CH A PTER ON E

!
Contexts

Differing Approaches to Sonata Form of the larger playing field. In influential English-
language scholarship today one might recognize
There is no consensus regarding the manner in four general trends: two broad musicological
which sonata form in the decades around 1800 lines and two broad music-theory lines. To be
is to be grasped. On the contrary, analysts are sure, the categories overlap — they are anything
confronted with a clutch of diverse approaches but airtight — and within each there are differ-
with differing emphases, interests, and termi- ences and varied accents in the way the gen-
nologies. This is contested terrain, particularly eral method is formulated. Still, musicology and
since the structure is basic to how we conceptu- music theory have often pursued distinct paths,
alize the Austro-Germanic art-music enterprise generating different questions and answers.
stemming from Haydn, Mozart, Beethoven, and The two broad musicological approaches,
Schubert. Our contribution, Sonata Theory, sometimes intersecting, are: (1) the style of
provides a via media among these approaches, eclectic analytical writing favored by Donald
remaining open to the positive insights that Francis Tovey and carried on (and varied) by
each has to offer and for the most part remain- such differing writers as Joseph Kerman and
ing methodologically compatible with them all. Charles Rosen and (2) the more strictly “his-
At the same time we propose new, genre-based torical-evidentiary-empirical” concerns of
perspectives, along with useful ways of formu- such diverse fi gures as William S. Newman,
lating analytical questions and moving on to Jan LaRue, Eugene K. Wolf, Leonard G. Rat-
productive hermeneutic endeavors — interpre- ner, and their successors. The two broad mu-
tations of meaning. sic-theoretical approaches are: (3) Schenkerian
Situating oneself within a confl icted field is and post-Schenkerian methodologies and (4)
a risky, fallible enterprise, in part because one is lines of analysis emphasizing motivic growth
obliged reductively to characterize the work of from small musical cells, as well as the iden-
others — and those others nearly always object tification of phrase-shapes and the patterns of
(often rightly so) to such characterizations. And larger sectional blocks — a style of analysis as-
yet it may be helpful to sketch out some rough sociated with Arnold Schoenberg, Rudolph
descriptions of viable approaches to the subject Réti, and Hans Keller, and including the work
of sonata form, if only to suggest an impression of Erwin Ratz and, most recently, William E.

3
 4 Elements of Sonata Theory

Caplin.1 At the risk of oversimplification (and section”) that needs to be resolved in the reca-
with apologies to those unmentioned), we pitulation.5 A central feature of Rosen’s writing
might characterize the interests of these four (as well as that of Tovey and Kerman) was the
categories by citing an example of an important description of individual compositional styles
text within each. and preferences, along with the pronouncement
1. Our fi rst-category illustration is Charles of cleanly-divided aesthetic judgments of the
Rosen’s Sonata Forms (1980, rev. 1988).2 Draw- works at hand — strong praise for the master-
ing on the analytical and prose style of Tovey works contrasted with tart dismissals of works
and grounded in a vast knowledge of the rep- deemed not to make the grade.
ertory, Rosen’s magnum opus stressed the va- 2. The second category is best represented
riety of procedures that one can encounter in by Leonard G. Ratner’s Classic Music (1980).6
the “texture” or “process” that we now call so- Somewhat parallel to the scholarly-inventory
nata form. (Hence his plural, “forms,” echoing work of William S. Newman and Jan LaRue,
Tovey.) 3 Rather than elaborating an intricate Ratner sought to reconstruct the concept of the
background plan for the form, Rosen preferred eighteenth-century style from the point of view
to demonstrate how difficult — or futile — it is of the eighteenth century itself. The book was
to provide a set of detailed expectations regard- to be
ing it because of the unique things that occur
in individual pieces by composers of genius. As a full-scale explication of the stylistic premises of
a matter of principle Rosen shunned the idea classic music, a guide to the principles according
of a “general practice” for the construction of to which this music was composed. . . . The ex-
sonatas — except for a few tonal requirements position of 18th-century musical rhetoric is found
and common textural choices — although there in theoretical and critical treatises. . . . [These
writings] point to what was current then, illumi-
were clearly better and more masterly solu-
nating our present view of the music. Coordi-
tions to the general set of problems at hand.4 nated with analysis of the music itself, the data
This somewhat intuitive approach, acute and gleaned from these writings make it possible to
invariably musical, also emphasized the con- determine the basic criteria of expression, rheto-
cept of tonal “polarization” (usually tonic and ric, structure, performance, and style that govern
dominant) in expositions and famously regarded classic music. . . . This book allows the student to
the expositional shift to a non-tonic key as an approach the music and musical precepts of the
“opposition[al]” move, a “large-scale disso- 18th century in much the same way a listener of
nance” (“structural dissonance” or “dissonant that time would have done.7

1. But even these broad categories are too limiting. In- University Press, 1944) [reissued in 1956 under the title
termixed throughout them all are the various traditions The Forms of Music], pp. 208 – 32.
passed on in the Formenlehre, the academic textbooks 4. Rosen, Sonata Forms, rev. ed., pp. 4 – 7. Cf. the differ-
of form, which seem to have a separate reception-life ing impression conveyed in W. Dean Sutcliffe’s review,
of their own. In addition, other influential European in Music & Letters 79 (1998), 601 – 4, of Rosen’s modest
perspectives that sometimes escape from or provide al- revision of his earlier work The Classical Style: Haydn,
ternative havens within the above four categories have Mozart, Beethoven, exp. ed. (New York: Norton, 1997
also proven provocative for current work — one thinks, [orig. ed., 1971]). This review, in part, calls attention to
for example, of the work of Jens Peter Larsen and Carl the earlier book’s apparent “emphasis on the normative
Dahlhaus. Moreover, in recent years differing scholars aspects of the style . . . stereotypes and formulas” — con-
have begun to seek new ways to blend together formerly cerns that raise a host of questions in these more skep-
differing methodologies. tical times and ones that Rosen himself had sought to
2. Rosen, Sonata Forms, rev. ed. (New York: Norton, clarify in the later Sonata Forms.
1988 [fi rst ed. 1980]). 5. Rosen, Sonata Forms, rev. ed., pp. 98 – 99, 229, 287.
3. Donald Francis Tovey, “Sonata Forms,” originally See also Rosen, The Classical Style, exp. ed., p. 33.
two different entries for the 11th (1911) and 14th (1929) 6. Ratner, Classic Music: Expression, Form, and Style
eds., the latter of which is reprinted in Tovey, Musical (New York: Schirmer, 1980).
Articles from the Encyclopaedia Britannica (London: Oxford 7. Ratner, Preface to Classic Music, pp. xiv – xvi.
 Contexts 5

Not surprisingly, Ratner paid close attention torically defensible musical “topics” (standard-
to the early theorists’ descriptions of what came ized musical gestures or types within phrases)
to be called (c. 1824 – 1845) “sonata form.” The and eighteenth-century conceptions of “rheto-
Newman-LaRue-Ratner projects (however ric” in this repertory.
they might differ in other respects) were ones 3. Moving to the music-theory side of things,
of data-gathering and recovery. One of their the touchstone of the third category is Hein-
features was to urge analysts to sideline nine- rich Schenker’s Der freie Satz (1935, translated
teenth- or twentieth-century views of sonata as Free Composition).11 For many music theorists
form in order to gain a more period-conscious interested in sonata form, no text is more cen-
conception of the form. 8 (In this regard these tral than this one. Opposed to traditional ways
interests are not without parallel to the perfor- of discussing musical structure, Schenker was
mance-practice movement and its quest for “au- convinced that he had discovered a new the-
thenticity.”) To varying degrees scholars within ory of form, “a new concept, one inherent in
this circle seek to describe sonata form (and the works of the great masters; indeed, it is the
other forms) from the perspective of late-eigh- very secret and source of their being: the con-
teenth-century theorists — favoring their termi- cept of organic coherence.”12 This theory was
nology and concerns and being cautious about to be grounded not in phrase- or section-rep-
going beyond them.9 Writers influenced by this etitions or in thematic manipulation but rather
point of view call upon the authority of late- in linear-contrapuntal views of the sonata as the
eighteenth-century or early nineteenth-cen- unfolding of a “fundamental structure” (Ursatz)
tury writers on the form (such as the important by means of more elaborate middleground and
statements of Heinrich Christoph Koch, Fran- foreground structures. Middlegrounds and fore-
cesco Galeazzi, Augustus Kollmann, and Anton grounds are understood as florid “diminutions”
Reicha). Several of them have also tended to of more simple, elemental background gestures
view harmony (modulations, key-areas visited, elaborated over the course of an entire move-
and so on) as the primary feature of sonata form ment. The method is highly sensitive to con-
in the years from roughly 1750 to 1820 — giv- trapuntal, linear voice-leading, long-range pro-
ing it the upper hand over thematic arrange- longations or descents of important individual
ment. In the mid-twentieth century Ratner fa- pitches, and the like. Here sonata form is un-
mously contested the earlier, thematic view of derstood as divided into two parts (exposition-
the sonata, which he regarded as discredited, an development || recapitulation) with a crucial
anachronistic, nineteenth-century (mis-)under- harmonic “interruption” (||) at the end of the
standing of the form as it had been originally development and a subsequent rebeginning at
grasped in Beethovenian and pre-Beethovenian the onset of the recapitulation, which restates
decades.10 Some writers influenced by Ratner’s and fi nally completes the fundamental structure
work are also concerned with identifying his- interrupted at the end of the fi rst part.13

8. See, e.g., Eugene K. Wolf, “Sonata Form,” in The 11. Schenker, Free Composition (German original, Der
New Harvard Dictionary of Music, ed. Don Michael Ran- freie Satz, 1935), trans. and ed. (with additional com-
del (Cambridge, Mass.: Harvard University Press, mentary) Ernst Oster (New York: Longman, 1979). Es-
1986), pp. 764 – 67. This essay outlines the rhetorical- pecially relevant is part 3, ch. 5 (“Form”), pp. 128 – 145.
tonal structure at hand and provides a historical over- “Section 3,” on “Sonata Form” (including Oster’s fa-
view of the origins and transformations of the form. mous footnote), is found on pp. 133 – 41.
9. In other respects Ratner-related styles of analysis 12. Schenker, Free Composition, p. xxi.
seem to be musicological variants of the well-estab- 13. Also to be noted in terms of Schenkerian and post-
lished sector of music theory, “history of music theory.” Schenkerian analysis is the summary of sonata form in
A more purely music-theoretical analogue is Joel Les- Allen Cadwallader and David Gagné, Analysis of Tonal
ter, Compositional Theory in the Eighteenth-Century (Cam- Music: A Schenkerian Approach (New York: Oxford Uni-
bridge, Mass.: Harvard University Press, 1992). versity Press, 1998), esp. ch. 11, “Sonata Principle,”
10. The locus classicus of this position is Ratner, “Har- pp. 303 – 59. Similarly, one should mention William
monic Aspects of Classic Form,” Journal of the American Rothstein, Phrase Rhythm in Tonal Music (New York:
Musicological Society 11 (1949), 159 – 68. Schirmer, 1989), particularly ch. 4, “Phrase Rhythm
 6 Elements of Sonata Theory

4. Our example of the fourth category is wake of Tovey’s similar assertions16 — has been
William E. Caplin’s Classical Form (1998).14 Its the repeated declaration that the “textbook”
opening paragraph proclaimed the need for “a view of sonata form is inadequate to deal with
new theory of classical form,” one that avoids the actual musical structures at hand. At best,
“ill-defi ned concepts and ambiguous terminol- such a scheme represents a conformist trap that
ogy derived from theories that have long fallen master-composers avoid falling into. In addi-
into disrepute.” Following the work of Schoen- tion, the implication has sometimes been that
berg and Ratz,15 Caplin viewed form as a group- to undertake any such “textbook” description
ing structure, and he set out to identify and classify of norms, however nuanced or sophisticated, is
the “formal functions” of smaller thematic/for- a mistaken enterprise. It is not difficult to fi nd
mal units. In practice, this entailed close at- conventionalized avowals on these matters.
tention to the structures and subparts of three Here is a strong version of the credo from Clau-
fundamental theme types: the sentence (consist- dio Spies, excerpted from an essay in a book of
ing, for Caplin, of presentation, continuation, Brahms Studies (1991):
and cadential functions; or basic idea [usually
repeated, perhaps with variation] + fragmenta- There is nothing new about “forms” with whose
tion + cadence); the period (antecedent + con- aid pieces of music are easily and lazily catego-
sequent); and the small ternary (A – B – A'). Much rized or typified, tagged, pigeon-holed, and con-
attention was also given to the anatomy of nu- veniently stored away without further — or even
merous “hybrids” that mix aspects of the more prior — hearing, and without further thought.
We were all initiated into the non-mysterious sto-
standard theme types (as defi ned by the author).
lidities of “form,” particularly the most fictitious
As the musical parts are assembled, they can one of all, “Sonata Form.” Nor is there, I has-
take on “framing functions,” “interthematic ten to add, anything new in the notion that such
functions,” “harmonic functions,” “initiating “forms” — and especially “Sonata Form” — are
functions,” “continuation functions,” and so fictions to whose specifications and proclaimed
on, often at more large-scale levels. One aim norms very few pieces of music worth any further
of analysis is to be able to recognize the theme thought actually conform in any appreciable way.
types (and hybrids) and to place them into a . . . It is almost as if Brahms had decided to com-
larger functional system of interrelated parts. pose [the Tragic Overture] as a potent rebuttal of
In the end, what was provided was an elabo- notions propounded by the tenets of Formenlehre,
rate taxonomy of different kinds of phrase-and- although [it] is by no means unique among his
works in this respect.
section juxtapositions.
The same point, put more gently — and after an
admirably detailed study of Brahms — may be
The War against the Textbooks
found from James Webster in the same volume:
One prominent feature of the study of sonata
form in recent decades — very much in the From examples like these it is clear that norms of
formal procedure, whether the bad old textbook

and Form: Some Preliminaries,” pp. 102 – 20. This is an menlehre: Über Formprinzipien in den Inventionen und Fugen
analytically sophisticated discussion of forms in general J. S. Bachs und ihre Bedeutung für die Kompositionstechnik
(including sonata form) and, in part, it seeks to blend Beethovens, 3rd ed., enl. (Vienna: Universal, 1973 [1st
some of the concerns of Schenkerians with the more ed., 1951]).
musicological (and often emphatically non-Schenker- 16. See, e.g., Donald Francis Tovey, “Some Aspects
ian) studies by Rosen, Ratner, and others. of Beethoven’s Art Forms” and “Musical Form and
14. Caplin, Classical Form: A Theory of Formal Functions Matter,” in The Main Stream of Music and Other Essays
for the Instrumental Music of Haydn, Mozart, and Beethoven (Oxford: Oxford University Press, 1949), pp. 272 – 73,
(New York: Oxford University Press, 1998). 160 – 62; and Tovey, “Sonata Forms,” pp. 210 – 12
15. Schoenberg, Fundamentals of Musical Composition, (“There are no rules whatever for the number or distri-
ed. Gerald Strang and Leonard Stein (London: Faber bution of themes in sonata form”).
& Faber, 1967); Ratz, Einführung in die musikalische For-
 Contexts 7

models or the numerical averages developed ear- was more complex than the reflex denunciations
lier in this paper, can never satisfactorily account suggested. The reiterated conviction that there
for the reality of individual compositions. In fact, was no single plan for sonata form in the later
when Brahms’s technique seems most paradoxi- eighteenth century, true enough in its narrow,
cal — as in the timeless, themeless, tonic retransi-
literal sense, rises to the level of an error when it
tion we have just analysed — the artistic result is
is naively taken either to dismiss the presence of
often the most poetic.17
substantially more complex systems of standard
practices or to discourage inquiry into those
Remarks along these lines could hardly be
practices. Is there a more effective way of exam-
more familiar. Even earlier, by midcentury, it
ining conventional musical gestures (or calling
had become a scholarly point of honor to de-
forth that which was conventional within indi-
clare war on the textbooks and, for some (again,
vidualized musical gestures) without producing
in varying degrees), on the often-wooden
ideas that were reductive, stiff, mechanical, pre-
limitations of classifying schemes in general.
scriptive? Is an aesthetically sensitive openness
Whether uttered in stronger or gentler versions,
to the study of convention within composition
such declarations advanced unswervingly or-
possible?
thodox late-twentieth-century convictions, and
The most strongly formulated arguments
they were caught up in the traditional philo-
against generalized principles of sonata practice
sophical dilemma of universals and particulars.
concealed a substantial weakness: in their inten-
For the most part — again, much as Tovey had
sity they tempted one to overstate the degree to
done — they took partisan positions on behalf
which such classifications were ever intended to
of the particulars, or at least on behalf of the
be equivalent to scientific laws. Within the hu-
ultimate noncapturability of the great masters.
manities norms, generic options, and more-or-
Apart from assessing this neonominalist argu-
less standard procedures are not laws at all. And
ment on its own terms, it would also be valu-
since they are not, there was no need to sup-
able to investigate the modernist assumptions
pose that the existence of numerous exceptions
that made such views possible: the mystification
or deviations invalidated the norm. Perhaps
of genius; the belief in the compulsion of the
the many deviations were purposeful dialogues
true artist to escape from confi ning, externally
with the background norm. But this would
applied rules or systems; the precept that what
mean, paradoxically, that the deviations helped
we most revere in music must not only be be-
to reinforce the socially shared norm that was
yond the grasp of academic minds and rational
being temporarily overridden. (Otherwise how
classification but must always be declared to be
could they be perceived as deviations at all?) But
so; and so on.
what is meant by a norm? And how could one
Studying and teaching musicology and music
come to an understanding of what such norms
theory in the 1960s, 1970s, and 1980s, the au-
might have been? We began to seek a way out of
thors of this book absorbed such views into the
the dilemma. The most profitable guidelines for
marrow of their bones. We also agreed — and
our solution lay within the domains of current
we continue to agree — that prior textbooks had
genre theory and hermeneutics.
invited a too rigid understanding of sonata form.
Given the flexibility found in the large-scale
So far as the gravamen of the charge goes, the
architecture of later-eighteenth-century com-
literal point is correct and has the added benefit
position, the main descriptive problem was the
of bringing caution to any new analytical in-
difficulty of positing convincing categories of
quiry. Still, the problem of determining the role
typical procedures. As scholars of eighteenth-
of convention within this “classical” repertory

17. Spies, “‘Form’ and the Tragic Overture: An Adjura- Analytical and Historical Perspectives, ed. George S. Bo-
tion,” and James Webster, “The General and the Partic- zarth (Oxford: Clarendon, 1990), pp. 391 (Spies) and
ular in Brahms’s Later Sonata Forms,” in Brahms Studies: 75 (Webster).
 8 Elements of Sonata Theory

century music perennially point out, surprising ible one. . . . What is needed, then, is a general
occurrences and variants abound — all the more theory of form that can account for conventional
so when one’s investigation takes a panoramic patterns and at the same time do justice to the im-
view, extending beyond Haydn, Mozart, and mense diversity that exists within the framework
of these patterns.19
early Beethoven to include the works of less-
explored composers. It is for this reason that at-
tempts to describe normative sonata procedures Thus the challenge: to articulate the implied
tend to bog down in trying to account for a host pattern-types that appear in some of the clear-
of seemingly unusual cases (of which there is an est or most notable exemplars and to do this
especially abundant supply in Haydn’s œuvre). with as much detail and specificity as the ma-
So much is evident, but the only alternative terial encourages. These heuristic norms need
to throwing up one’s hands in the face of such not be considered as literally existing “things.”
diversity (rallying around the cry, “Anything can Rather, they may be understood as what Dahl-
happen!,” which is obviously untrue) was to fi nd haus, following Max Weber, regarded as ideal
a reasonable middle ground between confiningly types or what we prefer to consider as regula-
rigid schemata and the claim of a near-total free- tive guides for interpretation. Moreover, these
dom. It was necessary to retrieve a workable her- norms would have to be defi ned neither by un-
meneutic space between the reductive textbook usual cases nor by expressive deformations of
models of the nineteenth and early twentieth more standard choices. Rather, they would de-
centuries and the unhelpful (though still fash- rive from the standard choices themselves, inso-
ionable) “lowest-common-denominator” har- far as the frequency of those choices (not their
monic models, whose claims to adequacy have inevitability) permit one, inductively, to infer
been challenged on both historical and concep- a background set of guidelines shared by com-
tual grounds in an important essay from 1991 by posers and a community of listeners at a given
Mark Evan Bonds.18 In that essay Bonds distin- historical time and place. As we constructed
guished between “conformational” and “genera- these models, then, we were concerned to iden-
tive” concepts of sonata form, traced the fortunes tify types or tendencies that (in retrospect) were
of these concepts historically, and submitted the influential generic participants in the eventual
mid-twentieth-century ascendancy of the gen- crystallization or early reification phase of the
erative models to a critique. Among his conclu- sonata in the mid-eighteenth century, when
sions: the preferred options became both clearer and
somewhat more consistent.20 The result was the
Few analyses [today] openly acknowledge the
system that we call Sonata Theory.
extent to which composers worked within the Our intention is not to lay down binding laws
context of formal conventions. . . . But it would or invariant rules concerning either the parts of
be ludicrous to argue that sonata form was not a sonata or the sonata as a whole. Instead, we
at least in part an a priori schema available to the are trying to sketch the outlines of a complex
composer. . . . Sonata form, for Haydn, was in set of common options or generic defaults. It
fact a point of departure, a mold, albeit a flex- is not that any attempt to recover standard pat-

18. Bonds,” The Paradox of Musical Form,” ch. 1 of related genres, works for chamber ensemble, and solo
Wordless Rhetoric: Musical Form and the Metaphor of the and accompanied sonatas in all but a few major cen-
Oration (Cambridge, Mass.: Harvard University Press, ters.” Similar observations regarding the increasing
1991), pp. 13 – 52. normativity of certain kinds of sonata procedures — es-
19. Bonds, “The Paradox of Musical Form,” p. 29. pecially those identified with the Viennese Classicism
20. E.g., as articulated in Wolf, “Sonata Form,” of Haydn, Mozart, and early Beethoven in the period
The New Harvard Dictionary of Music, p. 766: “By circa 1770 – 1800 — may be found in the writings of vir-
about 1765, however, full sonata form [i.e., with full tually every author who has investigated such things.
recapitulation] — though never the rigid textbook va- See, e.g., the many similar remarks in Charles Rosen,
riety — was rapidly becoming the norm in fast move- Sonata Forms, rev. ed., pp. 145, 153, 156 – 58, 161, and
ments and many slow movements of symphonies and 286 – 87.
 Contexts 9

terns is a flawed enterprise; rather, it is that prior lowed: symphony movement? overture? sonata?
attempts have been inadequately conceived. We chamber music? how long or “grand” a move-
offer Sonata Theory as a heuristic construct that ment? how complex? how “original”? how “in-
can help the task of analysis and hermeneutics. tense” or “challenging” to listeners? what is the
At any point, the method outlined here can be expected audience? for connoisseurs or amateurs
expanded or modified through criticism, cor- (Kenner or Liebhaber)? how “unusual” in its in-
rection, or nuance. Indeed, we invite this. The ternal language and manner of presentation? in
proposed construct is intended only as a begin- competition with whom? whom am I trying to
ning, as a work-in-progress — not as a fi xed set impress? for what occasion? and so on.
of fi nalized dicta. As an assemblage of separate Once these gateways had been determined
subparts, each of which should be subjected to and work begun in earnest — the task of cre-
constant testing and refi nement, the utility of ating an engaging musical pathway through
Sonata Theory as a whole does not rest on the pre-established, generically obligatory sta-
unexceptionable validity of any correctible sub- tions — the composer faced practical issues of
part. musical continuation from one idea to its suc-
cessor. (A succeeding phrase, even an utterly
contrasting phrase, would typically be heard
Sonata Theory: Introductory Remarks as “reacting to” what had been established up
to that point — moving outward to another
What follows lays the groundwork of a method branch of the musical ramification.) A sonata
of approaching analytically any sonata-form form required that certain audible goals be suc-
movement from the period of Haydn, Mozart, cessively articulated and secured, even though
and Beethoven. A central premise of this method the individual details of each sonata journey
is the conviction that we must seek to under- could differ remarkably. A composer’s choices
stand the backdrop of normative procedures involved not only varying senses of the propri-
within the different zones or action-spaces of ety of “what sorts of things could reasonably be
the late-eighteenth-century sonata. Much of this expected next” within the style but also how
book sketches out key technical features of those delectable surprises, even varying degrees of
norms as we currently understand them. seeming transgressions, might be folded into
At any given point in the construction of a the expanding network of ideas. Within each
sonata form, a composer was faced with an ar- compositional zone (action-space) or subsec-
ray of common types of continuation-choices tion these “internalized” features included such
established by the limits of “expected” architec- things as generically appropriate types of themes
ture found in (and generalized from) numerous and textures; reasonable lengths of individual
generic precedents. (To produce a keyboard-so- passages (which depended on the anticipated
nata or symphonic movement was to place one’s length and complexity of the whole composi-
individual achievement into a dialogue with a tion); dynamics; degrees of anticipated contrast;
community-shared pool of preexisting works, standard “topics” or thematic formulas; prop-
probably including some well-known ones, that erly placed cadences and/or cadential delay or
formed the new work’s context of understand- frustration; the handling of major- and minor-
ing.) This is not to say that any skilled composer mode coloration; boundaries of taste; and the
soberly pondered these choices, one by one, in limits of eccentricity.
the act of composing. Surely the most common The options available from compositional
decisions were made efficiently, expertly, and zone to zone existed conceptually within the
tacitly on the basis of norms that had been in- knowledgeable musical community as something
ternalized (rendered automatic) through expe- on the order of tasteful generic advice — en-
rience and familiarity with the style. Still, even abling and constraining guidelines (not invio-
before a sonata form was begun, a composer lable rules) within the “sonata-game” — given
might, consciously or not, confront an array of by a shared knowledge of precedents. Moreover,
initial questions acting as a fi lter for all that fol- the available guidelines for each moment (pri-
 10 Elements of Sonata Theory

mary theme, transition, medial caesura, second- implication is that not to choose the fi rst-level
ary theme, and so on) were not accessible in an default would in most cases lead one to consider
arbitrary, non-weighted fashion. Some choices what the second-level default was — the next
were virtually obligatory; others less so, some- most obvious choice. If that, too, were rejected,
times in discernible degrees. (For novice-com- then one was next invited to consider the third-
posers, one might wittily fantasize — provided level default (if it existed), and so on. Or perhaps
that the image is not taken too literally — some- at some point in this process a composer might
thing on the order of an aggressively complex decide to do something unusual by rejecting all
“wizard” help feature within a late-eighteenth- of the default choices altogether, in pursuit of a
century musical computer application, prompt- deformation of that compositional moment.
ing the still-puzzled apprentice with a welter of As might be imagined, the whole system
numerous, successive dialog boxes of general in- was highly complex, typically involving at any
formation, tips, pre-selected weighted options, compositional point more than two default lev-
and strong, generically normative suggestions as els of options. This is why it requires so much
the act of composition proceeded. What would time — and space — to reconstruct the back-
have been urged here were such things as the- ground system. But it is only through an un-
matic-modular shape, style, effect, and format derstanding of what the main options were that
appropriate to the relevant action-space mo- we can come to grips with the implications of a
ment — not literal content, the burden of which composer’s choices from moment to moment.
was still placed on the composer.) In confronting any individual composition we
Within the late-eighteenth-century style seek to determine which gestures in it were nor-
some of the options were much more frequently mative within the style, which were elaborate, ele-
chosen: To suggest the strength and pre-estab- gant, or strained treatments of the culturally avail-
lished hierarchical ordering of these options we able norms, and which were not normative at all.
call the more normative procedures first- and sec- Sonata Theory starts from the premise that an in-
ond-level defaults within the various zones.21 Most dividual composition is a musical utterance that is
simply put, composers selected (or adapted) set (by the composer) into a dialogue with implied
fi rst-level options more frequently than second- norms. This is an understanding of formal proce-
level ones, and so on. (Writers of minor-mode dures as dynamic, dialogic. Our conception of the so-
sonatas, for instance, more often modulated to nata as an instance of dialogic form is not accurately
the major mediant, III, in the exposition, than described as seeking to reinstate a bluntly “confor-
to the minor dominant, v — a less common op- mational” view of that structure (in Bonds’s origi-
tion.) As we use it, however, the term default nal sense of that category). Viewed more subtly, it
connotes more than a merely preferred option is not the obligation of a sonata to “conform” to
for otherwise detached consideration. First-level a fixed background pattern, which then, in turn,
defaults were almost ref lexive choices — the might be construed as an “ideal” or “well formed”
things that most composers might do as a mat- shape from which deviations might be regarded as
ter of course, the fi rst option that would nor- compositional errors or aesthetically undesirable
mally occur to them. More than that: not to distortions. Rather, the composer generates a so-
activate a first-level-default option (for example, nata — which we regard as a process, a linear series
to provide an expositional move to v instead of of compositional choices — to enter into a dialogue
to III) would require a more fully conscious de- with an intricate web of interrelated norms as an
cision — the striving for an effect different from ongoing action in time. The acoustic surface of
that provided by the usual choice. An additional any sonata form (what we literally hear) sets forth

21. At some level the literal, computer-defi nition con- sense of ongoing, strongly weighted advice, standard
cept of default — an assumption prebuilt into the large- choices, and normatively arrayed options). As men-
scale automatic (but alterable) decisions of a software tioned earlier, the metaphorical implication, if appli-
program at the moment of its initialization — is not fully cable at all, is to be worn loosely.
congruent with our free adaptation of it here (in the
 Contexts 11

the sonic traces of this individualized, processual of expectations, presumably in order to generate
dialogue, one that, from the standpoint of recep- an enhanced or astonishing poetic effect.22
tion, it is the task of the analyst to reinvigorate. Deformations — unusual or strongly charac-
The backdrop of norms against which a sonata or terized, ad hoc moments — are common within
any of its successive zones is placed into dialogue the works of many different late-eighteenth-
is no monodimensional, reified “thing.” On the century composers. Indeed, they are rampant
contrary, that backdrop comprises complex sets (or in Haydn, who delighted in producing surpris-
constellations) of flexible action-options, devised ing effects. Such occurrences, in dialogue with a
to facilitate the dialogue. Understanding form as norm, should not be regarded as redefi ning that
dialogue also helps us to realize that in some cases norm unless the composer continued to employ
standard procedures may be locally overridden for that idiosyncratic feature in other works (thus
certain expressive effects. These effects differ from customizing the norm for his own use) or un-
composition to composition: each needs to be in- less later composers picked up the deformation
terpreted individually. The more piece-specific as one of their more or less standard options.
one’s readings can be along these lines, the better. When this later occurrence happens, the origi-
In any analysis merely to assert that something is nal exception is no longer to be regarded as a de-
done “for expressive reasons” or “for reasons of formation per se but becomes one of the lower-
variety” is obviously inadequate. level defaults within the Sonata-Theory system.
Background norms and standard options are What was a deformation in Beethoven could be-
classifi able into common and less common se- come a lower-level default in Schumann, Liszt,
lections at different times and different places. or Wagner — part of a larger network of nine-
Within an individual composition, a mark- teenth-century sonata-deformation families.
edly exceptional procedure here or there is just The essence of Sonata Theory lies in uncov-
that — exceptional. We call such an occurrence a ering and interpreting the dialogue of an indi-
generic deformation: a stretching or distortion of vidual piece with the background set of norms.
a norm beyond its understood limits; a pointed This style of analysis considers every aspect of
overriding of a standard option. The term “de- the individual work: themes, harmonic and con-
formation,” in this specific context, is a narrow- trapuntal motion, large- and small-scale shapes,
defi nitional, technical one, grounded in prece- textures, dynamics, instrumentation, tempos,
dents in literary theory and other research areas. repeat conventions, and so on. The main re-
In its strictly limited, analytical usage within So- quirement for the application of the method is
nata Theory, “deformation” carries no negative to grasp the controlled flexibility of the implicit
charge, no negative assessment. On the contrary, underlying system of conventions. Elaborating
such deformations are typically engaging, aes- that system is the goal of the Elements.
thetically positive occurrences that contribute At every turn, our aim has been to focus on
to the appeal and interest of a piece. As we use the most basic features of the sonata and never
the term, it signifies only a purposely strained or to forget why we perform and listen to this mu-
non-normative realization of a musical action- sic in the fi rst place. To overlook fundamental
space, a surprising or innovative departure from things leads one’s analyses astray or renders them
the constellation of habitual practices, an imagi- sterile, bookish, or irrelevant. The best analyti-
native teasing or thwarting, sometimes playful, cal system is the one that seeks to reawaken or

22. It would be a mistake, therefore, to read into this “deformation” — to be able to perceive in it a genre-en-
technical usage any residual connotations of the evalu- abled, positive sense of strain, a deliberately manufactured
atively negative, such as the “deformed” (in its more tension set apart in this aesthetic-analytical, “artificial”
typical meaning), the “disfigured,” the “misshapen,” the context from any implication of criticism or (much less)
“abnormal,” the “poorly formed,” or the “ugly.” Those censure. These connotational points are revisited and
are not our connotations, and within the framework of amplified in the “Deformation” section of appendix 2,
Sonata Theory terminology we distance ourselves from which also offers further reflections on the concepts of
them as strongly as we can. The central thing is to be dialogic form and sonata-form action-spaces.
able to grasp the intended nuance of the technical term
 12 Elements of Sonata Theory

re-energize the latent drama, power, wit, and rently “orthodox wisdom” regarding sonatas to
wonder within individual compositions. When- radical questioning — comfortable trenches of
ever an analytical system diverts attention from thought that had long been part of our own re-
the impact of the music as real experience — or, flexive modes of approaching this music. From
even more, when it fails to heighten our own the beginning we sought to listen carefully to
experience of the music — then that analytical this repertory, trying to remain open to what it
system is in need of correction. We hope that seemed to want to tell us on its own terms, in-
Sonata Theory, in its practical application, will sofar as we could apprehend those terms in our
lead beyond the academic explanations and in- own, very different times. Before long we came
terpretations of the self-enclosed work into a to understand that everything that we had con-
larger reflection on the changing meanings of sidered to be established about sonata-analysis
this music within society. had to be rethought. If only for this reason, we
In part, we do this by redirecting analyti- realize how curious Sonata Theory might at fi rst
cal attention to those portions of the sonata that appear, especially to scholars habituated within
have been taken for granted or passed over in other modes of analysis and accustomed to other
relative silence in most preceding discussions. kinds of theoretical questions. The value of any
These include the composer’s treatment of cae- analytical system, however, lies in the robust-
suras (medial and fi nal), the textural drive to- ness of its interpretive power. It is that inter-
ward important cadences (including especially pretive adequacy that we have been seeking.
the moments of what we call essential exposi- Whenever existing terminology was adequate,
tional closure [EEC] and essential structural closure we have retained it; whenever it was misleading
[ESC]), the rotational aspect of the sonata move- or connotatively unhelpful, we have decided to
ment as a whole (its tendency to cycle repeat- change it; whenever it lacked a term for a cru-
edly through large, thematically differentiated cial concept, we have been obliged to devise a
blocks), and many other considerations. Al- new one.
though this was by no means clear to us when Readers might initially fi nd that the basic
we began this project, one result of our work concerns of Sonata Theory are learned relatively
has been to defamiliarize the sonatas of Haydn, quickly — like the moves of chess. These con-
Mozart, and Beethoven — permitting us to hear cerns may seem simple precisely because they
them in what we have found to be more re- are simple. At all points in the analysis of a so-
warding ways. To some extent, we discovered nata, we have tried to emphasize the most es-
early on that we often had to overcome our own sential features and dramatized musical goals.
patterns of habituation in analysis and under- Beyond the elementary principles of Sonata
standing “in order [to adapt the words of Viktor Theory, though, lies an elaborate network of
Shklovsky] to return sensation to our limbs, in possibility, nuance, flexibility, sophistication,
order to make us feel objects, to make a stone and detail that takes patience to master. As with
feel stony.”23 The idiosyncratic concerns — even chess, again, one may learn the moves rapidly,
the idiosyncratic terms — of Sonata Theory can but to play the game at a fully proficient level is
help in this regard. more difficult. Notwithstanding its many pos-
For the authors, one of the most challenging tulates and axioms, Sonata Theory is no me-
burdens in devising Sonata Theory has been to chanical system. Rather, in proper application
remain willing to submit all components of cur- it is an art that requires training, musical sensi-

23. Shklovsky, Theory of Prose [from second edition, of art makes perception long and ‘laborious’. . . . Art is a
1929], transl. Benjamin Sher (Elmwood Park, Ill.: means of experiencing the process of creativity. The artifact itself
Dalkey Archive Press, 1990), p. 6. In order to ac- is quite unimportant.” Sher defends his translation, “en-
complish these things, declared the Russian Formalist stranging” (as opposed to the more traditional choices,
Shklovksy, “[we have] been given the tool of art. . . . By “defamiliarizing” and “estranging”), on p. xix.
‘enstranging’ objects and complicating form, the device
 Contexts 13

tivity, and much experience with the repertory The goal and the course to the goal are pri-
in question. mary. Content comes afterward: without a goal
At the heart of the theory is the recognition there can be no content.
and interpretation of expressive/dramatic trajecto- In the art of music, as in life, motion toward
the goal encounters obstacles, reverses, disap-
ries toward generically obligatory cadences. For the
pointments, and involves great distances, detours,
present, we might only register the degree to
expansions, interpolations, and, in short, retarda-
which this concern resonates with Heinrich tions of all kinds. Therein lies the source of all
Schenker’s much-quoted description of musical artistic delaying, from which the creative mind
motion and dramatized process in Free Composi- can derive content that is ever new. 24
tion (Der freie Satz, 1935):

24. Schenker, Free Composition, p. 5.

 C H A P T E R T WO

!
Sonata Form as a Whole
Foundational Considerations

S onata form is the most important large struc-

ture of individual movements from the
“common-practice” tonal era. It sets forth and
allegro of the symphony [or sonata]” disposed
in “two sections” [zwey Theile] and three “main
periods” [Hauptperioden] (Koch 1793); within
resolves its musical discourse within a large-scale “larger pieces of music” a “well-conducted mel-
binary format. The term “sonata form” was al- ody [!] . . . divided into two parts, either con-
most surely unknown to Haydn, Mozart, early nected, or separated in the middle by a repeat
Beethoven, and their contemporaries: it seems sign” (Galeazzi 1796); “an elaborate movement
to have surfaced only in the 1820s and 1830s. In [or] a long movement . . . generally divided into
the late-eighteenth and early-nineteenth century two sections” (Kollmann 1799); “grand binary
this structure would have been grasped primar- form” [grande coupe binaire] (Reicha 1826); and
ily as the customary design of first movements so on.1 Still, “sonata form” (Sonatenform) seems
within sonatas, chamber music, and symphonies, to have been a familiar term by the mid-1820s,
although it was by no means confined only to at least in A. B. Marx’s Berliner allgemeine musika-
first movements (nor only to rapid-tempo move- lische Zeitung circle, where it referred both to the
ments). The varying descriptions from contem- multimovement cycle as a whole and, occasion-
porary theorists were more convoluted. There ally, to the form of an individual movement.2 It
the form was variously described as: “the first was only in 1838 and 1845, though, in technical

1. Heinrich Christoph Koch, Versuch einer Anleitung zur on Practical Musical Composition (London, 1799; rpt. New
Composition (Leipzig: Adam Friedrich Böhme, 1793; York: Da Capo Press, 1973), p. 4 [ch. 1, section 10];
rpt., Hildesheim: Georg Olms, 1969), pp. 304 – 5 (from Anton Reicha, Traité de haute composition musicale (Paris,
section 101), trans. Nancy Kovaleff Baker in Koch, 1826), discussed, e.g., in Ian Bent and William Drabkin,
Introductory Essay on Composition: The Mechanical Rules Analysis (New York: Norton, 1987), pp. 18 – 20, and es-
of Melody, Sections 3 and 4 (New Haven, Conn.: Yale pecially Peter A. Hoyt, “The Concept of développement
University Press, 1983), p. 199; Francesco Galeazzi, El- in the Early Nineteenth Century,” in Music Theory in
ementi teorico-pratici di musica, vol. 2 (Rome: Puccinelli, the Age of Romanticism, ed. Ian Bent (Cambridge: Cam-
1796), the relevant extracts of which were excerpted bridge University Press, 1996), pp. 141 – 62.
and translated in Bathia Churgin, “Francesco Galeazzi’s 2. In the journal’s fi rst year of publication (1824) the
Description (1796) of Sonata Form,” Journal of the Ameri- term ‘sonata form” appeared in both senses. The fi rst,
can Musicological Society 21 (1968), 181 – 99 (above quota- apparently initially the more common, was a descrip-
tions from pp. 189 – 90); A. F. C. Kollmann, An Essay tion of the entire multimovement cycle (used by Marx,

14
 Sonata Form as a Whole 15

discussions of the form’s particulars, that Marx ized action involving differing types of idealized
put the stamp of approval on the term “Sonaten- mid- and late-eighteenth-century personalities.
form” with regard to the individual-movement (Its potential for “extramusical” connotations
structure. 3 Throughout this book we use that and analogues is discussed in the final section of
term as a familiar quick-reference, even as we chapter 11, “Narrative Implications: The Sonata
realize that that designation was not current in as Metaphor for Human Action.”) Sonata form
the eighteenth century. emphasized short-range topical flexibility, grace,
Sonata form is neither a set of “textbook” and forward-driving dynamism combined — in
rules nor a fixed scheme. Rather, it is a con- both the short and long range — with balance,
stellation of normative and optional procedures symmetry, closure, and the rational resolution of
that are flexible in their realization — a field of tensions. By the mid-eighteenth century it had
enabling and constraining guidelines applied in become obligatory for the first movement of a
the production and interpretation of a famil- standard multimovement instrumental work; it
iar compositional shape. Existing at any given had also become a common, if optional, choice
moment, synchronically, as a mappable con- for the slow movement and the finale. Slow
stellation (although displaying variants from movements and finales sometimes also displayed
one location to another, from one composer to different adaptations of the form. Although the
another), the genre was subjected to ongoing guidelines in most of this book were written
diachronic transformation in history, changing predominantly with first and last movements
via incremental nuances from decade to decade. and single-movement overtures in mind (all en-
Haydn’s conception of what was customary ergetic “Allegro movements”), they are also ap-
within sonata form in 1770 differed somewhat plicable, occasionally with some modifications,
from Beethoven’s conception in 1805. However to slow movements.
such models might be said to have differed, they From the compositional point of view sonata
also shared certain crucial, genre-defining fea- form was an ordered system of generically avail-
tures that make them all recognizable as sonata able options permitting the spanning of ever
form. Here we are dealing primarily with the larger expanses of time. A sonata-form project
model that crystallized during the second half of was a feat of engineering, like the construc-
the eighteenth century and that reached a peak tion of a bridge “thrown out” into space. In
in the mature works of Haydn and Mozart and the eighteenth-century style this temporal span
the early works of Beethoven. was to be built from rather simple materials:
What we now call sonata form was developed trim, elementary musical modules whose brev-
as a response to aspects of the world view of the ity and small-scale balances seemed best suited
Enlightenment and the concomitantly emerg- to short-winded compositions. In the hands of
ing modernism. Considered generally, it could most composers, constructing a sonata-form
be understood as an abstract metaphor for disci- movement was a task of modular assembly: the
plined, balanced action in the world, a general- forging of a succession of short, section-specific

Heinrich Joseph Birnbach, and others), a usage that per- kurzes Allegro (A-moll) in der Sonatenform,” BamZ,
sisted throughout much of the nineteenth century, es- 1824, 410b). See the discussion of terminology and quo-
pecially in German-speaking regions. The other use of tation of sources in the entry by Hans-Joachim Hinrich-
“Sonatenform” referred to the structure of an individ- sen, “Sonatenform, Sonatenhauptsatzform” [1996], in
ual movement. It fi rst appeared in a casual, unexplained Hans Heinrich Eggebrecht, ed., Handwörterbuch der musi-
way — as if it were already a common label — in Marx’s kalischen Terminologie (Stuttgart: Steiner, n.d.), pp. 1 – 7.
1824 essay on the E-minor second movement (Prestis- 3. A. B. Marx, Die Lehre von der musikalischen Komposi-
simo) of Beethoven’s Piano Sonata in E, op. 109 (“Es tion, praktisch-theoretisch, vols. 2 and 3, 1st eds. (Leipzig,
bildet mit dem letzen Satze die eigentliche Sonate und 1838 and 1845), 2:482, 497; 3:195; cited in Hinrichsen,
ist auch in der Sonaten-Form hingeworfen,” BamZ, I, “Sonatenform, Sonatenhauptsatzform” [1996], pp. 6 – 7.
1824, 37b) and in Carl Loewe’s discussion of the fi rst See also Marx, Musical Form in the Age of Beethoven: Selected
movement of Beethoven’s Cello Sonata, op. 102 no.1 Writings on Theory and Method, ed. and trans. Scott Burn-
(“Hart und rauh, im männlichen Zorne, beginnt ein ham (Cambridge: Cambridge University Press, 1997).
 16 Elements of Sonata Theory

musical units (spaces of action) linked together priate for their specified location. These three
into an ongoing linear chain — pressing down spaces can be viewed as expansions of the three
and connecting one appropriately stylized mu- phases of the continuous rounded binary form
sical tile after another.4 One of the challenges (the rounded binary structure in which the first
facing the mid- and late-eighteenth-century part ends in a secondary key). We shall take
composer was to use a seemingly unassuming, up these spaces individually. (In figure 2.1 we
galant language, grounded in structural punc- have provided two diagrams of Sonata Theory’s
tuation and periodicity, to produce ever more conception of the most common type of sonata
spectacular spans for occasions of enhanced dig- form: 2.1a refers to the exposition; 2.1b to the
nity, prestige, or social importance. Ever-larger, whole sonata-form movement.)
thematically differentiated binary structures
(sonata forms, often with built-in repetitions of
individual sections), eventual accretions to the Exposition
structure (slow introductions and longer codas),
and multimovement conventions all had their As with all of the action-spaces the exposition
roles to play in this process of generic enlarge- is assigned a double-task, one harmonic and the
ment. And ultimately they led to the grandly other thematic-textural (“rhetorical”). Its har-
monumental, personalized structures of Haydn, monic task is to propose the initial tonic and then,
Mozart, and Beethoven. following any number of normative (and drama-
The most typical sonata forms (what we call tized) textural paths, to move to and cadence in
Type 3 sonatas) articulate an overall rounded a secondary key. In major-mode sonatas — the
binary structure. The two parts of this larger most common in the eighteenth century — this
structure are, in modern terminology: (1) the was the key of the dominant (which may be in-
exposition and (2) the development and recapit- dicated as V T, meaning “a V that is tonicized”),
ulation. As will be elaborated at the end of this thereby generating tonal tension. In minor-mode
chapter, both parts may be marked for repeat, or sonatas this was usually the key of the major me-
the composer may eliminate the repeat of part diant (III), although a less-often-selected choice
2 or, under some circumstances, both repeats. (second-level default) was the minor dominant
Notwithstanding its binary origins, the norma- (v). The differing psychological and structural
tive, Type 3 sonata consists of three musical ac- world of minor-mode sonatas is dealt with in
tion-spaces (again, the exposition, development, chapter 14. Here, for the most part, we shall fo-
and recapitulation), laid out in a large A||BA' cus on major-mode practice.
format. Hence the common observation that The exposition’s rhetorical task, no less impor-
the form consists of an originally binary struc- tant, is to provide a referential arrangement or
ture often arrayed in a ternary plan. Each of the layout of specialized themes and textures against
three spaces is usually subjected to thematic and which the events of the two subsequent spac-
textural differentiation. Each is marked by sev- es — development and recapitulation — are to
eral successive themes and textures, all of which be measured and understood. We refer to this
are normally recognizable as generically appro- layout as Rotation 1 or the expositional rotation.5

4. To be sure — and particularly in the hands of the mas- 11, subsection “Recompositions, Reorderings, Interpo-
ter composers of the period — certain passages within lations” (especially n. 2 and the text to which it refers),
individual sonata forms may from time to time give the and ch. 18, subsection “Haydn’s Treatments of Type 4
impression of a broader continuity of internal ramifi- Finales” (especially n. 49 and related text).
cation. This is especially the case with the startlingly 5. Sonata-form structures are centrally concerned with
original musical language of Haydn, who, even within the formal principle that we call rotational form or the
a generally modular and “sectionalized” concept of for- rotational process: two or more (varied) cyclings — rota-
mal practice, often favored passages of ongoing Fortspin- tions — through a modular pattern or succession laid
nung (a moment-to-moment “spinning-out” of modular down at the outset of the structure. Appendix 2 pro-
growth and elaboration). For brief characterizations of vides a broader introduction to this principle, which
Haydn’s often-“vitalistic” compositional style, see ch. pervades the discussion of sonata form in this book.
 a. Exposition only: the Essential Expositional Trajectory (to the EEC)

PAC final
MC EEC cadence

Continuation modules
(series of energy-gaining modules)
’
TR S C
Energy-gain + Relaunch Postcadential Appendix or
Acceptance of P set of “accessory ideas”
New key May be multisectional (C1, C2, etc.)
Often forte Usually piano and of varying lengths. Usually forte
or gaining in rhetorical force.
P
Often lyrical, etc.
Either modulatory or
nonmodulatory
Launch

Proposes the main

idea for the sonata
in V
(or, if P was in minor, in III or in v)
Tonic key Nontonic key
Exposition, Part 1 Exposition, Part 2

b. The entire structure: the Essential Sonata Trajectory (to the ESC)

final
MC EEC cadence interruption

’
Development
TR S C Often P- or TR-dominated
MC ESC
final
cadence
Perhaps rotational ’
Coda
P P TR S C
Restart Tonal resolution

Often recomposed
(emph. IV?)
V
I V or III as chord I I I
Exposition Development Recapitulation
One central mission: laying out the S, as agent, carries out the central
strategy for the eventual attainment of generic task of the sonata—securing
the ESC: a structure of promise. the ESC: a structure of accomplishment.

Figure 2.1 The Generic Layout of Sonata Form

 18 Elements of Sonata Theory

Because the exposition’s succession of events closure at the EEC. In performing or listening
serves, especially in its second half, to pre- to any sonata-form exposition one should sense
dict the plan and purpose of the entire third the broad drive of these generic vectors. When-
space — the recapitulation, which finally re- ever one hears the onset of S-space within any
solves the work — its layout may be understood exposition, one should listen with an alert sense
as articulating a structure of promise (indicating of anticipation for any subsequent PAC — how
how it proposes that “things work out” in the it might be approached, secured, delayed,
recapitulatory rotation-to-come). Because the thwarted, or deferred. One should experience
arrangement of rhetorical modules in Rotation any sonata form with a strongly “directed” pre-
1 provides the ordered set of events that articu- paratory set, pressing forward conceptually and
lates the uniqueness and specific personality of anticipating genre-defining events-to-come.
that piece, it should be kept in mind when as- Following the EEC one or more additional
sessing all of the later events in the movement. cadences (PACs) may follow within the closing
Within the expositional rotation the tonal zone or closing space (C). (Not all expositions con-
and rhetorical tasks unfold simultaneously, in- tain C-modules; it is possible for the S-conclud-
tertwined with each other in mutually reinforc- ing EEC to be delayed until the end of the ex-
ing ways. The exposition begins with a primary position, in which case there is no closing zone.)
theme or primary idea (P) in the tonic that sets the Whether or not C-modules are present, the final
emotional tone of the whole work. The most cadence of the exposition will generally be a
common layout for the remainder of the exposi- perfect authentic cadence in the secondary key
tion continues with an energy-gaining zone of (again, V:PAC, III:PAC, or v:PAC). This final
transition (TR) that leads to a mid-expositional cadence might not occur directly at the double
break or medial caesura (MC). This is typically bar. Frequently the final cadence is followed by a
followed by the onset of a specialized, second- C-module that prolongs the newly reinforced
ary-theme zone (S) in the new key. The generi- tonality by means of a pedal-point or some other
cally essential tonal purpose of the exposition device. Additionally, the final cadence is some-
is to drive to and produce a secure perfect au- times followed by a reactivation of V in prepa-
thentic cadence (PAC) in the new key (notated ration for a repeat of the entire exposition: if so,
as V:PAC in major-mode sonatas, III:PAC or this reactivating passage is the retransition (RT).
v:PAC in minor-mode ones). We refer to the
first satisfactory PAC within the secondary key
that goes on to differing material as the point Development
of essential expositional closure (the EEC): this is
one of the central concepts of Sonata Theory This action-space renders the established tonal
and one that is dealt with at length in other tension more fluid and complex. While the ex-
chapters.6 Producing the EEC is the generically position had split its tonal assertions into two
assigned task of the S-idea(s). The large dot- broad blocks or contrasting planes (I and V in
ted-line arrow in figure 2.1a suggests a broadly major-mode sonatas), the development typically
vectored trajectory from the start of the exposi- initiates more active, restless, or frequent tonal
tion to the EEC; the smaller dotted-line arrow shifts — a sense of comparative tonal instabil-
below it suggests a subordinate trajectory from ity. Here one gets the impression of a series of
the beginning of S to its own point of PAC- changing, coloristic moods or tonal adventures,

6. For the moment, we might emphasize that the fi rst ahead. Additionally, there are other ways of deferring
satisfactory PAC in the new key is often but not always the sense of a clear EEC (ch. 8). The clearest way of sug-
the first PAC in that key. A fi rst PAC, for instance, might gesting all of this in brief is to defi ne the EEC as the fi rst
be followed by a thematic repetition of all or part of the new-key PAC that proceeds onward to differing or contrasting
S-idea that we have just heard — which would automat- material — or, of course, that closes the exposition itself,
ically defer the EEC to the next satisfactory PAC further if there are no closing modules that follow that PAC.
 Sonata Form as a Whole 19

often led (in major-mode works) through the ordered thematic pattern established in the ex-
submediant key, vi, or other minor-mode keys position). Developments often refer back to (or
with shadowed, melancholy, or anxious con- take up as topics) one or more of the ideas from
notations. Any authentic-cadence attainment the exposition, most commonly selected, as
in a non-tonic key is to be understood as an it happens, from Rotation 1’s first half (P and
important developmental event — a cadential TR). More often than not, the modules taken
ratification of an attained tonal station. (A vi: up and worked through in the development are
PAC is especially common in major-mode so- presented in the order that they had originally
natas.) Ultimately, the standard development appeared in the exposition (even though sev-
culminates on an active dominant (VA , mean- eral expositional modules are normally left out
ing “a V that is an active chord, not a key”). entirely). Thus the modular succession encoun-
At this point the dominant from the end of the tered in the development — not only the exposi-
(major-mode) exposition is usually recaptured, tional events referred to, but also the possibility
detonicized, and reactivated. of an episode or largely new theme — is never to
This last point needs underscoring. In the de- be considered arbitrary. On the contrary, even
velopment the final cadence is usually a half-ca- within this more unpredictable, developmen-
dence in the tonic (I:HC), although a cadence in tal texture the thematic choice and arrange-
a related minor key, normally followed by a brief ment is of paramount importance and derives
reactivation of V, is also a possibility. In addition, its significance through a comparison with what
a I:HC is frequently followed by a prolonga- had happened in the exposition. The develop-
tion of dominant harmony, a “dominant-lock” ment is variable in length, although in the pe-
or “dominant preparation.” The typical I:HC riod 1760 – 90 one would normally expect it to
conclusion of the development — just before the occupy a smaller space than that established by
onset of the recapitulation — brings us to a har- the exposition. Longer, more elaborate develop-
monic interruption. (This crucial interruption is a ments in the 1780s, 1790s, and later decades are
defining feature of the Schenkerian conception monumentalized statements that invite special
of sonata form.) The VA at the end of the devel- attention.
opment is not resolved to the I that usually be-
gins the recapitulation. Rather, the phrase — and
the development section as a whole — is normally Recapitulation
“interrupted” on VA (notwithstanding any fore-
grounded or local, connective “fill” that might This action-space resolves the tonal tension orig-
bridge the end of the development to the reca- inally generated in the exposition by rebegin-
pitulation), and the next cycle of events is newly ning on the tonic (with the initial theme in the
launched with the opening of the recapitulation. most common Sonata Types, 1, 3, 4, and 5) and
True, this more fundamental interruption on usually by restating all of the non-tonic modules
the dominant may sometimes be masked on the from part 2 of the exposition (S and C mate-
foreground with an apparent V – I cadence (with rial) in the tonic key. For this reason — its largely
the I triggering the recapitulation). But the more referential retracing of the rhetorical materials
fundamental or background concept is that of laid out in the exposition (Rotation 1) — we also
harmonic interruption on VA. (Those unfamiliar call the recapitulation the recapitulatory rotation.
with the Schenkerian, linear-contrapuntal view (Exceptions and reorderings of thematic mate-
of things might notice that this interruption di- rial may be found in some sonatas.) Because of
vides the entire sonata form at the end of the de- its function in bringing tonal closure to the en-
velopment. This contrasts with the eighteenth- tire form, we refer to the S/C complex in the
century “binary” division of sonata form at the recapitulation as the tonal resolution. Its shape and
end of the exposition.) manner of unfolding had been established by the
In terms of their rhetorical strategies, devel- exposition’s structure of promise. Correspond-
opments may or may not be fully or partially ingly, we consider the recapitulation to articulate
rotational (that is, guided in large part by the a structure of accomplishment. Minor-mode sonatas
 20 Elements of Sonata Theory

that had sounded S and C in the major mediant exposition :||: development – recapitulation :||).
(III) in the exposition have the additional option This is the most formal and earliest norm. Many
of sounding them in either the major or minor late-century first movements, especially those
mode in the recapitulation. after about 1760, repeat only the first part (the
The recapitulation’s S, launching the tonal exposition), although in works prior to 1790 one
resolution following a recapitulatory MC, leads need not be surprised to see the second part also
to the production of a satisfactory I:PAC that repeated. After that date, repeating the second
goes on to differing, non-S material. This is the part is an uncommon gesture that invites analyt-
moment of essential structural closure (the ESC), ical interpretation. It is also possible to find both
most often a point parallel to the exposition’s parts unrepeated. This occurs in lighter works,
EEC. The ESC represents the tonal goal of the in some midcentury symphonies (some Stamitz
entire sonata form, the tonal and cadential point symphonies from the 1750s; some early Mozart
toward which the trajectory of the whole move- symphonies; and so on) and in some slow move-
ment had been driving: this is suggested by the ments (especially those in the format of the less
longest dotted-line arrow in figure 2 – 1b. From expansive, Type 1 sonata, lacking a develop-
the perspective of Sonata Theory, it is only here ment). The nonrepeated exposition is also a ge-
where the movement’s tonic is fully called forth, neric feature of the overture or sinfonia. (In other
stabilized as a reality as opposed to a mere po- words, expositional repeats will not appear in ei-
tential. As in the exposition, C-material will ther operatic or concert overtures; this is also
follow, now in the tonic. The recapitulation’s true of the overture’s mid-nineteenth-century
final cadence is generally a I:PAC (or, in minor, offspring, the symphonic poem).7 In this aspect
sometimes a i:PAC), although this too may be the lighter overture is to be distinguished ge-
followed by a prolongation of tonic harmony or nerically from the more formal first movement
by a transition leading either back to a repeat of a sonata or grand symphony, which at least
(of the entire development and recapitulation) had available the common option of expositional
or forward into the coda. repetition. Nonrepeated expositions within first
A coda (outside of sonata space) may or may movements do sometimes occur in more broadly
not follow the recapitulation. More information scaled and ambitious works after 1780, but when
about codas, along with a discussion of the other they do — as in Mozart’s Symphony No. 35 in
optional or parageneric feature of some sonatas, D, K. 385, “Haffner,”8 or in Beethoven’s Vio-
the introduction, may be found in chapter 13. lin Sonata in C Minor, op. 30 no. 2, his Piano
Sonata in F Minor, op. 57, “Appassionata,” and
his String Quartet in F, op. 59 no. 1 — they are
Repetition Schemes exceptional and need to be considered as con-
sciously expressive choices.9
Within eighteenth-century sonatas and sym- One curious (and rare) possibility is that
phonies one may find both parts repeated (||: of literally writing out an expositional repeat,

7. Thus the rule. Exceptions are extremely rare and dis- those of the Serenades in D, K. 320, “Posthorn,” and in
concertingly puzzling, such as the repeat of the exposi- E-fl at, K. 375. Such examples — perhaps related to ear-
tion in young Mozart’s Overture to Apollo et Hyacinthus, K. lier or existing concepts of repeat-convention options
38 (1767), labeled as the “Prologus/Intrada” to the opera. in overture-symphonies, in smaller-scale symphonies,
This piece is a Type 2 sonata (Chapter 17) whose first rota- or in some serenades — require individual attention.
tion (exposition) is provided with a repeat sign. Much later, Within the larger symphony it may be that during the
the odd “expositional” (?) written-out and slightly varied 1770s (though not, it seems, in the 1780s) Mozart was
repetition in Berlioz’s Overture, Le carnaval romain is also exploring the possibility of the omission of the exposi-
curious, suggesting that the form of this unusual piece is tional repeat as a lower-level default.
more purely rotational (or perhaps instrumental-strophic 9. The solution of Beethoven’s op. 59 no. 1/i, which
with fortissimo refrain) than a sonata per se, although it is initially suggests an expositional repeat only to abort it
also manifestly in dialogue with certain sonata norms. almost immediately in favor of development, is antici-
8. Other examples within Mozart’s major works in- pated in the fi rst movements of Mozart’s Serenade in E-
clude the fi rst movements of his Symphonies Nos. 31 fl at for Eight Winds, K. 375, and Haydn’s Piano Sonata
in D, K. 297, “Paris,” and 34 in C, K. 338, along with in D, Hob. XVI:51.
 Sonata Form as a Whole 21

normally including variants the second time Control, balance, generic identification, and for-
around. This occurred most famously in C. P. E. mal architectural splendor: these would appear
Bach’s unusual set of six keyboard Sonaten mit to be the central reasons why literal repetition
veränderten Reprisen, H. 136 – 39, 126, 140 (W. played such a prominent role in the style.
50/1 – 6, Sonatas with Varied Repetitions), Consequently, repeat signs should not be
composed in 1758 – 59 and published in Berlin taken for granted, passed over lightly in analy-
the following year. In Haydn’s works the proce- sis, or omitted in performance. Repeat signs are
dure surfaces only (and wisely, in Tovey’s view) never insignificant.11 Block-repetitions are an
in a few “purely lyric slow movements,” such as integral component of the style, and compos-
the Adagios of the Quartet in C, op. 33 no. 3, ers can work with this defining convention in a
“Bird,” and the Symphony No. 102 in B-flat.10 variety of ways. When previously obligatory (or
(Both slow movements are in F major; in the exceptionally strong first-level default) exposi-
quartet the Adagio is the third movement; in tional repeats began gradually to disappear — es-
the symphony it is the second.) pecially in the early nineteenth century, with
What are the purposes of large-scale repeats certain works of Beethoven (op. 30 no. 2, op.
within sonata form? Central to the concept of 57, op. 59 no. 1, and so on, and later with Men-
the grand sonata or symphony is a system of delssohn, Schumann, and others) — the genre
schematic repeat-conventions, balances, sym- itself was undergoing a major rethinking.12
metries and proportions that call attention to The familiar, current views — Schenkerian and
and help to define the genre. The emphatically otherwise — that propose that some repeats are
architectural construction calls attention to the structurally insignificant while others are more
genre’s ordered formality — and in the case of the important (because of the unfolding of certain
grand symphony, also to its grandeur and public structural tones or other significant events, per-
splendor. Repeats were an important feature of haps under a first-ending sign) miss the larger
a sumptuous, high-prestige display of grand ar- point of repeat signs as generic identifiers.13
chitecture, one to which large-scale repetitions Even when the structural-tone aspects might
were essential — especially that of the exposi- be convincing (but, perhaps paradoxically, only
tional repeat in the first movement. The styl- as local details), the gist of these claims seems
ized form thus celebrated the “Enlightenment” to be based on later-nineteenth-century prem-
(or “modern”) culture that makes such an im- ises, which came to look on all unaltered repeti-
pressive, moving, or powerful art possible. One tion as an aesthetic error. Such a conviction also
of the structure’s implications would have been came to affect performance in the omission of
that this culture had devised a rational, balanced repeats or in the insistence on an altered inter-
means to shape and contain the fluid, raw, el- pretation in the repeat. It may be, though, that
emental power of music. By extension, the pro- saying the same thing twice was what the com-
cess probably also represented the controlling or poser had in mind.
harnessing of those impulsive, instinctive, libidi- It is easy to object to our general argument
nal, or “uncivilized” elements within ourselves. here. One could strive to minimize the impor-

10. Tovey, “Sonata Forms,” Musical Articles, p. 214: 12. Curiously, in 1826 Reicha suggested — in passing
“Haydn saw that the only place for C. P. E. Bach’s de- and without explanation (Traité de haute composition mu-
vice was in purely lyric slow movements. Even there he sicale, p. 300) — that fi nales may lack an explicit repeat:
never had the patience to plod and pose (as C. P. E. Bach “When the fi rst part is not repeated, as in overtures and
did to the bitter end) through a repetition [recapitula- fi nales . . .” (“Quand la première partie n’a pas de reprise,
tion] of both parts. When his second part comes to reca- comme dans les ouvertures ou dans les fi nales . . .”).
pitulate the second group it combines both versions.” It may be that Reicha had sonata-rondos, Type 4 sona-
11. For the quintessential statement of that which the tas, in mind (ch. 18 ).
present argument opposes, see Douglass M. Green, Form 13. Cf., e.g., Jonathan Dunsby, “The Formal Repeat,”
in Tonal Music: An Introduction to Analysis, 2nd ed. (New Journal of the Royal Musical Association 112/2 (1987),
York: Holt, Rinehart, and Winston, 1979), p. 82: “HIS- 196 – 207.
TORICAL NOTE [sic:] Ordinarily the repetition of a
part is of little significance in formal analysis.”
 22 Elements of Sonata Theory

tance of the usual repetition schemes by an ap- However we decide this matter, we should
peal to history: deriving them step-by-step from note three things. First, the issue of notationally
the earlier binary forms, then asserting that the indicating a repeat of section 2 was still part of
persistent lingering of the repeat conventions the historical concept of “grand binary” form
into the 1780 – 1820 period of the grand sym- (within a symphonic first movement) around
phony was an outdated survival, vestigial, un- 1800, even when that repeat was notationally
necessary to the perception of the genre. The elided. Its conceptual presence remained there,
larger question, though, is why the conven- counterpointed against the given, simpler struc-
tion remained available into the later phases ture. It persisted as historical-generic memory,
of 1780 – 1820 period and beyond (particularly even when it was not made physically present on
after Beethoven’s occasional removals of the the acoustic surface of the music. Second, any
expositional repeat had occurred). The expo- retention of the second repeat toward the end of
sitional repeat must have persisted, however the eighteenth century should be regarded as ex-
sporadically, because it was not merely vestigial. pressively significant, especially since its stron-
It continued to be genre-defining, a sign of spe- gest composers — Haydn and Mozart — were
cial grandeur and formality — with an ear at- apparently coming to believe that repeat 2 was
tuned also to the grand tradition and historical not as obligatory as that of repeat 1. When the
lineage that had led to the mid- and later-nine- repeat was called for, it must have been placed
teenth-century sonata and symphony. there for a reason, as in the slow movement and
Of the two standard large-scale repeats, the finale of Mozart’s Symphony No. 41 in C, K.
second, longer one (development-recapitula- 551 (“Jupiter”), where formal processes and
tion) was the one more vulnerable to suppres- monumentalized grandeur are principal top-
sion. This second repetition was increasingly ics throughout the whole work. Third, given a
reduced to the status of an easily discardable nineteenth-century work lacking an indication
option in the 1780 – 1800 period.14 In some of that second block-repetition, any reworked
cases, concerns of absolute length or a sense of referencing back to this increasingly atavistic re-
redundancy in closing particularly dramatic so- peat 2 within a longer, discursive coda, as in the
natas twice might have overridden the genre- first movement of Beethoven’s Symphony No. 3
defining principle of long-range architectural in E-flat, op. 55, “Eroica,” should be viewed as
repetition and balance, at least with regard to such, not as an innovative addition or accretion
this development – recapitulation section. Per- to a previously postulated, differing symphonic
haps the logic of the situation suggested that the practice.
obligatory repeat of section 1 alone (the expo-
sitional repeat) was to be viewed as sufficient as
a genre definer.

14. See, e.g, Michael Broyles, “Organic Form and

the Binary Repeat,” The Musical Quarterly 66 (1980),
339 – 60.