,
518 and 519. The relative functionality of the device varies greatly, but stylistically
it almost always carries strong popular suggestions. In bars 1744 of K. 519, for
instance, it comes across as a natural but rather un-arty device for intensication. A
similar type of patterning may be found in keyboard works by Durante and Marcello
among others, suggesting that this is also a particularly Italianate syntax (compared
with, say, the more worked manner of musica tedesca).
Thus far we have considered the ways in which Scarlatti distorts or at least defa-
miliarizes received notions of opening and closure. He also treats warily that most
characteristic medial syntactical sign the sequence. The recognition of this in the
Pestelli writes of a collage technique in this sonata, but he is presumably referring to the larger-scale juxtaposition
of different types of material, separated by rests and pauses. Review of Fadini edition, Nuova rivista musicale italiana
23/3 (1989), 462.
A fairly precise equivalent of this gesture can be found near the start of K. 484, which later has a passage with
left-hand leaps (rst heard from bar 27) that resembles bars 42ff. of K. 221.
Only Ralph Kirkpatrick appears to have isolated this device as such; see Kirkpatrick, Scarlatti, 249.
See the second movement of Marcellos Sonata No. 1 in D minor, bars 711, or Durantes Le quattro stagioni
dellanno Sonata per cembalo, ed. Alberto Iesu` e (Rome: Boccacini & Spada, 1983). Le quattro stagioni was found
in the Biblioteca Nacional in Lisbon, dated 1747.
182 The keyboard sonatas of Domenico Scarlatti
literature is almost non-existent. Only Massimo Bogianckino touches on the matter,
noting that Scarlatti is reluctant to use the circle of fths.
This neglect shows the
difculties of assessing the stylistic mixture found in the Scarlatti sonatas. Sequence
is such a familiar form of patterning that the notion that it could carry a particular
signicance in a particular context might seem inconceivable. Presumably for most
writers on the sonatas any sequences observed were in effect stylistically neutral or
invisible. The neglect is also understandable because another Scarlattian absence is
at work. Study of the keyboard works of composers such as Marcello, Galuppi, Platti
and Seixas, to say nothing of Rodrguez, brings home how markedly Scarlatti simply
avoids the standard diction of the Baroque sequence. This is so strongly ingrained a
form of patterning that it can still be found relatively untransformed at the end of the
century, which is very striking in the context of a now widely practised mixed style
and periodic type of construction. Sequences are predictable and unitary in their
forward motion, while periodicity allows for sharp and unforeseen contrast. From
this point of view, the invisibility of sequence appears to be historically inbuilt.
It would be surprising indeed, in view of preceding discussions, were Scarlatti
not to apply a little Verfremdung to such an ingrained artistic habit. Bars 6
K. 120 (Ex. 4.11) pervert the Baroque sequence by means of hand-crossing they
make it (physically) unnatural. Sequence after all is normally the most self-evident
possible form of writing, without a marked inner content; as a medial sign, its job is
to move us from one harmonic or thematic area to another. In K. 120 Scarlatti gives
the mechanism an element of startlement and creative tension through the virtuosity.
Although this is apparently more visual than aural, the difculty of execution will
alter the colour and edge of the sound. The type of Verfremdung applied here must
be understood principally in the positive historical sense of the term; it is a way
of lending a renewed brilliance of effect to a very familiar device. The same might
be said of bars 525 and 5861 of K. 22. This also swaps sequential lines between
the hands, within a narrower range. A more negative physical disembodiment may
be found in bars 847 of K. 468, where right-hand glissandi are matched most
incongruously with a descending 810 linear intervallic pattern. An extra edge is
lent to this incongruity through the same means that we saw in the second half of
the Minuet of K. 379; the passage is in F major, but the glissandi, con dedo solo, can
only be realized by passing through Bs.
A stronger sense of estrangement from the device may be found in the Cats
Fugue, K. 30. This piece, sometimes regarded as an embodiment of the composers
respect for the old contrapuntal ways, as supposedly expressed in the letter to the
Duke of Huescar, is surely one of Scarlattis supreme gestures of disdain. The coun-
terpoint is intractable and rugged. There is a hidden creative virtuosity in creating
Bogianckino, Harpsichord, 66.
See the remarks by Charles Rosen concerning Schumanns use of the diatonic circle of fths in The Romantic
Generation (London: HarperCollins, 1996), 679. Of the sequence in general he notes its physical effect, a force
of motion, as composer and listener abandon themselves to it and allow themselves to be carried along by the
energy. As we shall see, abandoning himself to the sequence is just what Scarlatti generally avoids.
Syntax 183
Ex. 4.15 K. 293 bars 8495
what Kirkpatrick calls a magnicent tangle,
in so consistently avoiding the uency
of contrapuntal ways, in sustaining the awkwardness and dissonance, but it remains
hidden. At several points, the resistance to the natural gives way, and we are treated
to the most ironically mechanical of sequences. These are heard from bars 66 and
128. Given the surroundings, however, it is these sequences that form a blot on the
piece! The rst is certainly too long, and both feel creatively slack. Rarely is it so
obvious that sequence is being held at arms length. We can also nd examples from
beyond the world of the sonatas. Degrada cites a passage from the cantata Piangete,
occhi dolenti for its deliberately bizarre treatment of the voice, featuring two ris-
ing leaps of an eleventh. He ascribes this quite naturally to the text (scorning my
sorrow), but one might also note that this grotesquerie occurs in conjunction with
an old-fashioned sequence, made still more bizarre by huge offbeat multiple-stopped
chords in both violins.
The sense of disproportion to the rst sequence of K. 30 is writ large in the Sonata
in B minor, K. 293. This work has much in common with the modest sonatas in
spite of the fact that it deals in Baroque Fortspinnung rather than a galant idiom. Ex.
4.15 gives a avour of the sequential patterns that almost completely dominate the
piece. Given this dominance, we are forced to accept them as the primary thematic
material, not as a means to an end but as an end in themselves. This represents
defamiliarization on the largest possible scale. The sense of circularity is increased
by the fact that the second half quickly returns to a literal version of the rst half
Kirkpatrick, Scarlatti, 154.
See Degrada, Lettere, 299 and 302.
184 The keyboard sonatas of Domenico Scarlatti
(compare bars 64ff. and 10ff.). After this there is an almost literal transposition of
the rest of the rst half, which is another level of mechanical reproduction.
K. 293 offers a clear reductio ad absurdum of the Baroque sequence, yet it entices
precisely because it makes us listen so differently. How, it seems to ask, do we listen
to this device, which always connotes becoming, never being; what does it do to our
sense of musical time? Scarlatti exposes again the articiality of syntactical structures,
but not of course in the sense of wishing themdead. Indeed, it must be made clear that
not all Scarlattian sequences are as loaded as those mentioned so far at bars 6063
of K. 325, for instance, we hear a neutral use of the device, as a straightforward, ef-
cient way of returning to the tonic.
The passage at bars 5764 of K. 232 is another
case of sequence apparently being used straightforwardly, here as a natural intensi-
cation of the discourse a common rhetorical role. The fact that it is surrounded
by so many exotic scales, though, may also give it the avour of a quotation.
If the foregoing sonatas suggest a sense of Verfremdung through contextual manip-
ulation, there are a number of works where the internal diction of the sequence is
impaired. In the Sonata in G major, K. 314, the signicant moment occurs from bar
90 (see Ex. 4.16a). What precedes this is the Vivaldi-concerto-type gural pattern
that Sheveloff refers to as the source of many of the vamps. This passage, beginning
in bar 70, perhaps does not quite count as a pure vamp, given the relative clarity of
its stylistic origins. However, what emerges from it is just what one might expect in
such a stylistic context a linear intervallic pattern and melodic sequence of a type
commonly heard as the climax to a passage of animation. Just at the point when the
sequence would become fully established, at the start of its second rotation in bar 94,
it is broken off, and we are quickly returned to the more popular, outdoorsy mode
that has prevailed for most of the sonata. This popular mode is back in full command
from bar 100. The normal mechanics of the sequence have been interfered with;
Ex. 4.16b offers the expected continuation, which our stylistic competence tells us
should consist of at least three complete limbs before the arrival at the harmonic
goal. It is as if the Baroque Fortspinnung needs to be reined in before it consumes the
rest of the piece. A passage in K. 427 (bars 26
29) goes one better, presenting two
complete limbs and the beginning of a third before the pattern is sucked under, as
it were, by the wave of toccata-like animation.
K. 53, a typically broad-brush work in D major, contains another telling exam-
ple of an aborted sequence. The toccata-like ourishes settle down into a motor
rhythm in the right hand from bar 23 onwards, suggesting violinismo, while the left
hand crosses back and forth. The exact repetitions of two-bar units are ripe for a
broadening-out into a sequence before any cadence point can eventuate. Scarlatti
begins to full this syntactical expectation: at bars 312 a 98 linear intervallic pat-
tern is initiated. Not only is the pattern immediately denied (and sequence is the
most automatic form of patterning with the strongest implication of continuation)
but in its stead we get four identical arpeggiated units at bars 33 and 34. There is no
violent wrenching aside of the promised pattern; it simply fades away.
Other examples along these lines could include the passages found at bars 8083 of K. 252, bars 949 of K. 359
and bars 5865 of K. 520.
Syntax 185
Ex. 4.16a K. 314 bars 87102
Ex. 4.16b K. 314: expected continuation of passage from bar 90
In the second half this material is greatly extended and the sequential impulse is
now satised. The 98 is specically realized at bars 74, 76, 78 and 80 (following
for now the Gilbert edition) and is meshed inside a larger controlling 108 pattern,
indicated on the score in Ex. 4.17. Further, we are then treated to an ascend-
ing linear intervallic pattern, the 56 at bars 825. A good example of a typical
186 The keyboard sonatas of Domenico Scarlatti
Ex. 4.17 K. 53 bars 7295
non-congruence of patterning is found at 856: in bar 85 the a
the 6 of its pattern is met by an A in the bass which breaks the thread, although
such a means of bringing a pattern to a halt is quite common. At bar 86, though,
the melodic 56 continues, even though the bass line indicates that we have moved
on to a new phrase. Only at 889 does the right hand catch up with the bass, so
that harmonically there are four repeated two-bar cadential units (from bars 86 to
93), while thematically (including the precise shape taken by the bass line) there are
just three. This shows a considered management of phrase rhythm in the name of
avoiding square syntax, especially given the overt regularity of all the pieces basic
units. In addition, the exact repetitions of bars 8893 are thrilling in context, coming
as they do after so much sequential and manual ddling.
For all the sequential fullment of the second half, though, some sense of es-
trangement remains. This is strengthened very considerably when the full source
situation is considered. Gilbert and Fadini both do some tidying in different ways.
Syntax 187
On the second minim beat of bar 76 in the right hand, the fth quaver of the bar,
V and P both give g
. So does the new Lisbon source and all other sources. Gilbert
changes this to an f
so as to form part of the 98 succession previously discussed;
Fadini respects the V and P reading here but then changes the fth quaver of bar 80
to an e
so as to create symmetry at another level, bars 74/78 and 76/80 forming
matching pairs. (Only M and W support this change.) Perhaps the reading most in
the spirit of distance suggested above would be to follow just what is given by V and
P: in this way expectations are met but not to the letter. One is always treading on
thin ice in such instances, ascribing intentionality to details that may simply represent
a difcult source situation. Whatever the merits of individual cases, though, there
can be no doubt that the larger image of the composer allows one to defend seeming
anomalies with particular conviction.
Sequence is also used by Scarlatti in a fairly standard role as a means of rescuing
the sense of musical process, or as a sort of safety valve. We have seen how in K. 523
(Ex. 4.6) it was a good friend in a crisis. The associations with the technical re-
spectability of an older style are here exploited as eagerly by Scarlatti as by other
composers, but always with the proviso that his mixed style tends naturally to sharpen
the edges of its constituent elements. The Sonata in C minor, K. 116, is one of those
works that seems to contain clear approximations to amenco vocal technique. The
sequence from about bar 84 in the second half seems to be used as a means of re-
laxation by recourse to traditional technique, buying time before the next frenzy.
A sequence is also used to loosen the hold of the exotic in K. 242, in bars 737.
It responds to the primitive sequences of parallel fths heard earlier in the second
half by retaining the basic material and organizing it into a civilized 105 linear
intervallic pattern. Other uses of this device in extremis include K. 181, bars 6569
K. 429, bars 3640
, K. 371, bars 7884, and K. 57, bars 1468.
The Sonata in F major, K. 195, presents early in its rst half an extreme form of
opening insistence, a huge expansion of what was originally, in bar 7, a ller tag.
This gure is heard in twenty-one consecutive bars, during which the composer
plays around with the ne print of its diminutional structures to achieve a high
degree of ambiguity and dissonance. The long-winded linear intervallic pattern
that follows from bar 28, in simple parallel tenths, could be construed as a gesture
of mock frustration, ushering in a toccata idiom that dominates the rest of the
half. Its simplicity cleanses all the nagging complications of what went before. A
preposterously long sequence heard from bar 84 then easily outdoes that of the
rst half; it is just as outlandish as that examined in K. 39 (Ex. 1.1). Whether one
chooses to hear it as satirical exaggeration or sheer exuberance, there is no doubt
that the pattern outlasts its functional utility.
Although seemingly introduced as a
Such patterns are found in a number of sonatas. In K. 517 in D minor the second-half extension of the simple
sequence from the rst half, at bars 827 and 98103, turns an unremarkable three-bar pattern of descending
tenths into ve bars. The sequence now surely goes on for too long, but without apparent satirical import.
Rather, given the Prestissimo tempo, it seems to emphasize the irrational aspect of a speed that will resist any
rhythmic differentiation, that wants to consume all in its path.
188 The keyboard sonatas of Domenico Scarlatti
rescuing device, it becomes disproportionate in its own right. And so we return to
the Verfremdung of the sequence.
If many of the manipulations of phrase rhythm detailed above have been read as the
expression of a highly relativistic and critical creative spirit, there is another level at
which such operations may be understood. They form part of an all-encompassing
passion for musical movement in its own right, for the study of momentum, and for all
the patterns and mechanics of syntax. They involve an investigation of different ways
of experiencing time, space and movement. To claim that such a predilection helps
to dene an essential aspect of Scarlattis art seems unconvincing on the surface; is
not music in general and by denition naturally prone to dispense patterns in sound?
Even where composers seem to show no direct consciousness of such properties,
surely we are led to contemplate them quite independently of the particular manner
in which they are realized. What distinguishes Scarlatti in this respect is the sheer
intensity of his gaze. This intensity is aided by the conciseness of his structures. By
turning away from the possibility of more extended keyboard forms, the composer
was able to avoid the need to spread his invention more thinly; he could place
patterns under the closest of scrutiny. To identify this spirit of intense scrutiny, it
would be instructive to begin with works that do not appear to contain any of the
familiar distortions. The Sonata in G major, K. 14, represents a sort of music that
sets out to give pleasure through the neat, almost irresistible, symmetrical expression
of its shapes and phrases. This extends, as often in the Essercizi, to rhyming closes at
both ends of each half, so that between 18 and 19 we have a perfect mirror effect.
This is the dinkiest of many dinky moments in this sonata. The rst and last bars
also mirror each other. With all its matching patterns, K. 14 eschews surprise and
estrangement and instead delights in the pleasure of recognition.
Such pattern-making might seem hard to square with what we nd in most of
the sonatas. Scarlatti seems to move from an extreme of symmetry (or geometry) to
something nearer the other end of the spectrum. Yet if patterns are more commonly
broken than straightforwardly outlined, there nevertheless must rst be a conscious
recognition of their existence and a preoccupation with the way they unfold. In this
larger sense both K. 14 and its apparent opposites may t under the broader rubric
of intense syntactical exploration. After all, the neatness of a sonata like K. 14 also
shows an obsessive side.
Another work suggesting that sheer fascination with syntactical patterns weighs
at least equally with a critical realization of them is K. 257 in F major. Although
A good example of this would be the rotation dened by Farhad Abbassian-Milani in his study of the Essercizi,
Zusammenh ange zwischen Satz und Spiel in den Essercizi (1738) des Domenico Scarlatti, Berliner Musik Studien
9 (Sinzig: Studio, 1998). This circling movement using readily repeatable shapes is especially favoured in the
Essercizi but is hardly unknown elsewhere; compare the following discussion of K. 257. For a denition of the
term, see 145.
Syntax 189
a kinship with the toccata has been claimed,
the contained nature of its gestures
perhaps gives K. 257 more the avour of an invention, certainly in its initial phase.
In keeping with this generic suggestion, it continues to use the opening gambits as
a point of departure to a greater extent than is immediately apparent. The opening
leap up of an octave followed by the fall of a ninth is incorporated into the bass
line from bar 15 and is a constant presence thereafter; the tag is made to do service
as an agent of parallel sequential motion. The most fundamental shape, though, is
the falling third in the rhythm . It rst appears at bars 46
in falling sequence;
we might expect to hear a third sequential limb, but instead the need for space to
prepare a satisfying cadence asserts itself. Of course the sequence has done its job
harmonically after two bars by returning us to I, but the material has syntactical
implications that are not fullled. The same occurs at 1213, but with the right
hand rearranged to emphasize the parallel sixths/tenths with which the second basic
shape will henceforth be associated. Bars 1314 simply rewrite the preceding pair
of bars; the broken parallel fths in the right hand are not really to be thought of as
improper, since this is a fairly common type of keyboard guration in the eighteenth
Bars 1517 feature another rewriting, with the hands essentially swapping
parts, but now the sequence extends for a more natural three bars.
Indeed, K. 257 has strong circular tendencies. With its constant recycling of
material, we never seem to arrive anywhere, and all this material is connective and
sequential. This apparent lack of progress is reinforced in the rst half by the fact
that from bar 8 to the double bar we never leave V.
Ironically, the agent of this not-getting-anywhere is the sequence, the most di-
rected propulsive device there is. The sonatas obsession with the mechanics of
movement to the detriment of any marked inner content may be taken in the spirit
of fascination outlined earlier, but it might also suggest a droll parody of the art of
Fortspinnung, chopped up into small units. From bar 19 we hear a return of bars
1314 in the minor, but these now occur twice as if to prolong the pattern-making.
Bars 234 are certainly more distinctive, but more clearly than anything else heard
so far they represent a transition. This leads us on to more of the same, as bars 1518
are repeated directly at 258. In another context, bars 2931 would make an effec-
tive, unbuttoned closing unit, but they are heard here as another recycling. They
vary the material of 257, not just in the obvious thematic sense but also in pitch
structure. The two lines are simply swapped around. In addition, though, bars 29
to 32 correspond almost exactly to the pitch content of bars 1316, a relationship
that adds to the sense of circularity. The closing right-hand units then work in the
opening gambit note the rise of an octave from c
to c
followed by a fall of a
ninth to b
outlined at bars 334. Thus even this very typical closing phrase is of a
piece with the preceding material. We seem to be in a hall of mirrors.
See Chambure, Catalogue, 99, and Pestelli, Sonate, 169.
See Paul Mast, Brahms Study, Oktaven und Quinten u. A.: With Schenkers Commentary Translated, The Music
Forum 5 (1980), 545 and 11621, for examples, and 186 for an explanation as to how Brahms might have seen
such passages.
190 The keyboard sonatas of Domenico Scarlatti
At the beginning of the second half, in bars 389, the composer unusually com-
bines versions of two separate phrase units a version of bar 25 reverses into a version
of bar 24 as if demonstrating that they are even more alike than we thought they
were. This is made clear in the next two-bar phrase unit, which answers the rst with
a transposition of bars 256. Note too how he begins the second half, wittily, with
a rising octave shape, thus explicitly conjoining the opening gure with the later
material. The music is becoming still more uniform! It is also unusual for Scarlatti
to make the return to the tonic at this point a very common and underappreci-
ated part of eighteenth-century binary (and sonata) forms quite so prolonged and
secure; this again contributes to the deadpan avour of the work. Then we hear six
bars of the holding gure, a logical progression from the previous two (1314) then
four (1922). The fact that two bars of major are followed by four bars of minor here
replicates the order found in 1314 then 1922, in another conjoining of previously
separate events. The D minor version of the main melodic sequence of the rst
half, from bar 48, again hooks into earlier realizations all of the right-hand lines
in these passages occupy very much the same registral level, between about a
, so increasing the sense that we are endlessly revisiting familiar ground.
From bar 52 the original transition passage of bars 234 returns to its minor
coloration after the major-mode version at the start of the second half, and its
ensuing treatment at long last gives us some harmonic colour, a sense of progression
and a freer left-hand part. This is the one moment of freedom in the sonata, proving
by inversion that the repeated patterns found everywhere else are not as innocent as
they might appear. However, the complete passage from 52 to 58
is controlled by
another three-part descending sequence; there would seem to be no escape.
Bars 668 retain the right-hand pitches of bars 1517 and 257 when they return
to this material, instead of transposing them, thus making the circularity very clear.
The parallel phrase from bar 70 then does transpose the original material. However,
bars 7073 now generate their own matching unit. Bars 74ff. have no equivalent
in the rst half; they decorate the previous phrase, so that in the second half we
now have ve full or partial versions of the same melodic sequence (from bars 40,
48, 66, 70 and 74). The right-hand decorations at 74 and 75 and the breaking of
the pattern in the next bar suggest in their playfulness a small concession to our
need for some thematic variety. This is also an appropriate gesture of relaxation as
we approach the close of the sonata. The last three notes of the piece in the right
hand are a reminder of our basic shape; they are not present at the end of the rst
half. After the endless hearings of this falling-third gure, this nal version delivers
us from the prospect of a continuation. Making a cadence point thematic in this
way, with its consequent structural twist or correction of something heard earlier, is
a clear piece of structural wit. Although K. 257 uses Baroque stylistic features, the
playfulness, distancing and awareness of redundancy of speech articulate the concerns
of a supposedly later idiom. Indeed, it is in such a work that Scarlattis kinship with
Haydn is most plainly revealed. K. 257 works in the Haydnesque spirit of making
Syntax 191
something out of nothing, with the same popular tone that masks the wit of the
It is important to insist on the compositional and artistic integrity of a work like
K. 257, since it could easily fall victim, along with many similarly uneventful
sonatas, to a prevalent image of a shallow, digitally inspired vitality. Giorgio Pestelli
is one who has difculties with such mechanical works, those that do not exemplify
his theatricality or musical spectacle. If K. 257 recalls the issues raised in the
discussion of the modest sonatas, then Pestelli clearly feels boredom rather than
When nothing happens in a Scarlatti work, then it lacks his special poetry and is merely
a document of keyboard technique . . . Scarlatti was not able to be impassive, detached and
ascetic in the face of his musical material; unlike Bach, he did not have a passion for thought,
he was not a reasoner in music . . . Without musical spectacle, his most worldly art has very
little signicance and ends up running dry.
In other words, to quote Dales summary of this position, Scarlatti is temperamentally
incapable of writing abstract music for the keyboard and needs a strong outer
stimulus for composition.
This alleged incapacity for abstract thought is based on
a conception of the art of music that we reviewed at the start of this chapter. Depth
and abstraction, as exemplied by the talismanic gure of Bach, are to be realized by
harmonic and contrapuntal means; Scarlattis syntactical exploration cannot even be
conceptualized as a possibly equivalent category. Yet this exploration is both deep
in the concentration the composer brings to the task and abstract in that we are
provided with very little in the way of concrete thematic work or harmonic argument
or variety of texture that might interfere with our contemplation of the syntax. Of
course it is this particular type of abstractness, focussing on the wrong parameter,
that encourages such interpretations as Pestellis; it is all too easy to see only empty
guration and an apparent expressive indifference. The lightness of touch partly
issues from a certain disdain for high seriousness that was emerging as a modern
artistic stance.
This can also deect us from the intensity of musical thought, which
in Scarlattis case can be as much around as in the given work. This intensity is also
evident in the very fact that Scarlatti is able to abstract his music so exceptionally
from syntactical habit, those means that have become so ingrained they are often no
longer part of the conscious compositional process. All this is achieved, as Henry
Colles wrote of Scarlattis repetitions in general, with his eyes open.
Another way of yielding to the hypnotic effects of patterns while also being
distanced from them is to create a disjunction between implied and actual syntax.
We have already seen this in the opening unit of K. 523 (Ex. 4.6), which implied
Pestelli, Sonate, 198; Dale, Pestelli Review, 1867.
William Weber describes this as a sense of propriety that abhorred speaking in excessively serious terms. Did
People Listen in the [Eighteenth] Century?, Early Music 25/4 (1997), 683.
Colles, Sonata, 895.
192 The keyboard sonatas of Domenico Scarlatti
a symmetry that was withheld and then granted by degrees over the course of the
whole sonata. We have seen it too in the contradictory aspects of the modest sonatas.
An example of this is K. 323 in A major. It shows how even the most mundane
surface can conceal hidden terrors. In his edition of the sonata Howard Ferguson
counsels the player to note the irregular phrase lengths. All but the rst begin on
the half-bar, thus: 1st half, 5 1/2 + 3 + 4 + 3 + 3 + 2 + 5 + 5 + 2; 2nd half,
2 + 2 + 6 + 2 + 5 + 5 + 4 + 4.
In our syntactical terms, this is a really extreme
constructivist piece of writing; an idiom that promises to be light, airy and gratefully
divided into equal phrase units is treated both mechanistically and ambiguously.
Then there is the tension caused by the continual pull against the bar line, plus the
fact that K. 323 contains no rests whatever we nd a continuous texture from
start to nish. Certainly Gilberts suggested half-bar rest before the return to the rst
half, disappointingly conrmed by Ferguson in his edition, is undesirable from this
point of view indeed, anomalous by the terms of the piece.
What Ferguson does
not mention is the high degree of overlapping of phrases that creates this suffocating
syntax and texture. An instance of this may be found in bar 37, where the rst half
of the bar seems to end a two-bar unit as it parallels the three right-hand quavers
of bar 35. On the other hand, by analogy with the sequentialmotivic pattern that
unfolds in the ensuing bars, bar 37 is an indivisible melodic whole.
After all this ambiguity, there is a form of resolution at the end with two nal
four-bar units. Phrases is denitely not an appropriate term here, nor is it anywhere
else in the work. Arguably the sonata consists of just two phrases, if we bear in mind
the denition given by Roger Sessions that a phrase is articulated by a measure of
letting go.
If we then bear in mind the half-bar at the end of the rst half and
its effect on the performance, both in moving back to the beginning of the rst
half a beat too early and in moving immediately on to the start of the second, one
could easily conceive of the work as comprising just the one large phrase. This is
particularly remarkable, and radical, when we consider the miniaturistic nature of
the units that make up the language of the sonata. A work that might promise to
conrm all our worst prejudices about the impoverished nature of mid-century
style and its keyboard writing reveals a fundamental contradiction between syntax in
the small and in the large. Scarlatti denies the material its natural expression there
is something akin to Stravinsky about this process.
The Sonata in G minor, K. 111, suggests a very different style. It has a certain
Baroque darkness of tone; apart from a few bars of relative major early in the second
half, it is all in minor coloration. Incredibly, forty-one of its fty-ve bars feature
the same gesture, based on a falling arpeggio introduced in bar 1. Because of the
Scarlatti: Twelve Sonatas (Easier Piano Pieces No. 57, London: The Associated Board of the Royal Schools of Music,
1986), 28.
This is an example of the missing half-bar problem(K. 305 offers another example), which in turn has implications
for the missing-(whole-)bar phenomenon altogether. See the discussion of this feature in Sheveloff, Keyboard,
Cited in William Rothstein, Phrase Rhythm in Tonal Music (New York: Schirmer, 1989), 3.
Syntax 193
retention of this initial gesture, each repetition refers to the beginning, so that we
hear an endless series of openings. At the same time the bar 1 material is also clearly
a closing shape, as we can see from the cut of the bass line, the falling contour of the
right hand, and the cadential trill on the fourth beat. It would be very easy to imagine
bar 1 as the penultimate bar of an entire piece, being followed by unison Gs. This
reading is claried by the adaptation of the opening in such places as bars 11 and 37,
both suggesting a full close which is then denied. Because of its placement within the
whole structure especially in its most characteristic bar 5 form, where the left hand
takes over the arpeggio the material in fact also functions as a middle. Thus it is
caught between three possible syntactical functions, those of opening, continuation
and closing. This generates a mood that is both trance-like and distracted.
K. 111 is denitely a unied piece that is uneconomical to listen to; its rhetoric
may well derive from a twisted take on the Baroque exhaustion of an idea, but
parody is not necessarily suggested. The result is difcult to read; the effect hovers
between fascination and boredom, between pleasure in and disgust with the sonorous
material of the musical world. Much of the literature has tended to pass off all sorts of
repetitive practices in the sonatas as simple exuberance, but there is also an element of
compulsive, obsessive behaviour, particularly given the rather forbidding tone of this
particular work. This is most apparent in the mad voice leading of the parallel-fths
chords at 30, 32 and 34, very similar in form, sequential treatment and structural
placement to those found in the irrational K. 541 (Ex. 4.8).
The hypermetrical manipulation found in K. 323, K. 111 (not discussed above)
and so many other works is perhaps the key factor in creating that very distinctive
feeling for movement in the Scarlatti sonatas, evoked by many writers but rarely ana-
lysed. In this respect at least the composer may indeed be compared with Beethoven
in offering a very marked and readily recognizable rhythmic style. Although, as we
have seen, the composers syntactical awareness can take many forms, there is one
particular avour that stays in the mind. Cesare Valabrega described it as restlessness,
an agile and nervous mobility, while for Sacheverell Sitwell it consisted of an
alliance of rapidity and humour. Scarlatti, he wrote, has the alert nerves of someone
who is used to trafc. No one who has passed his life in the country could have
written the music of Scarlatti. He has no time to waste, and makes his points as sharply
and rapidly as a jazz composer.
The comparison with jazz, already suggested in
this chapter, is one of the best means available to grasp this rhythmic avour, full
of irregularities to an extent that few performers seem to realize. Kirkpatrick, who
also evoked this comparison,
gave some valuable advice to the player which rarely
seems to have been heeded. Of K. 105 in G major, for instance, he wrote that
it has
a supercial note picture that gives the impression of a predominantly homophonic style
(unfortunately borne out by Longos phrase markings), yet this sonata, like so many of the
others, has all the rhythmic polyphony of the Spanish dance. Almost nowhere in the piece
Valabrega, Clavicembalista, 213; Sitwell, Background, 152 and 1367.
Kirkpatrick, Scarlatti, 187.
194 The keyboard sonatas of Domenico Scarlatti
should accents fall simultaneously in both voices, nor has the bar line any function other
than that of indicating a basic meter that has already been established by the network of cross
accents between the two voices.
The supercial note picture may suggest not only homophony but also a regular
hypermetre, as we have seen with K. 323. Such features do not simply take care of
themselves in an accurate reading; they need conscious advocacy. It is this frequent
obliqueness of rhythmic style that makes jazz a good imaginative model for the
realization of such effects in Scarlatti.
One part of the avour of agile and nervous mobility produced by Scarlattis
treatment of patterns involves sheer speed. In K. 386 in F minor, discussed in Chap-
ter 3 as an example of stylistic fusion, the vivid sensation of speed is achieved less
by the Presto tempo marking than by the unpredictable manipulation of motive and
phrase. Understanding such manipulation helps us answer the perennial question of
why repetitions can sound so exciting in the hands of Scarlatti and yet can appear so
square in the hands of others. The working of a basic two-bar module from bars 8 to
19 illustrates this. Although bar 8 clearly begins a new section, delineated by the rst
cadence of the sonata, bars 89 function not just as a new idea but as a variant on
bars 67. This is most apparent in the near identity of the bass lines at 7 and 9, but
may be traced in all the material; the right hand in bar 8, for instance, elaborates the
same c
line as bar 6. Such a blurring of boundaries between sections already
aids the moto perpetuo feeling that is being developed. Bars 1011 make as if to
repeat 89, but halfway through turn into a transposition up a third, leading us to
the mediant. Bars 1213 then seem to present a complete mediant replica of 89,
although the rst right-hand note of bar 12 indicates that bar 10 is the model. The
complexity of cross-reference continues in bar 14, which seems to begin a repetition
of the previous two-bar unit in the same way that 1011 promised to. However, the
right-hand part of bar 15 departs from the expected shape. From this point of view,
the patterning seems to operate in two-bar cycles comprising bars 78, 910, 1112
and 1314, cutting across the two-bar hypermetre and demanding that the listener
process more information more quickly than would have been expected.
At the same time, the left hand in bar 14 has already departed from the anticipated
model; it takes its syncopated rhythm, and the ensuing stepwise descent, from bars
56 in the right hand. Bars 1617 then present the rst precisely aligned reiteration
of material, with their transposition up a step of 1415. The greater directness
of patterning here acts like an acceleration after the previous manoeuvres. Bar 18
presents a further sequential transposition up a step, but at the same time the left
hand reverts to source, transposing the original bass line of 89 to the dominant
minor. A twist from the last crotchet of bar 19 leads to an unexpected half-close at
the start of bar 20. Meanwhile the right hand has also broken the mould, rushing
towards this cadence point in undifferentiated falling steps. In other words, after the
Kirkpatrick, Scarlatti, 303; some instances of this network of cross accents in K. 105 are given in the following
discussion on 304.
Syntax 195
accumulation of nervous energy, making us edge ever further forward on our seats,
the music denies the gratication that would come from a rm cadence point. It
rushes us ever onward. The treatment of the chromatic scale that follows is also
telling. Scarlatti hardly ever gives us a complete chromatic scale collection in his
sonatas, and the omission of various steps in this example furthers the sensation of
impatient speed.
Even more in the second half of K. 386, the music feints in various directions.
Broadly, we seem to be hearing the same material as in the rst half, but the precise
direction of the journey cannot be foretold. Its reworkings offer an exhaustingly
rapid rate of events; they demand immediate readjustments of perspective on the
part of the listener. If we accept the relatively high speed of most of the Scarlatti
sonatas, then a work like K. 386 makes clear that this is not just a physical attribute
mental speed is just as much a determining factor, both for the composition and the
perception of such works.
Perhaps the most exciting moment of all arrives at bars 7981, when the second
limb of the second subject is reduced to pure pulsation, with undifferentiated quavers
in the right hand and minims in the left. The immediate repetition of a one-bar
unit, as at bars 478 and 5860, may be thought of as a holding action. However,
it also suggests the primacy of a pure rhythmic impulse over any of the localized
material which maintains it.
This exibility of pacing is a key element in Scarlattis kinetic art. A comparable
moment occurs in K. 96. After all the detailed inections of material earlier in the
second half and the panoramic changes of imagery throughout, bars 16580 clear the
air through a straightforward oscillation of tonic and dominant. The passage looks
nothing on the page but is brilliantly conceived in context. Although a variant of bars
7893 in the rst half, it stands apart through the consistency of its rhythm and tex-
ture. One might hear timpani strokes in the bass here among other possible references,
but the real topic here is propulsion pure and simple. Its all in the timing.
Such timing is also the hallmark of a comic art, an aspect we have hardly touched
on to this point. In the Sonata in D major, K. 45, a uent and easy toccata style is
interrupted in bar 12 by something very exotic (see Ex. 4.18a). The exoticism lies
in the scale forms used (with a descending tetrachord in the bass, extended by step
upon repetition in bar 14) and the alla zoppa rhythm caused by the strange dragging
imitation between parts. The voice leading is hardly ideal and the syncopations are
far from consistent note the very disconcerting and unnatural pause on the fth
quaver of bar 13. The passage is quite rewritten in the second half (see Ex. 4.18b); the
descending right-hand line from the rst half is reversed and becomes chromatic, for
example. On the second playing of this there is a further variation, with the hiatus
on the tenth quaver of bar 32 being even more awkward.
There is also very little
space between the two manifestations of the passage one beat compared with one
bar in the rst half. In addition, there are just two versions of the limping progression
Note that Fadini reads this differently, and there is also a problem with placement of the tenor a in her bar 31.
196 The keyboard sonatas of Domenico Scarlatti
Ex. 4.18a K. 45 bars 1215
Ex. 4.18b K. 45 bars 3033
instead of the total four from the rst half. Nothing could be less appropriate to the
character of the interruption than a literal repetition of the rst-half form. (K. 419
in F major has a similar feature, but one that is less disruptive.) So the composers
changes here, while seemingly perverse, are quite logical in a way, if we apply the
rules of comic timing. An interruption heard twice identically in each half becomes
an established feature rather than retaining its disruptive force. Also logical is the
fact that in bar 32 the interruption now interrupts or cuts short the intervening
normal cadential close we were expecting by analogy with the rst half (in bar 15).
Thus the surprise surprises anew in the second half.
K. 45 and all the works reviewed in this chapter so far demonstrate an inti-
mate understanding of the effects of syntactical patterning, whether wrought by
Syntax 197
under-, over- or non-repetition. How can we apply our awareness of these factors
to the vamp, the most upsetting and seemingly inorganic feature of Scarlattis style?
More than ever when attempting an overview of aspects of a composers style, the
very ordering of the following sonata sections under the category of vamp can dis-
tort their signicance. Every passage of this sort carries such a particular charge that
any label not only mutes their individuality but gives the misleading impression of
a more or less systematic stylistic feature. Any sense of collective identity must seem
especially weak when each vamp presents itself as such a unique, and often seem-
ingly inexplicable, interruption, as a possibly anarchic force. An obvious analogy
would be with the development section of a sonata form, when any recognition
of a distinct category conicts with the particular freedom of realization that is the
developments raison d etre. Yet although this comparison is appealing, as will be ex-
plored below, it skirts the central question, which is one of functionality. Whatever
their various freedoms, developments can be assigned various well understood roles
within the larger argument of a movement. With vamps, on the other hand, it is
often unclear whether they have any functional basis at all. Must they necessarily
relate to the specic context of the sonata within which they occur, or are they
rather self-satisfying, simply to be understood as aberrations from normal compo-
sitional service? They would often appear to be underdetermined by the particular
Such questions must be understood to involve rhetorical as well as structural co-
herence. The vamp of K. 193, for example (see Ex. 1.4b), may seem to have a clear
functional role in the structure of the work, but a close analytical reading could miss
the larger rhetorical point that such ends could surely have been achieved less ob-
trusively. Like all members of its putative species, this vamp seems disproportionate
in affect. Having found points of contact with surrounding material, one suddenly
draws back in realization of its disembodying qualities. What may become disem-
bodied is not just the surrounding material as when the vamps of K. 260 make the
normal seem unreal but ones whole sense of musical time. As has been suggested
already, such sections seem to live for the present, to know nothing of the reection,
distancing and control that allow for the generation of intelligible musical syntax.
They represent a species of what Jonathan D. Kramer calls vertical music, which
denies the past and the future in favor of an extended present, giving us the means
to experience a moment of eternity.
If it is a moot point whether a vamp may be
understood teleologically, such uncertainty must also encompass explanations as to
the internal form and extent of these sections. What are we to make of their often
grossly ungrammatical harmonic syntax? Are their proportions precisely calibrated
or, again, does such a question miss the point? Another difculty lies in assessing
the stylistic coherence of the vamp. Several possibilities have already been advanced:
that, as revealed by K. 532, such behaviour may derive from folk models, where
repetition of course carries a different signicance; that they take their cue from the
Jonathan D. Kramer, The Time of Music: New Meanings, New Temporalities, New Listening Strategies (New York:
Schirmer, 1988), 3756.
198 The keyboard sonatas of Domenico Scarlatti
free solo sections found in Vivaldi concertos; and, more radically, that their generally
athematic guration and decontextualized harmony place them outside the realm
of what is commonly understood by style altogether.
Those writers who have conceptualized the vamp as a category, or at least rec-
ognized it as a feature worthy of some comment, have normally sought stylistic
explanations rather than attend to the vamps troubling implications for structure
and rhetoric. A clear exception is Eytan Agmon. In his account of the vamp of
K. 319 he considers its instability to reect the higher[-]level instability of the dom-
inant prolongation that underpins this central part of the form.
Such a structural
interpretation might very reasonably be extended to many other vamp sections,
but Agmon then offers too ready and unsubstantiated an assurance of the stylistic
cohesion of the vamp with the rest of the work. Among those who seek stylistic
explanations in folk models are Ann Bond and Frederick Hammond. For Hammond
such passages have a clear choreographic analogue in Spanish dance, being animated
by a rhythmic pulse rather than by a directional movement;
this implies that our
functionality would be located more in the source than in the applied context.
Bond describes such sections as a peculiarly Iberian feature and likens them to
magical voyages through kaleidoscopic sequences of keys, in which our sense of
forward movement is suspended, under the trancelike inuence of these seductive
Although the type of inuence proposed here is more spiritual than
practical, Bond, like Hammond, suggests a lack of directional thrust and hence a
relatively weak sense of functionality. For Barry Ife, on the other hand, these sections
surely bear the mark of Scarlattis personal improvisatory style.
Even bearing in
mind the limitations of the concept of improvisation, as discussed in Chapter 2, it
is undeniable that vamps often give precisely the impression of being extemporized.
Yet they seem ultimately both too wild and too restricted to be accounted for un-
der this rubric. Who, after all, would improvise in this idiot fashion? Improvising
normally connotes variety of material and gesture rather than the monomania that
the vamps by denition display. Such an explanation also fails once again to account
for the place of the vamp in a wider rhetorical scheme. Why should the composer
choose to give the impression of an obsessive improvisation in the wrong generic
Pestellis account of the phenomenon combines Ifes rationale of improvisation
with a grounding in Baroque aesthetics. When he refers to the fatiguing experiment
that left its traces in Scarlatti, Pestelli surely has the vamp in particular in mind.
Such passages were not contrived, however; they owered under the composers
improvisatory ngers. Elsewhere the author suggests a more polemical slant to such
wandering expansions: they represent a return to the tradition of the toccata,
in other words, a denial of galant simplicity and sociability. This makes the vamp a
conservative feature both stylistically and even aesthetically, for all its extravagance
Agmon, Division, 4.
Hammond, Scarlatti, 178.
Bond, Harpsichord, 183.
Ife, Scarlatti, 21.
Pestelli, Sonate, 19 and 52.
Syntax 199
of affect. The Baroque model invoked by Sheveloff the exploratory solo section
of a Vivaldi concerto is not framed in the same manner.
As well as looking back, it is possible to look forward when trying to ground the
vamp historically. Rosen refers obliquely to the vamp technique in a discussion of the
slow movement of Mozarts Sinfonia Concertante, K. 364: although this movement
is written in archaic sonata form, meaning that the second half contains no distinct
development and recapitulation sections, a feeling of development is achieved as
in the sonatas of Scarlatti through the detailed intensity of the modulation.
have already indicated the difculties of aligning vamps with development sections,
but it remains an attractive comparison and one surprisingly little explored. The
link is especially plausible if we concentrate on the rhetoric of development sections
of the later eighteenth century, before an intensive reworking of thematic material
became the standardized procedure. Like most vamps, development sections of this
time offer a point of greatest rhetorical and technical freedom in the middle of
their structures; they are typically more repetitive and less obviously rational in
their syntactical organization than the framing material. Unlike nineteenth-century
development sections, they may well concentrate on pure harmonic exploration,
realized through free guration, so that in thematic terms they form an apparent
interlude. We should bear in mind that this middle section was often given some
such name as free fantasia by theorists of the time, without the moral imperative
to a careful husbandry of thematic resources implicit in the term development.
(In practice, such free developments may contain some thematic references or
residues, although these tend to remain around the edges of the section.) Although
examples of such an approach may be found in all genres in the eighteenth century
(for instance, in the rst movements of Clementis Sonata Op. 25 No. 6 and Haydns
String Quartet Op. 33 No. 4), perhaps the most ready association for many listeners
would be with the rst movements of Mozarts piano concertos, and the arena of
improvisation frequently found at the mid-point of the structure. This is led by the
soloist in non-thematic guration, often arpeggiated, and supported harmonically
by the orchestra. Indeed, it would seem to be concerto form itself which provided
the historical precedent for this type of developmental texture in sonata forms.
This in turn gives greater depth to Sheveloffs analogy with the solo sections in
Vivaldis concertos.
When we pursue the concerto connection, however, the analogy between the
vamp and this type of development starts to weaken. The guration found in vamps
can only rarely be understood as any sort of virtuoso display, even though it does
retain the physically effortful quality found in the concerto(-type) examples. Rather,
Another, more abstract, stylistic ingredient might be recitative. Although not making any direct connection
with Scarlattis practice, Michael Talbot suggests that Baroque recitative might have been the cradle of radical
techniques of modulation that did not nd general application until the development sections and transitions of
the Classical age. How Recitatives End and Arias Begin in the Solo Cantatas of Antonio Vivaldi, Journal of the
Royal Musicological Association 126/2 (2001), 174n.
Rosen, Classical, 215.
See the account in Rosen, Sonata, 8994.
200 The keyboard sonatas of Domenico Scarlatti
one returns to the contemplation of Iberian avours, as suggested by the vamps of
K. 193 (Ex. 1.4b) and K. 319. Even those vamps that do proceed from the basis of
concerto-like guration, as in K. 253 and K. 409, seem ultimately to transcend such
an expressive purpose.
If vamps are only awkwardly and partially assimilable into any historical or stylistic
context, they are at least as enigmatic when we try to account for the role they play in
individual sonatas. The following discussions attempt to determine some functional,
organic rationales for a vamps appearance in a particular context. We must always
bear in mind that the disruptive rhetorical force carried by such a passage may
render ineffectual any formal explanation. This contradiction was apparent in our
examination of the vamp in K. 193. Its apparent role as a sort of melting pot for
tensions exposed elsewhere in the sonata, or as a problem-solving device, can be
proposed for a number of other works.
The vamp which begins the second half of the Sonata in B major, K. 244, is one
of the more insistent members of the species, repeating twelve times a gure that is
specic enough in shape to seem thematic. However, it is new, although the context
of repeated two-bar units and the contours of both hands suggest bars 15ff. from the
rst half, a passage which itself almost carries the status of a vamp (its placement makes
it more akin to a stampede). The similarities in pitch of 1518 and 658 suggest
that both passages proceed from the same basis. Indeed, the vamp really usurps the
role of bars 1534 of the rst half, for when this material returns from bars 93 to
102, it is much more clearly directed and contained harmonically, outlining the tonic
minor by means of a fth-progression in the upper voice and a sixth-progression
in the bass (f
and b
respectively). It has become functional. There is a
clear irony in the fact that the vamp enforces a new, less disruptive character on the
rst-half material, but, in so doing, it in turn disrupts the larger structure. It solves
one problem and creates another.
K. 485 in C major seems to represent a clear case of the vamp coming to the
rescue. One would never guess from the galant opening, which uses the Couperin
pastorale schema, that this sonata would turn out to have the widest range of any
Scarlatti sonata: from F
to g
. One associates the galant with a narrow pitch range,
both of melodic and bass behaviour, yet the texture and sense of spacing here are
unmatched by any other Scarlatti sonata. The nearest equivalents, both also in C
major, are K. 356 and 357. The writing is full of wide intervals and couplings in
octaves, and there is generally a hole in the middle of the texture. This is summed
up by the extraordinary closing gesture, which revives the bass ller heard earlier
(every two bars from 5 to 13) and features both hands playing it two octaves apart.
The fact that it moves up two octaves, in the right hand, before moving back down
again to the same point, in both hands, increases the hollowness this is, as we have
seen, just the sort of cadential padding that the composer normally shuns at all costs.
There is also a lack of ne detail in the individual sections and the larger structure
everything is blocked out rather coarsely and, one suspects, parodistically. Indeed,
after the opening phrase of bars 15
, every phrase unit is repeated exactly until the
nal ourish. The harmonic plan also seems pointedly perfunctory.
Syntax 201
Beginning in bar 34 of the second half, the vamp then breaks down this mechanical
syntax, using the broken-octave gure from bar 13 that was perhaps the rst sign of
rebellion, in its anti-melodic nature after the previous sweet contours. This also gives
us a rare instance of a vamp section in which the repeated guration is quite explicitly
thematic. This repeated gure, driven on by harmony that is suddenly restless and
under-articulated, creates one very large phrase reaching from bar 34 all the way
through to bar 46
. The close-position chords in the left hand do something to
alleviate all the open sonorities heard before, although the gap in the texture mostly
holds. Perhaps the most impressive feature is at bar 36, where for the only time
in the vamp the right-hand rhythm is abandoned. The two arpeggio gures here
recontextualize the descending triadic gures heard so often in the rst half, giving
them an intensity and shape they never had before. After this the vamp grows more
and more vehement, a display of temperament to compensate for the lack of it in
the rst half.
A more detailed investigation of argument is appropriate for the Sonata in B
minor, K. 409 (Ex. 4.19), with its central black hole the longest and arguably
most extreme of all Scarlatti vamps and unusual explicit reprise of the opening
material. (Bear in mind that the opening material rarely returns in the tonic in the
second half of a sonata.) What forces, if any, hold such seemingly disparate material
together? What sort of sensibility informs the composers choice and manipulation
of material?
The principal strain in the argument of this piece may be said to concern hy-
permetrical manipulation and a concomitant struggle between regularity and ir-
regularity of internal organization. The rst half displays both extreme regularity
and ambiguity in its syntax. This process is set in train by the opening unit, which,
unusually for Scarlatti, may be described as a theme, having a clearly demarcated
boundary and containing several distinct thematic impulses within itself. Thus the
opening unit may be subdivided into groupings of three bars (a sequence cut short)
and a more or less indivisible ve bars. Sheveloff, on the other hand, believes that
the organization of this unit is essentially 4 + 4: the augmented second in m. 4
marks a phrase break in which the A closes the rst four-measure unit, while the
G and F serve as upbeats into the second unit.
Although this might seem an
attractive solution, I cannot bring myself to hear the passage in this way. There is
no question that in voice-leading terms the A and F of bar 4 are the necessary
continuation of the descending parallel tenths outlined in bars 13, but the marked
disparity of texture and rhythmic values between bars 3 and 4 and the fact that
the right hand of bar 5 simply continues the descending quavers of the previous
bar suggest that 45 constitute a single, indivisible impulse. There is therefore an
overlap of function at bar 4
, but those elements suggesting a fresh start at this
point make the stronger impression. The varied form of the right hands mate-
rial at bars 911 makes the break between third and fourth bars of the unit even
Private correspondence, 1994.
202 The keyboard sonatas of Domenico Scarlatti
Ex. 4.19 K. 409 bars 186
Bars 47 then spend much time circling around the dominant, in a manner that
seems quite distinct from the sequential drive of the initial gesture. Bar 8, a solitary
bar of tonic, has to bear the weight of all the contrasting earlier activity, and it
hardly seems long enough to ground the tension. The effect of the resumption of
the opening in bar 9 after this has something in common with our missing-bar
phenomenon. On a broad scale, therefore, a regular eight-bar unit exists, but it
contains some internal discomfort, however one perceives its subdivisions. This in
spite of the assertive nature of the theme; note how the energy of the sequential
Syntax 203
Ex. 4.19 (cont.)
descent is physically and visually manifested in the left hands extravagant leaps up
and down. Typically and necessarily, Scarlatti immediately repeats his formulation,
with the initial right-hand variant almost taunting the listener. The composer often
repeats immediately his most challenging pieces of invention, as if to assure the
listeners that they did not mishear rst time around.
As a counter to the somewhat schizophrenic theme, bars 1724 then feature an
almost excessive regularity of phrase rhythm. They combine the features of the two
As noted for instance in Sheveloff, Grove, 338.
204 The keyboard sonatas of Domenico Scarlatti
parts of the theme on several levels. The two-bar unit 1718 encapsulates the two
parts, bringing them side by side; thus 17 corresponds to 1 (more broadly the rst
three bars) and 18 to 4, in both hands. The left-hand pattern is then repeated while
the right hand of bars 1920 rhymes with 6 (or 7) then 8. The left hand at bar 7
is particularly signicant for the hasty attempt it makes at balance within the rst
eight-bar unit, inverting the initial rising octaves and following this with a vertical
octave, adding to the cramped feel at the end of the phrase. However, by this the
left hand shows it has a conscience, so to speak, which is then evident in its four
identical units of bars 1724. The attempt to dispel the tensions that arise through
ambiguity of phrase structure by means of grim reiteration is signicant given the
nature and role of the vamp to come.
Above this the right hand in 1720 acts as a sort of compression of bars 18, as we
have seen, and this is followed by a pseudo-sequence at 214. Bar 22 refers to bar
4 in a more direct way, however, but with A replacing the earlier A. This audibly
irons out the original awkward augmented second of bar 4, yet it also disrupts the
very square enunciation of B minor. While the A hints at the upcoming D major
and therefore acts as a sort of modulatory device, it more importantly develops the
principal sub-plot of the piece, the conict between A and A which lends an edge
to the primary syntactical problems. After all, the A in bar 4, which announces the
disruption to the sequence, is made additionally prominent by the fact that it has
been preceded in bar 2 by an A in the bass. This forms part of a melodic-minor
descent from B to F while the A at 4 forms part of a harmonic-minor version
of the same descending interval. After the difcult A at bar 22, A is reafrmed
in the following bar, by means of another awkward interval (the diminished fourth
DA) and a clash with B in the left-hand part.
This eight-bar unit, seemingly simple in intention but rich in associations, leads
to yet another with similar characteristics. Bars 256 are almost a transposition of
1718 but for the initial g
; this has to move upwards and, with another rst-beat a
suggests rather a parallel with 212. On the last quaver of bar 28 a precipitate shift
towards III occurs as the left hand for once breaks its conscientious pattern and A is
once again highlighted, in the right hand. This central event rather upsets the ideal
of a balanced eight-bar phrase which is I believe one subject of this music; further
confusion is created by the fact that bar 30 rhymes with 28, while bar 29 presents
the pattern in the opposite direction. The right hand at 312 then transposes the
equivalent bars 234, an identity obscured by the differing ornamental suggestions for
24 and 32 provided in the Gilbert edition. These complex relationships between all
the phraseal units act as a destabilizing force, undercutting the large-scale regularity
of the eight-bar phrases and ultimately demanding the cleansing properties of a
The conrmatory D major phrase from bar 33, beginning with a repeated a
, also
reworks many elements of the theme. The left hand reverts to a two-octave span in
its rising leaps with the arrival at III; it also mirrors the opening in its reversion to
stepwise intervals between pairs of bars after the VI alternations of the intervening
Syntax 205
passages. Octave displacement aside, the opening bass line consists purely of stepwise
movement until the VI of 78. Meanwhile, the right-hand pattern from bar 33
represents a new fusion of elements from the two parts of the theme, the dotted
crotchets from 1 and the stepwise quavers from 4 now being superimposed. This
time, however, the composer tries a different syntactical strategy, reverting to the
falling sequential impulse of the opening bars, but at half the speed. Nevertheless,
the sequence once more fails to complete itself, being lost again on the fourth
sequential degree as the expected C in the bass is replaced by A at bar 39, followed
by an awkward elision through to new material at bar 40. Both of the right hands
voice-leading components resolve, e
to d
and a
to the appropriate f
, and the
left hand moves to D, but the textural and thematic disruption jolts the listener. This
revives the situation found in bar 4, where what should be an overlap due to the
voice-leading continuity sounds more like an interruption. Thus while the whole
unit from bar 33 makes up a regular twelve bars, it continues the problematic internal
division of the theme, consisting of 7 + 5 bars.
Given the interweaving of thematic and syntactical features observed so far, it
should come as no surprise that Scarlatti bases the last ve bars of the phrase on
the latter part of the original theme. From bar 40 in the right hand we hear a pair
of descending units very similar to the falling shape at bars 45; in fact the second
of these is at the same pitches as its model save for the substitution of a
by a
Equally, the cadential bars 434 bear an obvious resemblance to 78, demonstrating
the composers extreme sensitivity to the nuances of cadential formulae. If we ignore
Gilberts ornamental suggestions at bars 24 and 32, then the effect of the appoggiatura
in bar 44 has considerable structural signicance, since it rhymes directly with bars
8 and 16. Also noteworthy here in terms of our sub-plot is the doubling of the A
at 42
. While this arises rst of all for reasons of registral management, it highlights
the triumph of A over A and also stresses the 2 + 2 construction of bars 4043;
therefore the following bar once more has no companion, just as bar 8 seemed to
require a breathing space after it. The repetition of the whole twelve-bar unit from
45 is structurally appropriate as a rhyme for the dual presentation of the opening
That Scarlatti recognises the problematical status of bar 44 is evident from the fact
that its equivalent does not appear at the end of the matching phrase. Instead it is
elided with the beginning of the next unit at bar 56. Thus the elision, normally a
device utilized to break up an overly square phrase structure, is here used to give
greater regularity to the hypermetre, to square matters up. The unit starting at bar 56
once more makes great play with a
, reinforced by the largest left-hand leaps so far,
up to a
at 60 and 62. Otherwise the phrase represents a perfect 4 + 4 construction.
However, this seems a hollow regularity. The alternation of tonic and dominant
harmonies the same strategy adopted in the left hand at 1732 is now taken up
by both hands. Although such reiterative directness is a common rhetorical gambit
at this point of a Scarlattian structure, conrming the arrival on the new key area, it
can hardly pass as a triumphant solution to the syntactical argument. There is none
206 The keyboard sonatas of Domenico Scarlatti
of the internal complexity which the previous phrases attempted to incorporate and
which seems to be the implied model for the syntactical action of the sonata. Bars
56ff. may be symmetrical but they lack the variety of shape to be thought of as
The problems of grouping begin once again with the extension of the phrase
at bar 64, matching bars 56 and 60 and undercutting the apparent symmetry of
the eight-bar phrase. Of course this problem is inherent in the phrase itself: its IV
alternations demand a following I, and the pattern has been so rmly established that
the thematic form it takes seems inevitable. Bars 645 are in fact difcult to align
within the larger structure. If bar 64 begins as a fth repetition of the two-bar unit,
bar 65 breaks with the expected continuation and instead seems to provide a link to
the new material of bar 66. Because of the new material which it ushers in and the
ascent in the right hand which takes us there, bar 66
sounds climactic, and a much
stronger hypermetrical downbeat than that found two bars earlier at 64. From this
point of view bars 645 almost function like an extended two-bar upbeat. At least,
this is true of the right-hand part; the total picture is more ambiguous still. While
the right hand begins a 2 + 2 pattern at bar 66, the left hand seems to have a 2 + 2
construction from bar 65, so that the downbeats of the two hands conict, a ne
state of affairs after the unanimity of the previous eight-bar unit. In the midst of
this, the right-hand ascent in octaves ABCD provides a textural and pitch
reminder of the very opening, now reversed. This time the nal bar of the phrase
cannot be hidden under the cloak of hypermetrical respectability; however one
chooses to subdivide it internally, the total phrase from bar 64 only adds up to seven
bars, a classic disturbing example of the missing-bar phenomenon. In retrospect this
provides a sting of hypermetrical tension to undercut the extreme regularity of the
vamp. How could any performer resist adding an invisible bar at this point? Not
just hypermetrically but also technically given the widely spaced writing and in
particular the very difcult leaping gures some breathing space seems essential.
Thus in the rst half all attempts to arrive at true regularity of organization
have been thwarted. All reasonable means of bringing the syntax under control
appear to have been tried, from melodic compression to a rather old-hat linear
intervallic pattern (another sequence that is not self-evident) to a more modern,
buffa-like reiteration of tonic and dominant sonorities. The syntactical play is in-
formed by the same duality of sequential and periodic impulses that we saw in
K. 523 (Ex. 4.6). More drastic action seems to be required if the ideal is to be
achieved. The vamp from bar 71 provides this by representing a hypermetrical sim-
plication (it is really in 12/8, entailing endless divisions into hierarchical groups
of four identical units), one so extreme that the concept of a phrase is lost in an
immediate sense. Also dispensed with is any real sense of melodic exposition, as the
composer concentrates purely on rhythmic properties.
If the vamp provides on the one hand a hypermetrical simplication, on the other
it represents a marked increase in harmonic complexity. It is as if the composer is
working with an ideal of balance of harmonic movement which has thus far been
Syntax 207
weighted to one side, and indeed many of the vamps in the sonatas appear to result
from the need to provide a richer sense of harmonic action than has previously
obtained. K. 485 certainly ts this bill too. On the other hand the section does build
on the one aspect of harmonic complication present in the rst half, the conict
between A and A, upon which the opening sequence had foundered. Thus the
lowest of the three voice-leading components of the endlessly repeated right-hand
gure hovers very much around the region of A/A/B; note especially the dramatic
movement of A to the enharmonic B at 11819. (The semitonal equivocation
around these notes is reected by other layers of the texture, for instance by the
DDE and EEF traced in the bass between 71 and 98.) The vamp is also cut
short in voice-leading terms on the A at bar 142, before we move to a ve-bar
phrase that effects a thematic retransition. With the vamp essentially nishing at bar
142, it is almost the same length as the rst half of the piece (seventy-two to seventy
bars)! The retransition utilizes broken octaves in the right hand to set up the return
of the opening, at the same pitches as the original rst two bars.
This ve-bar phrase obviously upsets the four-square units of the vamp and yields
another odd-bar-out at 147. However, this would be to overestimate the regularity
of the vamp itself. Once one has adjusted to the hypermetre, one perceives three
initial groupings of 4 12/8 (bars 7186, 87102 and 10318), from which point
the hierarchical organization breaks down. From bar 119 there would seem to be
two groupings of 3 12/8; the sense of demarcation between 130 and 131 is very
strong due to the anomalous right hand gure at 130 (signicantly using a
and its
lower octave and anticipating the gure at bars 1435) and the clear sense of return
from 131 to the material from bars 91ff. The fact that complications arise with the
fourth of the vamps very large units suggests an extraordinary afnity with the earlier
abortive sequences which also foundered on the fourth step. The vamp, one should
note, also plays a role of not just harmonic but also textural compensation, as it lls
in the largely unused middle registers of the instrument in close position.
This huge unruly section leads to a creative boil-over in the unusual formal
device of a reprise of the opening, one that can obviously be justied in the very
unusual circumstances. However, the exact return does not last. The vamp forces a
new, sweeping form of the opening bars this big sequence empowers the little
sequence from the start, which now proceeds down the whole tonic scale. In the
process the offending A from the fourth bar is smoothed out to an A. Not only
does the sequence nally realize itself in the fullest possible form, but the nal bar
of the unit does not come to a halt as have the endings of almost all the previous
eight-bar phrases; instead the momentum of the sequence sees the bar lled up by
a quaver gure. It would not do, however, to imagine that Scarlatti has now solved
his syntactical difculties. If there are no problems with the internal organization
of this eight-bar phrase, it is because there is none! By its nature such a sequence
is internally indivisible; it also lacks any true harmonic substance, given the parallel
movement of the parts throughout, and contains no cadential articulation. In fact
this apparently triumphant solution sidesteps the matter of phrase construction and
208 The keyboard sonatas of Domenico Scarlatti
articulation entirely. The decorated repetition of the unit seems to acknowledge as
much with its witty reintroduction of an A to decorate the B in 158, setting off a
chain reaction of similar gures.
From this point the ideal of a syntactically and thematically balanced eight-bar
phrase appears to be abandoned. At bar 164 a new discrete four-bar phrase is intro-
duced to ground the momentum and give some cadential balance to the previous
activity, with the left hand remembering its best manners. The unit ends with an
appoggiatura, a resolution not heard thus far. We then hear an equivalent
of bars 1724, which in this context does not sound so abrupt in its introduction;
it is almost as if the insertion at bars 1647 is compensating for all the previous
isolated single-bar phrase endings, as a sort of extended afterbeat. Where the ver-
sion of bars 1724 differs signicantly from its model is in the right hand at 1723.
The displacement of the a
back a quaver has several functions. As well as audibly
reminding us that the sub-plot is not yet resolved, it introduces further cross-phrase
confusion by echoing bars 1645 of the prior four-bar unit. The displacement also
means that the two four-bar units at 16871 and 1725 are more symmetrical than
the rst-half equivalent: bars 169 and 173 now match, and in addition 171 and 175
feature rhyming ornamental gures. Indeed, this symmetrical ornamentation has a
structural meaning (which is why the Gilbert extrapolation of the feature to the rst
half is misleading): we are more strongly encouraged now to hear 16875 as two
separate four-bar units rather than as a return to possible eight-bar organization.
This trend is continued in the nal eight bars, which twice outline the same bass
progression as 1647, with the left hand nally achieving some balance between
the leaping octave gure and its cadential responsibilities. Melodically, 1647 also
form the substance of 1769 to enforce the new four-bar tendency. In the nal four
bars attention is focused on nailing the sub-plot. The a
is heard for the last time in
bar 180, preceded uniquely by dominant and tonic scale degrees and thus put in a
context in which it cannot create ambiguity. The following g
then rises to a
in a
reversal of the augmented second of bar 4. This then leads to b
, and the nal a
in the penultimate bar is explicitly resolved, surrounded on both sides by the tonic
note so that it too can no longer act as a destabilizing agent. The last bar gives us
a unique and appropriate fourfold B, as if to underline the point that here is a nal
bar that scans.
The wit of this reprise is very compressed after the blazing vamp and requires
some quick aural adjustment if much is not to be missed. This plus the apparent
abandonment or sidetracking of the original premise raise once more the question
of the composers sense of an ending. The end to K. 409 is rhetorically weak, in
spite of the relatively strong gestures presented in the nal few bars. To imagine,
though, that with some more small adjustments or perhaps the addition of several
more phrases this could be remedied, if desired, is surely to miss the larger rhetorical
point. In this detailed reading of K. 409 the vamp has been shown to perform a
number of functional and corrective tasks. Yet there remains a gap between these
functional aspects and the sheer anarchic presence of the vamp in its own right.
Syntax 209
It is rough and inartistic, out of scale with the rest of the work and the sonata
was already rather rough in effect, perhaps especially given the prominence of the
wild left-hand leaps. We may conduct a discussion of K. 409 in terms of its main
thematic material, but surely the real main thematic material, both statistically and
affectively, is the right-hand gure of the vamp, repeated seventy times over with
just the two intervening bars of adjustment. It would plainly be misguided to hear
this gure as a direct relation of anything in the rst half, such as bars 18 and 26, or
even as an outline form of bars 45. Even if it were clearly and signicantly related
to any prior shape, its unbelievably excessive treatment would take it well beyond
the realms of developmental necessity.
How, then, do we listen to the vamp? Do we listen to it differently from the other
sections? Perhaps initially we listen to it without any cognitive adjustment, but as the
mixture of stubborn guration and unpredictable harmony continues on and on,
moving well beyond what seems reasonable and rational, we must surely lower or
raise our sights. The gestural excessiveness of this and all other vamps invites quite
opposed reactions. One might feel hypnotized, tuning out at an immediate level and
then tuning in on a higher level, so achieving Bonds trancelike state; it is as if, as
has been said of Ligeti, Scarlatti seeks to eliminate repetition through repetition.
Alternatively, one might feel browbeaten and nally agitated. Are such sections to
be heard as dynamic or static? On a larger scale, how does the vamp change the way
we listen to the following music, and, in retrospect, how we hear the whole piece in
our minds ear? It was suggested in the discussion of K. 260, with its multiple vamps,
that the normal music fades into insignicance. This must also be a possibility with
K. 409. At the very least, the vamp relativizes the status of the surrounding material.
Even if we set the greatest store by its functional, corrective aspects, its impact on
the following material in the second half can be judged in two ways. From a positive
point of view, the vamp has a sort of laxative effect, helping the opening theme to
solve its internal structural problems. On the other hand, one might maintain that
the would-be reprise collapses under the weight of the vamps example and that the
subsequent music loses its capacity to carry out detailed if ambiguous operations
over any span longer than four bars.
When trying to assess the place of the vamp in our conception of the whole, we
might bear in mind Sheveloffs denition that such passages sound like an improvised
accompaniment waiting for the entry of an important musical event.
made a similar suggestion. Quoting a passage from K. 260, he felt it seemed to be an
accompaniment to a song, a melody or a more precise line waiting to emerge above
it. Perhaps such a line was in the mind of Scarlatti and his listeners; an unheard line,
though none the less precise and expressive.
Such a melody of course never arrives.
The problem with Sheveloffs analogy is that invariably the vamp is the important
Alastair Williams, New Music and the Claims of Modernity (Aldershot: Ashgate, 1997), 85.
Sheveloff, Grove, 338.
Bogianckino, Harpsichord, 1012. The author also links such a feature with the emergence of a fortepiano style.
210 The keyboard sonatas of Domenico Scarlatti
musical event, at least in retrospect, in the amount of contemplation it engenders.
We may not grasp this initially, of course, precisely because of the lack of conventional
melodic behaviour which normally does so much to guide our memory of a piece.
Vamps impress themselves on our minds in different, less accountable ways.
In those specimens with clearer thematic relevance to their surroundings, such
difculties of comprehension can be less acute. The vamp in the second half of K.
511 is based on the surrounding toccata material (and prepared by a mini-vamp
heard in the middle of the rst half ). Since the sonata is effectively monothematic,
the insistent repetition of the same gure in the vamp stands out far less than usual,
although clearly given a less mobile registral treatment than elsewhere. It is thus the
changing harmonic background that makes the passage most memorable. Compara-
ble situations obtain in K. 438 and K. 469, which feature similar vamping guration.
A strong analogy between vamp and sonata-form development may be found in
the Sonata in E major, K. 216. Here the repeated vamp cell is not only thematic,
but it derives from the opening theme itself, from the gure heard at bars 25 (see
Ex. 4.20a and b). Of the three notes heard on the last three quavers of each bar
here, only the nal one is retained in the second half, leading to the same downbeat
appoggiatura motive. The fact that the whole gure is heard three and a half times at
the outset even provides some sort of syntactical precedent for the vamp. Cementing
the connection is the evident structural parallelism between the two passages. The
start of the second half presents a dominant version of the opening, a standard
gambit, and so the vamp is prefaced at 689 by a dominant version of bars 12. Both
the fragmentation and insistent repetition of a thematic module t the mould of a
conventionally understood development section.
On the other hand, the sense of purpose in Scarlattis development is less certain
than that. Alain de Chambure comments thus on the start of the second half: The
harmony is made to evolve in a hardly perceptible fashion, rather in the manner of
Schubert in some of his sonatas. On this occasion, the tense vocal improvisation is
turned into a melody.
It is difcult to agree that what we hear in the vamp of
K. 216 is a melody as such, but it undoubtedly does have a strong melodic character,
and this is central to how we might hear the passage. Although a fragmented version
of a thematic cell, the repeated gure is characterized above all by its appoggiatura, an
intense melodic device. In addition, after the rst two renderings at bars 6970 and
7071, which retain the repeated note across the bar line, all subsequent versions
This is the sort of passage Philip Radcliffe must have had in mind when he wrote that Scarlattis way of using a
short phrase as the foundation . . . of a string of modulations was prophetic of Haydn and Beethoven. Radcliffe,
Scarlatti, 33. Note also Leonard B. Meyers remark that harmonic instability tends, in Romantic as well as Classic
music, to be complemented by motivic constancy. With wider terms of reference than simply development
sections, this rhetorical/behavioural model offers another attractive way of comprehending vamps (and certainly
that of K. 216), except that any motivic denition is of course often difcult and that vamps seem excessive
in their dialectic of instability and constancy. Meyer, Style and Music: Theory, History and Ideology (Philadelphia:
University of Pennsylvania Press, 1989), 316.
Chambure, Catalogue, 89. As is frequently the case in this publication, the French original says something rather
different and perhaps less acute.
Syntax 211
Ex. 4.20a K. 216 bars 19
Ex. 4.20b K. 216 bars 6881
describe a falling third, yielding a more vocal sense of line. It therefore becomes
difcult to hear the passage just as guration; each repetition has its own specic
melodic intensity. From this point of view the comparison with Schubert is apt. The
sensation of each sound seems more important or at least more striking than
any organizing developmental force that arises from their totality. This is akin to the
magical voyage evoked by Bond.
At the same time two other possible models for an understanding of the passage
may be put forward. The shape of the repeated left-hand guration is very similar
to that found in the recercata movements of Alberos six three-part works entitled
Recercata, fuga y sonata. Compare the excerpt from the Recercata No. 5 in C minor
212 The keyboard sonatas of Domenico Scarlatti
Ex. 4.21 Albero: Recercata No. 5 (unbarred)
given as Ex. 4.21. Not only is the left-hand gure very similar in its own right (down
to the sustained initial bass note), but it is treated similarly too, tracing a pattern of
gradual stepwise descent. Note also that the gures are consistently conjoined with
appoggiaturas in the right hand. Such pronounced likenesses make one wonder
whether, as a historical principle, the recercata/ricercare lies behind the Scarlatti
vamp. Certainly Alberos realization of the genre seems to have some connection to
the vamp. As discussed in Chapter 3, these quasi-improvisatory preludizings evoke
an antique world, although one that was by no means dead in terms of contemporary
Spanish keyboard composition. More generally, they lean on a tradition of improvised
(harmonic) licence that obviously appeals as a source for Scarlattis practice. However,
leaving aside rhetorical differences between the two types of free writing (such as
differing placement within the larger structure and the more focused nature of the
vamp), there is a basic problem that we have encountered before. Scarlattis licence is
not put in a generically allowable context; nor does he acknowledge the apparently
aberrant nature of vamps by means of some sort of internal labelling, even if it were
only con licenza. The absence of either sort of framing to the invention presented
by the vamps suggests either the sort of studied elusiveness we have dened before
or that the vamps should be understood, as far as their appearance on the page goes,
as an organic feature after all.
A second possible historical ingredient in the form taken by the vamp of K. 216 is
offered by Karin Heuschneider. She observes the presence of a passacaglia-like bass
here and in the case of K. 260.
If we acknowledge this as a possible model for the
The Piano Sonata of the Eighteenth Century in Italy, Contributions to the Development of the Piano Sonata, vol. 1
(Cape Town: A. A. Balkema, 1967), 27.
Syntax 213
construction of the passage, it is clearly heard in applied rather than literal form. The
passacaglia bass is spread over a long time, comprising a single stepwise descent of a
fourth from V to its dominant, the F of bar 89. The bass in fact overshoots its goal,
moving down to F by bars 7982 then on to F and E. The F is then reinterpreted
as E, which moves up to the local dominant. This type of bass-line movement can
in fact be found in the majority of vamps. To gain the greatest historical purview
over this behaviour, though, we need to widen Heuschneiders terms of reference.
The tendency for the harmonic contortions of vamps to be founded on basses
that move by step, generally descending and often by an octave, suggests the regola
dellottava. (An explicit aligning of Scarlattis practice in the vamps with this precept
does not seem to have been made in the literature.) This widespread formula was
associated both with keyboard continuo playing and with improvisation (and hence
the fantasia). C. P. E. Bach, for instance, advocated organizing ones improvisations
around a bass line of rising and falling scales. What unies these various technical
procedures is the sense that they provide a frame for relatively free invention in
other musical parameters, that they hold the music together, and although the usual
reservations about differences in artistic realization and implication must apply, they
clearly offer a strong historical model for understanding the vamp sections. K. 319,
for instance, as demonstrated by Agmon, offers a vamp organized around a descent
of an octave from c
to c, taking the extraordinary form of an octatonic scale.
the vamp of K. 225, the bass begins on E and moves via D down to C before pushing
up by step to another C in bar 63
. Often the manoeuvres are more complicated,
as in K. 531, where the rst descent is diverted back to the starting point of B
before a more straightforward descending octave progression unfolds (bars 6785).
In K. 180, the structural descending octave twists back on itself several times before
completion. Sometimes other intervals are involved, as with the falling sixth from B
to D observed in K. 193. Often enough the technical basis of conjunct movement
is retained even when the total shape of the bass line cannot be so readily grasped.
This is the case with K. 409.
The vamp of the Sonata in G major, K. 124, also illustrates this. Preceded by
a ourish in D major and a pause, the bass simply hover[s] in mid-air around
this structural D,
moving between B and E, before it is reafrmed by another
arpeggiated ourish at bars 1023. Few sonatas are more frankly popular in tone than
K. 124. Its repetitions have such urgency that one listens beyond any symmetrical
syntax to the sheer physical energy they generate. The work is repetitive at all
points of its structure, not just at prime articulative moments. There is one section
apart, one which clearly builds to a climax rather than expressing heavy insistence:
the vamp of bars 83103. Exceptionally, it is built on several successive melodic
impulses rather than on a single repeated gure. In addition, its exquisitely painful
dissonances differ greatly from the highly diatonic language elsewhere; the rst half
See Agmon, Division, 4.
This phrase is used in Edwards, Iberian, 32. For her the repeated chords create a static, intense atmosphere.
This echoes the judgements of Bond and Hammond that such sections are not conceived dynamically.
214 The keyboard sonatas of Domenico Scarlatti
moves from a very clear I to a very clear V, with nothing else whatever apart from
a dominant minor enclave. As in a number of sonatas already mentioned, the vamp
contains all the harmonic ambition and bite.
If it was suggested earlier that vamps
often seem underdetermined by context, it may be that in cases such as K. 124
and K. 485 they arise in response to an overdetermination of harmony and phrase
structure elsewhere in the work. They function therefore as a sort of outlet. This must
be proposed fairly gently, since such a causal explanation is hardly binding; many
sonatas that seem limited in these respects fail to employ vamps. We could also hardly
maintain that there is anything inhibited about the rest of K. 124; expressively, the
vamp may function as much as a histrionic intensication of the rest as a contrasting
K. 253 in E at major resembles the case of K. 124 harmonically but not syntacti-
cally. When is a vamp not a vamp? The one here has such a stronger prole than the
surrounding material that, in retrospect, it is clearly the rst half that represents the
waiting for the arrival of an important event. Similarly, the resumption of the ofcial
material at the end of the second half, from bar 43, seems like a structural rump. The
vamp dwarfs the rest even more than in K. 409; here too it is exactly as long as the
rst half. This rst half seems to suggest a street band, amiably dishevelled in musical
conduct. A number of different, short-lived gambits are offered, held together more
than anything by all the similar linking and cadential phrases. Only with the fanfare
that nally declares itself properly from bar 14
does the music achieve any syntactical
comfort, aided by the antiphonal treatment. Harmonically, on the other hand, this
is all as straightforward as imaginable.
The vamp then offers the customary harmonic mobility and elusiveness, as if
to balance the harmonic equation of the whole. Syntactically, though, it presents a
greater rather than lesser degree of denition. The non-vamp material is consistently
written against the bar line; compare the very explicit lling of 12/8 bars by the
vamp, with each downbeat heavily stressed. Once the invention has settled after
the characteristic nerviness of bars 228, the repeated gurations and large-scale
sequential construction feel more comfortable than what was offered for much of
the rst half.
In stylistic terms this is the least elusive of vamp sections. It has a strong Baroque
avour, especially once it settles from bar 29, and is easily the most direct illustra-
tion of Sheveloff s proposed Vivaldian descent. The violin-like guration suggests
Arthur Haas notes that, with more than two thirds of the work utilizing nothing but I, II and V chords, what
Scarlatti does elsewhere justies this heavy dependence on tonic and dominant. La pratique de la modulation
dans les sonates de Domenico Scarlatti, in Scarlatti: 13 Recherches, 60. K. 124 is also discussed at 578.
The corrective sense of the vamp is emphasized by the fact that the composer recapitulates the start not the
opening bar, but bars 23 before the closing material returns. Thus recontextualized, it carries far more impetus
than on its earlier appearance. Note too the reworking in the second half of bar 44 compared to the equivalent
point in bar 3; the extra imitative entry by the bass gives a more transparent sense of organization and anticipates
the antiphonal treatment of the transposed closing phrase which follows. By then cutting from bar 3 to bar 14,
the composer also omits all the less uent material of the rst half. It is in effect replaced by the processes of the
Syntax 215
we are listening to a solo episode from a concerto. There is also a very clear em-
bodiment of the regola dellottava in the basss linear progression of an octave from
B to B, without detours and with the latter part emphasized by the scoring in
octaves. In K. 253 the vamp is quite patently a rhetorical match for the outer sec-
tions, given the clarity of its stylistic associations. The rhetoric of the whole clearly
embodies a topical opposition. The composer seems to be pitting a vernacular style
against high art, an echo of an Italian past, not unlike the plot suggested for K. 513
(Ex. 3.13). This reading would also promote the claims of the vamp, in certain
senses, to greater authority and coherence. As we have seen both with other vamp
sonatas and other instances of topical play, this is clearly an uncomfortable opposition
if we try to construct a sense of the whole; either the two are left to rub against
each other, or the vamp wins the day in our ears through its greater incisiveness
of invention. On the other hand, its harmonic mobility and subordinate structural
role (as a prolongation, no matter how memorable, of the dominant) may limit its
It is the pull between functional and non-functional rationales that makes vamps
both so fascinating and so upsetting. If it would be trivial to declaim in approved
current fashion that they offer nothing but rupture, it would also be trivial to imagine
that their functional aspects can constitute an entire explanation for their presence.
They are more and less than bleeding chunks or perfect servants of the larger form. As
much of the preceding discussion has focused, with epistemological inevitability, on
functional explanations, we will nish with an appeal to their more ineffable qualities.
Siobhan Davies, the choreographer of a number of Scarlatti sonatas, to whomwe have
already referred, writes of her experience: He must have been incredibly excited by
his imagination and the sheer thrill of letting go.
Although clearly not meant
to refer specically to our current subject, the notion of simply letting go offers a
wonderful translation of the sense in which vamps place themselves beyond the easy
reach of normal constructs. Whatever their possible historical roots, through their
sheer abandoned intensity they can seem indifferent to considerations of rhetoric,
style, form, even expression.
The notion of intensity can in turn be enlisted in an attempt to explain the
aesthetic moment of the vamp. Wim Mertens has invoked this in his account of
American minimalist music. Citing Jean-Francois Lyotard The intensity exists but
has no goal or content and Gilles Deleuze Each intensity wants to be itself, to be
its own goal and repeats and imitates itself Mertens makes this a central category
for the understanding of this apparently anomalous, unhistorical musical style. We
have already considered briey the link between the repetitive behaviour displayed
by vamps and that embodied by minimalism, and while there are obvious perils in
aligning musical phenomena from such different periods of history, the connection
is a useful working tool given the lack of anything very comparable in the music
of the eighteenth century. The strategy effectively treats Scarlatti according to the
Siobhan Davies, A Week in the Arts, The Daily Telegraph, 20 May 1995, A5.
216 The keyboard sonatas of Domenico Scarlatti
second category of modernism outlined in Chapter 2. Because there is no economy
or reserve of intensity, Mertens writes after Jacques Derrida, there is no historical
category, since intensity is totally outside time.
This has an obvious relevance to
the thoughts we might entertain about the ontology of vamp sections. It corresponds
to the sense that they know no economy (the thrill of letting go ), nor history,
nor any goal beyond replicating themselves. For all their possible functional and
organic attributes, they continually threaten to oat clear of them in an autistic
self-sufciency, a repetition without rationality or purpose.
This self-sufcient intensity allows for the two basic reactions to such passages out-
lined before. The vamp may be heard or felt as highly physical, a kind of music that
offers a tangible projection or articulation of bodily energy,
one which is unpre-
cedentedly direct because it is so relatively unmediated (by clear stylistic signals, for
instance). On the other hand, the intensity may through its very lack of differen-
tiation become abstracted, so that the vamp in fact invites a sort of out-of-body
experience. This is still rooted in a physical reaction, of course, but one which has
been relocated to the higher level mentioned before. Such an experience might
conceivably be connected to the realms of folk music, and amenco in particular,
suggesting the sort of abstract inuence postulated in the discussion of K. 277 in
Chapter 1. Timothy Mitchell has written that for real acionados of the form, a-
menco goes beyond the aesthetic in the direction of psychic cleansing, mysticism,
and even trance.
An abstract experience of a vamp section may indeed involve
such ecstatic possibilities. Whichever sensation predominates for each listener to each
vamp, these sections relativize the status of the material that surrounds them or
with which they surround themselves.
Wim Mertens, American Minimal Music, trans. J. Hautekiet (London: Kahn and Averill, 1983), 119, 121 and
Taken from Lawrence Kramer, Classical Music and Postmodern Knowledge (Berkeley: University of California
Press, 1995), 27. Citing the art historian Norman Bryson, Kramer reminds us that the distancing of the palpable
body has historically served as a cardinal sign for the condition of being civilized. By this measure vamps fall
conspicuously short of civilized values.
Mitchell, Flamenco, 224.
What Scarlatti does for syntax he also does for the elements of musical grammar. If
Scarlattis radically relativistic approach to rhythm and syntax has remained under-
appreciated, the same is less true of his harmonic and voice-leading peculiarities.
This is not surprising given our greater attunement to these elements we are
trained from an early age to spell our music correctly, as it were, and to avoid poor
grammatical relations between successive sounds. In these terms musical intelligence
and literacy are dened largely by the resourceful avoidance of such infelicities. Yet
even here, the extent of Scarlattis estrangement fromcommon practice the manner
in which the composer apparently goes out of his way to infringe the laws governing
the continuation and combination of voices is far from common knowledge. The
composers uncertain historical and stylistic position colludes with an uncertain grasp
of his anomalous language to lend him a marginal place in musical pedagogy, a
fundamental current function and means of dissemination for eighteenth-century
It is thus doubly no accident that Scarlatti does not gure much in the
teaching of musical rudiments, in the acquisition of harmonic and contrapuntal
skills in tonal music. Teachers will have enough difculty explaining to their charges
the aberrations found in Bach, Handel and Haydn without opening the Pandoras
box that Scarlattis sonatas represent. Edward Dent imagined the likely response:
If Domenico Scarlatti writes consecutive fths, why shouldnt I do so too?.
Although consecutive fths are far from the most frequent or disturbing of the
composers licences, such a question does allow us to turn the matter on its head.
Instead of asking why Scarlatti broke so many rules so often, we should rather ask
why most composers did not do so. Why the stability? What factors inhibit the
wider adoption of the relative free-for-all that the sonatas hold out as a possibility?
Donald Francis Tovey combined an acknowledgement of the crassly unacademic nature of the sonatas with a
marginal placement of the composer when he noted: Such work, taken by itself, seems as isolated as a dew-pond;
but Mozart, Clementi, and Beethoven assiduously pumped the contents of that dew-pond into their own main
stream. The Main Stream of Music, in Essays and Lectures on Music, collected, with an Introduction, by Hubert
Foss (London: Oxford University Press, 1949), 345.
Dent, Scarlatti, 177.
218 The keyboard sonatas of Domenico Scarlatti
Scarlatti was assuredly not the only composer of the common-practice era to
entertain critical thoughts about the immutable laws of music, nor the only one
to ignore or deviate from them as a consequence, but surely no one else offered
such an extreme practical response. Keeping more or less to the letter of the law,
as most composers have done most of the time, may be said to arise in the rst
place for reasons of social communication. The rationale is similar to that offered
to those who display faults of grammar and spelling in their prose writing that
the substance of their work will be judged harshly, whatever its intrinsic merits.
Errors undermine the authority of the whole and our condence in the control of
the writer. Similarly for composers, broadly following rules and precepts provides a
basis for comprehensibility. These are what make a language system possible at all;
communication of course needs constraints. Following these laws to the extent that
one is conscious of doing so at all allows for a smooth delivery of ideas, without
interference, without the reader or listener being distracted by faulty mechanics.
Scarlattis ideas, on the other hand, are to an unprecedented degree concerned
with the very delivery and articulation of material, precisely those inner mechanics
that allow competent utterance and promote competent listening. His invention, as
we have seen in so many circumstances already, is focused just as much on the edges
of an utterance as on its putative substance.
A second force for stability concerns social and professional status. Any analogy
with spoken or written language is weakened when we consider the demanding
nature of musical competence, how much sustained effort is required to achieve full
literacy and statistically how few are able to demonstrate this productively. An abil-
ity to move within accepted constraints is like a badge of professional competence.
Composers of pre-modern times need and want to demonstrate this ability in order
to belong, to be accepted by fellow composers, performers and informed listeners.
Indeed, why otherwise should products offered in a professional capacity be taken
seriously? Scarlattis apparent indifference to such concerns has been accounted for
in many ways, as we saw in Chapter 2. One of the explanations reviewed there con-
cerned his rm grounding in traditional techniques, in effect that learning allowed
liberty. If this does not account for the failure of other well-schooled composers to
follow a similar path, it does get us close to the technical spirit of many of Scarlattis
infringements and procedures. On many occasions, for instance, the learning goes
underground, as seen in K. 402; the ne grain of the music delivers more solidity
than is suggested by the big picture. Nevertheless, one must not miss the broader
sociological point: most composers want their learning to be an active presence, not
an absence. It should be heard and seen.
In another respect too, learning, or at least competence, is fundamental to the
technical spirit of the composers dissolute behaviour.
Scarlatti does not after all
Note in this regard William Webers idea that the codication of stile antico in the seventeenth and eighteenth
centuries served to counter the socially open-ended nature of the music profession and to create a musical elite.
The Contemporaneity of Eighteenth-Century Musical Taste, The Musical Quarterly 70/2 (1984), 189.
Compare Piero Santis characterization of sregolatezza in Santi, Nazionalismi, 51.
Irritations 219
abandon the premises or precepts of tonal language, which would be an impossibility.
What he allows us to do, and this is radical enough, is to glimpse a world beyond
these boundaries. He suggests the cultural contingency of the rules in the knowledge
that they are indispensable. This does not imply that they must be obeyed, since they
so often are not, but that they form the basis for comprehension and judgement. As
Loek Hautus has it in his discussion of Scarlattis licences, deliberate breaking of a
rule implies recognition of it; the exception to the rule must be projected against
a background of regularity.
Such recognition of the rules afrms their force at
the same time as we are encouraged to hear beyond them. They remain, in other
words, epistemologically active. For this process to have full effect in a Scarlattian
context, a relatively high degree of competence or learning from the listener must be
assumed. Such qualities must also be granted to the composer, given the frequency
and conspicuous nature of his offences. There is a certain condence implicit in
such rule-breaking, a sense that he can afford to disdain the outward appearance of
high art.
Nevertheless, such behaviour might wear rather thin if we were presented just
with a number of isolated infractions, as if a simple rebellious gesture were sufcient
to drive the point home. More instructive is to note the contexts in which such
infractions take place, to see how the composer conceptualizes them within the
larger discourse. This, after all, is the level of the operation at which learning must
be demonstrated, if what Giorgio Pestelli calls the game of complicity
composer, player and listener is to be sustained. Otherwise, communication really
will be weakened. At the same time, we must not neglect the instantaneous impact
of such features. As with the consideration of vamps, any attempted phenomenology
may easily nullify their unpredictability and individually upsetting qualities. There
is a danger, inherent of course in any attempt to dene a style, that we will become
too tolerant of them; they will no longer be seen as eventful but rather will take on
a systematic character.
A different sort of tolerance has been extended to Scarlattis offences by a number
of writers. There has been a tendency to minimize or even overlook them. The
campaign for Latin clarity outlined in Chapter 2 necessitates looking the other
way, since such features can contribute little to the guiding image of elegance and
lucidity. Ralph Kirkpatrick, in his inuential chapter on Scarlattis harmony, did not
look the other way but found rational explanations for many of the most aberrant
features. This was part of the campaign to give respectability to our composer,
quite understandable in the circumstances. After all, many of these features seem so
unaccountable that it would be quite easy to write them off as examples of artistic
mannerism, as the work of a sprightly buffoon. Too lurid a presentation can only
further marginalize their composer and discourage further enquiry. In the case of one
of the most celebrated passages of wrongdoing, the chain of parallel root-position
Hautus, Insistenz, 137.
Rosen, Classical, 163. This phrase, once again, refers to Haydn.
Pestelli, Music, 88.
220 The keyboard sonatas of Domenico Scarlatti
chords found in the second half of K. 394 (see Ex. 6.5, from bar 76), Kirkpatrick
provided a schematic reduction to demonstrate its essential orthodoxy:
Frequently a progression that is actually based on a simple enchainment of harmonies fullling
all the orthodox requirements for common tones or suspensions is realized by Scarlatti at
the harpsichord in terms of consecutive fths and apparently entirely nonvocal movement
of parts, as in [K. 394]. Yet regarded in terms of interchange and transposition of parts, such
a passage is seen to outline a progression of the utmost simplicity and orthodoxy, and to be
rich in common tones.
Kirkpatrick may be quite right to point to the learning and control which underpin
the progression, but he fails to explain why it is there at all, nor does he acknowledge
its freakish quality. This classic example of disdain, we may safely assume, will never
nd pedagogical use as an embodiment of simplicity and orthodoxy.
Nevertheless, Kirkpatricks explanation does raise one of the qualications that
must attend any study of Scarlattis dissolute behaviour. He reminds us that this is
instrumental music. Instrumental style was quite reasonably allowed to be freer in its
treatment of voice leading and texture than the vocal models that formed the assumed
basis of best compositional behaviour. The precise extent of the latitude remained
a subject of endless theoretical dispute throughout the century. With Scarlatti such
freedom is then pushed beyond what might have been thought of as reasonable
limits, as part of his keyboard realism, to be explored in Chapter 6. We must also
remind ourselves of the advent of the galant outlook on music. This, as we have seen,
entailed an antagonistic separation from the strict style and was, in theoretical terms,
associated especially with the free treatment of dissonance. A larger issue concerns
our collective image of the music of the eighteenth century. As stressed already, we
tend to view it from afar as an era of polished moderation, of exemplary harmony
and counterpoint; this is inextricably tied up with its pedagogical function, not just
in the classroom but in performing terms too. This tidy image is based on a selective
reading and understanding of the musical evidence, viewed through the pedagogical
abstractions that arose in the nineteenth century and that were maintained relatively
unaltered in the twentieth. For instance, despite the work of Heinrich Schenker
who stressed the more horizontal approach to harmony he believed was found in
the best contemporary teaching practices our general sense of the rules governing
the vertical combination of notes would seem to be much narrower than that which
obtained at the time.
Even allowing for this and the other qualifying factors, though, the Scarlatti sonatas
still tend to defeat such measures of historical sympathy. We may well acknowledge
the need for a more expansive view of the musical constraints of the time, but our
Kirkpatrick, Scarlatti, 225. An interesting take on this passage from K. 394 may be found in the arrangement by
Stephen Dodgson in Domenico Scarlatti Baroque Sonatas Arranged for Brass Quintet (London: Chester/Wilhelm
Hansen, 1982). He adds countermelodies on rst trumpet and horn which somewhat hide the bareness of the
voice leading.
Irritations 221
liberality surely has its limits.
In addition, it is not just the rules as such that are
subject to stress, but the wider notion of craftsmanship. The sonatas are full of messy
edges, whether syntactical or textural, quite apart from any evident solecisms. These
add up to a music of untidy excellence; they prompt Peter B ottingers twisted slogan
der unreine Satz, by which he characterizes an impure compositional style that
deals in irritations.
As suggested in the previous chapter, though, such untidiness may be as much
productive as destructive. If understood as an embodiment of Verfremdung, it may oc-
cur as much for positive historical and expressive reasons as negative, anti-normative
ones. These deforming details lend an edge to the routine of listening; they help
keep our hearing alive. The composers pronounced taste for discrepancy may be
most easily grasped, as we have seen, through noting the corrective efforts of editors
and performers. It may be no more than glancing irregularities that prompt such
corrections. For example, in his edition of the Sonata in E at major, K. 475, Muzio
Clementi tidies away many of the untidy details that help to enact the knockabout
comic sense of the work. In bar 10
(see Ex. 5.1a) he removes the rst left-hand
crotchet so that the shape of the whole bar matches the equivalent bars 13 and 16.
Then in the right hand of bar 16
he removes a minor infraction of voice-leading
rules, changing the b
to a g
so as to match the equivalent points of bars 10 and
13. Thus the preceding f
now rises properly to the local tonic as did the earlier
sequential equivalents. How does one counsel a performer who is uninterested in
Verfremdung and puzzled by untidy excellence to square up to the evidence of the
earliest sources? Such details after all will tend to niggle away during the early stages
of learning a piece, which involve breaking it down into units of invention as a means
of getting ones bearings. Here it is as if the units will not stand still for inspection
after the model provided by bars 911, each subsequent unit contains one irritating
difference. Persuading the player that such irritations are not only worth the trouble
of retaining, but worth trying to colour signicantly in a performance, might involve
pointing to their positive expressive function. These particular details exemplify a
restless, even hyperactive, creative sensibility that can by keeping the performer
alert generate a more dynamic style of execution.
Another apparently puzzling feature found in K. 475 is changed by Clementi. In
the closing theme (see Ex. 5.1b) he alters the right-hand part in bars 47 and 48 so
that it matches 43 and 44 in the previous phrase unit. The original version, although
it again seems so odd, is far more expressive; in writing his answering phrase to
425 from bar 46, which we would of course expect to match the previous unit,
Scarlatti effectively reaches the equivalent of bar 45 two bars early. Thus we now
hear three successive versions of what was set up as the closing pre-cadential bar.
This increases the sense of comic redundancy already inherent in the material. Such
As Peter Barcaba says, whatever our pretended liberality, the revolutionary aspects of Scarlatti will always
seem to be puzzling and against the rules. Barcaba, Geburtsstunde, 382.
Untidy excellence derives from Piero Rattolino, Scarlatti al pianoforte, in Domenico Scarlatti e il suo tempo, 113;
B ottinger, Ann aherungen, 756 and 81. See footnote 73 on p. 40 for further comment on this phrase.
222 The keyboard sonatas of Domenico Scarlatti
Ex. 5.1a K. 475 bars 916
Ex. 5.1b K. 475 bars 4050
buffa-style cadential repetitions can always carry an inbuilt sense of self-parody,
Scarlatti actually manages to trump this with his own level of reductive travesty. Again
here Clementi is so valuable because he shows just the expectations that Scarlatti is
working against, with or through.
The comedy in fact becomes even richer at the end of the second half (see
Ex. 5.1c). The second phrase unit of the closing material from bar 92, the equivalent
of bars 469 in Ex. 5.1b, moves down an octave but this time does provide a match
for the preceding four bars. Thus we hear three playings of the initial one-bar shape
Concerning the issue of whether such cadential repetitions must necessarily be heard as comically redundant
or whether they may in fact be more generously and less pointedly conceived, see the contributions by Wye
J. Allanbrook (Comic Issues in Mozarts Piano Concertos) and Janet M. Levy (Contexts and Experience:
Problems and Issues) in Mozarts Piano Concertos: Text, Context, Interpretation, ed. Neal Zaslaw (Ann Arbor:
University of Michigan Press, 1996), 75105 and 13948.
Irritations 223
Ex. 5.1c K. 475 bars 8598
before the right hand delivers the closing closing pattern in bar 95. Following this,
however, Scarlatti appends two further repetitions of the pattern, so that bars 957
once more present three consecutive playings. He therefore has it both ways now,
working with the listeners symmetrical syntactical expectations by means of the
rhyme of bars 925 with 8891 before restoring the anarchy, as it were. Of course,
this in its own right answers a symmetrical need, creating a rhyme across the two
Once more performers (and editors and listeners) might be encouraged to
look for the spirit that seems to animate such happenings to respond in kind to a
certain slapstick avour behind these particular discrepancies.
Voice leading
Alongside such features we also nd more specic offences, of which those against
the tenets of voice leading are often among the more conspicuous. This is certainly
the case with the parallel fths of K. 394, particularly disturbing since it is difcult
to place them in any sort of stylistic context. Often in the sonatas they have popular
connotations, although this can cover a wide range of affect. As found in works like
The nal repetition in bar 97 also has the more positive function of restoring the obligatory register of the
upper voice, allowing a more decisive nish.
224 The keyboard sonatas of Domenico Scarlatti
Ex. 5.2 K. 247 bars 8595
K. 96, 224 and 242, they represent an eruption of the primitive, with the rudeness
emphasized by immediate repetition; heard singly and in quite understated fashion
near the start of K. 208 and K. 415, on the other hand, they glance wryly at the
pastoral tradition. One of the most remarkable instances of parallel fths occurs in
K. 247 in C sharp minor. This begins as a nely wrought sonata in the Baroque
but eventually covers a great stylistic range. Rather like K. 263, discussed
in Chapter 3, it does so without any rupture. The dotted rhythm rst heard in
bar 3 recurs throughout, underpinning and softening any changes of style and affect.
Compare its appearances at bars 3, 12, 22, 32 and 39, where we move by degrees from
the clearly Baroque to the clearly Spanish a sort of stylistic modulation. Towards
the end, bars 8992 (given in Ex. 5.2) destroy this art of gentle transition. This
transposition of the second subject introduces very marked, even lurid, parallel fths
in the left hand, rst in one direction then the other. The material has always had a
plausibly Spanish character in its repetitions of a short, quasi-melismatic cell, but the
change here makes this suggestion startlingly explicit. It is as though a primitive
spell were being cast over the music. This is an odd place in the structure to unfold
such a meaning, and obviously this makes us reevaluate the tenor of the whole piece,
which has been relatively unied in tone and gesture. Bars 8992 are so exotic that
in the Johnson edition published in London in about 1757 and the two Vienna II
copies of the sonata there is some rewriting to avoid the crudity and incorrectness.
The semiquaver d
in 89 and 90 is changed to b, then both d
s in 91 and 92 are
replaced, by b and f
Scarlattis younger colleague Albero seems to use consecutive fths in the same
way, as a calculated artistic effect. The fths found in bar 20 of his Sonata No. 3
in D major act as a stylistic transition from the Arcadian pastoral manner of the
Pestelli notes the similarity of its idiom to that of the Essercizi; Pestelli, Sonate, 222. Compare also the writing
found in bars 34 with works like K. 69 or K. 147.
See Choi, Manuscripts, 78 and also 18081.
Irritations 225
Ex. 5.3a Albero: Sonata No. 12 bars 2024
Ex. 5.3b K. 301 bars 3944
opening to something more urgently rustic. A more ambiguous example is found
in the Sonata No. 12 in D major (see Ex. 5.3a). Are the parallels found in bar 23
accidental, incidental or deliberately bad? What follows is, as in Sonata No. 3, a move
to the minor, then some hectic dance steps, suggesting that the voice leading helps to
change the linguistic register. A comparable instance is found in bar 42 of Scarlattis
K. 301 in A major (Ex. 5.3b). These parallel fths seem to come out of the blue, in
a work of neat gestures that convey a rened populargalant avour. However, the
preceding two bars have offered a passing hint at something more exotic, so that our
fths could form part of the same stylistic moment. On the other hand, they might
also be conceived as a purer form of disdain, not so much referable to the particular
context as what could be simply described as a bohemian touch.
The Sonata in D major, K. 178, also contains a good example of what might seem
to be casual incorrect voice leading, rst heard at bar 31 (see Ex. 5.4a). This is clearly
not the worst of howlers and might not even register strongly with many educated
listeners, and so the question arises whether such parallel fths are anything more
than incidental. Both parts are simply enunciating standard cadential formulations
in principle, this is like the situation in the rather less harmless passage in K. 222,
to which we will shortly turn our attention. Yet such a manifestation must gnaw
away in the mind of any listener or player, even allowing for the stylistic context,
which is popular here. The offending bar is repeated twice more at 37 and 39 before
To use the term of Henry Colles found in Colles, Sonata, 896.
226 The keyboard sonatas of Domenico Scarlatti
Ex. 5.4a K. 178 bars 2840
Ex. 5.4b K. 178 bars 738
the rst half is over, so there is plenty of time to catch up with the problem. The
very fact of its repetition, that Scarlatti has allowed the incorrectness in a bar that
by denition we know must recur several times, increases the likelihood that this is
more than a passing whim. Perhaps we are being challenged to make sense of the
incident it has already been noted that the composer often repeats his errors in
this spirit. The nal version of the feature, though, seems to conrm any suspicions.
In bar 77 (see Ex. 5.4b) the offending parts are brought a literal fth apart, so that the
oddity is unmissable. This is a witty moment the composer owns up, as it were
but also rather disconcerting in its placement.
This is a conrmation of wrong-
doing and so in a certain sense represents a form of resolution, but it also presents
us with a stronger infraction of voice-leading conduct, just when nal closure is
In his edition, Longo does his best to mollify the problem. He leaves the rst-half
examples untouched, but takes advantage of the altered melodic conguration that
precedes those in the second half. He ties the d
over the bar line at 745 and then,
conrming the more explicit wrongness of the nal version, replaces the d
at bar
with a d
which is then tied over the bar. He thus avoids both the sudden landing
on an open fth on the rst beat of 77 and the explicit sounding of the parallel fths
on the second beat.
We may smile at such editorial contortions, just as we may smile at Hans von
B ulows charge that the composer took excessive pleasure in covert and overt parallel
Further wit arises from the fact that the change of octave which brings about these literal fths is a common
rhetorical device in Scarlattis cadential closes, used to bring about a stronger sense of nality through a shift in
registral colour.
Irritations 227
Ex. 5.5 K. 551 bars 3443
fths and octaves, and that the wider voice-leading conduct of his sonatas very
frequently offends eye and ear.
As already suggested, though, liberal tolerance has
denite limits in such cases. The moralistic air that surrounds such pronouncements
has never entirely cleared, as is evident in the continued exaltation of the strict
style at the expense of the galant noted in Chapter 3. And it can be the seemingly
more random moments of offence that give us the greatest trouble. It is notable
that most of the Scarlatti examples collected by Brahms in his study of the feature
involve wholesale parallel motion rather than fths or octaves out of the blue, in
contexts where they are harder to explain.
Often such contexts feature light, half-
heard collisions, as in Ex. 5.5, from the Sonata in B at major, K. 551. In bar 39
two scales, one travelling twice as fast as the other, are superimposed, leading to
all sorts of strange parallel intervals. The effect is particularly noticeable given the
straightforward obedient imitation between the hands in the previous three bars.
Indeed, it is this respectable procedure that brings about the trouble; the left hand
continues imitating the right at the distance of a beat into bar 39
and then presents
a logical continuation of the line while the right departs from the pattern. Com-
parable instances may be found in K. 17 (the piled-up fourths rst heard in bars
2021), K. 184 (bar 20), K. 212 (bars 3033) and of course K. 254 (see Ex. 1.3).
Another type of voice-leading irritation involves missing notes. This is often
found in conjunction with cadential unisons, when expected notes of resolution
fail to eventuate. In K. 132, for instance, the seventh found in the upper voice in
the penultimate bar does not resolve. K. 525 offers a typical lack of punctiliousness
in bar 23 (see Ex. 5.6a), where the a
suspension, prepared properly at the end of
22, does not resolve. In addition, there are parallel octaves between the third and
fourth quavers (EF). A g
on the third quaver of the bar (paired with an e
would solve both problems. Kirkpatrick, no less, and Horowitz both in fact play
this (also in the matching bars 29 and 31), by analogy with the equivalent points in
the second half.
Here the composer has himself provided an immaculate solution
to the wrongdoing of the rst half. The performers changes are perhaps motivated
B ulow, Klavierst ucke, ii.
See Mast, Brahms, 545, 11621 and 186.
Deutsche Grammophon: 439 438 2, 1971/1994 (Kirkpatrick); Sony: 53460, 1964/1993 (Horowitz).
228 The keyboard sonatas of Domenico Scarlatti
Ex. 5.6a K. 525 bars 214
Ex. 5.6b K. 466 bars 1421
as much by a desire to tidy up the discrepancies between the rst- and second-half
versions altogether as to correct the faulty voice leading. As it stands, this is a nice
game of discrepant details and errors corrected in the end, which it seems quite
unnecessary to interfere with.
Another striking hole in voice leading is found in K. 466 in F minor (Ex. 5.6b).
A fth-progression in the bass from C in bar 16 to the G in bar 20 is conjoined
with a four-part rising sequence in the tenor and a three-part pattern in the soprano.
The tenor, the most active and wide-ranging voice, seems to go missing at the
very moment of completion: there should be a minim g
at the start of bar 20. In
fact, subsequent events show that the effect of the missing note has been precisely
calculated. The g
found in bar 21
provides a delayed voice-leading gratication that
also helps to maintain tension between the two separate units of the larger phrase.
The means by which this delayed g
is prepared and quit are also signicant. It is
reached by means of an appoggiatura a
that forms a strong minor-ninth dissonance
with the bass and followed by a variant involving a
, strengthened by a sharpened
soprano note. It is as if this textural layer has become sensitized by the disturbing
absence at the start of bar 20, generating the dissonances and adjustments that follow.
If some of the features discussed above remain fairly localized in effect, there are
many cases where incorrectness casts a shadow over the entire sonata. K. 222 in A
major offers an extreme example. In his 1970 dissertation, Sheveloff introduced his
discussion of this piece with the thought that there are times when Scarlattis licenses
Irritations 229
Ex. 5.7 K. 222 bars 2940
remain unbelievable and almost inexplicable no matter how many times one studies
them. He pronounced himself honestly puzzled by the dissonances in the two-bar
unit of bars 323 (see Ex. 5.7) which include four consecutive sevenths in the
latter bar to the extent of approving of Longos creditable and still useful job in
correcting the passage.
Yet in terms of structural placement, this is just the point
at which the composer often introduces rogue or wrong notes in the run-up to
the nal cadence of the half, when the tonal sense is quite secure. We have already
noted examples of such cadential estrangement. Secondly, it is possible to make some
sense of the passage both harmonically and thematically. The basic harmony is clear
enough I in 32 leading (possibly through I
at 32
) to IV at 33
then V at 33
. The
upper voice has got out of phase with this; the a
at the end of 32 belongs with the
following IV and the b
in 33 belongs with the following V. The a
is a passing note
in a chromatic rising-third line, interrupted by the consonant skip down to c
. In
fact, we have heard almost exactly the same right-hand line already, in bars 78; the
only difference is that the fourth and seventh notes swap around. Nowb
leads to c
the reverse of the earlier alto progression. Its reappearance at 323 is connected
with a game played precisely from bars 78 with establishing the dominant and the
various degrees of oversharpening required or not, since Scarlatti takes us too far
sharpwards. The point surely of the haunting passage from bar 18 onward is that the
dominant attempts to settle into place by repetition. The common tone is placed
Sheveloff, Keyboard, 261 and 263.
230 The keyboard sonatas of Domenico Scarlatti
conspicuously at the top of the texture, giving some stability of contour after all the
previous see-sawing. The turn to minor at 212 is also a means of afrming V as
well as cancelling out the oversharpening.
The left-hand part at 323 has also been heard almost exactly before at bars 234
and, immediately preceding the unit under discussion, at bars 3031, thus providing
a big thematic overlap between the two separate parts of the structure. Loek Hautuss
principle of insistence
comes to mind when one considers the combination of
the hands both lines have been heard almost verbatim before, and so now neither is
prepared to give ground, as it were. The other nicety about 3031 (and 234) is that
the intervallic conduct is so blameless the hands move in parallel thirds almost all the
way. Thus both in specic thematic terms and given the play of harmonic indicators
note the AA in 33 the muddle at 323 has its place, although it does not lose its
unbelievable character; and for all the dissonance, this is more stable harmonically
than what has gone before. Not only that, but we have also already heard four
consecutive sevenths, if on a slightly different time-scale; see the two upper voices
at bars 11
! The unit that follows from bar 36 reects the events of the previous
one. The b
that lls in the fourth b
(compare the lled-in g
of 323) is
a witty but not wounding contribution to the oversharpening debate. Two of the
consecutive sevenths remain in bar 37, preceded by two consecutive fourths; in the
parallel place in 39 there is a thorough recomposition which solves all the problems.
The previous upper voice is placed in the alto and the bass rests on a dotted crotchet
(it has been in continuous quaver motion from its entry in bar 5). Most signicantly,
the soprano resembles the alto part heard frombars 18
onwards this was the motive
associated with the dominants attempt to articulate itself free from interference.
K. 123 in E at major offers even more screeching dissonances, involving parallel
major sevenths (at bars 318).
If it is any consolation, they sound worse than they
actually are. The c
and a
in bar 31 act as neighbour notes to the controlling b
but they relate to each other in the manner of a consonant skip. The parallel sevenths
formed by this and what the ear hears as a d
succession are not supported by
the notation, in which it is clear that the two notes belong to different voices. Thus
what looks harmless on the page and is in all voice-leading essentials unimpeachable
hurts the ear.
Such clashes as found in K. 222 and K. 123 can be rescued to an extent by an
appeal to contrapuntal process; they seem to be brought about by parts with their
own thematic integrity that move as if oblivious to each other. Such an analytical
gambit is quite common, as Janet M. Levy has suggested: When counterpoint or
See Hautus, Insistenz, especially 1389.
A milder version of the same pattern may be found from bar 57 of K. 364, while bars 25ff. of K. 154 offer a
very similar rhythmicmotivic conguration, there involving parallel fths.
Irritations 231
Ex. 5.8 K. 128 bars 1218
voice-leading can be invoked to explain the origin of a chord progression, then
everything from fussiness and complexity to ambiguity and peculiar dissonances can
be understood and legitimized.
Although Levy is referring primarily to approaches
to later nineteenth- and twentieth-century music, it is a measure of the strangeness
of Scarlatti that such measures might also be required when dealing with much of
his language. To exempt the approach taken here from such a charge, one might
point to the manner in which the surrounding material plainly seems to prepare and
tease out the sources of the ambiguity. The composer himself uses counterpoint as
a pretext for such a scrape, creating an ironic hidden respectability while denying its
overt manifestation.
After all, no one could maintain that counterpoint in its more respectable guise
can be invoked to deal with the sonatas of Scarlatti. He does not invest heavily in
the patina of craftsmanship by which most composers quite naturally signal their
authority it is applied technique rather than a pure display of it that animates the
composer. In many cases, of course, explicit resolution of a problematic feature is
not sought. Even where it is, the aberrations may come back to haunt us; as is the
case with vamps, their disruptive rhetorical force can easily outweigh their apparent
structural integration. The frightening specimen of voice leading rst heard in bar 14
of K. 128 in B at minor, for example, is provided with a correction almost imme-
diately, two bars later (see Ex. 5.8), and the phrase itself or its answering companion
are reworked on four occasions in the second half. The two nal corrective versions
in bars 59 and 68 are the most convincing apparent liquidation of the problem, but
by then the original offending unit has been heard so many times, in ever different
harmonic settings, that it has acquired a sort of strange stability. This disorientates our
sense of what is normal, lulling us into acceptance; in another instant, though, we
may snap back to musical reality, disengaging from any sense of trust in the whole.
Janet M. Levy, Covert and Casual Values in Recent Writings about Music, The Journal of Musicology 5/1
(1987), 20.
232 The keyboard sonatas of Domenico Scarlatti
The sort of hidden learning dened above is perhaps at its most striking when the
music itself makes a display of counterpoint before seeming to abandon it. This of
course is a very common pattern at the start of sonatas; it has been interpreted earlier
as a manifestation of difdence or disdain, but any such reading can generally only
be made after the event. There are exceptions, in which distancing is achieved by
the form of the imitation itself. K. 362 presents a laconic reductive parody; in K. 422
the ourishing right-hand opening suggests a grand style but the left-hand answer
is lopsided and the right hand strangely silent, making for a disconcertingly naked
In most cases, though, the imitation must be taken literally at the mo-
ment of its execution. It suggests organization, good technique, learning, control,
rhetorical certainty.
In K. 493 in G major a sort of galant counterpoint sets in once the opening strict
imitation has been abandoned. This is more extended than usual, with successively
smaller gaps between the imitation of each point, but surely there is something
pointedly pedantic about the procedure. It gives way in bar 10 to a more natural
phrase rhythm and a texture that is neither precisely polyphonic nor homophonic,
one that reuses the second bar of the opening point. This passage repeats itself with
slight variations each time, building up the momentum (the subtle changes of pitch
and scoring each time suggest that the ornamental differences are also positively
calculated). What is reached via an ascending scale that expands the repeated one-
bar cell (for the rst time delivered without any of the ornaments that accompanied it
in its opening learned guise) is a recontextualizing of the opening tag, now made the
start of a pre-cadential ourish; compare bar 20 with bar 1. This process encourages
the sense that the opening has been heard as not viable and in need of transformation.
What ensues for much of the rest of the sonata is relaxed polyphony, neither clearly
chordal nor formally contrapuntal.
In K. 224 in D major, on the other hand, an easy-going imitative beginning is
succeeded from bar 17
by something rather more strictly and earnestly contrapun-
tal. This presents us with bar after bar of overlapping entries of a standard tag (one
also found in K. 150), moving climactically ever higher in the upper voice. This is
followed by a return to a more casual form of note-spinning that is clearly related to
the opening material. In bar 44 we hear a single rhythmic reworking of the tag so in-
tensively treated before, made chic and decorative. Aside from this, the counterpoint
seems to have been exhausted by the earlier episode and disappears.
In the second half, however, the tag is reinterpreted in a decidedly primitive
context at bars 723, with rude parallel fths in the left hand. Of course, the stylistic
change is likely to blind us to this resemblance. The original tag itself, as seen from
bar 96
of Ex. 5.9, consists of a suspension prepared on the third beat of the bar,
restruck on the downbeat and then resolved down a step on the second.
That this texture should be heard as incongruously thin given the grand manner becomes clear at the start of the
second half, which contracts the distance between entries and adds counterthemes. Texture and style are made
more compatible.
The harmonic rhythm here and the diminutional ambiguity of the two-semiquaver gure mean that one may
also hear the resolution as occurring on the third quaver.
Irritations 233
Ex. 5.9 K. 224 bars 8198
third sequential exotic version seen from bar 81
of Ex. 5.9 clearly retains all these
attributes (and the rhythmic conguration is similar). At bars 914 there is a moment
of white heat which forms a climax to the non-functional harmonies of the second
half. It presents parallel E major and F major chords over an E in the bass, a classic
Phrygian progression, but this also brings us back remarkably to the learned world,
since three consecutive versions of the tag are embedded within the passage: thus from
the last semiquaver of bar 91 to bar 94
we nd BBA, AAG and BBA. Not
only that, but the clash with the chordal member a second above is also replicated;
thus the C clashes with the B at 92
just as the F clashes with the E at 98
The superimposition of primitive and civilized features in this passage encapsulates
brilliantly the polyglot versatility of our composer. In this sense it is no surprise
when the furore then returns. However, it lasts for only a fraction of the time it did
in the rst half. This makes sense given that the furore has already been presented
in several different guises from the start of the second half. K. 224 therefore offers
a classic instance of applied technique; the learning has not been shelved but has
gone underground.
Giorgio Pestelli cites the opening of K. 437 in F major (Ex. 5.10a) for its evocation
of a Frescobaldian canzona,
but the work as a whole seems to provide a purer form
of abandoned counterpoint than K. 493 and 224; there seems to be little attempt
to hold to the textural premises of the start. A more modern manner makes itself
felt almost immediately, and towards the end there is a marked change of tone to
Pestelli, Sonate, 256.
234 The keyboard sonatas of Domenico Scarlatti
Ex. 5.10a K. 437 bars 15
Ex. 5.10b K. 437 bars 1624
something akin to a popular song. Yet K. 437 is full of witty recontextualizations
of the opening point. This is especially true of its rst bar, consisting of a solitary
dotted minim. It is only too easy to embed this in the texture, as in bar 20, where
it is heard in both outer voices (see Ex. 5.10b), or, most charming of all, the nal
bar of the rst half the cadential resting point on c
also represents the rst note
of the subject, which will immediately become clear when bar 1 is repeated. If it
is objected that this hardly counts as real counterpoint or real learning, the answer
is that such cheating is fundamental to all contrapuntal art. The very prevalence
of tags in polyphonic writing arises after all precisely to allow for maximum con-
structive potential of the given material and hence maximum integration of texture.
Using a single note as a thematic binding agent obviously takes this learning to an
extreme of economy. Thus the bass at the start of the second half consists of a series
of dotted minims joined into a rising chromatic progression, technically a sort of
However, the second part of the two-bar opening theme is not altogether ne-
glected either. Its last three rising quavers are also found in the reworking at bar 20,
in the alto (and tenor). In bars 4950 the altos changing-note gures are very much
like those found at the start of bar 2 (see Ex. 5.10c), and the soprano features dotted
minims; thus the two limbs of the subject are superimposed. Something similar hap-
pens in bars 567, but with the added incorporation of the rising three-quaver gure
from bar 2, and the dotted minim now in the bass. Throughout the sonata the long
note seems to have been exploited for its sonorous value alone. This is certainly the
Irritations 235
Ex. 5.10c K. 437 bars 4957
case with the passage rst heard at bars 2021, which has the separate character of
an objet sonore, and is even more striking at 567, with the sudden registral plunge
of the bass and consequent textural gap. Several commentators have suggested that
bells are being evoked here.
It is Scarlattis triumph so completely to disguise a
polyphonic entry, turning counterpoint into colour.
For all this celebration of Scarlattis hidden art, we must remind ourselves how
important the more formal sense of counterpoint has been in the reception of
Scarlatti, and indeed all composers. One only need call to mind the disproportionate
attention and adulation given to the nales of Mozarts String Quartet in G major,
K. 387, and Jupiter Symphony, or the fugues in Haydns Op. 20 string quartets.
There is a denite sense that the critical community is more at ease with counterpoint
as a type than as a style, in other words with complete polyphonic entities that
traditionally connote the summit of creative and technical mastery.
The Scarlattian
literature has witnessed something of a battle along such lines. Thus Max Seiffert
opens his account of the Scarlatti sonatas by owning that Scarlatti was not much of
a fugue writer. As if to answer these charges, Cesare Valabrega devotes the last pages
of his 1935 book on the composer to a consideration of the Cats Fugue, K. 30,
in which Scarlatti gives a proof of [his ability with] the science of sound, in spite
of his general orientation against such a genre. Even so, he then nishes in an oddly
downbeat way by conceding that Scarlatti does not write Germanic fugues, that they
do not have the complexity of Bachs.
The same ideology is served by the views
See Kirkpatrick, Scarlatti, 203, and Livermore, Spanish, 115.
This ideological overbalance is also apparent, for instance, in Linton Powells discussion of Alberos keyboard
works, which devotes far more space to the fugues than to the other movement types. See Powell, Albero.
For example, a large part of Donald Toveys scorn for Clementis habit of including short canons in his larger
structures seems to arise from the implication that the composer did not have the courage or technique to execute
counterpoint on a larger scale. See Raymond Monelle, Toveys Marginalia, The Musical Times 131/1769 (1990),
Seiffert, Klaviermusik, 420; Valabrega, Clavicembalista, 30912.
236 The keyboard sonatas of Domenico Scarlatti
of Degrada and Pestelli, understandably keen to re-establish the composers serious
credentials by emphasizing counterpoint at every possible opportunity.
Of course, all this is not to suggest that every piece of counterpoint in a Scarlatti
sonata is loaded or skewed in a particular way. For instance, at bar 72 of K. 345 we
hear a brief contrapuntal linking passage in a work that is largely homophonic and
treble-dominated. A very similar one-bar passage, also placed near the start of the
second half, is found in another work in a mostly homophonic popular style, K. 314
(see bar 63). In neither case does the material have to suggest the pointed entry of
a learned style; rather such moments can simply be a manifestation of a technical
instinct or training that uses counterpoint to get around tight corners.
Cluster chords and dirty harmony
Another nicety derived from contrapuntal precept that is apparent in an overwhelm-
ing proportion of keyboard music of the time (and of many later times too) is the
tendency to keep to a similar number of parts throughout. Scarlatti offends most
conspicuously against this, and also against any sense of the limits of dissonance,
in the cluster or acciaccatura chords that have naturally aroused so much critical
interest. There is a tension between the point of view that they can essentially be
assimilated with various historical precedents and the point of view that they are
primarily a modernist feature. The use of dissonant, non-harmonic notes in chords
around cadence points was an established part of Italian continuo practice, and the
rst theorist to describe them in print seems to have been Francesco Gasparini,
possibly a teacher of the young Scarlatti.
On the other hand, to those who read them in a modernist light, any historical
precedents are peripheral, particularly given that in many works the clusters them-
selves are found in clusters, most famously in the case of K. 119 (see Ex. 6.14b).
In such cases the real dissonance comes less from the constitution of the individual
chord as such than from its insistent repetition or alternation with other impure har-
monies, so that there is an accumulation of harsh sonority. Commenting on B ulows
description of the K. 119 chords as ugly and horrible, Roman Vlad counters that
our ears are now used to more than this, since Le Sacre . . . to the extent that in order
to give back to old music its effectiveness and force, we need if anything to accen-
tuate the dissonances rather than remove them.
Indeed, although a comparison
with The Rite of Spring can easily be dismissed as anachronistic, it can be argued that
the sensational effect of Scarlattis clusters demands such extreme measures to do
For example Degradas assertion of the typically contrapuntal nature of [Scarlattis] compositional mentality;
Degrada, Lettere, 275. As the preceding analyses will have demonstrated, I do not dissent from this judgement,
but for Degrada and Pestelli this counterpoint generally has to be of the demonstrable (strict) kind and they do
not sufciently emphasize the ideological dimensions to Scarlattis and our own response to the whole issue.
Vlad, Storia, 25. B ulows reaction is perhaps preferable to a calm acceptance of these dissonances as part of
the style; similarly, the more recent complaint by Georges Beck about les dissonances inhumaines at least aids
Vlads restorative wish. Beck, R everies, 14.
Irritations 237
them historical justice. This is especially the case in connection with such a sonata
as K. 119, where the dissonant chords do indeed seem to be thumped out as in the
famous passage from Les Augures printaniers.
Such interpretations should also be related to performance practice. The counsel
from the theoretical sources of the time was that the acciaccatura notes should not be
held on. Indeed, this was well understood in the case of the so-called passing acciac-
catura often found in solo keyboard contexts. Thus we nd in works such as a Toccata
in F major by Galuppi and a Toccata in G major by Alessandro Scarlatti a notation of
block chords that include acciaccatura notes and the indication Arpeggio.
In such
contexts the harmonic notes might be held on after the initial ourish, but not the
acciaccaturas, which fullled a decorative function. In the case of the simultaneous
acciaccatura, the same principle is generally thought to apply. But, as has often been
pointed out, this is not manageable in works like K. 119 and K. 175; it is precluded
by the rapid repetition of such chords. Even in works where such advice might be
followed, it is not clear whether the Scarlatti performer should proceed thus.
any case, we should bear in mind that harpsichord damping was often so poor that
there is little sonic difference whether these extra notes are immediately released
or not.
The fortepiano sonatas of Giustini published in 1732 furnish an important con-
tribution to this debate from several points of view. They feature acciaccatura chords
notated exactly as in Scarlatti. Aside from the organological implications of this coin-
cidence Sheveloff believes that such chords add bite to the gentler sonority of the
piano for which Scarlatti also conceived most of his crush sonatas
they also bear
on their manner of performance and their contrasting usage in Scarlatti. Clusters
are found in the following movements: the Balletto and Sarabande of Sonata No. 1,
Andante, ma non presto of Sonata No. 3, Preludio of Sonata No. 4, Preludio of
No. 5, Allemande of No. 7, and the Allemande and Dolce of No. 11. These clusters
must presumably be held on for the full indicated duration, since passages in the
Preludio of No. 5 offer a counterexample. Signicantly, this is marked Adagio, e
arpeggiato nell acciaccature. At bars 1
and 2
the acciaccaturas are clearly marked as
small notes preceding the arpeggiated chordal notes. This occurs several times later;
elsewhere the dissonances are written as normal-size notes. The lack of such nota-
tion or any titular acknowledgement of their presence in the other movements surely
means that they are to be given full value elsewhere. As far as usage is concerned,
these clusters always occur at important points of harmonic articulation, either at a
cadence point or near the beginning of a phrase. This is substantially different from
The toccata is part of Sonata No. 6 in F major in Baldassare Galuppi: Sei sonate, ed. Iris Caruana (Padua: Zanibon,
1968) No. 5052; the Alessandro Scarlatti example is found in the opening section of his Toccata No. 9 in G
major from the Primo e secondo libro di toccate.
For example, Ann Bond writes that the added notes found in the left-hand chords in bars 8082 of K. 490
should be released quickly (without offering any rm musical rationale for this advice), while those found after
the double bar of K. 215 may be held on. Bond, Harpsichord, 199200.
See Sheveloff, Frustrations II, 96.
238 The keyboard sonatas of Domenico Scarlatti
Scarlattis use of them, where they are most commonly found in the middle of a unit
and less frequently at a beginning. Many of Scarlattis most striking uses of clusters
as for example in K. 115 or K. 490 cannot in other words be assimilated into the
traditional patterns of articulative or cadential delineation.
If we turn back to the source of this feature in continuo playing, it may be that
we do not in any case have the full measure of the historical evidence. Lars-Ulrik
Mortensen has recently drawn attention to the marked change in Italian continuo
style that had occurred by the beginning of the eighteenth century. Not only were
very full-voiced realizations common, but the doubling of dissonances was too, even
if it broke the rules. Mortensen maintains that the discretion and unobtrusiveness
in continuo playing so strongly advocated nowadays would have seemed no more
than a curious relic of the past to an [eighteenth]-century Italian musician.
tradition has an obvious relevance to Scarlattis practice, not just in terms of liberal
dissonance treatment but also in terms of full textures, and then more broadly in the
sense that such sonorities seem to be valued for their own expressive and sensuous
effect. However, Scarlatti does not in general aim for the marvellous fullness so
frequently noted of this style of continuo playing, and this reminds us of the limits
of such a parallel altogether: it does not really explain why the composer brought
such dissonances routinely into notated music. Although we have seen that they do
appear in other solo keyboard music of the time, this tends to be in more delim-
ited and far less striking contexts. In their exuberant excess, the continuo practices
reviewed certainly offer a closer match, but then the question arises: why should
Scarlatti wish to transfer such continuo technique onto the written page when its
whole raison d etre lay in being improvised? Indeed, such features were surely al-
lowable precisely because they were not committed to paper and hence beyond
close visual scrutiny. One other possible explanation for the clusters has been that
they reect guitar technique and, by extension, suggest an exoticpopular stylistic
On the whole, however, some conceptual gap remains. As with vamps, a
fairly rm historical context does not seem to be equal to what the sonatas present;
it is difcult ultimately to hear the clusters simply as an intensication of existing
It was stressed earlier in connection with cluster chords that the sensation of
dissonance often results as much fromaccumulation as the unorthodoxy of individual
harmonic entities. In the case of K. 64, for instance, the number of non-chordal notes
is relatively few, but their close proximity disorientates the listener. After the added
notes found in bars 28 and 30, for instance, the ear is easily persuaded that it is hearing
further clusters in bars 31, 32 and 34, yet these are simply chords of the dominant
seventh a dissonance so routine that we normally never even hear it as such. The
lasting impression of the whole is, to borrow a memorable phrase of Degradas from
his study of the late cantatas, of a deliberately dirty harmonization.
In other
Unerringly Tasteful?: Harpsichord Continuo in Corellis Op. 5 Sonatas, Early Music 24/4 (1996), 677.
See for instance Boyd, Master, 183, and Bond, Harpsichord, 182 and 199.
Degrada, Lettere, 303.
Irritations 239
Ex. 5.11a K. 150 bars 5762
Ex. 5.11b K. 198 bars 546
Ex. 5.11c K. 57 bars 96111
contexts, the dissonant sense can also accumulate through many small aberrations,
producing a sort of horizontal dissonance. K. 184, for instance, features so many
small clashes, near false relations and unusual scale forms that the whole work seems
to vibrate with dissonant sound. In many cases this seems to be in the name (or under
the pretext) of exoticism K. 179 in G minor offers one of many other instances.
Such dirty harmonic practice can take many different more localized forms.
In bar 58 of K. 150 (see Ex. 5.11a) the pedal c
in the alto, prolonged beyond its
harmonic function in the previous bar, illustrates a common means of generating dis-
sonance. This together with the spacing of the chord creates the harsh sound. In bars
and 55
of K. 198 (Ex. 5.11b) the right hands G and E imply a perfectly plausible
V6/4, only the left hand has already moved on to the (7/)5/3, another common
type of discrepancy. In bar 105 of K. 57 (Ex. 5.11c) we nd a disagreement between
I6/3 of F and a right-hand part that outlines IV with semitonal lower appoggiaturas.
240 The keyboard sonatas of Domenico Scarlatti
Ex. 5.12 K. 407 bars 198
Note that in the model for the passage, at bar 101, all is correct, but second time
around the right-hand material begins a bar ahead of itself, as it were, and this
causes the clash. The effective superimposition of F and B at major chords here
may be taken so much further in other works that one wants to reach for another
apparent harmonic anachronism bitonality. In bars 1012 of K. 214, for exam-
ple, the imitative counterpoint between alto and tenor takes precedence over the
harmonic sense and we consequently hear a mish-mash of harmonies that sounds
Irritations 241
Ex. 5.12 (cont.)
If most of these harmonic clashes need many notes to make their effect, the
Sonata in C major, K. 407, manages with a minimal texture (see Ex. 5.12). This
skittish work features the most apparently gratuitous of dissonances, the insistent
major seventh rst heard in bar 16, yet this is inspired by a less conspicuous piece of
misbehaviour found at the outset. The respectable device of imitation subtly misres,
setting up problems that are quite systematically worked through for all the apparent
eccentricity. Just after the left hand enters with a tonal imitation of the right, the
right hand strikes a C, which lends some aural confusion to the event. Although we
242 The keyboard sonatas of Domenico Scarlatti
are still in C major, the F of bar 2
being required for voice-leading reasons, what is
offered suggests a play of modulatory indicators. In any C major work F is the rst
important accidental we might expect to hear, as it indicates the basic grammatical
move to the dominant; C would be the next such accidental, in the typical process
of oversharpening which enables the subsequent settling on V to sound relatively
stable. What happens in bar 2 suggests an attempt to go to V and V/V simultaneously,
a crowding of the natural course of events. The too-close proximity of C and F
must therefore be teased out from bar 16 onwards.
The mini-consequent from bar 2
in the right hand makes as if to continue the
same textural process, but at bars 45 the hands suddenly play together, in contrary
motion. This much simpler form of counterpoint suggests a marked retreat from the
earlier complications. The behaviour of the two hands in relationship to each other,
in conjunction with the harmonic argument, becomes one of the main themes of the
piece. The very plain C major cadence that follows seems to expose the redundancy
of the earlier accidentals. However, just when we are reaching the equivalent point
of the second, matching phrase, the new F at bar 11 moves us toward a half-
cadence on V of V (using very standard phraseology compare bars 678 and
824 of K. 243, for instance). The whole phrase lasts nine and a bit bars from
this point all phrase lengths are deantly irregular except for those that nish each
Almost by way of compensation, the motivic construction of the sonata is very
clearly dened. If reduced to its lowest common denominator, motive (a) can be
dened as a descent of about half an octave followed by a second (in either direction).
This is heard more simply than it can be described; versions of it may be found at
bars 01
, 5
, 11, 12, 312, 44, 50
51, 623 and 6871. Motive (b) consists of a
scalic third; see for instance bars 1
2, 4, 1314, 16ff., 23, 43, 51, 73. An extension
of this third into a scale may be found at 2022, 2931, 434, 546, and 68ff.
(in both hands).
The Schleifer
gure that initiates the obvious wrongdoing at 16 is a version of (b).
According to the understood usage of this gure, the outer notes should receive
harmonic support and the middle one act as a passing note between them. Thus the
and f
should be consonant, but in fact the g
is, since it ts with the left-hand
harmony. However, it cannot be heard in this way; the rules of usage demand that g
be heard not just as subordinate, but as an embellishment of the embellishment (it is
a passing note from the consonant skip a
on the way to the primary pitch, the f
In other words not only is the f
dissonant, but it receives diminutional support to
double the dissonant effect. Suggestions that this passage represents a village band,
or even out-of-tune bugles should not be dismissed, but they divert attention from
the radical aspects of K. 407s harmonic argument, substituting an amiable pictorial
See footnote 22 on p. 11 for an explanation of this term.
See Chambure, Catalogue, 139, and Kirkpatrick, Scarlatti, 202.
Irritations 243
One should also note that the Schleifer pitches are set up by the right-hand pitch
activity throughout bars 1315, precisely the standard formula that enunciated V
of V; this reinforces the sense that bar 16 represents a superimposition of V and its
dominant (just like bar 2). Thus while the left hand moves properly from the cadence
point on D onto the dominant G, the right hand continues to express D through
the triad members F and A; having originally spurted ahead, it now lags behind.
The f
dissonance does not even resolve properly, to the g
for which it so painfully
substitutes; it moves in bar 20 to f
(the wrong harmonic direction!), becoming
part of V
of IV of V. This is followed by witty augmentations of the Schleifer twice
over at 21 then 23, the texture thins, momentum slackens and we nish back on V
of V. This is exactly the point reached in bar 15, so that the harmonic argument
has failed to advance. In order to reach the desired end of a properly articulated G
major Scarlatti must therefore transpose by a fth, so that we start with V of V and
its dominant.
Afurther complication should be pointed out, in that although the left hand at bars
256 seems to move between I and IV of D (and at 1617 between I and IV of G),
its activity may be read in another way, as a move between V and I of G (and V
and I of C in the previous phrase). In this latter reading, while the right hand pulls
sharpwards, the left hand in fact pulls atwards, so that both are a step away on the
circle of fths from where they should be. Thus not only is there an implicit bitonal
clash between triads of A major and D major at 25, for instance, but bar 26 hints at a
clash of G and A majors. Over and above all this, the dissonant note is now C, the
other over-eager accidental of bar 2. This at least has been successfully disentangled
from its bar 2 companion.
The repetition of this unit from bar 34 may seem unbalanced (since the sonata has
been moving in paired phrases) but also makes sense; it leads to another close on V so
that we have two on V/V and two on V. The closing unit brings relief in the form of
very clear patterning. A contrary-motion form of (b) is heard in both parts in bar 43,
then (a) follows in the right hands next bar while the left continues down to form
a scale. In fact, much about this material specically recalls bars 45. Not only does
this introduce an eight-bar unit, but the internal divisions of that unit are as clear as
they could be. Further, it provides at long last a proper dominant equivalent to
the single tonic cadence of the half. The closing phrase also has a specic textural
and indirect registral signicance. The fact that the hands nally make sweet music
together acts as a sort of (temporary) textural resolution. Registrally, the coverage of
the whole keyboard in this phrase forms an antithesis to the previous sense of being
stuck in a groove which accompanied the repeated dissonance. The expansiveness
of tessitura helps to signal the harmonic relaxation.
C is immediately reintroduced after the double bar in a manner that matches
bars 258 (tied Cs heard two bars apart). This seems rather cruel after its effortful
eventual removal from the rst half. In addition, the vertical C/G clash of bar 2
is revived by the g-c
of bar 51
, this being further dramatized by a new insistent
inner voice. It is placed in the context of a diminished triad, formed with the B
244 The keyboard sonatas of Domenico Scarlatti
heard in the bass. At bar 54 this tritone is given a satisfactory harmonic context:
of D minor. More fundamentally, the C is at last allowed more straightforward
generative powers, as it leads to a tonality from which the F is excluded. As a further
layer in the directional harmonic game, the seven-bar phrase nishes at 57 a fth on
the sharp side from bars 15 and 24.
Unlike the model, the following Schleifer in bar 58 is consonant in context and
offers a proper voice-leading resolution of the preceding elements: f
a D minor I after the previous V, and the pair of d
s at 58 and 60 answer the pair of
s at 51 and 53. Not only that, but the Schleifer gap-lls the tritone, with both the
and g
from the start of the second half moving impeccably in by step. The left
hand from bar 58, which reuses part of the opening point (compare bar 1 in the right
hand and bar 2 in the left), is harmonically ambiguous, though. The FA dyads look
back to the previous phrase, forming a D minor 6/3 with the upper voice, while
the alternating EG dyads look forward to the following brief tonicization of A
minor. The introduction of G forms part of the game of harmonic balance as it is
a further step sharpwards on the circle of fths; it also rubs against the surrounding
Bs. The B then takes over in an attempt to cancel out all the too-prominent and
awkwardly managed sharps.
Bar 62 is hypermetrically ambiguous; it seems really to function as an extended
upbeat to 63 using another version of (a) the c
traced at 62
. The
Schleifer with which it overlaps once more has a possible functional relationship with
both third pairs, either of which could be the prolonged harmony. Now, however,
the order is reversed; the second dyad AC ts with the previous A minor harmony,
while the initial GB moves us toward F major. So for all the relative consonance
there is still an element of ambiguous overlap. We should note too that, alongside
the F major, D and A minor are both relatives of at-side keys (C major counts as
at in this notional context of prolonging V). Another ve-bar unit follows from bar
68, leading to a V of C version of 1415 at bars 70
72; this sets up the expectation
of a return to the material of bar 16.
The problem material from bar 73 is much less dissonant than its rst-half equiv-
alent, due to a completely different left-hand part; instead of using the material of
16ff. the composer inverts the two left-hand parts from bar 51. The end result is a
completely clear V
of C. This harmonic clarication is aided by a topical relaxation
into a clearly popular mode, as can be heard in the insistent drone fths of the left
hand. Surely it is only now that we can truly hear the village band. Even then
there is a tweak of the tail in the barely manageable left-hand ornament in 76. At
bars 778, though, we nd a real twist having sorted out the rst part of the origi-
nal offending phrase, the composer now complicates the second part. Thus Scarlatti
follows the cleansed equivalent of 1620 (more accurately 258) with a more dis-
sonant version of 2930, prompted by the need to have more at-side emphasis to
counter the CF complex. The offending note is the left-hand b
in bar 78; this
creates very clear bitonality between hands, more explicit than anywhere else in the
piece. The B is necessary so as to break the literalness of transposition, otherwise
Irritations 245
the phrase would end in F major. Of course, having rewritten bars 736 and 7980,
Scarlatti could have done the same with bars 778! It is all part of the game.
The closing unit returns intact, almost exactly transposed. This is a necessary piece
of absolute symmetry given the continual adjustments that take place elsewhere, and
once more there is some sense of topical relaxation; the exact repetition of short units
has the avour of comic opera. Except towards the end of these units the tessitura
of the piece is high; the lack of low bass registers accords with the lack of security
in harmonic movement. K. 407 offers a skit on harmonic properties, rejoicing in
an uncoordinated execution of the expected tonal plan. Its real subject concerns the
question How does one modulate?, with the movement to V dramatized through
the most glaringly dissonant of means. All the expected moves are there, as indicated
by the sequence of accidentals, but they are radically disembodied through being
isolated, the normal harmonic background being withheld.
The wit of Classical composers, of whom Scarlatti is perhaps to be regarded
as the rst, is rather like that of the metaphysical poets they couldnt help it,
it was simply a natural way of thinking and writing. It is based once more on
Subotniks supreme condence of a style in which. . . tonality was so secure. In this
style, the weight and power and articulation of tonal areas were exciting in their
own right and were sufcient in themselves to concoct a rousing story, as K. 407
illustrates. The modulation to the dominant in particular was literally an art form.
This need not of course be problematized, as it is here, in order to be effective; the
very act itself was assuming a harder creative edge. Those who miss the harmonic
complexity of Baroque and nineteenth-century language often fail to grasp the
visceral excitement of tonal articulation that is found in what Carl Czerny called
an art then at the height of its youthful powers.
Scarlattis condent harmonic practice is also unusual in less sensational ways.
His modulations may be marked by some peculiarity of modal or registral treat-
ment, or may even be surplus to formal requirements.
Although such habits may
be understood as the sort of clever playfulness discussed above, they can also be
understood more hedonistically. In other words, colour seems to outweigh the de-
mands of grammar. K. 223, with its ungrammatical chord progressions (discussed
further in Chapter 6), seems to offer an extreme example of this, but many of the
aberrations considered in this chapter may be contemplated in such a light. The
notion of a sensuous approach that transcends grammatical meaning or function has
produced many comparisons with music of the twentieth century, especially with
the treatment of harmony and texture by Debussy and Ravel.
Cited in Villanis, Italia, 169. Hans von B ulow endorsed Czernys assessment in the preface to his edition;
B ulow, Klavierst ucke, i. One should not overlook the fact that such remarks indicate the growth of an idealized
view of the eighteenth century, all pre-lapsarian purity and innocence, to which I have referred a number of
times; nevertheless, this perception of fresh power seems to me essential to an understanding of post-Baroque
eighteenth-century harmony.
Sheveloff describes Scarlattis modulations as militantly individual; Sheveloff, Grove, 341. Haas, Modulation
and Talbot, Shifts also contain thoughtful discussions of the composers modulatory practice.
246 The keyboard sonatas of Domenico Scarlatti
Ex. 5.13 K. 188 bars 10423
Just as remarkable, though, as the features that prompt such comparisons are
the many subtly unusual touches captured so well by Kirkpatrick when he wrote
that in the sonatas of the middle period, Scarlatti succeeds in making conventional
harmony sound even stranger than before.
In many cases this can be achieved from
without, through the disembodying implications of surrounding unconventional
material, or it may arise through unusual textural or rhythmic gestures. On many
other occasions, though, it seems to be simply the harmonic expression in its own
right that is suffused with an undemonstrative strangeness. Often this is connected
with a subtle, barely glimpsed modal avour. It may be found, for instance, in the
three-card trick heard frombar 20 of K. 183, in which several diatonically ambiguous
notes lend an unusual avour to a harmonic process that is in any case somewhat
opaque. Sometimes this ambivalence is connected with the establishment of a new
key, as with the eetingly unsatisfactory c
heard in bar 18 of K. 125, surrounded
by Cs which denote a smooth transition toward the dominant.
Here any modal
avour is a by-product of basic tonal manoeuvres. Ex. 5.13, from K. 188 in A minor,
is an exemplary case of subtle oddity. This sonata is dominated by the minor mode,
save for a brief account of C major in the rst half and the return to C promised by
bars 10910. With the D minor of bar 108 doubling as II of C, the next two bars
outline IV and V, and although the bass I is articulated in bar 111, the inner-voice
A here cuts strangely across the expected chordal completion. Similarly in bar 114,
an inner-voice D lends ambiguity to a harmony that ought surely to be F major.
Kirkpatrick, Scarlatti, 1645.
Sheveloff perfectly describes this C as a very special note, a vague partial negation of the motion towards the
dominant that, while insufcient to arrest it, adds considerable spice; Sheveloff, Keyboard, 417.
Irritations 247
The unusual parallelism of the left-hand voices is hard to account for. It may well be
heard as exotic, but most of the rest of the sonata accomplishes this far more overtly,
and the entrance of a new, distinctive melodic line from bar 111 suggests that we are
hearing a relieving episode amidst the popular reiterations.
All of the strange effects or irritations considered so far, no matter how certainly we
might think we can grasp them, continue to nag away in ones mind; Scarlatti would
presumably approve of the collective critical neurosis they have induced. A number
of global explanations have been advanced for his unreiner Satz. The learning
to liberty equation already discussed can be further inected by considering two
Spanish cases of the earlier eighteenth century. The Missa Scala Aretina written in
1715 by Francisco Valls caused a famous controversy; its Miserere nobis features
a second soprano part introduced in dissonant intervals of a second and ninth. A
censure published by Joaqun Martnez de la Roca of Valencia Cathedral began a
pamphlet war that lasted for ve years until 1720, with some seventy-eight being
published altogether. Valls defence was: If in the pursuit of beauty a rule of the
ancients is temporarily disregarded, what evil is there in that? Even Alessandro
Scarlatti was invited to comment, and did so in a 1717 Discorso di musica sopra
un caso particolare in arte.
Given the participation of his father and the fact that
the affair took place just a decade before his arrival in Spain, and a few years before
his relocation to Portugal, we may well assume that Domenico was aware of such
polarized feelings. Any easy critical movement from the learning promoted by the
conservatives to the beauty that may result from infractions of the rules must be
reconsidered in such a light.
A similar controversy that took place in 1756 and 1757 between Jaime Casellas of
Toledo and Josep Duran of Barcelona has recently been uncovered. The polemics
began with the criticism by Casellas of a madrigal by Duran for its offences against
the rules of contrapuntal science. In reply, Duran proposed another kind of knowl-
edge, less rational and more sensible and artistic. In support of his freer treatment
of dissonance, Duran listed a number of illustrious Italians, noting the emphasis
placed on originality and inspiration in Neapolitan conservatories. As well as citing
his teacher Durante, he also mentioned Scarlatti in justication for his freedoms.
(In the context of such a polemic it seems doubly odd that Scarlatti himself should
seem to claim the contrapuntal high ground in his 1754 letter to the Duke of
Huescar.) Such theoretical and aesthetic disputes make one wonder whether some
of Scarlattis licences were informed by a consciousness of this particularly (although
hardly exclusively) Spanish debate. (Recall in this connection the world of K. 402,
See Hamilton, Spain, 21823,

Alvaro Torrente, A Critical Approach to the Musical Historiography of
Eighteenth-Century Spanish Music (Cambridge: unpublished, 1995), 1213, and Zuber, Blumen, 16.
Anna Cazurra, The Polemics between J. Casellas and J. Duran Regarding Italianism in Spanish Music of the
Eighteenth Century, paper read at the conference Music in Eighteenth-Century Spain, Cardiff, July 1993.
248 The keyboard sonatas of Domenico Scarlatti
discussed in Chapter 3.) On a larger scale, any notion again of Scarlatti work-
ing from the most respectable of technical bases runs counter to such historical
An associated rationale for Scarlattis liberties is implied by Durans allegiance to
a new Italian school, but has rarely found voice in the more recent past. This is to
understand the liberties as a sort of Italian pragmatism, a cousin of the shoddy work-
manship that stands in implicit contrast to the Austro-German technical world.
Thus Ann Bond writes that Scarlattis writing is full of loose ends unresolved
discords, parts that disappear, and so on. Like all Italians, he writes for immediate
effect and does not worry about academic detail in situations that pass too quickly to
be observed.
Although the suggestion of an anti-academic orientation is sound
enough, the implication that such loose ends arise quite innocently or are simply
culturally determined seems inadequate to the scale and nature of the operation. As
we have seen, it is precisely in the conception and manipulation of such features that
the composers learning does appear.
Associated with this rationale in turn is the appeal to continuo practice so elo-
quently advanced by Kirkpatrick. In this interpretation the loose ends reect the
almost unlimited [liberties] that can be taken in the conduct, in the omission of
parts, or even in the occasional introduction of doubling consecutives in the inner
parts. Perhaps, he wrote in an appeal to insider knowledge, only the experienced
continuo player and harpsichordist is prepared to understand it.
Even if we accept
the terms of this argument, we must ask What kind of continuo playing? The
assumption that continuo practice is a monolith, outside time, style and country, has
been nicely punctured by the work of Mortensen cited earlier. More broadly, we
must again wonder why this explanation should hold more for Scarlatti than any
other keyboard composer of the time, all of whom we may assume also had plenty
of continuo experience.
Another explanation too issues directly from the keyboard. Luigi Villanis, noting
Czernys complaints about the incorrectness of some passages, averred that these
were liberties often granted to the virtuoso.
Are virtuoso gestures exempt from
the rules of good conduct? In bars 379 of K. 56 (see Ex. 5.14) the left-hand sevenths
on the second beat move up a step on the fourth beat. The right hand meanwhile
features correct resolution of the sevenths. This may be a joke on our perceptions,
since with the urry of hand-crossing by the left hand, such crudity of voice lead-
ing may pass unnoticed. In such a case the virtuosity almost acts as a pretext for
the infraction rather than a simple causal explanation, so that again any sense of
Libby, Italy, 15. For a fuller quotation see Chapter 2, p. 59.
Bond, Harpsichord, 182.
Kirkpatrick, Scarlatti, 238. This global explanation has been enthusiastically endorsed by Roberto Pagano.
Kirkpatricks intuition of basso continuo practice as the stylistic matrix of Scarlattis keyboard writing would
alone be enough to make him the true interpreter of Scarlattian poetics; his text continually refers to the
experienced continuo player to resolve problems that continue to be insurmountable obstacles for musicolo-
gists with a less rened . . . and complete critical armoury. Pagano, Dizionario, 634.
Villanis, Italia, 169.
Irritations 249
Ex. 5.14 K. 56 bars 379
innocent departure from the rules is compromised. The difculty with all these sug-
gestions is that they are rather blunt instruments. None can conceivably apply only
to Scarlatti. If we accept their explanatory force, we have to ask, once again, why
such factors did not allow for more Scarlattian ventures from other composers.
Of course it is not just the modern critical community that struggles to come to
grips with such features. Even once the disputes between ancients and moderns, as
illustrated by the Spanish cases considered above, had lost some of their force later in
the eighteenth century, there was still the difculty of how to come to terms with
the freedoms found in the new instrumental style. In an English context, as Simon
McVeigh comments, it was only towards the end of the century that there was an
attempt to explain the whimsical contrasts of modern instrumental music, which
accorded neither with the sublime nor with the beautiful. He points to the new
aesthetic category of the picturesque developed in 1794 by Uvedale Price. Although
this could carry its literal meaning, as found for instance in Haydns folk material, its
more important attributes were capricious contrast and lack of symmetry. Price, in
An Essay on the Picturesque, highlighted sudden, unexpected, and abrupt transitions,
a certain playful wildness of character, and an appearance of irregularity in the work
both of Haydn and of Scarlatti.
The phrase playful wildness evokes the spirit of
many of Scarlattis adventures most aptly, and indeed the category picturesque may
be usefully invoked in both its senses. Of course, the literal sense of the term must be
treated with some reserve, and even the applied sense may lend too friendly a face
to many of the composers misdemeanours. Nevertheless, Prices concept reminds
one that Scarlatti and Haydn can be protably linked both aesthetically and also
in a sense historically, given the warm reception of the music of both in England.
An equivalent term, the ornamental, was coined by William Crotch in the early
1800s. For him, Scarlatti was the originator of such a style, in sonatas in which all
is calculated to amuse and surprise, to create a smile if not a laugh.
A further assessment of the spirit that such freedoms seem to serve comes from
another sphere, Barbara Trapidos novel Temples of Delight. The mother of Flora
Fergusson, a friend to the books central gure, was at the time of her marriage a shy
young music student with. . . a graceful, gliding carriage bearing witness to many
Simon McVeigh, Concert Life in London from Mozart to Haydn (Cambridge: Cambridge University Press, 1993),
See Annette Richards, The Free Fantasia and the Musical Picturesque (Cambridge: Cambridge University Press,
2001), 110. She notes that Scarlatti was also often paired with C. P. E. Bach in English criticism (113).
250 The keyboard sonatas of Domenico Scarlatti
years at the exercise bar in ballet classes. Mr Fergusson, on the other hand, was a
miser, an educated and scholarly man of the drier and dustier sort[;] . . . it distressed
him to part with money. After marriage Floras mother threw herself into domestic
duties that left her with little time for music:
Her Scarlatti scores languished, leprous with neglect, in a damp gas cupboard from which
they emerged only with the move to the prime locality four years later . . . So the house
was devoid of music. It went without saying that the elderlies, who regularly banged on
the ceiling with broom handles at the sound of a footfall on the oorboards, would have
considered Scarlatti sufcient grounds to petition for the Fergussons eviction. . . She had
assumed, for the rest of her days, a kind of greyish camouage which worked its way deep
into her being . . . She held her mouth permanently drawn into a tight, disgruntled little knot
like an anal sphincter.
In a nice variant on the game of ancients and moderns, Floras mother was a dancer
and she marries a man with an accountants mentality. Her abandonment of the
music of Scarlatti is equated with a loss of vitality, grace, generosity and colour,
made even plainer when we read later: Youll starve, my girl, her mother said,
and she drew up her mouth in that mean, pinched little gesture, born of all those
decades of repressing Scarlatti in the gas cupboard.
Scarlatti becomes the symbol of
a rich and authentic life. He is also, to adapt this to our particular current purposes,
very unclerical in his creative work quite the opposite of everything that is mean,
dry and pedantic.
The uncertain status of some of the tempo markings given to sonatas forms part of
the universal set of ambiguity surrounding so many Scarlattian operations. A small
number of writers have picked up on this difculty: that many Andantes and Allegros
seem to approach each other in actual speed.
Andantes often seem to be quicker
than we might expect, and the ubiquitous Allegro marking seems susceptible of very
different interpretations.
Naturally one could not claim that this is an ambiguity
unique to Scarlatti; just to take several examples from within his orbit, Alberos
Sonata No. 18 in B minor is marked Andante but seems to demand a quick and
aggressive approach, while the rst movement of Seixass Sonata No. 31 in D minor
(1965) has material of a pronounced Allegro cast yet is marked Largo. We also noted
earlier in this chapter a Giustini movement, from his Sonata No. 3, that was headed
Andante, ma non presto! Similar apparent ambiguities are in fact frequently found
Trapido, Temples of Delight (London: Penguin, 1990), 513.
Temples of Delight, 99.
See Kirkpatrick, Scarlatti, 293, and Pestelli, Sonate, 218.
Note Hermann Kellers remark that, for Scarlatti, Allegro seemed to be an almost neutral, exible concept;
Keller, Meister, 64. See also Howard Schott, Playing the Harpsichord (London: Faber, 1971), 115 (the sonata
K. 24 is misidentied as K. 27).
Irritations 251
in the music of the rst part of the eighteenth century.
Contemplating such cases,
and the extent of them, can suggest that there has been an irrevocable slippage of
meaning and usage in many tempo designations. On some counts, though, we can
be sure; it is quite evident that an Andante marking denoted a considerably quicker
speed in the eighteenth century than it came to do subsequently. Thus the Allegro
andante appended to K. 343, for example, should not be seen as problematic in
itself; nor apparently the Andante allegro given for K. 151, except that the work
with which it is paired, K. 150, also in 3/8, is marked Allegro yet seems to require
a much less lively one-in-a-bar execution. When we nd that the primary sources,
V and P, sometimes disagree on tempo indications, we might feel that such a matter
was not even conceptualized in the eighteenth-century mind, so that it was treated
with what looks to us like relative indifference. Finally we must acknowledge that
tempo in any era is a fraught business, that it often nds a relatively low level of
intersubjective agreement, as we all insist on the integrity of our personal taste, or
the correctness of our body clocks. Georges Beck, for example, asks why Scarlatti
places Andante at the start of K. 86 when it is clearly an Allegro,
yet the given
indication seems to me to correspond quite adequately to the owing character of
the music and its proper performing speed.
What lends this issue a keener edge in the case of Scarlatti is the celebrated lack
of slow movements. As already noted, the overwhelming majority of sonatas carry
designations of Allegro or quicker, while tempo indications slower than Andante are
almost unknown. This is not just a question of markings on the page, however; it is
more crucially one of affective character. Scarlattis slower movements, his Andantes,
do not by and large appear to deliver those qualities of solemnity, lyrical warmth,
concentration, respite and inwardness that we variously expect to nd in a good
proportion of slower music of his and other eras. Indeed, it sometimes appears that
the composer does not even recognize or allow the distinct affective character so
cherished by listeners and other composers. Thus a number of his Andantes seem to
offer passages of misplaced Allegro music. Bars 1418 of K. 213 in D minor show
one example of this, in a work that denitely ranks among the composers slower
specimens of Andante tempo. This passage could easily be felt as one-in-a-bar gu-
ration, so unlike the heavy crotchet harmonic rhythm that predominates elsewhere
in the 4/4 metre. Although it seems gesturally thin in this context, this is not to
say that it cannot be justied or made effective in performance; one could main-
tain that its very bareness creates a type of tension that ts well in a work that
contains many harsh angularities and strong dissonances. Bars 212 of K. 259 in
G major also seem to lack sufcient tension in context, but this is a rather different
case from K. 213. All the material from this point to the end of the half is conceivable
at an Allegro tempo indeed, in his recorded performance Mikhail Pletnevs tempo
See for example Peter leHuray, Authenticity in Performance: Eighteenth-Century Case Studies (Cambridge: Cambridge
University Press, 1990), 368.
Beck, R everies, 16.
252 The keyboard sonatas of Domenico Scarlatti
is frankly Allegro
so that K. 259 appears to offer an example of an Andante
marking that is hard to come to terms with. However, the opening material of the
sonata, all Arcadian innocence, is clearly of an Andante typology. Ralph Kirkpatrick
recognized this difculty when he wrote that harmonic progressions that knit well
and sound simple and clear in fast passages sometimes seem to lose their momentum
at a slowtempo, unless heard in terms of the long span of tonal structure.
the diagnosis seems more convincing than the suggested adjustment of perception.
Although one must consider whether Scarlattis Andantes can even be conceived
as a category given the implications of the tempo ambiguity discussed above, many of
them do in fact seem to form a race apart. They qualify as irritations not necessarily
on the technical grounds covered earlier but on two other counts. They often
suggest a listless and uncentred expressive character, and this is turn can act as an
irritant given the affective expectations we bring to slower movements. A common
perception, for instance, has been the difculty faced by the performer planning a
Scarlatti programme when there are so relatively few works that can offer the right
sort of respite or variety.
One rationale for this perceived absence that must be
entertained lies in the fact that Scarlatti wrote almost entirely a series of separate one-
movement sonatas. Given such self-sufciency, considerations of inter-movement
balance need never have arisen. Indeed, the attractiveness of the pair theory to those
who believe it was a creative rather than clerical matter surely lies in the way that it
overcomes this disconcerting aspect of Scarlattis sonata production.
The question of expressive character has occupied Pestelli in particular. He writes
that slow movements do not adapt well to the Scarlattian art, suggesting an inability
to relax. This incompatibility of character between Scarlatti and the slow move-
ment, however, did not prevent [him] writing beautiful specimens in which rhyth-
mic restlessness becomes the principal poetic motive. Pagano takes what he believes
to be the harmonic orientation of the slower movements as the basis for an intriguing
characterization of melodic style: Even if many of the melodies of the slower sonatas
show stylistic connections with the most characteristic features of Italian vocal style,
the choice of harmony as the basis of the poetics lends melodic elements a role that
is often decorative, sometimes nostalgic, in certain cases parodistic.
Paganos commentary presupposes the central role of melody in slow movements,
as the prime focus for the heartfelt expression to which we are accustomed. It will
not do to suggest that such an affective expectation is anachronistic; to take another
example from Scarlattis immediate orbit, the slow movements in the sonatas of
Seixas have much greater expressive immediacy.
The melodic tendencies proposed
Virgin: 5 45123 2, 1995. On the other hand, Christopher Headington describes K. 259 as being like a stately
and melodious minuet; notes to recording by Dubravka Tomsi c (Cavalier: CAVCD 007, 1987), [ii].
Kirkpatrick, Scarlatti, 223.
See Rousset, Statistique, 79, or Kirkpatrick, Scarlatti, 3223.
Pestelli, Sonate, 218; Pagano, Dizionario, 637.
They are described by Brian Allison as more dramatic and expressive than those of Scarlatti. Carlos Seixas:
The Development of the Keyboard Sonata in Eighteenth-Century Portugal (DMA dissertation, North Texas
State University, 1982), 18.
Irritations 253
by Pagano together with Pestellis rhythmic restlessness help us to approach and
dene the markedly unsentimental character of many of the Andantes. They do
have intensity but they do not have warmth, at least not of a straightforward sort.
The relentlessness with which we found Cecil Gray expressing unease in Chapter 2
is nowhere more tangible than when we contemplate the affective properties of these
works. Perhaps this is yet another area of accepted relaxation or creative automatism
where our composer shows constant vigilance.
But this is not so much a binding denition of expressive character as a hint at a
avour conveyed by so many of the Andante sonatas. They are certainly not lacking
in lyricism many give the sense of a well-dened individual lyrical voice that we
noted early on with K. 277 (Ex. 1.2) but this often tends to be somewhat passive.
The greatest lyrical fervour is often in fact found in faster or livelier pieces. One
instance of this passive conduct is the habit of concluding each half of a slower
sonata with successive downward couplings of a short phrase unit, which seem to
allow the music to drain away rather than nish cleanly. Examples may be found in
K. 158, 197, 234 and 481. On a different plane we have already dened the passive
attitude to time embodied by a sonata like K. 404. Indeed, the intense expressive
austerity discussed in that connection offers another conceptual category that we
may protably explore. Acertain sense of fatalism, of a melancholia ritually expressed,
imbues many of our Andantes, in such works as K. 234, 426 and 546. This often
arises once more from repetition. The very contained syntactical sense of K. 234, for
instance, is created by the use of just two basic ideas, which are repeated internally as
well as recurring in various forms in each half. This yields a certain grave formality
which is reinforced by a relatively austere harmonic language. Rafael Puyana remarks
that this austerity derives from an old Spanish tradition. The intense loneliness
which Jane Clark evokes as an essential element in the sonatas is also dened in
relation to Spanish tradition, if through the very different agency of folk music.
This quality might seem quite opposed to those outlined above, but the composite
Andante avour we are pursuing derives much of its fascination from the tension
between personal and impersonal expressive modes. One of its by-products is the
restlessness already mentioned.
Many of our Andantes contain pronounced old or archaic elements, which tends
to reinforce the terms of Puyanas austerity. K. 185, for instance, begins in the manner
of a chaconne. The opening of K. 296 in F major presents a typical Baroque gambit,
one associated with Corelli, in which sustained upper voices are set against a falling
bass line.
Scarlatti makes the held top voice(s) of the trio sonata model idiomatic
to the keyboard through repetition, and the combination of falling stepwise motion
and repeated notes is felt in many subsequent passages. Yet for all the surprises and
odd features that follow this model opening, the sonata lacks dynamism. The many
Clark, Boyd Review, 209.
See Mortensen, Continuo, 672. Compare this opening material with that found at the start of Marcellos Sonata
No. 8 in B at major or Seixass Sonata No. 6 in C major (1965).
254 The keyboard sonatas of Domenico Scarlatti
repetitions are curiously lacking in cumulative effect; they seem to exist for their
own sake rather than for functional purposes. The music seems to hover rather
than to unfold with a sense of clear direction. It is as if the composer is trying
to write a piece of music without any ideas in the accepted sense; instead, more
abstractly, the notes dene space and time, a concern that is reminiscent of the vamp
The only really sharp edge to the structure of K. 296 is encountered in the build-
up to and climax of bars 512. This is one of the most frankly Spanish passages in
Scarlatti, a rare open acknowledgement of source. It shows that the Andante quality
we are trying to dene may not obtain through an entire structure. Andante for
Scarlatti seems to be cognate with a certain expressive groundlessness, difdence
sometimes, that is quite unlike the energetic certainty of gesture that informs many
of the quicker sonatas. This of course can be a virtue it produces the poetic
motive of ambivalence and restlessness. Sometimes, however, as here in K. 296, the
music snaps with varying degrees of violence. This may involve outright rupture
although this is more likely in those idyllic works that lie at the edge of our current
concerns, such as K. 215 or K. 277 or what I dene as a lyrical breakthrough, to
be discussed in Chapter 7. In this case, as found in sonatas like K. 426 and K. 408,
there is a strong, but always brief, suggestion of the emotional frankness we expect
to nd in many slower movements.
The Sonata in D major, K. 534, shows all the elusive qualities of its species. This is
certainly a piece that fails to declare itself, whose expressiveness lies in its uneasiness
and ambiguity. It contains several ourishes that hint at the French overture, as
with the imitative points at bars 12, 56 and 1011. Interspersed with the Baroque
posings are many Spanish touches; the chains of acciaccatura gures heard throughout
might be galant in another context, but the guitar-like harmonies (as in bar 12) push
them in another direction. The interrupted progression to IV
(instead of VI) at bars
18 and 34 certainly sounds exotic, although what follows up to the cadence point is
standard galant cadential diction.
The many imitative and contrapuntal touches during the Spanish passages are
difcult to read are they simply to be taken as part of the unfocused rhetoric of
the sonata as a whole? This is certainly not a democratic mixture of elements as
found in K. 96; rather, it sounds thematically restless. The events at the start of the
second half are typical of this strain. The Baroque ourish leads directly into an
exotic descending scale in sixths (sounding like a lament) above a repeated low A,
easily the lowest note of the piece. This singular event is cut off by a return to the
opening ourish in the bass. The right hands imitation, the rst not at the octave,
is in turn cut off by an abrupt shift to the minor and a return of Spanish diction.
The subsequent half-cadence is reached by means of a tenor suspension gure heard
on a number of occasions through the sonata a strangely disembodied reference
to a learned style. The continuity of thought is fairly consistently tenuous in this
Irritations 255
Like K. 534, K. 544 in Bat major is marked Cantabile. For Massimo Bogianckino
this sonata seems caught up in the threads of an indenable malaise suggesting a sort
of tedium that musical expression had most certainly not known before.
A sense
of malaise is indeed palpable, as in K. 534, although the present work is clearer in
its expressive contours, with a long climactic passage after the double bar and two
very long silences. The initial material is heard four times in the rst half, starting
twice on the tonic and twice on the dominant. The phrase from bar 7 has a more
overtly pathetic shaping, with its repeated sighs and the build-up of textural and
tessitural intensity. Yet from bar 12 this music dies away (just how graphically will
depend on how the performer takes the Arbitri instruction applied to a brief urry
of semiquavers). The appearance of the transposed opening material from bar 14,
especially after such a long silence, might suggest a retreat from the previous shaping.
Its exact repetition from 18 furthers the feeling of unexpansiveness.
From the start of the second half the head motive nally leads to something
more expansive, introducing a phrase of sustained intensity. With the transposed
forms found after another long silence from 33ff., which also of course refer to the
opening, we realize that this is a piece that starts again and again. It seems weighed
down with gestures, realized in desultory fashion. Concentration is achieved only
with the lyrical blossoming in the rst part of the second half.
But how can it be desultory in spite of the minimum of material used and the
frequent repetitions? There is an odd temporal perspective inherent in this sonata.
On the one hand, K. 544 consists of just a handful of phrases, with a good deal of
internal repetition. In this sense the work is almost miniaturistic in the manner of
K. 431, yet there seems to be a disproportion in the relation of part to whole. The
dominant area of the rst half, bars 1422, consists only of one phrase, repeated with
the customary overlapping. One might normally expect such a passage to be merely
a part of a larger section it might function as a closing theme, for example, or the
start of the second-subject group. On the other hand, the piece seems interminable
in its stops and restarts (repeats need to be taken for the full effect). Thus there is a
sense that the piece is both too short and too long.
A small-scale embodiment of this elusive, enigmatic temporal sense is found in
the Arbitri indications. These also seem strangely proportioned. They are far too
slight to represent some sort of release after the intensity built up prior to their
appearance. They seem more throwaway gestures than the resolving ourishes which
the rhetorical situation would seem to demand. To elaborate them, perhaps even
into the pause bar, would surely destroy the effect, which is that the real release is
provided by the silences. Time, not music, is the healer, as it were. The unexpansive
freedom of the Arbitri shapes must surely stand as it is. Andr as Schiff lls in the
pause bars after the Arbitri indications on the second playing of each half with
cadenzas based on written-out trill gures.
This is plausible and stylish enough,
Bogianckino, Harpsichord, 96.
Decca: 421 422 2, 1989.
256 The keyboard sonatas of Domenico Scarlatti
but it masks the radical bareness of the conception of the piece; in being historical,
Schiff obscures the real historical moment of the silences. Where else at this time
does one nd such loaded non-sound?
The inconsistency of ornamental indications found in the principal sources for the
sonatas needs to be examined from two angles: it is a matter both of performing
principle and of compositional purpose. We have already noted a number of instances
where performers and editors unquestioningly tidy up such inconsistencies, and it
has been suggested that the apparent untidiness may serve particular or more general
compositional ends. It is this inconsistency that concerns us here rather than how
Scarlattis ornaments are to be realized, on which subject there have been a number
of studies.
As with other of the composers peculiarities, his ornamental practice
can be partly but not fully rescued by an appeal to historical context. Imprecision
and inconsistencies of ornamentation, and of notation altogether, abound in music
of the eighteenth century, in spite of any number of treatises on the subject not that
notation can ever exactly be precise. As what we call the work concept crystallized
in the following century, alongside changes in the dissemination and reception of
the musical product, the status of the score changed. As scores came to exist no
longer just for immediate use but also for continued contemplation, composers
were moved to provide tidier, more painstaking, written versions of their work.
The libertarianism of eighteenth-century ornamental notation and practice, which
has vexed and sustained many scholars through their careers, may thus reect this
different cultural dynamic. It is also quite logical in its own terms there was no
reason not to be relaxed about something whose precise realization was by denition
in the gift of the performer.
In Scarlattis particular case the status of the score is of course yet more provisional,
in the absence of autographs which can lend greater authority to claims about
notation. However, it would be too easy to use the source situation as a smokescreen
for the ornamental aberrations we encounter, magically tidying up all on the basis of
perceived uncertainties in the chain of transmission. A certain cultural imperialism
may even play a part in such judgements, with the works having been copied in
Fadini, La graa dei manoscritti scarlattiani: problemi e osservazioni, in Domenico Scarlatti e il suo tempo, 183206,
offers a good overview; the virtual chapter Ornamentation in Scarlatti, found as Appendix IV in Kirkpatrick,
Scarlatti, 36598, needs circumspect handling, since it is now thought to rely too heavily on the treatise of
C. P. E. Bach. See, for example, the glancing remark by Kenneth Gilbert C. P. E. Bach is surely irrelevant for
Scarlatti in his Preface to Domenico Scarlatti: Sonates, vol. 1 (Paris: Heugel, 1984), ix.
This suggests that the very term inconsistency is inappropriate, since it is surely loaded by a more recent
preference for uniformity. A comparable case, raising comparable matters of principle, is given by James Webster
in The Triumph of Variability: Haydns Articulation Markings in the Autograph of Sonata No. 49 in E Flat, in
Haydn, Mozart, and Beethoven: Studies in the Music of the Classical Period. Essays in Honour of Alan Tyson, ed. Sieghard
Brandenburg (Oxford: Clarendon Press, 1998), 3364. He states that Haydns articulative variability is consistent
with fundamental aspects of his musical style (33), something we might also claim for Scarlatti.
Irritations 257
Spain by unknown scribes (They didnt know what they were doing out there in
Madrid). Yet those who have looked most closely at the main sources reiterate a
belief in the care of their notation, certainly in the case of the scribe who copied the
sonatas of the second layer (from K. 148) in P and V. Emilia Fadini writes, citing
Kirkpatrick in support, that Scarlatti notated ornaments with extreme care.
however, stops short of directly confronting the most unsettling feature: the absence
of an ornament altogether when it has already featured in a parallel passage or when
our stylistic sense leads us to expect one. Can such an absence also be carefully
While such absences are far from unknown in other cases, the Scarlattian picture
is characteristically more extreme. It is thus no accident that Howard Ferguson offers
the following reasoned summation precisely during a discussion of Scarlatti in his
book Keyboard Interpretation from the Fourteenth to the Nineteenth Century: As is usual
in [eighteenth]-century music, ornaments are sometimes missing when consistency
would lead one to expect them. In such places the player must decide whether this is a
copyists slip which should be remedied, or whether there is perhaps some reason for
the omission.
The open-mindedness that Ferguson advocates is, though, slightly
less liberal than it seems. The occasions on which a clear musical reason exists for
an omission will be few. In most cases instinctive musicianship will take over, and
the natural reaction will be to create uniformity. After all, once furnished with an
ornament, a cadential or motivic conguration will generally sound incomplete, at
or featureless without it. This is what Howard Schott implies when he writes of
Scarlattis ne notational variations that are often internally inconsistent within a
composition and frequently at odds with the players musical feeling.
such issues can and ought to be debated as a matter of general musical principle
one persons inconsistency is anothers variety in the particular case of Scarlatti
it seems to be just the players musical feeling that the composer is making sport
with. Ornaments may disappear and reappear with disconcerting irregularity, in a
fashion that can seem precisely calculated to invite a perplexed reaction from the
player or score reader. Yet this ornamental practice has its own consistency with
the creative ethos we have dened elsewhere. The studied carelessness, the almost
aggressive detachment from routine should come as no surprise. Indeed, perhaps we
may conceive of an ornamental aesthetic rather than just an ornamental practice.
To repeat a point made in other contexts, though, what is being asked of the
performer who would like to trust the evidence of the sources is and should be
hard to swallow. It is easier to talk in grand abstractions of the composers variety and
Fadini, Graa, 195.
Sheveloff is just about the only writer to square up to the issue of missing ornaments: Scarlattis potential for per-
versity in such matters seems unfathomable, he is as likely to avoid a trill at exactly the point at which every listener
expects one; his jesting with art often includes such reverse ornamental effects. Sheveloff, Frustrations II,
Keyboard Interpretation from the Fourteenth to the Nineteenth Century: An Introduction (London: Oxford University
Press, 1975), 136.
Review of Fadini edition, The Musical Times 129/1748 (1988), 539.
258 The keyboard sonatas of Domenico Scarlatti
informality than to translate this even only occasionally into ornamental practice.
Thus Christophe Rousset states that taking liberty with the composers [ornamental]
suggestions would t with the tone of the preface to the Essercizi and the general
ambience of the sonatas. Agreed, as long as this does not simply mean liberty
to standardize the form and appearance of ornaments, as Rousset the performer
resolutely seems to do.
Equally, in a discussion of that familiar topic, whether trills
(in Scarlatti) should begin on the main or upper note, Kenneth Gilbert warns against
imposing on Scarlatti [a] uniformity of practice which everything we know about
his art would tend to deny,
yet as an editor he loses few opportunities to add
ornaments in square brackets by analogy with parallel places earlier or later in the
same piece. Indeed, the Fadini edition, which almost never inserts such suggestions,
has been criticized for failing to do so.
What makes the spirit of Scarlattis practice difcult to grasp is that different
sources may disagree on the notation, or, more relevantly here, non-notation of
The new Lisbon source provided by the Libro di tocate, for instance,
often differs signicantly in this respect from V and P, which differ from each other
often enough. This apparently unsystematic approach, the possible logic of which
has already been stressed, might easily suggest to the positivist that we must return
all evidence to the larger frame of eighteenth-century liberalism, that there is no
case to be constructed for Scarlattis exceptional usage of ornament. Yet, although
the ornamental indications and absences of any particular sonata might thus be
open to correction or completion, globally there is more than enough evidence to
encourage the performer and scholar to take such inconsistencies seriously. In any
case, the point of this exercise is not to encourage complete delity to V and P or any
other reading of a single sonata, nor is it to deny that in some contexts the addition of
parallel ornaments is a good solution; rather, it is to suggest that even ornamentation
should be subject to constant vigilance. Where does the great eighteenth-century
shibboleth of good taste t in with this? The very notion of taste implies freedom
of choice, and performers do of course in the act of tidying reveal their own taste
a predilection for symmetry and naturalness that happens to be universally shared.
It would be nice, though, to hear some who did not simply provide the customary
well-trained chorus of matching ornaments, who were prepared to lose some of this
freedom in the name of another one.
The most persuasive indicators of Scarlattis perversity are those situations where
the manipulation of ornament can be shown to have a structural impact on the work
at hand. Such readings have been proposed for a number of works already, such as
K. 409 (Ex. 4.19) and K. 493 (discussed earlier in this chapter). Many more examples
of inconsistency do not, however, appear susceptible to a specic rationale. An
Rousset, Statistique, 78, and compare Roussets practice in his recent recording (Decca: 458 165 2, 1998).
Gilbert, Preface, ix.
See Hammond, review of Fadini edition, Music and Letters 69/4 (1988), 565, and Pestelli, Fadini Review, 463.
To offer one simple example, see the different readings of bars 1011 of K. 450 offered in Choi, Manuscripts,
Irritations 259
Ex. 5.15 K. 515 bars 4756
instance of this may be found in bar 54 of K. 515 (see Ex. 5.15). In the Gilbert
edition shown here, the trill has been shifted to the rst beat to correspond to that
found in bar 50 of the parallel phrase, yet, as is noted in the editorial commentary,
both P and V place their trill on the second beat in the right hand. It would be easy to
assume, as Gilbert presumably has done, that this is a simple and not very momentous
case of scribal error; but since we are very unlikely to uncover evidence that will
conrm this, it is just as defensible to accept the reading and try to understand
its implications. Such a discrepancy seems to exist for its own sake, simply in the
immediate jolt that it gives to our perceptions. It might therefore be viewed as one
more tiny piece of information towards the composite picture of Scarlattis creative
malpractice. In other words, it is purposive aesthetically if not structurally. However,
its effect need not be wide-ranging in this sense alone; in enlivening our conception
of the whole sonata in which it is found, it may indeed have an intrinsic structural
role, if one that is difcult to quantify.
Such a situation is no different in principle from similar cases of inconsistency
found in the notation of other composers works. What makes it less innocent
is our knowledge of more conspicuous and loaded aberrations in other sonatas,
and realistically, if there is to be any reassessment of performing habits, it is these
aberrations which must be addressed and interpreted. The opening four bars of the
Sonata in C major, K. 461, offer a ne instance of the structural implications of
non-parallel ornamentation (see Ex. 5.16). It is difcult to imagine any performer
not amending bar 2, adding a trill so as to match what the left hand does at 4.
one level this may be taken, like the example in K. 515 above, as the sort of messy
detail that enlivens our perception of the whole, both individual works and the
entire corpus. There are, however, several more specic arguments in favour of just
what the sources transmit. Simply in terms of colouring, the added ornament in the
This is what both Christophe Rousset and Trevor Pinnock do. Decca: 458 165 2, 1998 (Rousset); Archiv: 419
632 2, 1987 (Pinnock).
260 The keyboard sonatas of Domenico Scarlatti
Ex. 5.16 K. 461 bars 17
left-hand echo individualizes the lower registral space, suggesting a more active and
vivid eld of sound. The answering echo in fact poses a question rather than simply
completing the pattern. In addition, this ts with a principle specic to this sonata,
in that it sets up a textural topic of opposition between the hands, as we saw in the
case of K. 407 (Ex. 5.12). The forms this takes the most obvious being the frequent
use of contrary-motion scales will be discussed further in Chapter 6. This plot
suggested for the added left-hand trill at bar 4 is simply but wonderfully conrmed
by the fact that it is the left hand which continues the phrase from bar 5; having
taken the ornamental initiative, it now assumes thematic leadership. This time there
is no answer from the other hand; the left hand simply repeats its unit at bars 78.
The less articulate right hand is reduced to a two-note cadential commentary.
K. 446, a Pastorale in F major, explores the effects of non-parallel ornamentation
in a more playful way. This is found in the second subjects left-hand gure from
bar 13
, in which a typical siciliana rhythm in the tenor register alternates with
single low bass notes. In the rst phrase the thrice-repeated dotted gure is always
ornamented; in the second from 15
this ornament disappears, only to reappear on
the third repetition to witty effect (as if to say only kidding). Observing what appears
in V and P (and in the Fitzwilliam Cambridge copy too) adds enormously to the life
and character of the passage. It individualizes the sense of line and register, giving
a simple accompanying gure a mind of its own, so to speak. This is particularly
signicant given the generic basis of the sonata. In a simple pastoral style, we would
not expect an accompaniment to be at all self-conscious; it should be purely and
plainly functional.
That this is a conscious playing with expectations, on a level at which we do not
expect surprises, might be conrmed by what happens in the equivalent passage
in the second half. This time the pattern is the same until the sixth hearing of the
gure, where, in a double bluff, the ornament is not revived. A performer may
well nd the evidence of the sources too irritatingly sporadic to be taken seriously:
isnt this a typical example of scribal shorthand, the addition of the ornamental
complement to the dotted gure being left to the musical intelligence of the player?
What weakens such a claim, though, is the reappearance of the trill on the third unit
of the second phrase in the rst half. Without this, there would be every justication
for matching the ornamental pattern of the complete phrase to the preceding model.
With it, though, there is the strong implication that a simple embellishment has left
the sphere of executive discretion and is subject to precise authorial control. Again,
Irritations 261
Ex. 5.17 K. 212 bars 6177
though, this need not mean that the performer should feel constrained to replicate
this exact sequence of ornamental hide-and-seek. A number of other realizations
that retained the spirit of the ornamental enterprise would be possible. The one
unstylish solution, it should be clear, would be to inect the dozen appearances of
the gure identically each time. With the observance of repeat marks, the potential
for perversity in an imaginative performance is then exponentially increased.
For a nal example we will turn to a passage where all the ornaments are indicated
but conspicuously fail to rhyme with each other. In K. 212 Fadini and Gilbert both
systematize the ornaments of the rst three parallel phrases of the second half (shown
as Ex. 5.17). Gilbert changes the appoggiatura g
given by both P and V in bar 68 to a
so as to match the V reading of bar 72. On the other hand, Fadini retains this g
68, but then for 72 she chooses the e
given by P rather than the g
given by V. Thus
while Gilbert make the two bars match by upper rather than lower appoggiaturas,
Fadini does the opposite! If Fadinis is the more respectable editorial procedure,
consistently following the P