The orchestral consensus

What we know about eighteenth-century orchestras comes courtesy of a long paper trail: payrolls, concert reviews, travellers’ accounts, performance treatises, sheet music and more. They reveal what John Spitzer and Neal Zaslaw dub an ‘orchestral consensus’ in their definitive The Birth of the Orchestra. Encompassing ‘what an orchestra was, what instruments it comprised and in what proportions, and how those instruments were organized’, it lasted from the 1740s through the early nineteenth century – or, put in terms of the symphony, from about a decade after the genre emerged through to the middle works of Beethoven.² Differences between countries, cities and individual orchestras remained significant during this time, inspiring (or being inspired by) distinctive styles of orchestral scoring. Still, enough remained constant that music and musicians could circulate internationally and encounter similar conditions of performance.

At a minimum, every eighteenth-century orchestra could field four-part strings: two violins, viola and a bass line or basso covered by cellos and double bass. That force alone, joined by keyboard continuo according to local preference, sufficed for many symphony performances through at least the early 1770s. During those years, numerous works were composed for strings alone in four or even three parts, and the wind parts of many other symphonies were advertised as optional. At the same time, by 1740 most orchestras had winds, initially one or more bassoons, which at first doubled the basso; two or more flutes or oboes, sometimes played by the same, ‘double-handed’ musicians; two or more horns or trumpets, or both; and timpani. Clarinets joined some ensembles as early as the 1750s, but they did not become a standard in symphonies until very late in the century. Much more significant was the eventual creation of independent parts for all the winds. Around the middle of the century, most symphonies employed what were known in sheet music and music catalogues as six-, eight-, or eleven-part scorings. The strings counted as four parts regardless of how many different instruments played the basso, and the wind complement could amount to a single pair of oboes, flutes, horns or trumpets (six parts), one pair each of woodwinds and brass (eight parts), or two woodwinds, two horns, two trumpets and timpani (eleven parts). By the 1790s, the standard had grown to fourteen parts or more, the ensemble most typically associated with the modern term ‘classical orchestra’: four-part strings plus one or two flutes, two oboes, two clarinets (sometimes), one or two bassoons (now playing their own parts rather than doubling the basso), two horns, two trumpets and timpani. While expanded and unusual scorings do occur, most symphonies abided by these instrumentations, which in turn matched the most common disposition of orchestras.

Consensus did not extend to the number of players.³ Depending on finances and needs, performance spaces and taste, orchestras could have as few as ten members and as many as a hundred or more – though few standing orchestras had more than seventy (those assembled for special occasions were often larger). Large or small, most ensembles grew over the course of the century, at least until the 1780s when Spitzer and Zaslaw note a levelling off. On the other hand, during any given period they could differ radically from one another in size. The orchestra of the Concert spirituel in Paris, a leading venue for symphonies from the 1740s to 1790, went from around forty to around sixty members in that time. By contrast, that of Prince Eszterházy had only about fourteen members when Haydn became director in 1761, and about twenty-four when he departed in 1790. Both ensembles got larger, but a Haydn symphony played for Eszterházy by twenty musicians might have been performed in Paris, in the same year, by two or three times that many.

Balance varied as well, though less dramatically. In 1776 Haydn had seven violins, two violas, one cello, one double bass, eight winds and timpani. The high proportion of winds, nearly 50 per cent of the total, was typical for smaller orchestras, while larger ensembles had more strings: at the Concert spirituel in 1774, there were twenty-six violins, four violas, ten cellos and four double basses versus fifteen winds and timpani. Despite that difference, the examples share a strong opposition between treble and bass. The violin sections are large, 35 to 45 per cent of the ensemble, and while the fourteen low strings of the Concert spirituel reflect a French preference for especially strong bass, if one counts bassoons (still largely doubling the basso in the 1770s), then the bass constitutes 20 to 30 per cent of both totals. That proportion decreased over time, presumably diminishing somewhat the power of the bass; at the same time, regional variations such as the French enthusiasm for bass also diminished, such that by the early nineteenth century the proportions of balance were largely standardised across Europe.

By that time the instruments themselves had also begun to change, becoming more like their modern counterparts thanks to venting and key systems for the winds and the concave Tourte bow for the strings. For most of the eighteenth century, however, orchestra players stuck with older technologies.⁴ Apart from the double bass, which was found in three-, four- and five-string varieties of differing size and resonance, the strings retained a design established in the late seventeenth century and characterised by a flatter profile and lower bridge than the modern instrument. They were strung with gut and played mainly with convex bows. The woodwinds had fingerholes and generally only two or three keys, and the brass remained ‘natural’ instruments, capable of playing only those notes that belong to the overtone series of the key in which they were pitched.⁵ The last point is worth emphasising since it had a formative influence on orchestration. Brass players could change the key of their instrument by inserting extensions into its sounding tube – the longer the extension, the lower the pitch – and a system developed in the 1750s allowed horn players to do so between and eventually within symphony movements. Still, neither horns nor trumpets played much beyond the tonic triad and first five degrees of the scale in symphonies, whose composers mostly avoided the instruments’ highest registers, where the overtone series offers more notes. Hence the abilities of the brass shaped the whole symphonic sound, determining when they could be brought to bear and with what kinds of musical material.⁶

Venues and quality

Orchestras played in theatres, churches, a variety of concert rooms and outdoors. The acoustical results would have varied considerably depending on the nature of each space and the size of the ensemble. Spitzer and Zaslaw suggest that the orchestral style of the period was optimised for a dry sound characteristic of theatres and the outdoors, which allowed common effects like abrupt dynamic changes and repeated notes in the strings to be heard clearly. As they point out, however, a new generation of small, hard-surfaced concert rooms increasingly put those same effects into very live spaces that must have made listeners feel like the orchestra was all around them. This more visceral experience gained ground over the course of the century, as orchestras playing in large spaces took on ever more players to achieve comparable immediacy.⁷

Audience behaviour would further have affected orchestral sound, or at least its perception. Used in theatre pieces and church services as well as concerts, symphonies encountered differing degrees of attentiveness. In concerts they served mainly to open or close programmes of instrumental and vocal solos, and descriptions abound of audiences arriving, departing, chatting or playing cards during symphony performances.⁸ This does not necessarily mean that the orchestra was ignored; only that its sonorities probably drifted in and out of consciousness, perhaps according to some commonly encountered patterns of orchestration discussed below. Other descriptions, meanwhile, tell of audiences listening closely to symphonies, reacting to specific moments and calling for movements to be encored. Such circumstances presumably favoured the orchestra’s capacity for nuance.

A favourite topic of eighteenth-century observers, the quality of orchestral performance turned on many variables, among them money, stability and leadership. The best orchestras resided in the urban capitals and the courts of the German and Austrian nobility, which could muster the considerable resources necessary to underwrite even a small ensemble.⁹ In the cities, financing came from ticket sales and, in many cases, core groups of aristocratic sponsors, while in the courts it came from single patrons like Prince Eszterházy. Cities and courts also had the institutions that could keep ensembles together for a season of concerts or longer – sometimes much longer. In the cities, theatres employed musicians for comparatively long seasons and in turn provided most of the players for concert series such as the Concert spirituel (which had much shorter seasons) and for the many individual concerts organised by composers and soloists. The chances were, then, that any given performance in Paris, London, or another major city involved some proportion of musicians who played together regularly. Court orchestras enjoyed still greater stability. The most widely praised ensemble of the century, that of the Elector Palatine in Mannheim, enjoyed such continuity that by the time most of the encomiums appeared in the 1770s, many of its members had decades in service as well as sons and nephews who trained and performed beside them. Together with the skill and longevity of Christian Cannabich, director since 1757, the players’ long experience with one another helped to produce a legendarily unified manner of execution.¹⁰

Leadership had several dimensions.¹¹ Directors guided the ensemble in performance, typically with the violin from the concertmaster position and sometimes in tandem with a keyboard player. They ran such rehearsals as there were; concerts often had only one, which made the stability of the ensemble and its performance style all the more critical. Directors hired the musicians and they usually composed at least some of the music, another point of particular significance for the symphony. All of the century’s most influential composers of symphonies spent time directing orchestras: Giovanni Battista Sammartini in Milan; Cannabich and his predecessor Johann Stamitz in Mannheim; Johann Gottlieb Graun in Berlin; Haydn, Carl Ditters, Antonio Rosetti and others at smaller German and Austrian courts; Johann Christian Bach in London; and François-Joseph Gossec in Paris. Their experience allowed them to capitalise on the existing strengths of specific ensembles and even individual musicians, as well as to develop effective new sounds. More broadly, their insider’s knowledge formed the basis for the reciprocity of composition and performance noted above, without which symphony and orchestra alike might have remained colourful accompaniments to other things – their role throughout much of the century. Instead they moved from the background to the foreground of concert life, thanks in no small part to the interaction of musical style and ensemble capability.

Scoring and style

Although symphonies could display an orchestra’s abilities through striking effects like the premier coup d’archet (‘first stroke of the bow’, that is, a unison attack in all the strings), the relationship between style and ensemble ran much deeper. From its earliest years, the symphony stood for a modern and essentially comic idiom in which composers changed musical ideas and moods quickly rather than elaborating them at length as in some genres of the first half of the century.¹² The orchestra’s capacity for contrast made the changes vivid, an advantage that composers began to exploit already in symphonies for strings alone. Although lacking the colour and volume of winds, these works compensate with an astonishing array of textures, some of them inherited from antecedents of the symphony such as the concerto, trio sonata and dance suite. In four-part writing, the default is often what might be called ‘2+2’: that is, the two violins play in unison, octaves, or close harmony with one another, and the violas and basso do the same. Accommodating two-, three- and four-part harmony but only two layers of rhythm, the pattern affords flexibility and clarity and may dominate slow movements and finales, where lyricism and dance set the tone. First movements employ many textures in addition to 2+2, above all 1+1+2 (violin 1 and 2 in imitation or dialogue, viola and basso together); 2+1+1 (violins together, viola and basso independent); 1+3 (melody in violin 1, remaining parts together); and 1+2+1 (melody in violin 1, inner parts together, independent basso). In addition, the viola, basso, or both instruments regularly drop out, leaving reduced textures of two or three upper strings. The options can succeed one another with dizzying speed: the first movement of Sammartini’s Symphony in C (Jenkins-Churgin 7, c. 1730) registers twelve shifts of texture in only 112 bars.¹³ Demarcating the arrival of new musical ideas and harmonies, the shifts also create a drama of their own, a comedy of instrumental role switching for which the ideas and harmonies are themselves a vehicle.

Ensemble and style connect further in the so-called Trommelbass (‘drum bass’), the repeated crotchets or quavers in the basso that pervade many symphonies through the 1770s, including those with winds. Accommodating a new emphasis on homophonic, melody-and-accompaniment textures and an associated slowing in the rate of harmonic change, repeated bass notes invested long-held chords with a sense of rhythmic motion. They also highlighted one of the characteristic strengths of the early orchestra, namely its powerful bass: even in a small ensemble, the combination of cellos and basses, frequently doubled by bassoons, continuo and viola, would have given repeated notes a forceful resonance magnified further by the stringed instruments’ use of open strings in common keys such as C, G and D. In soft dynamics, the scoring is equally effective, producing an urgently pulsing hum. Frequently criticised as monotonous, the Trommelbass was all the same not simply a harmonic crutch but rather another means of conveying orchestral splendour.

Early symphonies with winds show the influence of another symphonic antecedent, the three-movement Italian opera overture or ‘sinfonia’. Neapolitan opera in particular cast a long shadow over orchestral technique, beginning in the 1730s with composers such as Giovanni Battista Pergolesi and Leonardo Leo, and continuing through the 1750s with Niccolò Jommelli and Baldassare Galuppi. Pushing contrast and homophonic clarity to new heights, they injected fresh colour through the addition of woodwinds and brass and streamlined the string textures into predominantly 2+2, Trommelbass-heavy homophony. They also pioneered widely imitated effects such as fanfare openings taking advantage of the ‘natural’ range of the brass, quiet passages featuring woodwind solos or the strings alone, and crescendos. The first two actually enjoyed longer lives in the symphony, but the last, generally attributed to Jommelli, came to rival the premier coup d’archet as the quintessential symbol of orchestral virtuosity. More than simply an increase in volume, the mid-century crescendo involved a coordinated intensification of melody (which rises), instrumentation (which grows) and rhythm (which subdivides; e.g. crotchets become quavers become semiquavers). All of this happened within four to twelve bars, leaving listeners if not breathless, as one writer claimed, then at least with a vivid impression of the orchestra’s entire sonic gamut.¹⁴

To the extent that sinfonias and concert symphonies can be distinguished (they were frequently marketed together), the latter tend to adapt Neapolitan conventions to a more mercurial and texturally diverse environment.¹⁵In the 1740s and 50s, both Sammartini and the leading symphonist in Vienna, Georg Christoph Wagenseil, used one or two pairs of winds to double the strings, play brief solos, or support the harmony, either in short punctuating chords or in long tones that sustain behind the strings. They also silenced the winds periodically to expose the contrasting colour of the strings alone. Thus far their works recall Leo or Jommelli, as does the use of slow movements, and the second key areas of fast movements, to feature wind solos or the strings by themselves. Yet the concert symphonies often have more changes of colour than the sinfonias, switching many times between the full ensemble or ‘tutti’ and various smaller groupings. Despite a predominance of homophony, moreover, the textures frequently grow quite complicated, whether due to counterpoint between the violins (as in many works of Sammartini), or the mixing of strings and winds in different roles. In an elaborate example from the first movement of Wagenseil’s Symphony in G (Kucaba G3), the first violin dialogues with the oboes on the principal melody of the second key area, while the second violin holds a pedal tone with the horn, and violas and basso play two distinct accompanimental figures.

Still more elaborate are Johann Stamitz’s symphonies, especially the late works of c. 1754–7, by which time he was leading an already excellent ensemble at Mannheim and had experienced the orchestras of Paris as well.¹⁶ He was indebted to Jommelli on several counts, including the crescendo, which he used frequently.¹⁷ He also explored new role divisions among the strings, activated the viola and basso with scales and melodic figures, and featured the winds – two woodwinds and two horns – with an enthusiasm matched in the 1750s only by other Mannheim composers such as Franz Xaver Richter and Anton Fils. The woodwinds (oboes, flutes or clarinets) always have at least one, and more often two featured episodes in the fast movements, and both woodwinds and horns regularly double the principal melody in the strings, which incorporates plenty of triad- and scale-based material suitable for the brass. In addition, though they frequently sit out the slow movements, the winds dominate the minuet-trio movements found between the slow movement and the finale in all the late works. Indeed, while Stamitz’s addition of a fourth movement has been much celebrated as a formal innovation, it was no less consequential for orchestration. Symphonic minuets returned again and again to the brassy, top-heavy sound Stamitz achieved with horn-ready melodies and heavy wind doubling, and likewise innumerable trios served, like his, as oases of pastoral lyricism scored for winds alone or with subdued string accompaniment.

By 1760 the eight-part scoring of strings and two pair of winds, sometimes augmented by trumpets and timpani, had become standard. It would remain so for two decades and even longer among some composers, although performance practice accommodated variations such as the playing of woodwind parts by different instruments in different orchestras and the omission of trumpets and timpani (or their addition to works that lacked them). The parts themselves settled into consistent roles, with the violins presenting most of the melodic ideas, the lower strings playing a lot of Trommelbass and other harmonic support, and the winds doubling, sustaining, punctuating and soloing. Equally consistent were contrasts in scoring that lent many symphonies a similar orchestral shape. First and last movements used the full ensemble, slow movements omitted the winds or featured them in solos, and minuets (in four-movement symphonies) performed the same switch in miniature, from tutti scoring in the minuet to strings alone or wind solos in the trio.

Contrast remained a basic principle within movements as well, where the order of colours and textures varied greatly but within common parameters that may have affected audience reception. Most importantly, the opening sections of many first movements incorporated some or all of four common ingredients: ‘melodic’ tuttis, which were homophonic and used for principal musical ideas; ‘concerted’ tuttis with counterpoint, motivic action in the basso and sustained chords, used for harmonic transitions and cadences; reduced textures for the strings alone or strings with a wind solo, used again for principal ideas; and either a crescendo or quick alternations between tutti and reduced scoring, used for all purposes. Sometimes arranged so consistently as to suggest a kind of orchestrational ‘form’ (for instance in the Op. 3 symphonies of J. C. Bach), and elsewhere presented in many diverse orders (as in works by Haydn, Gossec, Ditters and other leading figures of the 1760s), these ingredients furnished a precis of orchestral effects likely to be heard later in the symphony. Their appearance near the beginning may further have suited those occasions, mentioned above, in which the finer nuances of style were lost to conversation and other distractions. The opening sequence of scorings primed audiences to perceive the swelling and diminishing of volume, the addition and subtraction of voices and other broad strokes of a performance that now mirrored, now contrasted with the hubbub of social interaction.

Orchestration as variation

Like Stamitz before him, during the 1760s and 1770s Haydn capitalised on his association with a virtuoso court orchestra to create some exceptionally rich orchestrations. The abilities of Eszterházy’s players are most evident in the showy pieces Haydn wrote within a few years of becoming their director in 1761, symphonies nos. 6–8, 31 and 72.¹⁸ They include difficult solos for winds and strings along with sounds he would not use again for some years: a wind choir accompanied by pizzicato strings (No. 6); duets in the tenor range for bassoon and cello (nos. 7 and 8); and four soloing horns (nos. 31 and 72). In less extravagant works, the orchestra must still negotiate unusually dense counterpoint – only Mozart and C. P. E. Bach ask for anything comparable – and its wind players must perform solos well beyond the conventional contexts of slow movements, trios and second key areas of fast movements.

The reliability of the musicians undoubtedly encouraged Haydn’s signature orchestral technique as well, the use of colour and texture for variation. Theme-and-variation movements offer the most obvious examples, often emphasising orchestral as much as thematic or harmonic development. Two of the early showpieces (nos. 31 and 72) end with finales that review the soloists and orchestral sections in successive variations. Later examples, particularly slow movements from the later 1770s and early 1780s, instead offer systematic, almost textbook-like illustrations of orchestral technique. While alternating between minor- and major-key versions of its theme, for example, the Allegretto of Symphony No. 63 offers the wind section both alone and accompanied; the tutti in dialogue with three-part strings; a grand concluding tutti; and no fewer than four different homophonic string textures, including one in which the flute doubles the melody in the first violins (a favourite sound of Haydn’s). Haydn is not always so didactic, but orchestration figures large right through his last, alternating variation set in the second movement of Symphony No. 103, where the strings and winds trade and redecorate both themes.¹⁹

An aspect of Haydn’s broader tendency to develop musical ideas continuously, his use of colour and texture as means of variation also occurs outside actual variation movements. Any thematic repetition is a candidate, whether immediate, as in the parallel phrases of antecedent-consequent pairings, or delayed, as in the reprises and recapitulations of sonata, rondo and minuet forms. The second possibility takes dramatic form in several first movements from the 1760s, where the openings of the recapitulations depart significantly from the movements’ opening bars: strings take over what were originally themes for winds, or vice versa (nos. 8, 24, 31 and 72), or the winds interject previously unheard solos and elaborations (nos. 6, 13, 30 and 35). Often encompassing domains beyond orchestration (e.g. nos. 24 and 31 both shift from major to minor), the changes help figure the recapitulation as a new event in a still-unfolding drama.

In the 1770s and 1780s, orchestral variation becomes less dramatic but more pervasive, especially following Haydn’s oft-remarked turn towards a more popular style around 1774. During this time his orchestration grows more transparent and consistent, with less counterpoint, more Trommelbass and textures of longer duration to match more expansive tunes and forms. The new expansiveness means that melodies are often played twice in succession or divided into antecedent and consequent phrases, in which case the first or antecedent hearing is frequently given to the strings, and the second or consequent to the winds (e.g. nos. 68, 74 or 84). Something similar happens in sonata- and rondo-form reprises, where themes played initially by the strings return doubled by one or more wind instruments – or, if already doubled the first time around, with a different doubling (e.g. nos. 56, 66 and 82). Still another possibility involves themes distributed among multiple instruments, like a swaying minuet tune in the first movement of Symphony No. 76, whose first two bars are played by the horns and flute, the next two by the oboes, and the remaining four by all the winds. Itself an increasingly common way to create contrast within themes, dialogue of this kind practically begs to be varied, and the subtlety of the recombinations belies the apparent simplicity of Haydn’s ‘popular’ manner. By the end of the first movement of No. 76, the wind section has shared the tune in four different ways, each time to a different string accompaniment. Much as in Sammartini’s textural changes of several decades before, the switching of instrumental roles becomes a drama all its own.

In the 1770s and 1780s Mozart used colour and texture in comparable ways, though with fewer effects on his recapitulations and reprises. More often he altered immediate thematic repetitions, especially by reconfiguring ever more elaborate dialogues between strings and winds (e.g. K 183 in G minor, K 200 in C, the ‘Paris’ Symphony K 297 in D, or K 550 in G minor). Variations like these, or like Haydn’s of the same period, are common coin among prominent symphonists of the following decades such as Adalbert Gyrowetz, Paul Wranitzky, Anton Eberl, Ignaz Josef Pleyel and Beethoven: witness Beethoven’s Symphony No. 1 in C (1800), where all the principal themes of the first movement are rescored in the recapitulation. The new generation did not, however, follow the example of Haydn’s own symphonies of the 1790s, in which variation achieves an unparalleled complexity amidst increasingly developmental forms.²⁰ Whether in variation, rondo, or sonata movements, scarcely a melody returns in the ‘London’ symphonies without the texture being changed or new instruments adding doubling and countermelodies. In first movements, moreover, where recapitulations routinely depart from the expositions, orchestration takes on quasi-narrative qualities, for example when opening themes are heard twice or more in the course of an exposition, and different scorings raise the question of which version will return later on. The answer typically provides a kind of colouristic ‘resolution’, as multiple possibilities are combined (nos. 96, 98 and 104), a missing option filled in (No. 100, where a tutti replaces competing string and wind scorings), or other expectations set up and fulfilled. In Symphony No. 102, back-to-back renderings of the opening theme juxtapose a vigorous tutti with a reduced texture featuring solo flute. The second version returns as a ‘false recapitulation’ midway through the development section, then disappears at the beginning of the real recapitulation. Yet rather than let the flute represent a mistaken alternative or formal disorder, Haydn has it play the theme again just before the final cadence. A colouristic reprise thus settles the last unfinished business of the movement.²¹

New contrasts

The nuance of Haydn’s variations, and of colouring generally in the last two decades of the century, was facilitated by modifications to wind scoring. In the 1770s wind instruments began to appear in all slow movements, not just those in which they played solos. In addition, by 1780 many composers had introduced one or two independent bassoon parts as well as multiple upper woodwinds, most commonly one or two flutes with two oboes. With more voices and a more flexible bass than the horns could provide, the wind section became more self-sufficient and appeared more frequently by itself, unsupported by strings. Since the woodwinds alone could now muster harmonies in five or six parts, the section was also more frequently split in two, most commonly in passages where the winds sustained and the brass and timpani punctuated. The separation of the bassoons from the basso, finally, together with a rise in independent cello parts and more ambitious writing for the viola, brought an activity to the tenor range that eventually transformed the whole orchestral sound. The tenor instruments became the most versatile in the ensemble, playing solos, furnishing harmonies, doubling melodies at the octave and providing light basses in various reduced textures. While they also united with the double basses to reconstitute the old basso, their primary effect was to mitigate the familiar opposition between bass and treble. Already in the 1770s and early 1780s, symphonies by C. P. E. Bach (Wq 183), Gossec (Brook 83–5), Ditters (now von Dittersdorf; see Grave Am1, and the symphonies on Ovid’s Metamorphoses), Cannabich (nos. 46, 57, 59 and 63) and others showed the effects, their sound growing fuller and more diffuse, and their gradations of colour ever finer.²²

Mozart was populating his middle ranges even in the early 1770s, writing extensive sustaining parts for the winds and using the inner strings as much to fill in the harmonies – often with double stops, tremolo and arpeggios – as to reinforce the outer parts. He placed particular confidence in the violas, making them equal partners in contrapuntal passages and frequently splitting them into two lines (‘divisi’) that double violin or woodwind pairings an octave below (e.g. K 84, 114, 162 and 200). The last remained a popular device among many composers through the end of the century, and Mozart himself found additional ways to expand the viola’s role: the opening of the Symphony in G minor (K 550), with its divisi violas murmuring an accompaniment to the violins, is an emblematic example. In the later symphonies as in the piano concertos and operas, Mozart also greatly expanded the responsibilities of the winds. In addition to playing without the strings, they reverse roles with them and accompany melodies that are no longer harmonised within the string section itself – for example, an exchange between bass and treble in the second movement of the ‘Jupiter’ Symphony (K 551 in C), where strings in bare octaves play against pulsing chords in the bassoons and horns. The winds also acquired enhanced solo roles, especially in melodic dialogues between multiple winds or between winds and strings.²³

The last innovation is especially significant inasmuch as many other symphonies of the 1780s – Dittersdorf’s and Cannabich’s, for example, and even Haydn’s to an extent – continued to ‘feature’ the winds, assigning them self-contained and often extended melodies in favourite places (expanded from slow movements, trios and second themes to include variations, rondo episodes and closing themes in sonata forms). Where shorter solos and dialogues occurred, the context tended to be developmental, as when fragments of a theme are sequenced in different keys. Mozart was more likely to build such exchanges into the themes themselves. The Allegro of the ‘Prague’ Symphony (K 504 in D) opens with a back-and-forth between string antecedent and wind consequent, followed by a repetition of the antecedent with oboe countermelody. At closer range, strings and winds divide the antecedent of the second key area of K 550, with the order of entries reversed on repetition. Closer still, violins, horns and bassoons alternate nearly bar-by-bar in the opening theme of the Symphony in E flat (K 543), constructing a virtual ‘tone-colour melody’ that later shifts to bass strings, clarinets and flute. No longer only a means for distinguishing parts of a form, or even phrases in a melody, orchestral contrast could now define the very identity of the musical idea.

The force of Mozart’s later symphonies opened another dimension of contrast, a familiar but vastly magnified distinction between tutti and reduced scoring. It was inherited, at least in part, from symphonies and orchestras he heard in Mannheim and Paris during 1777–8, which influenced the formal and expressive designs of several of his works.²⁴Most striking, passages like the opening of the ‘Paris’ Symphony reveal a new insistence on the sheer grandeur of the tutti. Three bars of a D major chord, scored for full orchestra and articulated only by minim rhythms, sound all the more massive by comparison to an answering figure for violins alone, and they also postpone any real musical action; the movement gets underway only after the chord and violin figure have been repeated and a contrapuntal development begun. The power of what precedes is due only in part to Mozart’s expanded instrumentation (seventeen parts: strings, timpani and six pairs of winds). Appropriately likened to slow introductions, such openings occur also in eight- and eleven-part works written for Mannheim and Paris in the 1770s, whose orchestras must have sounded impressive even without the extra winds (e.g. in Gossec’s symphonies in F and C, Brook 83 and 85, or Cannabich’s nos. 55 and 57).

After 1780, ever grander tuttis reinforced a long-standing association between symphonies and the sublime. The interwoven textures, chorus-like agglomerations of instruments, and indeed the dynamic and colouristic contrasts of symphonies had already been deemed exalted and awe-inspiring.²⁵ Now the last of these became epochal in scope. Openings like that of the ‘Paris’ Symphony became a regular feature of actual slow introductions, sometimes expanded as in Wranitzky’s Symphony in D (Op. 36), where a rolling tutti stretches several bars before giving way to an elegant string melody; and sometimes reduced as in many of Haydn’s late works, where short bursts of sound alternate with various quiet responses. Other examples engender some of the most memorable fractures in late eighteenth- and early nineteenth-century symphonies. In the first movement of Mozart’s ‘Jupiter’, a tutti blast of C minor answers the trailing off of the strings in the second key area; in the second movement of Haydn’s ‘Military’ Symphony (No. 100), an even bigger blast of A flat (seventeen parts plus triangle, cymbal and bass drum) counters the dwindling to pianissimo of the main theme – ‘a climax of horrid sublimity’, according to one observer.²⁶And in the first movement of Beethoven’s Eroica Symphony (No. 3 in E flat), which adds only a third horn to the standard seventeen-part ensemble, the infamous dissonant tutti of the development section collapses into a familiar if updated reduced texture, woodwind melody accompanied by strings. The narrative weight these moments bear should not obscure their role as orchestral effects, successors to a long line of dramatic contrasts. Once again they broadcast the discipline, breadth and visceral impact that made orchestras famous, and that symphonies absorbed into the very heart of their style.