1 See, for example, Wilson Coker, Music and Meaning: A Theoretical Introduction to Musical Aesthetics (New York, 1972); David Osmond-Smith, ‘The Iconic Process in Musical Communication’, VS: Quaderni di studi semiotici, 3 (1972), 31–42; and Jean-Jacques Nattiez, Music and Discourse, trans. Carolyn Abbate (Princeton, NJ, 1990).

2 Raymond Monelle, ‘Music and the Peircean Trichotomies’, International Review of the Aesthetics and Sociology of Music, 22 (1991), 99–108.

3 Robert S. Hatten, Musical Meaning in Beethoven: Markedness, Correlation, and Interpretation (Bloomington, IN, 1994); Naomi Cumming, The Sonic Self: Musical Subjectivity and Signification (Bloomington, IN, 2000). Further use of Peircean semiotics is found in W. Jay Dowling and Dane J. Harwood, Music Cognition (San Diego, CA, 1986), 202–24; Richard Cohn, Audacious Euphony: Chromaticism and the Triad's Second Nature (Oxford, 2012), 22; and Stephen Rumph, Mozart and Enlightenment Semiotics (Berkley, CA, 2012), 7–9.

4 See Christopher Hookway, Peirce (London, 1985), 115.

5 See Charles S. Peirce, ‘On a New List of Categories’, Proceedings of the American Academy of Arts and Sciences, 7 (1867), 287–98 (p. 293), repr. in The Essential Peirce: Selected Philosophical Writings, 2 vols. (Bloomington, IN, 1992–8), i: 1867–93, ed. Nathan Houser and Christian Kloesel, 1–10 (p. 6); ‘Some Consequences of Four Incapacities’, Journal of Speculative Philosophy, 2 (1868), 140–57 (p. 149), repr. ibid., 28–55 (p. 42); ‘On the Algebra of Logic: A Contribution to the Philosophy of Notation’, American Journal of Mathematics, 7 (1885), 180–202 (p. 180), repr. ibid., 225–8 (p. 226); and ‘Sundry Logical Conceptions’, in ‘Syllabus’ (unpublished pamphlet, 1903), repr. ibid., ii: 1893–1913, ed. The Peirce Edition Project, 267–88 (p. 272).

6 Hatten, Musical Meaning in Beethoven, 244.

7 Nattiez, Music and Discourse, 7–8. For a useful overview of this framework and Nattiez's application of Saussure's notion of the paradigmatic to music, see Raymond Monelle, Linguistics and Semiotics in Music (Reading, 1992), 32–4, 90–126.

8 In a 1903 typology, Peirce attempts to define the relationship between these two trichotomies in quite narrow terms. Here he conceives icon–index–symbol as the trichotomy of the relationship between a sign and its object, while two other trichotomies apply to other parts of the sign complex. I see this 1903 typology as an interesting experiment in relating these trichotomies and in developing others. However, it is important to recognize that this typology is not laid out in earlier work and is quickly replaced by sign systems that are more elaborate still. Key to understanding both trichotomies is their articulation of the categories. Their intimate connection can then be grasped by considering the way in which a sign or thought will form part of a series or network of interpretants governed by rule or convention (and therefore involving symbols), which in turn must relate to the actual world of objects (thus calling for indices); once we take away these symbolic and indexical dimensions we must be left with something, and this is a sign as pure form or potentiality (that is, as an icon). For details of the 1903 typology, see Charles S. Peirce, ‘Nomenclature and Divisions of Triadic Relations, as Far as They Are Determined’, in ‘Syllabus’ (unpublished pamphlet, 1903), repr. in The Essential Peirce, ii, 289–99.

9 See Charles S. Peirce, ‘On a New List of Categories’, 294, repr. in The Essential Peirce, i, 7; ‘On the Algebra of Logic’, 180–1, repr. ibid., 226; ‘Of Reasoning in General’ (unpublished chapter, 1895), repr. ibid., ii, 11–26 (p. 13).

10 See Vladimir Karbusicky, ‘The Experience of the Indexical Sign: Jakobson and the Semiotic Phonology of Leoš Janáček’, American Journal of Semiotics, 2/3 (1983), 35–58; and Grundriss der musikalischen Semantik (Darmstadt, 1986).

11 See Monelle, ‘Music and the Peircean Trichotomies’.

12 This tendency to focus attention upon Peirce's proliferated sign types at the expense of the broad sweep of his philosophy is particularly evident, I would suggest, in Cumming's The Sonic Self. For a detailed critique, see Jairo Moreno, ‘Review of Naomi Cumming: The Sonic Self’, Music Theory Spectrum, 27 (2005), 283–354, and Ben Curry, ‘Reading Conventions, Interpreting Habits: Peircian Semiotics in Music’ (Ph.D. dissertation, University of Cardiff, 2011), 120–79.

13 See further Ben Curry, ‘Time, Subjectivity and Contested Signs: Developing Monelle's Application of Peirce's 1903 Typology’, Music Semiotics: A Network of Significations, ed. Esti Sheinberg (Aldershot, 2012), 149–61.

14 See Max H. Fisch, ‘Peirce's General Theory of Signs’, Peirce, Semeiotic, and Pragmatism: Essays by Max H. Fisch, ed. Kenneth Ketner and Christian Kloessel (Bloomington, IN, 1986), 321–55.

15 Charles S. Peirce, ‘How to Make our Ideas Clear’, Popular Science Monthly, 12 (1878), 286–302 (p. 293), repr. in The Essential Peirce, i, 124–41 (p. 132).

16 In this article Peirce's labels for his categories – firstness, secondness and thirdness – are avoided in favour of iconicity, indexicality and symbolism, as the latter set of terms indicates more overtly the close-knit relationship between sign types and the categories. For an introduction to Peirce's categories, see Hookway, Peirce, 80–117.

17 See ibid., 89–90.

18 There is a potential source of confusion here because Peirce will tend to emphasize the indexicality of such a connection on every level. Thus the pointing finger and pronoun are his most common examples of signs that saturate the bonds of an icon like ‘(__) loves (__)’. But clearly the symbolic is also in play here. Similarly, the iconic might have a part to play as well. For example, in propositions containing onomatopoeias such as ‘I heard a meow’, there is an iconic connection between the index ‘meow’ and the type of sound signified, especially if the word ‘meow’ is stated with an approximate impression of a cat. This point becomes more important in applying this model to music.

19 Peirce alludes to this point in his Lowell Lectures of 1903 when he notes that ‘I say of a stone that it is hard. That means that so long as the stone remains hard, every [future] essay to scratch it by the moderate pressure of a knife will surely fail. To call the stone hard is to predict that no matter how often you try the experiment, it will fail every time. That innumerable series of conditional predictions is involved in the meaning of this lowly adjective.’ Charles S. Peirce, ‘What Makes a Reasoning Sound?’ (first of eight Lowell Lectures given in 1903 under the title ‘Some Topics of Logic Bearing on Questions Now Vexed’), first published in full in The Essential Peirce, ii, 242–57 (p. 254).

20 This connection might, in turn, be extended by suggesting that the idea of the cuckoo's song can subsequently engender further association. See, for example, Raymond Monelle, The Sense of Music (Princeton, NJ, 2000), 17.

21 This point is derived from Peirce's conception of the category of ‘representation’ (later ‘thirdness’, or what I am terming ‘symbolism’) presented in his early but highly important essay ‘On a New List of Categories’ of 1867. Peirce demonstrates this category by considering how the process of comparing two things, such as the letters ‘p’ and ‘b’, requires a mediating representation that ‘represents one of them to be (when turned over) the likeness of the other’. Peirce, ‘On a New List of Categories’, 293, repr. in The Essential Peirce, i, 6.

22 I follow Cooper and Meyer here in using the term rhythmic structure to encompass a range of temporal factors such as pulse, tempo, metre, stress, rhythm and grouping. See Grosvenor Cooper and Leonard B. Meyer, The Rhythmic Structure of Music (Chicago, IL, 1960).

23 Patrik N. Juslin and Renee Timmers, ‘Expression and Communication of Emotion in Music Performance’, Handbook of Music and Emotion: Theory, Research, Applications, ed. Patrik N. Juslin and John A. Sloboda (Oxford, 2010), 453–89. See also Michael Spitzer, ‘The Topic of Emotion’, Music Semiotics, ed. Sheinberg, 211–23.

24 It is important to note here that I use the term pure form in the Peircean sense of a logical possibility, which I would distinguish from the notion of essence. Pure form or a pure icon can be conceived by considering that from any thought of an object (symbol) we can prescind the act of thinking to leave only the object (index). From this object we can prescind its existence leaving only the possibility of the qualities that object embodies. This is an icon proper or pure form – it does not reside anywhere because it does not exist.

25 I have emphasized the way in which an icon can articulate a valency of 1, 2 or 3. There may appear to be an awkward conflation of quality and quantity here, but it is important to note that Peirce does not draw an opposition between quality and quantity in conceiving his categories. These categories can be identified as concerning quality, actuality and rule or in quantitative terms as relations of 1, 2 or 3, but there is no straightforward opposition here between quantity and quality. Furthermore, any sign (for example, an icon) can be trichotomized in accordance with the categories. Thus the trichotomy of the icon in accordance with valencies of 1, 2 or 3 is consistent with other aspects of Peirce's approach to the categories.

26 A gestalt should be understood here as a musical unit perceived as an integral whole and not as a collection of parts.

27 This is only one of Caplin's criteria for identifying the period. The other is the pattern basic idea–contrasting idea (articulating a weaker cadence), followed by basic idea–contrasting idea (articulating a stronger cadence). I follow Allanbrook in classing these 12 bars as a (non-parallel or contrasting) period, whereas Caplin might conceive them in terms of a hybrid theme (hybrid 1 perhaps). See William E. Caplin, Classical Form: A Theory of Formal Functions for the Instrumental Music of Haydn, Mozart and Beethoven (New York, 1998), 265 note 1. Wye Jamison Allanbrook, ‘Two Threads through the Labyrinth: Topic and Process in the First Movements of K.332 and K.333’, Convention in Eighteenth- and Nineteenth-Century Music: Essays in Honor of Leonard G. Ratner, ed. Wye Jamison Allanbrook, Janet M. Levy and William P. Mahrt (Stuyvesant, NY, 1992), 125–71 (p. 131).

28 On the face of it, of course, the icon of the linguistic proposition (e.g. ‘(__) loves (__)’) is fundamentally different from that in music, because the former involves a symbol (the verb, in this instance), whereas the latter consists only in juxtaposition. It is important to remember, however, that Peirce conceives formulations such as ‘(__) loves (__)’ as icons only in so far as they map out relations between two objects. Words, after all, are essentially symbols, so the word ‘love’ is not an icon, but such words can function iconically in the mapping of relations and, in this sense, correspond to the mapping processes of musical form.

29 Hookway, Peirce, 131.

30 Evidence for this abounds, but two examples will have to suffice here: in a recent issue of Music Analysis, René Rusch discusses the very different interpretations of a striking shift in Schubert's Moment musical in A♭ major; and in a representative example from music semiotics, Eero Tarasti notes that the ‘cross’ motive in Bach's Fugue in C♯ minor from Book I of the Well-Tempered Clavier has acted for some listeners as a representation of Christ but for others as a representation of ‘absolute’ music. Rusch, ‘Rethinking Conceptions of Unity: Schubert's Moment musical in A♭ Major’, Music Analysis, 30 (2011), 58–88; Tarasti, Signs of Music: A Guide to Musical Semiotics (Berlin, 2002), 6.

31 One such context might be that of a lookout who is allocated three tunes to whistle by a thief to indicate the observation of particular different objects (for example, a police officer), or a comparable children's game in which specific actualities might be communicated through distinct melodic ideas. Such instances are clearly not representative of the way in which music is generally used in the West.

32 The sharing of qualities between two entities is inevitable, but ‘overtly similar’ refers here to a situation where the qualities shared are more obvious. To suggest that an index can operate iconically may seem contradictory, but recall that any index will involve an icon (see again Figure 1 on p. 407). Thus, returning to the example of Beethoven's cuckoo, we can observe that the mapping out of the cuckoo's song in time and pitch space is iconic, and that the saturation of this map with remembered experience of actual cuckoo song is indexical; but the indexical function here is articulated, in part, by the qualitative dimension (in this case the timbre) of each of the notes of Beethoven's cuckoo. Thus, an indexical function will always involve a more or less overt role for the iconic. In music, it is often particularly overt, but in language it is not – the exception being onomatopoeia.

33 Topics were first introduced in Leonard Ratner's Classic Music: Expression, Form, and Style (New York, 1980). Important subsequent studies include Kofi Agawu, Playing with Signs: A Semiotic Interpretation of Classic Music (Princeton, NJ, 1991); Raymond Monelle, The Musical Topic: Hunt, Military and Pastorale (Bloomington, IN, 2006); and, most recently, Rumph, Mozart and Enlightenment Semiotics.

34 Allanbrook, ‘Two Threads through the Labyrinth’, 132. Some of the fuzziness that characterizes topic theory is evident here. Ratner notes that in a singing style ‘presumably any of the familiar dance rhythms could be used’. In this instance a minuet is arguably used from the outset, but a minuet character is more noticeable in bars 9–12. This is perhaps owing to the character of the music in these bars, which appears to sit particularly comfortably with those terms identified by Ratner in his account of late eighteenth-century descriptions: ‘noble, charming, lively, expressing moderate cheerfulness’. Ratner, Classic Music, 9.

35 These claims highlight an important contrast between my thinking (and perhaps my interpretation of Peirce) and that of Monelle. In The Musical Topic Monelle claims that we should not assume that the ‘signified [of the musical topic] was ever part of the social and material world’. I follow Peirce in conceiving all thought as ultimately derived from experience of the material world. Insight into Peirce's position can be gained in his approach to the problem of mythical, non-existent objects. In 1905 he notes that ‘although no phoenix really exists, real descriptions of the phoenix are well known to the speaker and his auditor; and thus the word is really affected by the Object denoted’. Monelle, The Musical Topic, 26; Peirce, ‘Nomenclature and Divisions of Triadic Relations’, repr. in The Essential Peirce, ii, 289–99 (p. 295).

36 The other conclusions are as follows: ‘We have no power of Intuition, but every cognition is determined logically by previous cognitions. We have no power of thinking without signs. We have no conception of the absolutely incognizable.’ Peirce, ‘Some Consequences of Four Incapacities’, 141, repr. in The Essential Peirce, i, 30.

37 Karl-Otto Apel, Charles S. Peirce: From Pragmatism to Pragmaticism (Amherst, 1981), 96.

38 Robert S. Corrington, An Introduction to C. S. Peirce: Philosopher, Semiotician, and Ecstatic Naturalist (Lanham, MD, 1993), 57.

39 The later bar is stated first here, as it is only on hearing the later bar that it can be paired with the first. In this sense, we always work back from the present to form V2s.

40 See Caplin, Classical Form.

41 An intramusical level is conceived in this system as an aspect of the intermusical level. Intermusical level, then, will generally involve reference of one part of a work to another, but it may also involve reference from one work (or part of a work) to another work (or part of a work).

42 This terminology is encountered in Caplin, Classical Form. For a more recent summary of these functions, see William E. Caplin, ‘What Are Formal Functions?’, Musical Form, Forms and Formenlehre, ed. Pieter Bergé (Leuven, 2009), 21–40 (pp. 25–7). Caplin notes the similarity between his functions and the ‘beginning–middle–end paradigm’ of introversive semiosis (ibid., 25). See also Agawu, Playing with Signs, 51ff. Melanie Lowe deploys a similar taxonomy using the terms opening, closing and continuing functions. Lowe, Pleasure and Meaning in the Classical Symphony (Bloomington, IN, 2007), 30–54.

43 It is also important to note here that the extensive use of V2s in Example 3 is possible because the label V2 is not limited to instances of ending functions. Such ending functions are perhaps the paradigm case of the V2 because they draw attention most powerfully between a present moment and music that has passed. But V2s of a weaker sort will occur far more regularly whenever there is a sense of one musical gestalt being opposed to another that has passed.

44 This notion of a character in itself is clearly problematic, a point pursued by Peirce early in his career in ‘On a New List of Categories’. However, for Peirce (especially the later Peirce) we can prescind habit (the symbolic) and actuality (the indexical) in order to arrive at a quality (an icon) in itself as a pure possibility.

45 For a description of ‘Mozart loops’, see James Hepokoski and Warren Darcy, Elements of Sonata Theory: Norms, Types, and Deformations in the Late Eighteenth-Century Sonata (Oxford and New York, 2006), 80–6. Hepokoski and Darcy refer to bars 1–5 of K.332 as an ‘initial circular stasis’ and as exemplifying ‘tonic overdetermination’ (p. 92). More generally, they claim that ‘savouring the single- or double-cycle of stasis before proceeding onward is central to any expressive or hermeneutic understanding of the P[rimary] theme’ (p. 91).

46 In closing a unit, a V2 will simultaneously set up a V1 at a higher hierarchical level.

47 Caplin, Classical Form, 63.

48 James Hepokoski, ‘Comments on William E. Caplin's Essay “What Are Formal Functions?”’, Musical Form, Forms and Formenlehre, ed. Bergé, 41–5 (p. 41); James Webster, ‘Comments on William E. Caplin's Essay “What Are Formal Functions?”’, ibid., 46–50 (pp. 47ff.).

49 Extended scholarly considerations of the form of the Allegro of the ‘Prague’ Symphony are found in Charles Rosen, Sonata Forms (New York, 1980), 201–24; Susan McClary, ‘Narratives of Bourgeois Subjectivity in Mozart's “Prague” Symphony’, Understanding Narrative, ed. James Phelan and Peter J. Rabinowitz (Columbus, OH, 1994), 65–98; Elaine R. Sisman, ‘Genre, Gesture and Meaning in Mozart's “Prague” Symphony’, Mozart Studies 2, ed. Cliff Eisen (Oxford, 1997), 27–84; Ratner, Classic Music, 27–8; and Lauri Suurpää, ‘The First-Movement Exposition of Mozart's “Prague” Symphony: Cadences, Form, and Voice-Leading Structure’, Tijdschrift voor Muziektheorie, 11/3 (2006), 164–77. Further insights on the Allegro are also found in Hepokoski and Darcy, Elements of Sonata Theory, 152–5, 162–3. The slow introduction is considered at some length by Sisman, op. cit., 33–47, and Agawu, Playing with Signs, 17–25.

50 Sisman, ‘Genre, Gesture and Meaning in Mozart's “Prague” Symphony’, 57; Hepokoski and Darcy, Elements of Sonata Theory, 378. ‘TR modules’ refer to thematic units in the transition. It is difficult to say exactly where Hepokoski and Darcy might place the beginning of the ‘TR zone’. Bar 55 seems a likely place, as this is where ‘energy gain’ and an increase in dynamic level occur overtly.

51 Hepokoski and Darcy, Elements of Sonata Theory, 47.

52 Ibid., 172.

53 See Suurpää, ‘The First-Movement Exposition of Mozart's “Prague” Symphony’.

54 Meyer's implication-realization model is developed first in Leonard B. Meyer, Emotion and Meaning in Music (Chicago, IL, 1956). It is developed further and applied more systematically in Meyer, Explaining Music: Essays and Explorations (Chicago, IL, 1973). It is important to note that while the theory of musical valency draws on Meyer in positing the V3 with dotted arrow, valency theory is not exhausted by Meyer's work. In particular, valency theory avoids the more prescriptive aspects of Meyer's theory, such as the need to identify continuation types (e.g. linear, triadic and gap-fill) whilst admitting their relevance for certain listening strategies. My emphasis, in the section that follows, upon social and ideological forces in the development of meaning indicates a marked distinction between my position and the absolute expressionist position developed by Meyer.

55 Rosen, Sonata Forms, 221. Rosen, it seems to me, has picked up upon a particularly important aspect of the Allegro of the ‘Prague’ with this statement. However, despite a number of further insights into how and to what end the main theme is varied, he does not provide a clear argument as to what the significance of this theme – elucidated by alteration – might be beyond the very general statement that ‘the motif articulates structure [and] emphasizes the most crucial points’ (ibid., 224).

56 Caplin, Classical Form, 199.

57 Ibid., 198–9.

58 Ibid., 40. Notable also is the point made by Caplin, in personal correspondence with me, that he does not always label ‘relatively compressed cadential functions’. Caplin suggests that his theorizing a continuation phrase as containing a cadential function could be a confusing way of thinking. It strikes me that we simply need to recognize the operation of functions at different levels. This approach is pursued more overtly in Caplin's ‘What Are Formal Functions?’ and taken further, I would suggest, in my theory of musical valencies.

59 Hepokoski and Darcy, Elements of Sonata Theory, 80.

60 Caplin follows a comparable train of thought in his application of the beginning, middle and end functions (or initiating, medial and ending functions) onto what he terms main theme, transition and subordinate theme. See Caplin, ‘What Are Formal Functions?’, 26–8.

61 Hepokoski and Darcy make the important point that Mozart's Piano Sonata in F, K.332, and the Allegro of the ‘Prague’ can both be analysed in terms of the standard structural melodic underpinning 8–♭7–6–♮7–8. There is, of course, some truth in this observation, but the ‘Prague’, I would argue, draws out the potential ambiguity of this pattern by: (1) avoiding the tonic pedal (heard in the opening bars of K.332); (2) avoiding a root-position tonic until the end of the phrase; and (3) repeating the ♭7–6 progression, thereby allocating a higher proportion of bars to the flatward-leaning component of the line. Hepokoski and Darcy, Elements of Sonata Theory, 92.

62 Rosen, Sonata Forms, 221.

63 Another valency analysis might convert the dotted arrow to a solid arrow, as the C♮ in the oboe might be taken as insufficient cause to doubt the cadence in D and, as this turns out to be the case, there is certainly good reason for the solid arrow. However, given the repeated play with expectation around this theme that occurs subsequently, it seems appropriate to indicate at least some possibility of thwarting expectations for a cadence in D. One further objection here might concern the point that repeated listening to musical works can intensify their apparent meaningfulness. My theory might be taken to suggest that knowledge of what, in terms of form, will occur next will undermine the meaning of the music. My response to this point is that repeated listening to the same work inevitably does change our reading. I conceive the ‘Prague’ differently every time I hear it. But, importantly, this does not render later experiences of the work less meaningful even though, so to speak, arrows that were once dotted might become more solid. Greater predictability might reduce the intensity of one's experience in some respects, but it will intensify it in others. Once one can predict musical form it will acquire a valency reading more dominated by V3s, and this will, of course, enrich our experience in certain respects.

64 Rosen, Sonata Form, 221.

65 Rosen's assertion that the augmented chord on A pushes towards the dominant is perhaps derived from the fact that the harmonic device V♯⁵ of IV to IV leading to a PAC is quite common in Mozart (in the ‘Linz’ Symphony, for example, we find this device used in all but the second movement: at bars 67 and 213 of the first movement; bar 93 of the third movement; and bars 400–1 of the last movement). However, the simpler raising of the fifth of a more straightforward chord V is also a common device (see, for example, bar 4 of the second movement of Mozart's ‘Haffner’, K.385). At bar 72 of the Allegro of the ‘Prague’ we have just heard a half-cadence in D; thus the augmented chord, I would suggest, functions in a similar manner to that in the slow movement of the ‘Haffner’. It intensifies a perceived pull towards the resolution chord, which (along with the implied dominant-seventh chord on A that follows in bar 73) brings a greater sense of the opening note of the theme (in this case A) functioning as a dominant rather than a tonic.

66 For a more detailed valency analysis of the development section, an analysis dominated by V3s, see Curry, ‘Reading Conventions, Interpreting Habits’, 335.

67 See Hepokoski and Darcy, Elements of Sonata Theory, 378.

68 This is first argued in one of Peirce's most important essays: ‘Some Consequences of Four Incapacities’. Peirce argues that introspection and intuition are impossible. Instead, mental activity consists of inference from external facts by way of signs. Furthermore, the ‘entire phenomenal manifestation’ is considered a sign and ‘the phenomenal manifestation of a substance’, according to Peirce, ‘is the substance’. It follows, Peirce argues, that woman/man is a sign (ibid., 156, repr. in The Essential Peirce, i, 54). See also the conclusion of this article.

69 Charles S. Peirce, ‘Issue of Pragmaticism’, The Monist, 15 (1905), 481–99 (p. 499), repr. in The Essential Peirce, ii, 346–59 (p. 359).

70 Clearly the choice of late eighteenth-century repertory is significant here. It is my hope that valency analysis may play a role in developing a means of applying form-functional analysis and related approaches to later and earlier musics of the West, as well as music in non-Western contexts.

71 Terry Eagleton, The Ideology of the Aesthetic (Oxford, 1990), 19.

72 See Joseph Kerman, ‘Mozart's Piano Concertos and their Audience’, On Mozart, ed. James M. Morris (Cambridge, 1994), 151–68, and Harold S. Powers, ‘Reading Mozart's Music: Text and Topic, Syntax and Sense’, Current Musicology, 57 (1995), 5–44.

73 Rumph, Mozart and Enlightenment Semiotics, 111.

74 McClary, ‘Narratives of Bourgeois Subjectivity in Mozart's “Prague” Symphony’, 72–3.

75 This central claim highlights an important difference between my theory of musical valency and sonata theory. In sonata theory the focus upon norms, types and deformations is such that any question of a listener's subjectivity is almost completely eclipsed by questions concerning formal patterns and deviations from them. Thus in considering the end of the exposition of the Allegro, Hepokoski and Darcy pay no attention to any significance the main theme may develop for a listener as the exposition nears its completion and the music drives vigorously (unambiguous V3) towards the more confident (tonally relatively unambiguous) declamation of the main theme. Instead they suggest that a reading will (or should) be dominated by questions such as whether the PAC at bar 121 ‘is permitted to function as an EEC, ending S [the secondary theme area]’. Hepokoski and Darcy, Elements of Sonata Theory, 162.

76 Peirce, ‘Some Consequences of Four Incapacities’, 156, repr. in The Essential Peirce, i, 53.

77 Ibid., 54.

78 Ibid.

79 Eagleton, The Ideology of the Aesthetic, 9.