When scholars turn to the question of expression in Rameau's music, we generally do so in the exegetical mode: a particular musical gesture corresponds to a particular textual element, describing it, depicting it or otherwise creating an analogy between music and text. As a theorist, Rameau encourages us in this hermeneutic activity. From the handful of remarks in the Traité de l'harmonie (1722) to the more detailed assertions in his Observations sur notre instinct pour la musique (1754) and on to a chapter devoted to expression in the Code de musique pratique (1760), he encourages readers to regard musical gestures as fundamentally expressive and, more importantly, to regard their expressivity as an attribute of the properly functioning fundamental bass and corps sonore. His arguments are both prescriptive, urging composers to pay attention to texts when setting them to music, and descriptive, relying on a handful of examples to which he returns continually – Lully's ‘Enfin, il est en ma puissance’ from Armide (1686), as well as his own monologue ‘Tristes apprêts’ and chorus ‘Que tout gémisse’ from Castor et Pollux (1737).

Nevertheless, despite the important role expression played in his work, Rameau spent rather less time on an element of considerable interest to his contemporaries, namely, how music acted on listeners to accomplish its expressive feats. This was no small matter, given his routine assertions that the elements of his music theory were perceptible to connoisseurs and ordinary audience members. The present study argues, first, that modern readers can glimpse his beliefs regarding this process by attending to certain scenarios – commentaries on expression, descriptions of listeners, accounts of music-theoretical discoveries – that appear across his writings, especially after the Nouveau systême de musique théorique (1726). These scenarios rely on a type of physical and emotional response that by this time possessed a long literary and philosophical pedigree, rooted in the notion of wonder (admiration).Footnote ¹ Now a number of theories regarding this process were in circulation in France from the early seventeenth century to the end of the eighteenth century, and many of them, such as Boileau's and Montesquieu's, incorporated wonder to some degree.Footnote ² The second part of my argument shows that the experiential model Rameau employed bore a strong structural resemblance to Descartes's as codified in Les passions de l'ame (1649). I do not necessarily regard this as reflecting a programmatic commitment to Cartesian thought, in the manner recently criticized by Benjamin Straehli,Footnote ³ but rather as emanating from common beliefs and practices in the seventeenth century influenced by Descartes and continuing into the eighteenth.

Some provisos are in order. First, what I am tracing here is not an integral part of Rameau's theories; rather, it is an epiphenomenon, something that arose before and during the more pressing task of theorizing music. This means, on the one hand, that we must seek for clues to his beliefs in the margins of his writings, while acknowledging, on the other, that these clues reveal something important to our understanding of his claims and, more generally, of the claims underlying theories of mimetic expression. What we find will enhance, and in some cases confirm, our exegetical studies of his operas. Second, this project requires us to acknowledge that, intellectually, Rameau was a magpie who collected the ideas of others, adopting those elements that supported his theoretical concerns, only to relinquish them when more suitable ideas appeared. Thomas Christensen notes that ‘at one time or another, Rameau cast his theory of the fundamental bass in the varied rhetorics of neoplatonism, Cartesian mechanism, Newtonian gravitation, Lockean sensationalism and Malebranchian occasionalism’, while Erwin Jacobi points to Rameau's continually faltering friendships with younger colleagues – Castel, Briseux, Diderot, d'Alembert – who contributed to his theories but failed fully to embrace his ideas.Footnote ⁴ What this means for the present study is that Rameau's patchwork adoption of new theoretical formulations – say, empirical descriptions of the kind valued by the Encyclopedists – did not necessarily lead him to overhaul his views of how music acted on listeners, any more than it did ideas like the fundamental bass or the corps sonore. His conception of how musical expression functioned remained fairly consistent throughout his career because it suited his purposes well. A survey of some key historical points will help us appreciate how he arrived at his beliefs regarding expression.

By the time Rameau embarked on his career, the question of how external stimuli entered the body to provoke physical, emotional and intellectual responses had long been a matter of interest. Discussions became especially lively around 1600 as thinkers turned away from the cosmologies, sympathies and affinities of Renaissance thought to study the tangible relationships linking beholders with actual perceived objects:

This reorientation of method towards the experiencing subject has important consequences for the present study. The beholder now existed simultaneously as the model for a new kind of observing subject – the inquiring philosopher par excellence – but also as the subject of phenomenological inquiry, while the objects studied were treated as both fully autonomous and yet caught within the toils of scientific inquiry. As illustrated in Descartes's famous dictum, cogito ergo sum, researchers studied the world by studying their own encounters with it. Rameau would rely on this orientation through the course of his theoretical career, and its ramifications are worth considering briefly.

First among these was a new emphasis on sensory experience, and Descartes's first philosophical work, the Compendium musicae (1618), already exemplifies this trend with respect to music.Footnote ⁶ Written privately for his friend and mentor Isaac Beeckman, the Compendium bears little resemblance to the philosopher's more famous treatises, which began appearing in the late 1630s, and yet Jairo Moreno notes that his characterization of sound as an object of perception – as opposed to, say, a manifestation of musica universalis – reveals a new kind of materiality in music discourse: ‘Uncompromisingly programmatic, the statement not only proclaims the centrality of the external, physical and empirical manifestation of sound to his inquiry about music, but also, albeit indirectly, makes of perception the fundamental issue about knowledge of music’.Footnote ⁷ In a break with earlier writers like Zarlino, the listening subject is now an active receptor who hears, measures and orders sounds in the act of perceiving them.Footnote ⁸ Kate van Orden similarly stresses the physical responses of the beholder, noting that Descartes's anecdotes and references in the Compendium dwell on moments where sound is heard and processed: the percussiveness of cannon fire, the pleasant sensations accompanying certain musical gestures, the repetitions built into both Jesuit instruction and the drilling of soldiers.Footnote ⁹ And the ordinariness of these experiences, their basis in mundane experience, is precisely the point – hearing, repeating, understanding and recalling determine the course of critical reflection and ultimately indicate which musical properties are of interest, without recourse to specialized forms of intellection.Footnote ¹⁰ Descartes proposes that the sensual pleasure of listening to music derives from distinguishing among its simplest parts and comparing them, much as his admiration for a well-made astrolabe responds to its housing, assembly and decoration, apart from its ability to perform complex calculations.Footnote ¹¹

What matters for my argument, then, is not that Descartes established in the Compendium a lasting model for interactions between mind and body – his later works are more important in this respect – but rather that he exemplifies a view wherein sound acts physically on listeners: this would remain commonplace well into the eighteenth century. We observe something like it, for example, in the Traité du beau (1715) of the Swiss philosopher Jean-Pierre de Crousaz, who begins his discussion of musical beauty by first explaining the propagation of sound, its transmission through air and its subsequent capture within the anatomy of the ear.Footnote ¹² The trajectory of the experience is crucial; sound's physical properties lead one to consider the emotional responses it evokes. While differing in details, Rameau effects a similar orientation of sound and subject whenever he reminds readers that even inexperienced listeners respond to the fundamental bass and corps sonore. As David Cohen observes, Rameau's listeners instinctively understand (sous-entendu) the theoretical phenomena he describes because these are attributes of sound as it exists in nature: ‘The selfsame Nature that operates objectively in the corps sonore as a physical phenomenon also appears within the human subject as the faculty correlative to that phenomenon’.Footnote ¹³ Further, in a society that valued social decorum, the effects of sound on listeners carried significant moral implications. When Bossuet described how opera corrupts audiences, he employed essentially the same phenomenology that Descartes had used:

Arnauld, Boileau, and Saint-Évremond all provided similarly phrased commentaries.Footnote ¹⁵

We should note, too, how this epistemological framework prioritizes the responses of the individual; as Pascal Dumont puts it, once musical experiences are decoupled from the music of the spheres, they become taste-based judgments rooted in subjectivity.Footnote ¹⁶ This has historical consequences. Descartes may present himself in the Meditationes de prima philosophia (1641) as an idealized version of the inquiring subject, writing in comfortable seclusion before a fireplace, but in social practice his isolation quickly gives way to a host of autonomous subjects recording discrete experiences. Thus in his correspondence with Mersenne, when he rejects the notion that certain intervals evoke similar responses in all listeners, he is making the case for his radical subjectivity:

Similarly, when Crousaz introduces music as a topic in the Traité, it is precisely this attribute that fascinates him:

Inevitably, the question of personal taste becomes a major theme in seventeenth-century literature. It is a vast topic; we can for the present observe its influence on popular entretiens literature, where dialogues carefully stage the opinions of multiple characters. Well-known examples include Honoré d'Urfé’s L'Astrée (1607–1627), Madeleine de Scudéry's Artamène ou le grand Cyrus (1648–1653), Fontenelle's Entretiens sur la pluralité des mondes (1686) and Charles Perrault's Parallèles des anciens et des modernes (1688); in music, Jean-Laurent Le Cerf de la Viéville's Comparaison de la musique italienne et de la musique française (1704) provides an obvious example. I will argue that Rameau resists predicting the effect certain musical gestures have on listeners for effectively the same reasons. He may believe, to be sure, in the expressive force of isolated musical gestures, but he nevertheless does not make detailed claims for how they will dispose listeners to respond.

Finally, we should note that later writers could easily adapt this discourse of experience, body and mind to Locke's claim that nothing exists in the intellect without first existing in the senses.Footnote ¹⁹ Crousaz again provides a useful example. Trained in the rational philosophy of Descartes, Malebranche and Arnauld, his popular treatise on logic, first published as Systeme de reflexions (1712), nevertheless acknowledges the influence of Locke two years prior to the publication of the Traité du beau (1715).Footnote ²⁰ In passages like the following, it is difficult not to see old and new arguments rubbing shoulders:

On one hand, the sensory stimuli acting on the subject resemble the kinds of experience we have seen writers describing for the past century; on the other, however, we observe subject formation transpiring in terms of Locke's corpuscular mechanics, expressed in the characteristic notion of impulse. Later generations of French thinkers were quick to criticize the ease with which Crousaz shifted between Cartesian and Lockean models of experience.Footnote ²²

The stronger French-language adaptation of Locke's arguments occurred seven years later, in Dubos's Reflexions critiques sur la poesie et sur la peinture (1719). For Dubos, who had met Locke, the nature of sensory experience, its relation to the body and its influence over moral character might well exist, but they were irrelevant for understanding responses to artworks:

The rejection of geometry makes clear Dubos's rejection of Cartesian proof, but his argument is more potent than that, because he is also rejecting the role of a priori reason implicit in the cogito. What matters is what the individual feels, and feelings, for better and worse, arise through mimetic identification with the object portrayed:

Dubos's Reflexions was reprinted throughout the eighteenth century, and we find his arguments for mimetic art and emphasis on fully interiorized responses to stimuli echoed, for example, in Charles Batteux's equally popular Les Beaux art reduits a un même principe (1747). As Cynthia Verba has observed, mimesis and, to some extent, Cartesian reason still played significant roles in later thought, especially in the earliest efforts of the Encyclopedists, but the overarching trend was towards ever more extreme forms of Dubos's position.Footnote ²⁵ One notices this already in d'Alembert's Discours préliminaire to the Encyclopédie (1751), where he rejects Cartesian a priori knowledge in favour of the senses:

For Diderot, writing in the Lettre sur les sourds et muets (1751), this is what makes music the most expressive of the arts:

The move away from rationally conceived exterior processes is complete. Stripped of language altogether, music shakes and jolts (secousses) the senses to elicit responses from the beholder.

What this summary allows us to see is something that both Christensen and Cohen have pointed out, namely, that the intellectual basis for Rameau's theories was largely conservative.Footnote ²⁸ But it also shows us that he was not alone in this. D'Alembert, again in the Discours préliminaire, acknowledges Descartes as a model for philosophes, but also acknowledges the persistence of a philosophical method he regards as outmoded: ‘Let us always respect Descartes, but let us readily abandon opinions that he himself would have combatted a century later’ (‘Respectons toûjours Descartes; mais abandonnons sans peine des opinions qu'il eût combattues lui-même un siecle plus tard’).Footnote ²⁹ By emphasizing rational deduction over physical sensation, Rameau's theories remained rooted in the past, closer to Crousaz and his seventeenth-century predecessors than to the composer's immediate contemporaries.Footnote ³⁰ What concerns us here, then, are the assumptions on which Rameau bases his discussions of expression, and fortunately we have some evidence for what he believed.

Rameau's beliefs about how music acted on the body sound disappointingly simple, but this is because, in crucial ways, he believed it was a simple process. In the Traité de l'harmonie (1722), he wrote:

His purpose here was to praise the fundamental bass, and he intended his remarks about expression and sentiments to support this point. More often, however, the passions sat in uneasy balance with Rameau's theoretical goals. Thus in the Generation harmonique (1737) he wrote that ‘music, like other aspects of mathematics, embraces the knowledge of relationships, but it can also, like eloquence and poetry, boast of exciting and calming the passions’ (‘[La musique] embrasse, comme les autres parties des Mathématiques, la connoissance des rapports, mais qu'elle peut encore se vanter, aussi bien que l'Eloquence & la Poësie, d'exciter & de calmer à son gré les passions’).Footnote ³² Though mentioned in the same breath as theory, expression remained discrete from it. Elsewhere, Rameau seemed concerned with theoretical problems that were ontologically prior to expression: ‘music is a physical and mathematical science – sound its physical object, and the relationships between different sounds its mathematical object; its goal is to please and excite in us diverse passions’ (‘La Musique est une Science Phisico-mathématique, le Son en est l'objet Phisique, & les rapports trouvés entre différens Sons en font l'objet Mathématique; sa fin est de plaire, & d'exciter en nous diverses passions’).Footnote ³³ Expression in this instance transpires at an experiential stage consequent to his theoretical concerns, effectively absolving him from explaining the relationship between music and passion.

When Rameau did acknowledge expression theoretically, it was usually in the context of modulation, by which he meant harmonic progressions in general, whether or not they remained in a single key.Footnote ³⁴ Thus in the Traité (1722) he drew a correspondence between modulation and text:

Throughout his career, Rameau bound musical design to subtleties in the poetic message, and in this sense, text determined the course of the musical composition in an arguably expressive way: shifts in the music's unfolding coincided with shifts in the text's unfolding. But this also meant that music transmitted little on its own; the listener knew what the music expressed because she heard the words to which it was set. In the Nouveau systême de musique théorique (1726), Rameau returned to the infinite diversity celebrated in the Traité, illustrating harmony's expressive potential with eighty harmonizations of an ascending fourth. Crucially, however, he didn't tell readers what each harmonization expressed; his point, rather, was diversity itself, the surfeit of expressive possibilities generated by his system: ‘if it is not absolutely impossible to determine the melodies and modulations . . . that best agree with the most marked expressions, it is nevertheless an enterprise that would demand most of the lifetime of a single individual’ (‘S'il n'est pas absolument impossible de déterminer les Chants, & les Modulations en consequence, qui conviendroient le mieux aux expressions les plus marquées, c'est d'ailleurs une entreprise qui demanderoit peut-être plus que la vie d'un seul homme’).Footnote ³⁶ Music thus brims with expressive potential derived from its simplest elements, but for this very reason one cannot specify the connotations of each harmony.Footnote ³⁷ One can almost hear Descartes's response to Mersenne in this passage; music is expressive, but its expressions are indeterminate with respect to individual listeners. Like Descartes, Rameau discussed only what he could demonstrate.

As a result, it feels like a breakthrough when we arrive at the Observations sur notre instinct pour la musique (1754). There, goaded by Rousseau's Lettre sur la musique françoise (1753), Rameau analyses Armide's monologue, ‘Enfin il est en ma puissance’, to show that traditional French opera was rife with expressive potential. It is as close as we get in this period to the kind of exegetical commentary discussed previously, focusing in minute detail on the proper musical phrasing of Quinault's poetic text and the correct operation of the fundamental bass. And yet in this, Rameau's most summary statement on expression, he offers little in the way of expressive technique:

Most discourse on music from this period was capable of arriving at some form of this conclusion – that harmonic and melodic direction constituted expressive gestures – and so the passage does not, at first blush, strike modern readers as particularly revealing.Footnote ³⁹ Moreover, Rameau's peculiar numbering of his points indicates that he is, once again, reinforcing a more general theoretical principle, in this instance the primacy of the corps sonore: he first emphasizes harmonic progressions, giving each its own numeration, but then conflates them into a single point after proclaiming melodic activity a secondary attribute of expression. Harmony, to the surprise of no reader familiar with Rameau's writings, generates music's relevant expressive features. But there is something else noteworthy in this passage, namely that results can only be described with reference to each other:

As in Saussurean linguistics, connotation is established by the choices made, but also by those rejected.Footnote ⁴¹ The composer conveys humiliation by moving to the flat side, but also by avoiding the sharp side, ‘if one doesn't wish to torture [listeners], by drawing on an expression contrary to the sentiment to which this flat gives rise’ (‘À moins qu'on ne veuille s'y donner la torture pour en tirer une expression contraire au sentiment que ce Bémole fait naître’).Footnote ⁴²

Two points require emphasis. First, I would draw attention to the manner in which Rameau pairs emotions. I have in the past viewed this as a musician would: tonal music can only move higher or lower, to the sharp side or the flat, and it follows that these gestures promote passions working in opposition to each other. (We have thus far encountered sadness and happiness, humiliation and pride.) I now believe this picture is incomplete. If musical practice provides the composer only with opportunities for depicting passions in contrast, then we should also acknowledge that the tradition of viewing emotions in opposition extends back at least to Aristotle's Rhetoric.Footnote ⁴³ The theoretical discussions of passions Rameau would probably have known, those of Descartes and Malebranche, likewise dealt with passions as closely related opposing pairs. For Descartes, four of the six fundamental emotions, from which all other emotions derived, were pairs reflecting the opposition of good and bad: love and hate, joy and sadness. Malebranche, less concerned with the structure of emotions, made the same point, evoking the same Saussurrean associations as Rameau: ‘we hate only because we love, and the external evil is judged evil only in relation to the good of which it deprives us’ (‘Car en effet on ne hait, que parce que l'on aime, & le mal qui est hors de nous n'est jugé tel, que par rapport au bien, dont il nous prive’).Footnote ⁴⁴ This suggests a different way of thinking about musical expression. It is not simply that music's movements are conveniently analogous to contrasting passions, but rather that musical movement disposes itself in the same way as the principal passions. I will return to this idea.

My second and larger point relates to the first: there is a central organizing Cartesian passion that Rameau also finds a place for in his scheme. Consider this passage from the Generation harmonique (1737) treating the enharmonic genus:

Rameau uses the word admiration here in its older sense of experiencing astonishment at an extraordinary event.Footnote ⁴⁶ The listener in this situation admires neither the composer nor the compositional device, but the experience she undergoes.

This kind of wonder had a rich history. Its roots were traceable to Aristotle's Metaphysics, where it was treated as the basis for philosophical discovery: ‘Now he who wonders and is perplexed feels that he is ignorant . . .; therefore if it was to escape ignorance that men studied philosophy, it is obvious that they pursued science for the sake of knowledge, and not for any practical utility’.Footnote ⁴⁷ For Descartes, who generally associated Aristotle with the Scholastic philosophers he distrusted, wonder was similarly ‘the first of all the passions’, because it occurred ‘before we know whether or not the object is beneficial to us . . . . It is only knowledge of the thing that we wonder at’ (‘Et pour ce que cela peut arriver avant que nous connoissions aucunement si cet objet nous est convenable, ou s'il ne l'est pas, il me semble que l'Admiration est la premiere de toutes les passions . . . . Dont la raison est que, n'ayant pas le bien ny le mal pour objet, mais seulement la connoissance de la chose qu'on admire’).Footnote ⁴⁸ He echoes Aristotle's use of wonder, but adds to it the characteristic seventeenth-century phenomenology described at the outset of this article. His concern is for the manner in which stimuli act on the beholder to produce passionate responses. Malebranche, influenced by Descartes, likewise believed that wonder acted only on the mind and not the heart to produce intellectual experiences:

Through wonder, the strange becomes familiar, the unknown known. Like Aristotle, both philosophers regarded it as valuable because it provided the thrill that made learning worthwhile.Footnote ⁵⁰

Though Descartes's views on the passions connected him with a venerable philosophical tradition, his application of these principles was characteristically original. Indeed, the passions held an important place in Descartes's later thought because they allowed him to introduce what he called ‘substantial union’ into his famously dualistic treatment of mind and body.Footnote ⁵¹ The passions, in other words, illustrated how mind and body worked together in the course of practical experience. In experiencing external events, the body acted on the soul and the soul, in turn, on the body, creating the emotional states and responses necessary for leading a good life:

What distinguished Descartes from earlier commentators was the way in which he used wonder to study other, more complex emotional states. He argued that all emotions derived from six primitive passions (passions primitives): wonder, love, hatred, joy, sadness and desire. Placed first among these and treated at greatest length, wonder was fundamental to understanding how all passions functioned. Even though it acted directly on the mind, it possessed ‘considerable strength because of the element of surprise, namely, the sudden and unexpected arrival of the impression that changes the movement of the spirits’ (‘Ce qui n'empesche pas qu'elle n'ait beaucoup de force, à cause de la surprise, c'est à dire, de l'arrivement subit & inopiné de l'impression qui change le mouvement des exprits’). And this experience of surprise, and through surprise wonder, was fundamental to nearly all other emotions: ‘Such surprise is proper and peculiar to this passion, so that when it is found in the other passions – and it normally occurs in and augments almost all of them – it is because wonder is joined with them’ (‘laquelle surprise est propre & particuliere à cette passion: en sorte que lors qu'elle se rencontre en d'autres, comme elle a coustume de se rencontrer presque en toutes & de les augmenter, c'est que l'admiration est jointe avec elles’).Footnote ⁵³

The enthusiasm Rameau so evidently felt in hearing and contemplating the enharmonic genus was therefore not misplaced. Whether or not he was a close reader of the earlier philosophers – and I believe he possessed at least the popular awareness of Descartes common in this period – he was guided theoretically, by wonder, to consider music intellectually, a scenario he re-enacted throughout his theoretical writings. In the Traité, he mused on the vexing question of why music affected ancient audiences more profoundly than modern audiences: ‘Could it be that the more familiar things become, the less they cause surprise, and that the wonder they arouse when they are new degenerates imperceptibly as we grow accustomed to it, turning ultimately into simple amusement?’ (‘Seroit-ce en disant, que plus les choses deviennent familieres, moins elles causent de surprise; & que l'admiration où elles peuvent nous jetter dans leur origine, dégenere insensiblement à mesure que nous nous y accoûtumons, & se tourne à la fin en un simple amusement?’).Footnote ⁵⁴ We know that Rameau's audiences, too, initially wondered at his music, appreciating it only after repeated hearings.Footnote ⁵⁵

Similarly, in the Nouvelles réflexions sur le principe sonore (1758–1759), he imagined a researcher studying sound propagation by means of a monochord and experiencing wonder on realizing that the proportions of 1/2 and 1/4 were just as important as those of 1/3 and 1/5. (At issue here was the difficulty of perceiving the first octave of the overtone series.) Rameau had described a similar personal experience in the Démonstration du principe de l'harmonie (1750), narrating his own discovery of the overtone series in terms reminiscent of Cartesian self-examination. Having placed himself in an artificial state of musical innocence, he reacted immediately: ‘My search was not long. The first sound that struck my ear was a beam of light. I suddenly perceived that it was not [a single sound], but that the impression it made on me was compound’ (‘Ma recherche ne fut pas longue. Le premier son qui frappa mon oreille fut un trait de lumiere. Je m'apperçus tout d'un coup qu'il n’étoit pas un, ou que l'impression qu'il faisoit sur moi étoit composée; voilà, me dis-je sur le champ, la différence du bruit & du son’).Footnote ⁵⁶ The language here – with its image of a burst of light – sounds remarkably like his description of hearing the enharmonic genus.Footnote ⁵⁷ Once one is attuned to the rhetoric of wonder, its presence can be observed throughout Rameau's writings, as for example when he uses words like choc and surprise to describe hearing a cadence rompue.Footnote ⁵⁸ Rameau wanted readers to appreciate his own wonder in studying music, but he also expected listeners to feel a similar constitutive wonder in hearing music's smallest and grandest gestures.

Equally interesting in Rameau's discussion of the enharmonic genus is the way he related it to modulation. Harmonic motion runs parallel to the physical sensations induced in the listener: ‘surprise soon turns to wonder at seeing oneself . . . transported from one hemisphere to the other’. In perceiving the transport of harmony, the listener too feels transported, and this strikes me as useful for understanding how music instils more complex passions.Footnote ⁵⁹ A cadence rompue is scarcely as striking as the trio des parques from Hippolyte et Aricie, but if wonder is an adequate response to modulation in general, then the difference between the two is one of degree rather than kind. As in Descartes, the experience of wonder permeates all aspects of research and discovery.Footnote ⁶⁰ It is central to understanding how Rameau believes music moves the passions.

Moving the passions: the commonplace of an external force practised on the body to inspire emotion winds through Rameau's treatises like a well-worn adage. In the Traité, for example, he writes: ‘It is certain that harmony moves different passions in us according to the harmonies employed’ (‘Il est certain que l'Harmonie peut émouvoir en nous, differentes passions, à proportion des Accords qu'on y employe’).Footnote ⁶¹ Nevertheless, he offers little more there than general rules to explain how this happens – consonant chords suggest magnificence, chromaticism suffering – confessing once more that exhaustive description is beyond his reach. Elsewhere in the Traité, when writing about measure and rhythm, Rameau intends movement literally. Concerning dance types, he writes: ‘each character and each passion have their peculiar movement, though this depends on taste more than rules’ (‘Chaque caractere & chaque passion ont leur mouvment particulier; mais cela dépend plus du goût que des Regles’).Footnote ⁶² He returns to this idea in his Preface to the Observations. There he begins by attributing the moving of passions to harmony alone, stating that expressive melodic motion is a secondary effect derived from it. (Above, we encountered his argument in support of this proposition, from later in the treatise.) He then dismisses purely imitative movement as characteristic of Italian music:

It is a deft move, aligning his theoretical tenets with France's long-standing anxieties over its music, but we should also recall that for Rameau, as noted above, harmony itself, qua modulation, entailed a form of movement in its harmonic successions. In the Code, he described harmonic motion as engendering emotional responses by moving and altering the listener's disposition. Presumably, at this stage of Rameau's argument, these responses could be explained in a Cartesian manner by virtue of the animal spirits, as a form of mimesis more consistent with Dubos's thought or simply as sous-entendu. He chose not to dwell on such details.

What was important to Rameau was that music act on the listener in a particular manner. Explaining his theoretical principles to a readership of non-musicians in the relatively late Code, he took the unprecedented step of devoting a chapter to expression, in which he described listeners as ‘passive harmonic bodies’ (corps passivement harmonique) naturally influenced by the corps sonore.Footnote ⁶⁴ Alone, the corps sonore did nothing to affect a listener's emotional state; she felt no surprise at hearing bass movement by a fifth beneath an unchanging harmony, so long as the fundamental bass remained the same.Footnote ⁶⁵ Inexperienced listeners grasped melodies performed on a horn or trumpet in precisely this way. The situation changes, however, when Rameau introduces harmonic motion away from the initial overtone series: ‘Let us be no more astonished if we pass to [a harmony] that sounds false [on a natural horn or trumpet], the ear being surprised, and surprised in various ways, depending on how closely related the [chord] successions are’ (‘Ne soyons donc plus étonnés que si l'on passe à l'un de ceux qui s'y trouvent faux, l'oreille n'en soit surprise, & surprise de différentes façons, selon qu'il y aura plus ou moins de rapport dans cette succession. Sans doute que la Nature agit sur nous de même que sur ces corps’).Footnote ⁶⁶ This passage is worth reading closely. As we know, Rameau already considers it astonishing that a horn or trumpet should function according to the overtone series. Departures from that initial series are now equally astonishing because they surprise the ear, and the more abrupt the departure, the more surprising they are. The wonder here is Rameau's, but it encompasses musical effects he has described elsewhere with the same sense of wonder: the mysteries of the overtone series, the harmonic motion of lowly cadences rompues and extravagant enharmonic progressions.

To summarize, I think musical expression operated for Rameau much as wonder operated for Descartes. Descartes believed that primitive emotional states – happiness, sadness; love, hate; desire – derived from an initial experience of wonder that incited an intellectual, rather than physical, response. Physical responses followed instantaneously as the subject contemplated the object of wonder, liking or disliking it. For Rameau, the source of musical wonder in performance was modulation. Theoretically, this wonder derived from any harmonic succession the listener hadn't anticipated, but quickly gave way to relatively primitive emotional states similarly arranged in pairings of good or bad; anything beyond these primitive emotions, however, could not be predicted solely on the basis of musical content.Footnote ⁶⁷ In the same way that Descartes's wonder gave way to happiness and happiness to subtler inflections like pride, so too Rameau's wonder gave way to positive or negative emotions, revealed first by the harmonic progression and then, more subtly, by its attendant poetic text.

Consider, for example, how his music for the transformation scene in Pigmalion (1748) conveys wonder. Thomas Christensen alerts us to Rameau's symbolic scoring of the overtone series at this moment – as though the principle animating Rameau's theoretical system is likewise animating the sculptor's inert statue – but he also stresses that the modulation leading up to this moment is itself noteworthy: the preceding harmony, G major, shifts abruptly to E major for the statement of the overtone series.Footnote ⁶⁸ In Rameau's epistemology of wonder, this modulation would have been as remarkable as the overtone series itself, because the listener does not anticipate this shift in the fundamental bass. In terms of musical expression, the logic would run as follows. The character Pigmalion hears the chord of nature and expresses wonder:

Similarly, the audience member experiences wonder mimetically at hearing the fundamental bass descend by a minor third and, aided by Pigmalion's words, feels something akin to the character's emotional state. It is telling that the sculptor's expression of wonder corresponds not to the transformation, but to harmony and light, indices of wonder in Rameau's theoretical writings. The statue becomes animated only in the moment of silence immediately following.

A similar case can be linked even more closely to Rameau's description of enharmonic wonder in the Generation harmonique, namely the famous trio des parques, ‘Quelle soudaine horreur’, from Hippolyte et Aricie (1733). Downing Thomas observes that what concerned Rameau was the effect of the enharmonic modulations on the listener, and he argues plausibly that they should be understood physically, even viscerally, as expressing a violent and disturbing passion, horreur.Footnote ⁷⁰ Given what Rameau says about the enharmonic genus and wonder, however, Thomas's case could be extended to understand the passage in terms of wonder as well. Pluton warns Thesée that the parques foresee his dreadful future, and in the moment of their utterance, Thesée might have experienced his emotions in a Cartesian manner – an initial stroke of wonder which, with dawning realisation, turns into a more nuanced form of revulsion. In this same way, Rameau's listeners hear the scene from a vantage point analogous to Thesée's: they first experience wonder on hearing the music of the parques, which quickly turns to horreur at the utter strangeness of the enharmonic modulations.

What I'm proposing, then, serves as a useful complement to modern hermeneutic approaches to analysing Rameau's music, because these tend to draw on Rameau's harmonic practices.Footnote ⁷¹ Verba's recent work may be taken as an example. Undertaking a fundamentally exegetical project – ‘focusing on [Rameau's] concept of musical expression and examining how and where this concept resides in the dramatic settings of his musical tragedies’ – she stresses the manner in which the composer used surprising harmonic progressions to disrupt tonally stable passages.Footnote ⁷² Writing of the trio des parques, for example, she observes:

Verba's argument here strikes me as coextensive with Rameau's in the Code cited above: a surprising shift in the fundamental bass induces wonder in the listener; indeed, it must induce wonder in the listener to create the response described. The listener, immersed in the musical performance, responds to distinctive passages without necessarily grasping the harmonic progressions or their technical basis.Footnote ⁷⁴

Wonder, then, is less a skeleton key for unlocking Rameau's secrets than an informative detail allowing us to appreciate how he organized his beliefs about musical expression. He treated it in much the way Descartes did and for some of the same reasons, to describe a primal cognitive experience leading to more complex reactions. This need not constitute a material appropriation of Descartes, though it is interesting to note that an edition of Les Passions de l'ame was published in Paris, along with La Géométrie, the year Rameau published the Nouveau systême.Footnote ⁷⁵ Wonder, rather, represents one set of beliefs commonly held during Rameau's lifetime.

At the same time, Rameau's dependence on wonder signals the old-fashioned strand we often observe in his thought, because, as we have seen, for much of the century, discourse on the passions had moved towards an affirmation of sensory experience. Dubos's Reflexions critiques sur la poesie et sur la peinture, for example, showed little interest in wonder or opposing categories of passions; the word sentiment, with its overt sensory connotations, was as prominent as the more Cartesian passion, and to the extent that wonder was present, it appeared in references to works by others: a description of a figure in Raphael's St Paul Preaching in Athens (1515), an emotion underlying a description by Lucian, a passage from Montaigne.Footnote ⁷⁶ Dubos, like the Encyclopedists who followed, was less interested in describing a phenomenology of experience than in validating the sensorium as a basis for judgment:

What made his position radical was its redistribution of emphasis. Anyone undergoing an aesthetic experience would have an emotional response – a point Descartes would readily concede – but Dubos attributed little practical importance to knowing how those responses came into being or inhabited the intellect. They simply did; they were, first and foremost, corporeal responses to corporeal experiences. Closer to Rameau's time, Pierre Estève echoed Dubos's point that aesthetic experiences were measured by the senses. When wonder appeared in his L'Esprit des beaux arts (1753), he dismissed it as a purely intellectual exercise:

This mediation of ancient extremes by modern European culture was important to Estève's argument, allowing him to pass over intellectual forms of aesthetic experience however they originated. He carried this characterization of wonder, qua intellectual experience, over into his discussion of music:

By this line of reasoning music appealed to the senses and, in this respect, functioned much like the other arts. By mid-century, at least, Rameau's Cartesian wonder may have struck some readers as a bit dusty.Footnote ⁸⁰

The question, then, is why Rameau would persist in this particular conception of wonder. On other occasions, as Christensen has pointed out, when intellectual fashions changed Rameau readily substituted Lockean descriptions of harmonic motion for Cartesian ones. And it seems likely that by mid-century Rameau knew and understood the implications of Dubos's and Estève's work: Dubos's Réflexions had been widely distributed, with five editions appearing in France before 1750, while Estève was one of the better-known critics of Rameau's music theory, challenging him in 1751 on the identity of the octave, the possibility of transferring intervals from one octave to another and the characterization of the major triad as occurring in nature.Footnote ⁸¹ (A physicist and mathematician himself, Estève may have had Rameau's theoretical writings in mind when dismissing the importance of intellection for musical experience.) Moreover, Rameau had worked with Diderot on the Démonstration du principe de l'harmonie (1750) in the late 1740s, so he should have been familiar with emerging theories of sensibility.Footnote ⁸²

Several factors may have influenced Rameau's persistence in conceiving the passions similarly to Descartes. First, Dubos's, Estève's and later epistemologies were poorly suited to his needs precisely because they focused on the sensory experiences of the beholder, failing to stress the role of the artwork qua artefact in producing emotional responses. This isn't to say that someone like Dubos avoided matters of technique. In discussing theatre, he treated proper declamation and the selection of appropriately moving stories; in poetry, the selection and ordering of words; in painting, the assemblage of component parts into a compelling whole; and in opera, the depiction of natural events in instrumental music. But he regarded the purpose of technique solely as one of moving the beholder. The successful artist applied technical mastery to this end, but the relative merits of individual painters or techniques were irrelevant: ‘the greatest painter is the one whose works afford us the most pleasure’ (‘Le plus grand Peintre pour nous, est celui dont les ouvrages nous font le plus de plaisir’).Footnote ⁸³ There is little distance here between Dubos's position and that of d'Alembert or the later Diderot. Rameau's purpose, to the contrary, was to describe a rationally ordered musical system shaped by natural laws, the technical details of which were audible, generalizable and iterable for anyone who composed or listened to music. The effects of music on the beholder were relevant to his arguments, but not a principal concern because they occurred at a later experiential stage, signalling the successful application of the musical principles he advocated. The closest Rameau came to addressing contemporary theories of musical affect was in the Observations, where he treated music's effects on the listener as a matter of ‘instinct’ – a clear nod to emergent theories of sensibilité – but as we have seen, even here he needed to treat experiences not as something existing solely within the beholder, but rather as the products of natural laws rooted in harmony.Footnote ⁸⁴ For Rameau, what was important was that composers and listeners entered into intentional relationships with actual musical objects, the composer writing to elicit responses appropriate to the dramatic situation, the listening subject judging the adequacy of the composition to its task. This was a Cartesian scenario. Wonder grounded it epistemologically in a way that sensibilité did not.

Another advantage to avoiding the new sensibilité involved music criticism: simply put, responses to music drawing on opinions first formed in the seventeenth century persisted throughout Rameau's career. Indeed, negative reviews of Rameau's operas typically reflected these views in some manner. For example, an anonymous allegory from 1734, almost certainly aimed at Rameau's first opera, Hippolyte et Aricie, invoked a remarkably old-fashioned suspicion of musical pleasure:

When music impressed, it also seduced. Where Estève would reject intellectual exercises in music as distracting from sensory experience, this author regarded them instead as leading the listener into the worst kinds of sensuality. Rameau encountered this kind of criticism continuously, both as theorist and composer, and his response was inevitably the same, to argue that all music was subject to laws observable in nature and was thus admissible to personal enjoyment. Beginning with his second treatise, the Nouveau systême, he became increasingly dependent on perceived geometrical relationships as evidence of these natural laws.

This placed Rameau in an interesting position: we know that his music shocked audiences, an experience easily subsumed under the notion of Cartesian wonder, but we are also familiar with his protestations that his theories and compositions were simple and straightforward because they were rooted in nature. In light of the theorization of wonder undertaken here, those protests can sound disingenuous. His effects were rooted in nature as he perceived it, or at least in a system he had created by observing nature; nevertheless, with each surprising harmonic progression, he must have imagined thrills moving through the beholder's body, knowing full well that some audience members would resist them. We can go further. Rameau typically limited his theoretical observations to harmony, but considering contemporary criticism, we should note that every complex chorus, every richly orchestrated dance induced similar experiences. Ultimately, I believe this has less to do with disingenuousness on Rameau's part than with his subtle rearrangement of the musical experience. If every listener is, potentially, a corps sonore, every listener a passive body perceiving and receiving the corps sonore, then wonder organizes the physiology of the listening experience in a particular manner. It is transmitted by the senses, but in the tradition of Descartes and Malebranche, it ultimately addresses the intellect in a manner that treats the senses as means to an end. This, I think, was important to Rameau, because it permitted him to treat music as a serious topic for discussion, divorced from questions of morality and whims of taste. Where his critics, conservative and progressive alike, assigned musical experiences to the body, a treat for the senses alone, Rameau heard musical effects great and small and experienced wonder, an intellectual emotion.