We know from Rousseau's Discourse on the Origin and the Foundations of Inequality among Men that the question of origin of political society is finally the same as the question of the origin of reason in human beings. Reason, in turn, is inseparable from language; both are signs of our fall from our natural state into civil life. Accordingly, with the hope of unearthing the cause of this fall, we turn to Rousseau's Essay on the Origin of Languages, a book that bears the curious subtitle “where melody and musical imitation are spoken of.” Music becomes an explicit issue in the last third of the Essay (chapters 12–20), but why? Our first question then is why the origin of political life necessarily leads us to music.Footnote ¹

It is a regular feature of Rousseau's writing that he presents what seem to be logical relations as temporal movements. The Second Discourse contains multiple examples of animals that fall just short of being human in the state of nature (consider, e.g., the pongo, orangutan, quojas morros, beggo, and mandrill of note j) and multiple examples of human beings who live in a state that falls just short of natural (consider the Hottentot of note j and the Carib of paragraph 63), but there is not a single example of a human being in this presumably natural state. Because the pongo lacks speech, we are in no position to say whether it has that capacity to acquire speech characteristic of natural man. The pongo is natural, but it is not a man. Rousseau seems at first to mean to describe the Carib as having a “soul that, agitated by nothing, delivers itself to the sole sentiment of its own present existence without any idea of the future,” but he cannot mean what he seems to mean, for he goes on to say that the Carib “sells his cotton bed in the morning and, weeping, comes in the evening to buy it back, failing to have foreseen that he would have need of it for the next night.” In order to show that the Carib lacks foresight, Rousseau endows him with language and a notion of monetary gain, both requiring foresight and so neither of which can be present in natural man; the Carib is a man, but he is not natural.Footnote ² For this and other reasons, over time one is gradually led to conclude that the state of nature is rather a logical foundation of our understanding of human beings than their temporal origin.Footnote ³On the Social Contract is supposed to provide us with an account of the original agreement that stands at the beginning of political life, and yet every attempt to get at such an agreement seems to presuppose an already prior agreement—assembler and unir are always rassembler and reunir, idiomatically synonyms that betray the deep problem involved in giving any account of what comes first.Footnote ⁴ And, while The Reveries of the Solitary Walker seems at first to provide examples of reverie as that state in which we timelessly experience the sweet sentiment of our own existence, here too each such example is either a leading up to or a falling away from the perfect contentment of such a state.Footnote ⁵ We are initially led to believe that Emile will be an account of the imaginary education of a child over time. Rousseau adopts a baby at birth and brings him to manhood in such a way that he will avoid the amour-propre that ordinarily causes us to live outside ourselves. Emile will be educated “according to nature” so that his present is not sacrificed for his future. To the greatest extent possible, his desires and powers will be in harmony. On the surface, the first serious threat to this equilibrium is the onset of sexuality in adolescence. Emile, therefore, appears to be an account of the fundamental transformation of human nature over time. On further examination, however, we discover the seeds of alienation already present in infancy.Footnote ⁶ Our second question, then, is why Rousseau repeatedly begins by presenting as though they unfolded in time relations that in the end must be understood as logical.

In a way that seems at first altogether unconnected to these questions, the two issues—time and music—are brought together in Rousseau's account of the relation between melody and harmony in the Essay on the Origin of Languages. Melody is more fundamental than harmony; in particular, it distinguishes itself from harmony by unfolding in time. Is it possible, then, that this primacy of melody is connected first to the nature of language, and thereafter, by way of language, to the nature of thinking? And by first understanding the essentially melodious character of human thinking might we thereafter also understand why Rousseau finds it necessary to unfold the logical as though it were temporal?

Rousseau begins his account of music with a distinction.

The difference between articulation and sound is grounded in a split within passion. Anger breaks the sound continuum into discrete parts; tenderness alters the continuum in another way—pitch. Yet the two are no more really separable from one another than are articulation and communication, or the languages of the north from those of the south.Footnote ⁸ All speech must have a tempo and a pitch. Accordingly, not only were “to say and to sing … at another time the same thing” (12.2), in some measure, they must always be the same.

If language implies thought and thought language, then Rousseau suggests here that thinking must be poetic. All language is born of passion, even if the two aspects of language—articulation and communication—issue from different passions. This distinction between articulation born of anger and communication born of tenderness gives birth to another within music between melody and harmony.

Now, it is certainly true that Rousseau is serious about his understanding of melody as more fundamental than harmony and about his criticism of the music—especially the French music—of his time as mistakenly privileging harmony over melody. He repeats this criticism elsewhere at length.Footnote ⁹ And it is also true that Rousseau believes it to be a modern mistake not to pay attention to the power of music as morally imitative.

The argument continues by way of an analogy to painting—more specifically a comparison of drawing to melody. While sounds by themselves may please us as do colors, melody is like drawing (le dessein may also mean purpose or intention—design) in moving us far more powerfully. Drawing “is the imitation that gives these colors life and soul” (13.1). Rousseau informed us in chapter 2 of the Essay that the origin of language is not need but “moral need”—a need that places us in relation to another as a subject and not simply as object (2.3). That we take sensations to be “signs or images” means that we take them to point beyond themselves—to have a design, intent, or purpose. Drawing does this in a way that color by itself does not (although colors do seem to present themselves always as having shapes). Individual colors may please us, but the pleasure they offer is altogether static, fixed, and isolated—it is, so to speak, objective. In putting what is colored in a context, drawing connects one color to another by way of dessein—it makes a poetic image.

But what exactly does this mean? In describing the origin of language, Rousseau says that we do not first sense things in their true form; we sense them through sentiment or passion. Awakened in the night by enemies and frightened, we see them as bigger and stronger than they are. We mistakenly call them giants. It is a natural metaphor. We see not merely objects but also our own fear. Original language thus gave us fearsome “giants” rather than neutral “men” (3.3). Similarly, drawing gives us not unconnected objects in a world but rather these objects suffused with our powerful reactions to them. Melody has this character as well. It may at first seem simply a “succession of sounds” (13.2), but it yields much more than the sum of its parts, for one note sounded in a melodic sequence makes one anticipate the next; it draws us on—makes us want what will follow. Its power is inexplicable apart from the desire that it engenders in us. Once placed in time, sounds thus generate in us something of “the emotion that beautiful paintings [or scenes, descriptions: tableaux] cause in us and of the charm of being moved before a suffering [pathétique] subject” (13.3). Of course, these scenes may, and frequently do, contain human subjects, but they may also simply be drawings of recognizable objects. Rousseau means to indicate that in presenting us with images of things in our world, things important to us, les tableaux imitate not only the world, but tacitly ourselves, too, as subjects in the world. As Nietzsche (following Schopenhauer) will later say, music as “an immediate representation [or image: Abbild] of the will itself” lets “every painting, indeed every scene of real life and of the world, straight away come forth with higher meaning, to be sure even more so insofar as its melody is analogous to the inner spirit of the given appearance,” displaying “the metaphysical—the thing itself.”Footnote ¹⁰

With typical wit, and disdain, Rousseau opposes melody understood in this way to harmony.

Rousseau understands harmony as originating in the following way. As wrenched from a living being as a cry or a plaint, sound has a mood attached to it. This mood is expressed by a melody—a tonal continuum or fluid movement—which is then made calculable and codified into a series of discrete notes. These notes, when taken together, imply a scale—a series of intervals that as deriving from a melody that pleases us “fit together” (the original meaning of the Greek verb harmozein) or harmonize. As fitted to one another, these notes seem always to go together. So for example, perhaps we notice that the discrete notes we have designated from the tonal continuum at first number eight before their pitch repeats. Further examination shows us two pairs of notes that seem closer to one another than the others. To reflect this closeness, we divide what we now call the octave into twelve intervals and notice that between each note and the next, with two important exceptions, there are two intervals. Between the third and the fourth notes of our scale and between the seventh and the eighth, there is only one interval. The shape taken by this fixed relation among the notes of our initial melody is a mode (in this case what we call the major mode). As fixed it becomes a compositional tool for us; we invent or notice other melodies in the same mode, and by so doing, tacitly exclude other possible notes (the scale does not have to be divided into twelve intervals) and other possible relations (the minor mode has single intervals between the second and third and fifth and sixth notes in its sequence). We then see that the notes fitting into our mode need not occur in a temporal sequence but can be sounded simultaneously in certain combinations—perhaps thirds or fifths. These are chords. To sound them, of course, we must abandon our first instrument—the apparently univocal human voice from which the initial melody was wrenched—and so also sever melody from the language of its initial expression. This atemporal combination of sounds in a chord, that is, harmony, thus sanitizes melody by cleansing it of the expression of desire that originally gave rise to it. Because the musical expression with which we began is no longer intimately connected to the poetic expression in language in which it was born, we must add lyrics to make song, but the result is hopelessly artificial. The result is similarly artificial when we attempt in the manner of the French music of Rousseau's time to restore to song the emotion of motion by way of a temporal progression of chords.

Harmony, as presenting us with relations between sounds understood as objects, gives us no access to the subject. In melody notes come together not mathematically as parts of a system, but as an imitation of the inflections of the voice in response to real situations—“plaints, cries of sadness or of joy, threats, moans.”

Accordingly, with a few exceptions “the art of the musician consists not at all in immediately painting objects, but in putting the soul in a disposition similar to that in which their presence puts it.”Footnote ¹¹ The difference is that music, unlike reality, frames itself so as to call attention not only to sound understood as noise, but to sound understood as a sign or image.

In the immediate sequel, Rousseau remarks that when he imitates a meow, his cat first sits up and takes notice, but once recognizing that it is not really the sound of another cat, it sits back down and relaxes. It is thus not the sound that moves the cat but the initial recognition of another cat in the meow. And with the recognition that the source of the sound is a false cat, the cat sits down. This example does not seem to add appreciably to the account until we notice how much more it appears to indicate than Rousseau's argument can bear. The first sentence of the Essay informs us that “speech distinguishes man among the animals.” Shortly thereafter, we learn that speech only arises once one recognizes that one has been recognized by another as “a sentient and thinking being similar to him” (1.2). Is Rousseau's cat such a being? Are all animate beings capable of that “moral need” or passion that consists in the recognition of others as subjects? Is there a music for cats?

Perhaps Rousseau is not altogether serious about his cat, and it is rather meant to serve as an introduction to his account of the obstacles that thwart one human being from recognizing another as a subject similar to him and that prevent human beings from sharing one music. Caribs respond to European music as little as the cat responds to Rousseau's meow. Chapter 14 leads us to believe that melody is somehow more natural than harmony, which only comes to sight when music has been made a mathematical system. But chapter 15 makes clear that melody too is conventional. So passion, that is, moral need, is natural in the way that language is natural. Both are available only by way of signs, but signs are conventional. It may be the case that the sting of the tarantula is curable only by frenzied dancing, but this requires music that moves us—that is, our own music. The cure requires les airs (tunes or airs), and must understand la langue (language or tongue) if one is to be put in mouvement (emotion or motion). The double entendres of Rousseau's own language suggest the problem. Our bodies respond physically to causes that are not simply physical. The nerves yield only insofar as they are disposed by l'esprit (mind or spirit), “for it is not so much the ear that bears pleasure to the heart as the heart that bears it to the ear” (15.6). On the one hand, Rousseau seems to argue that the superiority of melody to harmony is that melody makes possible the sensing of the subject in the uttered sound. This is what transforms the pleasure that all men feel in beautiful sounds to the utter delight (volupté) we experience when these sounds are animated or ensouled by melodious inflections (14.1). And yet “the songs we like as most beautiful will always touch only indifferently an ear of one not accustomed to them; it is a tongue [langue] for which it is necessary to have the Dictionary.” This is the serious version of the problem of Rousseau's cat. Melody animates sound only by way of convention. Music is a language, and as itself a language cannot be the source of the sensing of the subject that makes language possible in the first place. It is true that melody is closer to nature than harmony, which in regularizing and systematizing sound treats it as a pure object and detracts from what it is that wrenches sound from us. Melody, on the other hand, does not cut the sound continuum but seeks to reflect it as a continuum; it thereby seeks to reproduce not sound as pure object but sound as wrenched from a subject and so expressive of our interest in the world. And yet, Rousseau cannot deny that melody too is conventional. It is an already mediated representation of the subject.

There is a further curiosity in Rousseau's account. In chapter 13 we came to understand melody on the basis of an analogy between music and painting according to which sounds detached from the subject would be like colors without drawing—“melody does precisely in music what drawing [does] in painting” (13.2). How odd then that having completed his story about the superiority of melody to harmony in chapter 15, Rousseau should title the following chapter “False Analogy between Colors and Sounds.” The terms of the two chapters are not, however, perfectly in sync. Perhaps painting and music are analogous even though colors and sounds are not, for “the effect of colors is in their permanence and that of sounds in their succession” (16.1). The model for taking in color, then, is looking, contemplating, and wondering “all at once” (16.2). This is not true of sound, which involves a succession in time—in music, a series of different notes that must be synthesized or put together at each moment. This is what melody is. Chords, in giving one all the notes at once, conceal this fundamental connection between music and time. Melody, on the other hand, is not what the spatial image of musical notation makes it seem to be. Its notes are not really spread out over time as though on a line, but rather at each moment all of what came before is consumed in forming an anticipation of what is about to come.

Perhaps Rousseau's distinction between color and sound anticipates Kant's account of space as the form of our outer intuition and time as the form of our inner intuition. Still, for Kant, even objects in space are experienced in time. Rousseau's account abstracts from the extent to which to experience a painting we must work our way through its details sequentially (as it had earlier abstracted from the extent to which colors, always present to us in surfaces, always have shapes—which, like clouds, we regularly imagine as having desseins). Insofar as this is the case, is the distinction between the experience of color and the experience of sound so pronounced as Rousseau would have us believe? While the musician may have the great advantage of the painter (for example, in knowing how “to paint things that one cannot hear” [16.8]), nevertheless Rousseau begins the next chapter, “Error of the Musicians Harmful to their Art,” with the sentence “See [voyez] how everything brings us back without cease to the moral effects of which I have spoken” (17.1; emphasis mine). We have been enjoined to “see” a movement. So painting is not so atemporal as it first seems, and music not so temporal?

To see what this might mean, it is useful to return to the beginning of Rousseau's account of harmony.

Our pleasure in sounds is thus not simply a pleasure in hearing a single pure note, for

The unity of sounds given to us in nature is actually a “natural harmony.” If one thinks of a sound as the result of a vibrating cord of a certain length, when plucked it will generate not only a sound with a frequency the length of this vibrating cord, but also overtones—concomitant harmonics. These are the sounds generated by the vibrations of aliquot parts of the original cord—the parts perfectly divisible into the cord by a whole integer with no remainder. Accordingly any “single” sound will be accompanied by natural thirds, fifths, twenty-thirds, one hundred thirteenths, and so on—presumably to infinity. What we call “unison” is thus in reality an ordered, but infinite, multiplicity. This multiplicity affects us pleasantly, but we are nevertheless generally unaware of its existence. “We give too much and too little empire to sensations” (13.1).

Rousseau announces the origin of music in melody; it and not harmony is what is first wrenched from us by nature. But what exactly does this wrenching consist in? What is it that moves us to sing? We are told that melody imitates plaints and cries, but of course plaints and cries are expressions of pain, but even when expressing the painful, melody gives rise to a sweet sentiment, to volupté. This is what it means to say that it is an imitation. Rousseau has made very clear how we give too much empire to the senses insofar as we make sound by itself the source of the pleasures of music that can arise only out of the wedding of sound with passion understood as moral need—a need that connects us to another subject. But how is it that we give too little empire to the senses?

If “verses, songs, and speech have a common origin” (12.1), the Essay on the Origin of Languages is about this origin. In the part of the Essay explicitly concerned with music, Rousseau seems to suggest that music, like language, is subject to a natural decay. Language—initially spoken, poetic, and tonally accented—has a natural tendency to aim at a precision of articulation that so celebrates the “objective” that it ends by rendering the subject invisible and thereby, ironically, becomes less objective. Music moves from continuous melody (in which the subject is in its way the object), to melody (analyzed into discrete units of pitch—i.e., notes), to scales (as the systematic temporal sequence of discrete units implied by these notes), to keys (understood as their atemporal collection), to harmony (as the pleasing atemporal relations of notes within a key), and finally back to an all but denatured melody (whether understood as a progression of chords or as a combination in time of now discrete units of pitch). Rousseau's account of the tendency in his time to value harmony over melody is therefore the history of a decline.

As Rousseau goes on to locate the date of this decline in the fifth century BC, one is forced to wonder about his scathing indictment of contemporary French music. Could the decay of music, and for that matter of language, be all but coeval with the origin of music? And just as Rousseau relies on written versions of Homer's sung poems to judge the spoken word more poetic and more musical than the written word (6.1–2), perhaps it is possible that we are less removed from the Greeks than he first seems to indicate. In the manuscript of the Essay, the title of chapter 18 is “That the Musical System of the Greeks had no Relation to Ours.” Editors regularly change the tense of “had” so that it reads “has.” But, of course, after Rousseau finishes his chapter, he has articulated a fairly clear relation between the two systems. Because the intervals of the Greek scale are smaller than ours, they can more closely approximate the intervals of speech. This seems to make it possible for song to remain closer to its origin in speech, and so, more natural. In his rather complicated comparison, Rousseau must indicate how to translate the notes of the Greek system into something available to a modern reader—into the system of modern music. Otherwise his account would be altogether unintelligible. Perhaps, then, he means to indicate by his title that prior to this chapter the two systems had no relation, but that now they do. If so, he would have indicated that, while the music of one people may be all but unrecognizable to another people, this “all but” is a rather important qualification, one pointing to some ground that the two share. Is there some common ground where music is, on the one hand, always in decay, and on the other hand, never altogether so corrupted as simply to be severed from its origin?

But what exactly is this origin? Without our “initial” experience of overtones, of the concomitant harmonics of each sound, without the harmony of unison, melody would be impossible. If passion wrenches cries from us, we must still recognize their musicality. We delight in the way sounds go together. Ironically, however, if we treat this “togetherness” as a static relation among fixed objects—that is, if we treat it as harmony is conventionally treated, as though it were the music of the spheres—we have no way of experiencing the true power of the togetherness of these “objects.” For that we need to experience how one note wants another, which in turn wants another. To understand the true “empire of sensation” it is necessary for us to put these “isolated moments” into time, for otherwise we cannot experience the power that attaches them to one another, and hence cannot experience them as they are. Grasping them in their true atemporality requires that we grasp them temporally; otherwise the longing embedded in them is closed to us.

It is true that melody is more fundamental than harmony, but only because genuine harmony would require grasping the truth of sound that is only revealed sequentially in melody. Without this original atemporal harmony of unison, melody would be inconceivable; yet this atemporal harmony of unison can only be experienced temporally. It is rather like the sweet sentiment of one's own existence, without which our passion to preserve ourselves is unintelligible, but which nevertheless can only be experienced as a longing to return to a lost perfect natural state. The volupté of melody has precisely this character of a unity unfolding in time, complete at every moment insofar as the past movement of the melody is always assimilated and made sense of in the present moment, and yet incomplete at every moment insofar as the anticipation of what will come next is also built into the present moment, and yet “complete” only insofar as it is incomplete, for there would be no unity of the melody without the yearning present at each moment. It is, therefore, a perfectly understandable blunder that in an attempt to articulate this unity, one would isolate its parts, understand their mutual relations, and articulate the principles that bind them together. Such musical cosmology, call it harmony, is bound to fail, however, because in articulating what wrenches melody from us as an object, its character as wrenching fades from view.

Rousseau all but begins the Essay on the Origin of Languages with the following sentence:

A man wants to communicate with another as soon as he has been recognized by the other as a subject; this seems straightforward enough. But what happens when we look at the matter from the point of view of the other? For him to recognize the first man, he would already have to have been recognized by him. So, while we expect Rousseau to say that recognizing another as similar moves us to attempt to communicate, in fact, the origin of language involves being recognized by another or—since being anonymously recognized would not affect us at all—sensing that we have been recognized.Footnote ¹² Rousseau's almost opening remark thus suggests that for language to originate we must be somehow capable of sensing that we are being sensed—of experiencing a subject as a subject and not simply as an object. Put differently, language, like music, assumes an extralinguistic context that on the one hand must be assumed and, on the other, can only be articulated temporally even though it cannot finally be understood as temporal. This is what it means for language, and so reason, to be necessarily poetic, and so musical.

The argument of Rousseau's Essay on the Origin of Languages is intimately connected to the argument of his Discourse on the Origin and the Foundations of Inequality among Men. We are political only insofar as we are rational, and our rationality is inseparable from language. The character of the origin of political life thus depends on the origin of speech, which in turn depends on the origin of music. In the Discourse the meaning of this origin depends on the status of the state of nature, a status, of course, much disputed. Did the state of nature ever really exist, or is it only a logical construct? Is the book about the temporal origin of inequality or about its logical foundations? And if the latter, why does Rousseau persist in creating the illusion of a temporal unfolding? This issue is clarified somewhat by the note to the first sentence of Rousseau's preface. Having declared the most useful, but least advanced, human knowledge to be self-knowledge, Rousseau glosses his claim with a quotation from Buffon, who argues that we are not designed for self-knowledge, but rather know everything else better than we know ourselves. We are meant to know objects, not subjects. Nature designs us to “seek only to be spread without and to exist outside ourselves.”Footnote ¹³ Our “internal sense,” what separates us from all that is not part of us, is responsible for such self-knowledge as is available to us. What, however can this internal sense sense? If we are essentially beings who extend ourselves beyond ourselves, if we are to sense ourselves we will have to sense ourselves in the act of sensing other things. Only then will we grasp ourselves as subjects. Our internal sense, as necessarily derivative, can never sense our sensing purely. Any sensing of sensing will always be reflective; self-knowledge will always be secondhand—indirect.Footnote ¹⁴ Rousseau argues that self-knowledge is so little advanced because the self (here the soul) is like the statue of Glaucus, so long in the sea that it was disfigured by various accretions. In the passage of Plato's Republic to which this seems a reference (611b–d), it is soul's association with body that obscures its true nature. Rousseau indicates this too when he refers to “the changes happening to the constitution of bodies.”Footnote ¹⁵ He therefore agrees with Buffon, whom he has just quoted as saying,

If body is what is responsible for obscuring the soul, then mustn't body be stripped away in order to make self-knowledge possible?

At the same time, Rousseau identifies the corruption of the soul with the fact that it no longer acts “always by certain and invariable principles.”Footnote ¹⁷ Apparently our corruption is connected to our unpredictability—our freedom. The mixture of body and soul that leads to the corruption of reason by passion and of understanding by delirium is at the same time the birth of our humanity. This is confirmed in what follows.

Now, when exactly is it that all men, dogs, and so forth are equal by common avowal? Rousseau seems to imply that equality exists within species only when they are pure. Yet Rousseau admits of a “natural or physical” inequality among men and other animals.Footnote ¹⁹ Equality within species would seem to exist only prior to their members having bodies—that is, logically, in the way all common nouns refer to the same thing by “common avowal.” Accordingly our “natural equality” is something we discover only in the act of losing it, and we lose it upon being embodied. We lose it at the moment of birth.

Like the Essay on the Origin of Languages, Rousseau's Discourse on the Origin and the Foundations of Inequality among Men must make an appeal to a logical state as though it were temporal. This is necessary because our natural state must be one in which we are both untroubled by the feverish living outside ourselves characteristic of amour-propre and yet, curiously, aware of our own contentment. It is not an accident that in the Discourse (as opposed to the Reveries), this awareness is called amour de soi-même—love of oneself; it is difficult to conceive of it as love of myself without its turning into amour-propre. By temporalizing our fall out of the state of nature, Rousseau makes it possible to present this state as, on the one hand, a state of perfect contentment, and, on other hand, an object of deep longing. Yet the contentment is only intelligible in light of the longing. The object in view is not temporal, but it is not sufficiently animated until it becomes part of a temporal sequence, where it can be seen not as an object, but as an object of desire. This presentation of a state of perfect harmony that only comes to sight as genuinely harmonious by way of longing is of course what Rousseau means by melody. Melody is the sign of the necessarily temporal, musical, and poetic character of thought.Footnote ²⁰ Rousseau writes as he does, regularly presenting logical movements as temporal, because he understands this to be the very nature of thinking—what one might call the music of reason.