1 William E. Caplin, Classical Form: A Theory of Formal Functions for the Instrumental Music of Haydn, Mozart, and Beethoven (New York: Oxford University Press, 1998); James Hepokoski and Warren Darcy, Elements of Sonata Theory: Norms, Types and Deformations in the Late-Eighteenth-Century Sonata (New York: Oxford University Press, 2006). See also William E. Caplin, James Hepokoski and James Webster, Musical Form, Forms and Formenlehre: Three Methodological Reflections, ed. Pieter Bergé (Leuven: Leuven University Press, 2009).

2 A start has been made in a separate article by William E. Caplin, ‘On the Relation of Musical Topoi to Formal Function’, Eighteenth-Century Music 2/1 (2005), 113–124. But it remains half-hearted, Caplin concluding that ‘the results are somewhat discouraging’ (124).

3 On the possible rapprochement between analysis and hermeneutics see, for instance, Lawrence Kramer, ‘Haydn's Chaos, Schenker's Order; Or, Hermeneutics and Musical Analysis: Can They Mix?’, 19th-Century Music 16/1 (1992), 3–17, and Scott Burnham, ‘How Music Matters: Poetic Content Revisited’, in Rethinking Music, ed. Nicholas Cook and Mark Everist (Oxford: Oxford University Press, 1999), 193–216.

4 More specifically, the article puts into practice the call for greater historicization of ‘Sonata Theory’ in Matthew Riley, ‘Sonata Principles’, Music & Letters 89/4 (2008), 590–598. It is also informed by two other British reviews of Hepokoski and Darcy's work: Michael Spitzer, ‘Review: Sonata Dialogues’, Beethoven Forum 14/2 (2007), 150–178, and Paul Wingfield, ‘Beyond “Norms and Deformations”: Towards a Theory of Sonata Form as Reception History’, Music Analysis 27/1 (2008), 137–177.

5 Hepokoski and Darcy, Elements of Sonata Theory, 603.

6 Hepokoski and Darcy, Elements of Sonata Theory, 252.

7 Hepokoski and Darcy, Elements of Sonata Theory, 251.

8 Hepokoski and Darcy, Elements of Sonata Theory, 252.

9 For examples see Scott Burnham, Beethoven Hero (Princeton: Princeton University Press, 1996), chapter 1.

10 Donald Francis Tovey, ‘Normality and Freedom in Music’, in Essays and Lectures on Music (London: Oxford University Press, 1949), 183.

11 This article uses ‘phrase’ rather informally, more in the manner of Caplin than, say, William Rothstein in Phrase Rhythm in Tonal Music (New York: Schirmer, 1989).

12 Donald Francis Tovey, ‘Harmony’, in The Forms of Music: Musical Articles from the Encyclopaedia Britannica (London: Oxford University Press, 1957), 59.

13 Caplin, Classical Form, 17.

14 Tovey, ‘Harmony’, 58–59, and ‘Beethoven's Art Forms’, in Beethoven (London: Oxford University Press, 1944), 73; Charles Rosen, The Classical Style: Haydn, Mozart, Beethoven (London: Faber, 1997), 70–71, and Sonata Forms (New York: Norton, 1988), 98–99.

15 Caplin, Classical Form, 17. From the perspective adopted here, James Webster's analysis of this movement is hindered by an inattention to functional types. He calls bars 21–24 (my ‘caesura prolongation phrase’) the ‘opening theme’ or ‘main theme’. This means he cannot record the constantly shifting functional meaning of the phrase throughout the movement. Webster, Haydn's ‘Farewell’ Symphony and the Idea of Classical Style: Through Composition and Cyclic Integration in His Instrumental Music (Cambridge: Cambridge University Press, 1991), 167–173.

16 Caplin, Classical Form, 17.

17 Caplin, Classical Form, 37.

18 Caplin, Classical Form, 39.

19 Caplin, Classical Form, 97. It might be possible to read a subordinate theme in bars 57–72. After all, subordinate themes, unlike main themes, occasionally start with continuation function or with dominant harmony (Caplin, Classical Form, 111–115). But in the context of the strategies of this particular exposition those ways of hearing the passage, or indeed the ascription of an initiating function to the caesura prolongation phrase itself, seem unconvincing.

20 Following the (American) analytical language of the two main theoretical models I discuss, I use the term ‘perfect authentic cadence’ to refer to a V–I cadence in which both chords are in root position and the highest voice ends on the tonic. ‘Imperfect authentic cadence’ is used to describe a V–I cadence in which one or both of the chords are presented in inversion, or in which the highest voice does not conclude on the tonic. A cadence ending on V is a ‘half cadence’.

21 See, for instance, A. B. Marx, Musical Form in the Age of Beethoven, ed. and trans. Scott Burnham (Cambridge: Cambridge University Press, 1997), 94.

22 Caplin, Classical Form, 16.

23 Caplin, Classical Form, 15.

24 Until Symphony No. 90 (1788) Haydn's symphonic slow introductions are always in a different metre from the following fast movement. Thereafter a shared metre is more common than not. Symphonies Nos 90, 91, 92, 93, 96 and 97 use 3/4 metre for both introduction and fast movement. Ethan Haimo, Haydn's Symphonic Forms: Essays in Compositional Logic (Oxford: Clarendon, 1995), 209–210.

25 Hepokoski and Darcy, Elements of Sonata Theory, chapter 3. Their work was trailed in an earlier article: James Hepokoski and Warren Darcy, ‘The Medial Caesura and Its Role in the Eighteenth-Century Sonata Exposition’, Music Theory Spectrum 19/2 (1997), 115–154.

26 Hepokoski and Darcy, Elements of Sonata Theory, 40.

27 Hepokoski and Darcy, Elements of Sonata Theory, 37.

28 Hepokoski and Darcy, Elements of Sonata Theory, 51–64.

29 Jens Peter Larsen, Handel, Haydn, and the Viennese Classical Style (Ann Arbor: UMI, 1988), 274–275.

30 Tovey's terms. See, for instance, his discussion of a similar passage in Mozart's Piano Concerto k503/i. Donald Francis Tovey, Essays in Musical Analysis, volume 3: Concertos (London: Oxford University Press, 1936), particularly 18.

31 Haydn avoids harmonies with G♯ or F♮ that would enable a definite modulation.

32 Hepokoski and Darcy, Elements of Sonata Theory, 45.

33 Hepokoski and Darcy regard the grand antecedent / dissolving consequent pattern as a relatively common technique that is merely ‘in dialogue with the principle of medial caesura declined’, or a ‘very mild instance of it’ (Elements of Sonata Theory, 45). In this movement, though, the applied dominants in the previous bars and the emphatic cadential rhetoric make the possibility of a genuine medial caesura fairly credible.

34 Hepokoski and Darcy, Elements of Sonata Theory, 58.

35 Hepokoski and Darcy, Elements of Sonata Theory, 55. Extreme ‘bait-and-switch’ is thus almost indistinguishable from certain cases of medial caesura declined, as here. See also Hepokoski and Darcy, ‘The Medial Caesura’, 144.

36 Hepokoski and Darcy, Elements of Sonata Theory, 58.

37 Hepokoski and Darcy, Elements of Sonata Theory, 34.

38 Hepokoski and Darcy, Elements of Sonata Theory, 40–41.

39 Hepokoski and Darcy, Elements of Sonata Theory, 41.

40 Hepokoski and Darcy, Elements of Sonata Theory, 136. There are few examples of expanded (or even regular) caesura-fill in Haydn's later symphonies. See Symphony No. 84/iv, bars 52–63, and Symphony No. 98/iv, bars 62–75. Neither involves material from earlier in the movement, however, as does No. 92/i.

41 The cadence on or in vi and the retransition are among the few obvious conventions of the classical development section, making this part of the development suitable for witty strategies. These became common around 1800. See Rosen, The Classical Style, 466–482, and Sonata Forms, 271–274. Webster's decision to call the caesura prolongation phrase the ‘theme’ means that he labels bars 112–114 ‘retransition’. This seems unhelpful, as these bars do not follow a caesura; indeed, they are the final bars of a tutti. Again (see note 15) the analysis does not record the shifting functional meanings of the caesura prolongation phrase. See Webster, Haydn's ‘Farewell’ Symphony, 171.

42 Rosen, Sonata Forms, 289–296.

43 Webster counts the recapitulation as, proportionally, Haydn's longest. His bar-counts differ from those given here, however, as he includes the caesura prolongation phrases at bar 21–24 and 125–128 within the exposition and the recapitulation respectively. See Webster, Haydn's ‘Farewell’ Symphony, 171.

44 For instance, the long interpolation in the recapitulation of Symphony No. 99/i (bars 162–179).

45 In characteristic fashion, Haydn shuffles his materials: the approach to the recapitulation's perfect authentic cadence is derived from the middle section of the exposition, between the two declined potential medial caesuras (bars 51–55), where it had been part of the dissolution of the ‘dissolving consequent’. The approach to the exposition's perfect authentic cadence used different material altogether.

46 A. Peter Brown regards bar 205 as the start of a third ‘reprise’ within the recapitulation. He suggests that the movement could be heard as five ‘structures’ (akin to Hepokoski and Darcy's ‘rotations’): exposition, development and three reprises (starting at bars 125, 166 and 205 respectively), each of the five being shorter in duration than the last. A. Peter Brown, The Symphonic Repertoire, volume 2: The First Golden Age of the Viennese Symphony: Haydn, Mozart, Beethoven, and Schubert (Bloomington: Indiana University Press, 2002), 241.

47 Rosen, The Classical Style, 83, and Sonata Forms, 288.

48 James Hepokoski, ‘Beyond the Sonata Principle’, Journal of the American Musicological Society 55/1 (2002), 91–154.

49 On Rosen's concept of necessity see Sonata Forms, 287, 296 and 342, and, for commentary, Riley, ‘Sonata Principles’, 594–597.

50 Rosen, The Classical Style, 49.

51 Edward T. Cone, Musical Form and Musical Performance (New York: Norton, 1968), 77. This concept is criticized by Hepokoski and Darcy, Elements of Sonata Theory, 245.

52 Symphony No. 85/i, bar 62.

53 Webster aims for such explanations, but is unconvincing. His analysis of the symphony ends with a rhetorically exaggerated series of ‘why’ statements. See Webster, Haydn's ‘Farewell’ Symphony, 172–173.

54 Hepokoski and Darcy, Elements of Sonata Theory, 18.

55 Hepokoski and Darcy, Elements of Sonata Theory, 20.

56 Hepokoski and Darcy, Elements of Sonata Theory, 232.

57 Hepokoski and Darcy, Elements of Sonata Theory, 11.

58 Hepokoski and Darcy, Elements of Sonata Theory, 9.

59 Michael Spitzer points out that Hepokoski and Darcy's understanding of musical form shares much with that of Schenker, who regarded the goal of a tonal composition as predetermined, and form as arising from the disruption of motion towards it by means of interruption, detours and other retardations. See Spitzer, ‘Sonata Dialogues’, 151. From this perspective, the recapitulation of Symphony No. 92/i could be regarded simply as taking an especially circuitous path to its (inevitable) goal.

60 Hepokoski and Darcy, Elements of Sonata Theory, 15.

61 Gretchen A. Wheelock, Haydn's Ingenious Jesting with Art: Contexts of Musical Wit and Humor (New York: Schirmer, 1991); Mark Evan Bonds, ‘Haydn, Laurence Sterne, and the Origins of Musical Irony’, Journal of the American Musicological Society 44/1 (1991), 57–91.

62 Wheelock, Haydn's Ingenious Jesting with Art, 26.

63 The thought of the composer's mind being the ‘theme’ may seem unlikely, but see the quotation referenced in note 65 below.

64 Hepokoski and Darcy, Elements of Sonata Theory, 16.

65 Morning Chronicle, 12 March 1791. Cited in H. C. Robbins Landon, Haydn: Chronicle and Works, volume 3: Haydn in England, 1791–1795 (London: Thames and Hudson, 1976), 49.

66 Morning Chronicle, 19 March 1791. Cited in Landon, Haydn in England, 60.

67 London Chronicle, 10–12 March 1791. Cited in Landon, Haydn in England, 50.

68 Diary; or, Woodfall's Register, reviews of a concert of 2 April 1791. Cited in Landon, Haydn in England, 64.

69 Morning Herald, 18 February 1792. Cited in Landon, Haydn in England, 134.

70 Diary; or, Woodfall's Register, 18 February 1792. Cited in Landon, Haydn in England, 135.

71 Oracle, 7 April 1795. Cited in Landon, Haydn in England, 300. Another example: ‘In the minuet, the trio was peculiarly charming: but indeed the pleasure the whole gave was continual; and the genius of Haydn, astonishing[,] inexhaustible, and sublime, was the general theme.’ Morning Chronicle, 19 February 1794. Cited in Landon, Haydn in England, 236.

72 Christian Friedrich Michaelis, ‘Ueber das Humoristische oder Launige in der musikalischen Komposition’, Allgemeine musikalische Zeitung 9/47 (12 August 1807), columns 725–726. Cited in Bonds, ‘Origins of Musical Irony’, 77–78.

73 Bonds, ‘Origins of Musical Irony’, 78.