The Sign and the Song

Meaning, Melos, and the Appearance of Necessity

Meaning is not what music says. Meaning is what becomes necessary while music lives.

I. The Question Mislaid

    There is a question that the whole modern science of meaning has asked with such persistence, and answered with such ingenuity, that we have forgotten it was ever a choice to ask it. The question is what does this mean?, where mean is understood as a relation, a bridge thrown across a gap, a rope tying a sign to something on the far side of it. What does this word refer to, this sentence assert, this gesture signify, this passage evoke, this cadence connote? The question presupposes that meaning is a possession, that things have meanings the way bodies have weight, that meaning is a content lodged somewhere, behind the sound, beneath the surface, inside the sign, in the mind of the speaker, in the convention of the tribe,  waiting to be extracted, decoded, interpreted, retrieved.

    I want to say, as flatly as I can at the outset and then spend the rest of this essay earning, I very much feel that this is the wrong question, and it has been the wrong question for a very long time. Not false, wrong. It arrives too late. It begins after the decisive thing has already happened. It is a question about the sediment, asked as though the sediment were the source.

    For most contemporary philosophy of music, meaning is something music has. For the ontology I have been trying to building in Spanish, what I have elsewhere called a humanismo trágico de la comparecencia encarnada, a tragic humanism of embodied appearing, meaning is something music does. And more radically than that, meaning is not even a doing that music performs upon a pre-existing world of objects and hearers. Meaning is an acontecimiento, an event, and what happens in that event is the most difficult thing to say plainly because our whole vocabulary of "meaning" was built to say something else. Let me try the plainest formulation I have, and then defend it against the entire tradition that will find it naïve. 

    Meaning is the appearance of necessity. It is the lived event in which something gives itself as unable to have been otherwise, and gives itself so to a body, in time, before another. Meaning is not information transferred. It is not reference secured. It is not a proposition made true. It is the moment at which necessity becomes perceptible, the moment at which this could not have happened differently stops being a thought and becomes an experience one undergoes.

    To defend that sentence, I have to go back before music, to the general problem of meaning as the twentieth century bequeathed it to us, because the errors I want to name in the philosophy of music are only the local dialect of what I have come to consider a much older and grander error, and unless I show the grammar of that error I will seem to be quarreling with musicologists when I am really quarreling with a whole picture of what it is for anything to mean anything at all.

II. The Semiotic Axiom

    Open the standard account. It tells you, before it tells you anything else, that meaning is "a relationship between two sorts of things, that is, signs and the kinds of things they intend, express, or signify." Notice what has already been decided in that opening clause, decided so early and so quietly that it looks like a definition rather than a decision. Meaning is a relationship. Meaning is between two sorts of things. There is the sign, and there is the something-else the sign reaches toward, and meaning is the reaching. Call this the semiotic axiom, that is, to mean is to stand for. Everything downstream, the whole magnificent architecture of analytic semantics, is the working-out of that axiom under successive pressures, and the astonishing thing is how much the great disputants agree, precisely where they think they disagree.

    Gottlob Frege (1848 - 1925) splits the sign's relation in two, because he has noticed that "the morning star" and "the evening star" pick out the same planet and yet do not mean the same thing, that Hesperus is Phosphorus teaches us something, whereas Hesperus is Hesperus teaches us nothing. So there must be, alongside the Bedeutung, the reference, a second thing, that is, the Sinn, the sense, the mode of presentation under which the object is given. A brilliant repair. But look what it preserves. Meaning is still aboutness. Sense is a second thread thrown across the same gap; it is a more delicate way of standing-for, not an escape from standing-for. 

    Bertrand Russell (1872 - 1970), who could agree with Frege about nothing, throws out the sense/reference distinction and re-founds meaning on logical form and acquaintance, and preserves the axiom perfectly, that is, for him, to mean is to denote, and philosophy's task is to purify the language of denotation until the sign and the world lie flush against each other. 

    Alfred Tarski (1901 - 1983) disciplines truth into a recursion, that is, 'snow is white' is true if and only if snow is white, and Donald Davidson (1917 - 2003), in the boldest stroke of the century, proposes that to give the meaning of a sentence simply is to give its truth-conditions, so that a theory of meaning for a language becomes a theory of when its sentences would be true. Meaning collapses into the conditions under which the sign would correspond to the world. The axiom is not questioned, it is crowned.

    Even the revolts against this picture revolt within it. Ludwig Wittgenstein (1889 - 1951) turns from the Tractatus, where the proposition was a picture of a fact, the sign a mirror of the world, the purest form of standing-for, to the Philosophical Investigations, and declares that the meaning of a word is its use in the language. This is presented, rightly, as an earthquake, and it is one, because it relocates meaning from a static correspondence into a living practice. But ask what "use" rests on and Wittgenstein answers with a term he lays down like a foundation stone and never digs beneath, that is, Lebensform, form of life. Meaning is use, use is a language-game, a language-game is grounded in a form of life, and there the analysis stops, at the very threshold where my own tries to begin.

    J. L. Austin (1911 - 1960) cracks the truth-conditional monopoly more decisively than anyone, that is, he notices that "Hello!" has no truth-conditions, that most of what we say we do rather than describe — we promise, warn, name, christen, apologize — and he calls this the illocutionary, the act performed in the saying. Here at last meaning becomes act before content, doing before standing-for. And yet Austin, too, ties the illocutionary act to convention, that is, its force is conventional force, its felicity depends on accepted procedures correctly executed. 

    Paul Grice (1913 - 1988) distinguishes natural meaning ("those spots mean measles") from non-natural, speaker's meaning, and grounds the latter in a reflexive intention, that is, I mean something by an utterance when I intend you to recognize my intention that you believe or do something. 

    John Locke (1632 - 1704), far upstream, had already made meanings into mental representations, ideas provoked by signs. Eleanor Rosch (1938) and George Lakoff (1941) refine the ideas into radial categories organized around prototypes. And inferential-role semantics, at the far end, says the meaning of a term is its role in the web of inference.

    Set these theories side by side and you see that they are not so much rival answers as rival placements of the same relation. Where does the far end of the sign's rope attach? To the object (reference), to the mode of presentation (sense), to the truth-condition (Tarski, Davidson), to the use (Wittgenstein), to the conventional act (Austin), to the recognized intention (Grice), to the mental representation (Locke), to the prototype (Rosch), to the inferential role. Every one of them keeps the rope. Every one of them keeps the axiom, that is, that to mean is to stand for, to point across a gap toward something that is not the sign.

    What I would very much like to do here, if at all possible, is a tentative to philosophically cut this rope. Not to deny that signs stand for things. Of course they do. That is what a sign is, a thing whose vocation is to stand for another. My claim is stronger and stranger, that is, that standing-for is not the origin of meaning but a late crystallization of it. That the sign is a cooled and hardened form of something molten and prior. That before anything ever stood for anything, meaning had already happened — as appearance, as address, as the necessity of a living gesture that pointed toward no object at all but reached toward another body across the shared air. The semiotic axiom describes the geology of meaning after it has petrified. It does not describe the eruption. And music (this will be my whole burden) is the art that keeps us nearest the eruption, because music is the human practice in which meaning most stubbornly refuses to petrify into the sign.

    Note, too, what the entire tradition quietly assumes about the axis along which meaning is measured. It is the axis of truth. Meaning is bolted to truth-conditions, to correspondence, to assertion, to the true and the false. The five great theories of truth — correspondence, coherence, constructivist, consensus, pragmatic — are marshalled precisely so that meaning can be derived from truth, or truth from meaning, They turn on the same hinge. But a musical phrase is not true. It is not false either. To ask whether the second theme of a Gabriel Fauré Nocturne corresponds to reality, or coheres with a system, or is agreed upon by a community, or works in John Dewey's sense, is to have missed what a phrase is by so wide a margin that the miss is instructive. The axis of musical meaning is not true/false at all. It is necessary/arbitrary, or, more deeply, incarnate/inert, alive/dead, comparecido/ausente, that is, appeared-before-us versus merely present. This reorientation is not a small adjustment within the theory of meaning. It is a change of the fundamental coordinate. Meaning's opposite, on my account, is not falsehood but arbitrariness, not error but inertia.

III. The Appearance of Necessity

    Everything now turns on the word necessity, and I must protect it at once from a confusion that will otherwise destroy the whole argument, a confusion with the necessity the philosophers already own.

When Saul Kripke (1940 - 2022) tells us that Hesperus is necessarily Hesperus, that a name is a rigid designator picking out the same object in every possible world, he means a modal, metaphysical necessity, that is, true in all worlds, discoverable a posteriori but binding a priori upon the identity. This is the necessity of the concept, the necessity you deduce, the necessity that holds whether or not anyone is there to feel it. It is cold and it is total and it belongs to logic. It is not what I mean, and it is nearly the opposite of what I mean.

    The necessity that is meaning, the necessity whose appearance is the event I am calling meaning, is not deduced but undergone. It is not true across all possible worlds. It is lived in this one, now, in a body. And here is the paradox that I take to be the very structure of meaning, and not incidentally the very structure of the tragic, the phrase that appears as necessary is, as a matter of fact, entirely contingent. Enescu could have written the melody otherwise. He did not have to write it at all. The performer could have shaped it a hundred other ways. The silence before the return could have fallen a beat sooner or later. Nothing in logic, nothing in physics, nothing in the score compels the inevitability we experience. And yet, when it is alive, when what I call melos has succeeded in incarnating itself, the phrase gives itself as something that could not have been otherwise. It arrives wearing the face of destiny while remaining, in its metaphysics, pure gift and pure accident.

    This is the crucial thing, and I want to state it as a principle, that is, that meaning is a phenomenological necessity standing on a metaphysical contingency. The necessary that did not have to be. The inevitable that could have been avoided. The gift that appears as fate. This is why I say the structure of meaning is tragic, and why an ontology of meaning must be, at bottom, what I call in Spanish a humanismo trágico. Saul Kripke's necessity abolishes contingency. The necessity of meaning appears against contingency, on its ground, and needs it. Remove the contingency — imagine a phrase that truly could not have been otherwise, mechanically determined, generated by rule — and you have not intensified its meaning. You have destroyed it. You have the necessity of a multiplication table, which no one has ever wept over. The pathos of the necessary phrase is precisely that it chose itself out of an infinity of the possible and now cannot be unchosen. That is what we hear when we say a passage is inevitable. We are not hearing a law. We are hearing a freedom that has hardened, before our ears, into fate.

    So there are two necessities, and the theory of meaning has hold of the wrong one. There is the necessity of the concept — timeless, worldless, deduced — and there is the necessity of the flesh — temporal, embodied, undergone, and haunted by the contingency it overcomes. Meaning belongs entirely to the second. And the second is invisible to a semantics built on the first, because a truth-conditional or referential or inferential account can register only what is the case or what follows. It has no instrument that can register the difference between a phrase that had to happen and a phrase that merely did. That difference is inaudible to the theory of the sign. It is the whole of what I mean by meaning.

IV. Presentation Before Representation

    If meaning is the appearance of necessity, then meaning does not lie behind the sounds waiting to be uncovered, and interpretation is not the excavation of a buried content. Meaning lies in the mode of appearing itself. It is comparecencia — a coming-forth, an appearing-before — and comparecencia is not a veil over a hidden depth but the very event of presence. The modern aesthetic reflex is to ask what does this passage mean? as though the sounds were a cipher and meaning a plaintext locked inside them. My ontology asks a different question, and the difference is total: not what does it mean? but when does meaning appear? Because meaning is not concealed behind the music. It is the music's manner of coming into presence.

    Here the oldest and surest analogy is the human face, and I take it not as a metaphor but as the literal paradigm. I take it as Emmanuel Levinas (1906 - 1995) took it, as the site where meaning is originary. You do not first perceive a configuration of features, infer from the configuration a set of inner states, and then, by a second act, arrive at the meaning "a person is here, addressing me." The face does not represent a person who is elsewhere. The face is the person appearing. It comes already meaningful; its meaning is not decoded but encountered; and it precedes every interpretation you will ever offer of it. A smile does not stand for happiness the way a red light stands for "stop." A smile is happiness appearing. The meaning is not behind the smile. It is the smile, in the mode of coming-forth.

    This is what I mean by presentation before representation, and it is the exact point at which I try to break with the whole tradition of musical meaning that treats music as depiction — music representing grief, joy, forests, storms, death. Music does not, in the first instance, represent grief. It presents a living way of existing which we may afterward, and secondarily, and often wrongly, name "grief." The naming comes after. The mode of being comes first and comes whole. And this ordering — appearing first, naming second — is not a peculiarity of music; it is the general structure of meaning, which the theory of the sign inverts by making the name the primary bearer and the appearing a mere occasion for it. The infant does not decode the mother's face. The lover does not infer the beloved. The mourner at the graveside does not interpret the absence into significance. In each case meaning has already appeared, in the flesh, before the first sign is drawn. What I am proposing is that music belongs to this order — the order of the face, the order of what I call in Spanish comparecencia — and not to the order of the cipher.

    I have been developing, in recent work, the idea of el rostro de la obra, the face of the piece, the notion that a work of music, rightly encountered, does not present a text to be construed but a countenance that regards us, that comes toward us with the frontal, addressing, irreducible quality of a face. This is not mysticism. It is the refusal to accept that the only alternative to formalism (music means nothing, only itself) is semiotics (music means by standing for). There is a third possibility, older than both, that is, that music means the way a face means — by appearing, incarnate, before another, in time, as necessity.

V. Address Before Aboutness

    The face brings with it a feature that the semiotic axiom cannot accommodate and that I take to be decisive: meaning, in its origin, is not about something. It is toward someone. Before aboutness there is address.

    The whole theory of the sign is built on aboutness — on intentional directedness toward an object or state of affairs. Even Paul Grice, who comes closest to relationality with his speaker's meaning, models the toward-someone on a prior aboutness: I intend that you believe something, the content comes first and the addressee is its delivery point. But listen to the primal scene of meaning and you hear the order reversed. The infant's cry is not about hunger; it is a calling-toward, an address flung into the air before there is any concept of what is lacked. The mother's soothing voice is not a proposition about safety; it is presence given as address, a toward-you that carries no about-what. The lament laments to someone — to the absent, to the dead, to God, to no one, but always toward. Song, before it says anything, is vocative. It is the case-form of the second person made audible. And this is why I locate the origin of meaning not in reference but in what I will call the melic address — the reaching of a living voice toward another across the shared and breathable air.

    This is the deepest reason I place myself nearer to Martin Buber (1878 - 1965), to Emmanuel Levinas, and to Jean-Louis Chrétien (1952 - 2019), to the philosophers of the between and the call and the response, than to any semanticist. For Buber, the primary word is not a sign that names but a relation that is spoken — the I–Thou that is not about the Thou but toward it, and in which meaning is not extracted but lived between. For Levinas, signification begins in the face of the Other as address and command, prior to any theme; the saying precedes and exceeds the said. For Chrétien, the origin is the call and the response, the voice that has always already been addressed before it speaks. My ontology of melos is the musical rendering of this priority of address over aboutness, of the vocative over the descriptive, of the between over the about.

    And this lets me say precisely where meaning is located, which I have the feeling the tradition has never satisfactorily answered. It is not in the score — the score is a trace, a set of instructions, a fossil, historical through and through, never neutral, never the music. It is not in the performer alone, nor in the listener alone, nor in the notes considered as objects. A single C has almost no meaning; meaning begins only when sounds begin needing one another — when one note leans and another answers, one resists and another remembers, and a relation of necessity is born between them. Meaning is relational in the strong sense: it does not belong to the terms but to the between. And the ultimate between is not note-to-note but voice-to-voice, the between of the addressing and the addressed. Meaning exists only in the encounter, as a kiss has no meaning lying in a textbook and all its meaning in the act, as a conversation means nothing transcribed and everything while it is happening between two faces. Music reaches meaning by belonging to this order of the between — the order of dance, of conversation, of love — and not to the order of the freestanding object with a content inside it.

    Here one can be generous to Austin and also disagree him in the same breath. Austin was right, more right than the truth-conditional tradition around him, I think, in saying that meaning is act before content — that we do things with words, that the illocutionary force of a saying is not its describing but its performing. He saw that "Hello!" has no truth-conditions and is not thereby meaningless. But Austin tied the force to convention, that is, for him, the performative works because an accepted procedure is correctly followed. My claim would be that the illocutionary is downstream of the melic — that beneath the conventional performative lies the unconventional one, the cry that calls without any procedure, the lullaby that soothes without any accepted ceremony, the lament that mourns before any rite. Song is the original performative, and it needs no convention to have force, because its force is not conventional but incarnate. Convention will come and stabilize it, cool it, ceremonialize it, turn the cry into the formula and the formula into the code. But convention explains the recognition of the force, never its origin. The origin is the breath reaching toward the other before any procedure exists to be followed.

VI. The Event and the Trace

    Necessity unfolds, therefore meaning cannot be frozen. This follows with the force of a theorem from everything said so far, and it is the point at which my ontology stands nearest to a single great musical mind — Sergiu Celibidache (1912 - 1996) — and, standing nearest, sees most clearly where I must go further.

    Meaning is temporal not incidentally but constitutively. A photograph of a dancer is not dancing; the dance exists only while time is becoming itself. Likewise a score is not meaning, and a recording, however precious, is not the event. The recording preserves the sounds. It does not preserve the acontecimiento — the appearing of necessity to a body, in a shared present, before another. This is why recordings, valuable as they are, never exhaust music: they capture the trace and lose the incarnation. They give you the sediment with extraordinary fidelity and cannot give you the eruption at all.

    Celibidache spent a lifetime fighting the reduction of music to a thing, and here our agreement is nearly complete. The score is not the music; the recording is not the music; the work is not the music; only the sounding event, constituting itself in lived duration, is music. He rejected musical Platonism, rejected the ideology of Werktreue understood as fidelity to a fixed object, rejected performance conceived as reproduction, insisted that music exists only in becoming, that the C-major chord means nothing until a second chord arrives and a third, until tensions and expectations and memory begin to form and the music constitutes itself in time. He distrusted the studio precisely because recording destroys the phenomenological conditions of the event. On all of this I say, yes, and yes, and yes again. If I have an ancestor among the practitioners, in the matter of the event against the object, it is he.

    And yet Celibidache begins where I would like to show that music has already begun. His question is how does musical meaning appear? — a phenomenological question, asked of sounding phenomena, from within the field of what already resounds. My question would be prior, that is, where does music come from?, and the answer to that question lies before sound. It lies in breath, in the voice, in gesture, in walking, in the body already orienting itself toward expression before a single tone has been produced. For Celibidache, meaning arises in the temporal constitution of the sounding phenomenon. For me, the sounding phenomenon is itself downstream of melos — of the embodied, vocal, gestural, addressing impulse that precedes notation, precedes analysis, and precedes even the acoustic event. Celibidache gives us a phenomenology of musical appearance. What I am after is an anthropology, and finally an ontology, of musical origin, not only how music comes into presence, but why there is music at all, and the answer to that is not "organized sound" and not "consciousness constituting an object in time," but a human need: the necessity to sing, to call, to lament, to soothe, to celebrate, to invoke. Phenomenology describes how this appears once it has appeared. It does not generate it. The origin is not perception. The origin is need, the irrepressible pressure of a mortal, breathing, relational creature toward the address that is song. Celibidache's central category is becoming. Mine would be melos, and becoming unfolds from it.

VII. Melos Made Audible

    Now I can say, with the ground prepared, the deepest and most specifically musical of my formulations. Meaning, in music, is melos becoming audible. Meaning appears whenever melos succeeds in incarnating itself — and melos is necessity before organization, the living impulse of voice and breath and gesture and rhythmic orientation that is prior to notation, prior to structure, prior to the sign, and from which all of those subsequently crystallize. The score is one possible trace of it; technique one possible vehicle; analysis one possible clarification. Meaning belongs to none of them. It belongs to the emergence of melos itself.

    But I must not leave melos as a single undifferentiated word, or it will look like a mere honorific for "the good stuff." Melos has an internal architecture, and I have come to conceive that architecture as a hýphasma — a woven cloth — of three strands that the Greeks already named and that no analysis can finally separate without unmaking the cloth: λέξις (lexis), ἁρμονία (harmonía), and ῥυθμός (rhythmós). These are not three parameters to be measured. They are three modes of the one incarnation, three directions along which necessity becomes perceptible, and meaning is not lodged in any one of them but in the weave.

    λέξις is the saying, the diction, the physiognomy of the phrase — the way a musical utterance speaks, its articulation, its consonants and vowels, its address, its face. This is the strand of comparecencia, of el rostro de la obra: the dimension in which the music turns toward us with a countenance and addresses us in the vocative. When I say a phrase must speak — not that it must be clear, not that it must be correct, but that it must say itself to someone — I am pointing to λέξις. Meaning as address lives along this strand.

    ἁρμονία is the fitting-together, the joining, the relation of tensions — not "harmony" in the narrow chordal sense but the whole economy of leaning and answering, resisting and resolving, by which one sound comes to need another. This is the strand of relational necessity: the between in which a single note, meaningless alone, enters into the obligations that make it inevitable. Meaning as relation — the C that must fall to the B, the suspension that must resolve, the phrase that must return — lives along this strand.

    ῥυθμός is not meter and not the barline. It is flow, the shape a thing takes as it moves, the way the phrase carries itself through time — nearer to the original sense of ῥυθμός as the form of what flows than to any grid of pulses. This is the strand of temporal necessity: the dimension in which inevitability unfolds, in which the appearing of necessity requires duration to appear at all, in which meaning cannot be frozen because ῥυθμός is precisely the refusal of the frozen. Meaning as the unfolding of the inevitable lives along this strand.

    And the hýphasma — the woven cloth — is the figure that appears only in the weaving, present in no single thread. This is the exact sense in which meaning is emergent without being constructed, relational without being arbitrary, incarnate without being reducible to any one of its dimensions. Pull out λέξις and study it alone and you have prosody without music; pull out ἁρμονία and you have grammar without speech; pull out ῥυθμός and you have motion without a mover. The meaning is the pattern in the cloth. It is what appears when saying, joining, and flowing are woven into a single act of address — when the phrase speaks, and coheres, and moves, all at once, toward someone, in time, as necessity. That figure in the weave is melos made audible, and melos made audible is meaning.

    And I must not pretend that the weave is my discovery, or that the three strands are a scheme I spun. They are older than any of us. When Plato has Socrates say, at Republic 398d, that melos is composed of three things — λόγος, ἁρμονία, and ῥυθμός, word and harmony and rhythm bound into one — he is already naming the hýphasma, already refusing to let music be any one of its threads; and Aristides Quintilianus, in the De musica, gives that intuition its fullest ancient body, the treatise to which every serious recovery of melos has since returned. Wagner reactivated the same unity, under other names and to other ends, when he demanded in Oper und Drama that drama restore what opera had torn apart; and in our own century Thrasybulos Georgiades — in Der griechische Rhythmus, in Musik und Sprache, in the posthumous Nennen und Erklingen — reconstructed the ancient welding of tone, word, and rhythm, time as logos, with a philological rigor no one has matched. I try to stand in that line, and I have changed one thread deliberately, that is, where Plato writes λόγος I have written λέξις, because I want the saying and not the reason, the diction of the phrase and not its argument. 

    But the debt I must name above all the others, the one without which this essay simply would not exist, is to Vicente Chuliá. In his Manual de Filosofía de la Música (Pentalfa, 2018) and, above all, in his Tratado de Filosofía de la Música (Pentalfa, 2022) — the fruit of his doctoral reconstruction of the espacio melológico, and of the years of lectures and recorded conferences he has given alongside them — Chuliá did what no one since the Greeks had fully dared, that is, he took melos out of the museum of dead terms and made it a category again, an ontological and anthropological determination of what music is, and of what the human being is who cannot help but make it. Almost everything I have said here about melos I owe, at its root, to him. Maybe, I might have carried the idea into a country he might not choose — into the phenomenology of comparecencia, into the flesh and the face and the tragic, rather than the categorial architecture in which he grounds it — but it is his gift I am carrying, and without his recovery of melos there would have been nothing in my hands to carry at all.

VIII. Four Families and a Funnel

    Let me now situate this against the century of thought that took musical meaning as its problem — not to award marks, but because my ontology is not eclectic, and the surest way to show that it is a single position and not a collage is to show exactly where it converges with each tradition and exactly where it inverts it.

    Seen from the ground I have been describing, the modern study of musical meaning falls into four families.

    The first is structuralism, and its great and honest exemplar is Heinrich Schenker (1868 - 1935). For Schenker, meaning descends, that is, the Ursatz generates the middleground, the middleground generates the foreground, deep structure produces surface. His ontology is the exact inversion of mine, and I honor him as a worthy antagonist precisely because the inversion is clean. Where he has structure generating surface, I have living surface — breath, gesture, melos — crystallizing into organization. His arrow points down, from the fundamental structure to the sounding note. Mine points up, from the sung and gestured necessity to the structure that later reads it back. This is not a quarrel about analytic technique. It is a quarrel about the direction of the causal order of music itself, and it cannot be split.

    The second family is semiotics and hermeneutics — Jean Molino (1931 - 2024), Jean-Jacques Nattiez (1945), Eero Tarasti (1948), Raymond Monelle (1947 - 2004), Leonard Ratner (1916 - 2007), Kofi Agawu (1956), Lawrence Kramer (1946), Yuri Lotman (1922 - 1993), the whole enterprise of meaning-as-sign, meaning-as-convention, meaning-as-interpretation, meaning-as-culture. With these I have real and important convergences, and equally real limits. For instance, Molino I honor for a genuine breakthrough — his tripartition of the poietic, the neutral trace, and the esthesic, and above all his insistence that the score is not the music. But his neutral level is exactly where I cannot follow him, because to me, there is no neutrality, that is, to me, the score is already historical, the recording already historical, the gesture already historical; everything already belongs to lived embodiment, and where he posits a neutral trace I posit an embodied event. Nattiez develops Molino into a full semiology and destroys, usefully, every naïve notion of objective musical meaning — but he remains within signs, codes, interpretants; meaning becomes interpretation, and I must answer that before signs there is singing, that meaning begins before semiotics. Tarasti is fascinating because his existential semiotics reaches toward Merleau-Ponty and Sartre, toward identity and presence and lived time — and still it is semiotics, signification, and I would invert him, that is, I would posit existence first, signs second. The topic-theorists, Ratner and Monelle, and the schema-theorists after Robert O. Gjerdingen (1954), teach us truths about cultural memory, about the pastoral and the military and the galant — but topoi and schemata explain recognition, not birth. They tell us which cultural topic a passage evokes. They cannot tell us why the melody had to exist. Kramer and Rose Rosengard Subotnik (1941) rightly demolish autonomous formalism, but everything in their hands becomes interpretation, ideology, history — and my objection is single and constant, in other words, yes, history shapes meaning, but cannot exhaust it; culture teaches us to recognize meanings but is not what first makes music capable of meaning at all.

    The third family I call embodied expressivism, and here are my nearest kin: Boris Asafyev (1884 - 1949), Mikhail Bakhtin (1895 - 1975), Edward T. Cone (1917 - 2004), Robert S. Hatten (1951), Thomas Turino (1957). Asafyev's intonatsiya — the idea that music consists not of notes but of living intonations, that music is action and process and breath, that meaning arises through intonation — is among the greatest ideas of twentieth-century musical thought and one of my true ancestors; I part from him only where he lets intonation historicize too quickly into cultural language, whereas I insist that melos precedes culture and culture shapes it afterward. Bakhtin, astonishingly close, gives me voice, answerability, the living utterance always addressed, dialogue and polyphony as ontological and not merely literary — and I extend his dialogism beneath language into the prelinguistic singing from which language itself descends. Cone's musical persona, the composer's voice, the intuition that music speaks through agency, I take almost whole — maybe prudently correcting only his residual sense that the persona is a fiction, where for me the voice is not metaphor but existential fact. Hatten's phenomenology of gesture, markedness, expressive agency, I embrace — insisting only that gesture is literally embodied and not merely analytically useful. Turino's participatory music, embodiment, performance as social existence, is one of my closest alliances — deepened only by my claim that participation itself flows from melos rather than from anthropology.

    And the fourth family is the one I am discreetly trying to found. It says that music is neither primarily structure, nor primarily sign, nor even primarily expression, but the incarnation of necessity — that breath, voice, gesture, rhythm, and orientation-toward-another become a living event from which structure, signification, narrative, style, form, and cultural meaning subsequently crystallize. Structural analysis, on this view, describes the crystallization of melos; semiotics describes the interpretation of melos; hermeneutics describes the historical worlds that melos can open. Each is true of its object. None explains how music first becomes alive.

    So the genealogy of my position is not a line but a funnel. Asafyev pours in the idea of intonation as living action; Bakhtin the ontology of voice and responsive utterance; Cone the intuition that music speaks through agency; Hatten the phenomenology of gesture; Turino the participatory, embodied nature of music-making; Molino the refusal to identify the music with the written object. And the single move that turns this inheritance into a position rather than an anthology is the move that makes melos — not intonation, not gesture, not sign, not structure, not expression, but melos, that is, he embodied, vocal, gestural, addressing impulse itself — the primordial source from which every other musical category derives.

    Two figures deserve a last word here, because they mark the two ways of starting too late, and between them they define the error I am trying to escape. Schenker starts too late by starting with structure, that is, he takes the crystal as the origin and never asks after the solution from which it precipitated. And Philip Tagg (1944 - 2024) — whom I honor for asking the right question, how can music communicate without words?, and for refusing to despise popular music, film music, the ordinary musical life of ordinary people — starts too late by starting with culture. Tagg's model is, at bottom, communicative, that is, sender, musical code, listener, decoded meaning; musemes function like morphemes, style-indicators like a shared lexicon, and consensus among listeners becomes the evidence of meaning. But convention explains recognition, never origin; a code is sediment and melos is source; and if eighty percent of listeners hear "sadness," that is excellent evidence and still does not explain why sadness became audible in the first place. To me, consensus is not ontology. Tagg's own project, turned right-side up, would read not music communicates through codes but music becomes codified because living gestures become culturally stabilized — which is almost the reverse of his direction. If Schenker starts too late with structure and Tagg starts too late with culture, then both structure and culture are downstream, and the source they overshoot is the same, that is, the pre-cultural, embodied emergence of voice, breath, gesture, and rhythm that I call melos, which is what first makes music capable of meaning at all.

IX. The Voice Before the Word

    I have claimed, again and again, that the sign is a late crystallization and that song is older than it. I want now to make that claim as concrete as it can be made, because it is the empirical and existential heart of the whole ontology, and because it grounds a conviction about pedagogy that I hold as firmly as anything I hold.

    The infant recognizes the mother's voice before understanding a single word. This is not a charming detail; it is a datum about the order of meaning. The voice, before it carries any proposition, carries presence. It does not communicate a content; it communicates a toward-you. Meaning begins there — in the recognition of an address that has no about yet — and everything the semiotic tradition studies is built on top of that recognition, never beneath it. Singing is already meaningful before language; the lullaby means before it narrates; the cry calls before it refers. Recognition — the primal knowing-again of a voice, a presence, a face — is the earliest form of meaning, and it is recognition of a someone, not construal of a something.

    From this I draw a conviction about how music ought to be transmitted, and it cuts against the reflex of a listening culture. Musical education should begin not with listening but with singing. Listening is receptive; it places the child before an object to be taken in. Singing is participatory; it places the child inside the event as one of its makers. And meaning, as I have argued through this whole essay, does not arise from observing life but from inhabiting it — not from standing before the between but from standing in it. To teach a child to listen first is to teach the semiotic posture: here is a thing, decode it. To teach a child to sing first is to teach the melic posture: here is an act, be it, address and be addressed. The one produces interpreters of objects. The other produces participants in events. This is also the deep sense of filiation as against school: a school transmits a technique, a code, a stabilized set of conventions to be recognized and applied; a filiation transmits a living manner of singing, an unconscious mastery passed body to body, breath to breath, the way a voice is learned before a grammar — the way meaning itself is first received, in the flesh, as address, before it ever hardens into the sign.

X. Against the Merely Possible

    I have kept for last the most practical face of this ontology, because it is where I began, years ago, before I had any of these words — in the act of criticism, of judgment, of hearing a performance and knowing, with a certainty I could not yet justify, that something had or had not happened.

    When I criticize a performance and find it wanting, I am very rarely saying that was ugly. Ugliness is a category I can almost always forgive. Sometimes I even like it, if it is born from necessity. What I am saying, when I reach for the harshest judgment I have, is something else: nothing needed to happen. Everything in the performance was merely possible. Every choice was defensible, every note in place, every marking observed — and nothing was necessary. No breath, no silence, no accent, no turn of tempo, no shading of a vowel emerged as though it had to emerge. The performance was a sequence of the arbitrary wearing the costume of the correct. And this, and not ugliness, and not error, and not inaccuracy, is the true opposite of meaning. Meaning is the opposite of arbitrariness. A performance becomes meaningful not when it is beautiful and not when it is faithful but when every breath, gesture, silence, tempo, articulation, facial expression, and sound seems to emerge from the same living source — not because it was planned, but because, having happened, it could not now be imagined otherwise. The mark of meaning is the disappearance of the alternative. While the arbitrary keeps all its alternatives visible — it could have gone this way, or that, and it went one way for no reason that reaches the ear — the necessary consumes its alternatives as it passes, so that afterward we cannot conceive the phrase having gone any other way, though we know, with the other half of our mind, that it could have.

    This gives the critic a criterion that no semiotics and no structural analysis can supply, because both of those measure along the wrong axis. Structural analysis can tell me whether the reading is coherent with the deep structure; it cannot tell me whether anything had to happen. Semiotics can tell me whether the topical and stylistic conventions were legible; it cannot tell me whether the phrase appeared or merely occurred. The axis I judge along is necessary/arbitrary, incarnate/inert — and it is the axis of meaning itself. This is why my criticism has often looked, to those who measure along the axis of accuracy, either mysterious or unfair. I am not measuring accuracy. I am asking whether melos incarnated — whether necessity appeared — whether the rostro de la obra turned toward us and addressed us, or whether we were left with a well-executed corpse.

    And this is finally what the rostro de la obra (the face of the piece) names: the locus at which a work's necessity becomes legible, the countenance in which its inevitability shows itself. A performance is meaningful when it lets the face of the piece appear — when λέξις speaks, ἁρμονία coheres, and ῥυθμός carries the whole toward us in a single act of address, so that the woven cloth shows its figure and we are met, across the air, by a necessity that had every right not to exist and now cannot be unmet. The performer's task is not to represent the work and not to interpret it in the sense of decoding a hidden content. It is to let necessity appear — to become so transparent to melos that the arbitrary falls away and what is left could not have been otherwise. Not absent, this transparency; transparent, in Celibidache's sense and further: the personality does not vanish but becomes the window through which necessity comes forth. That is the whole ethics of interpretation, and it is an ethics and not a technique, because it is answerable to the between — to the other toward whom the address is flung — and not merely to the object on the page.

XI. Coda: Life Recognizing Itself in Sound

    I can now gather the whole argument into the single sentence it has been reaching for, and then let it open rather than close — for this is an ontology sin bloque, without system, a thinking that refuses the premature block into which philosophy loves to freeze its own becoming.

    Meaning, generally, is the lived event in which necessity appears — the coming-forth, to a body, in time, before another, of something that gives itself as unable to have been otherwise, and that gives itself so against the ground of its own contingency, which is why the structure of meaning is tragic and why an honest humanism must be a tragic one. And meaning in music is the highest and least petrified instance of this, because music is the art that keeps nearest to the eruption and furthest from the sign: it is melos made audible — the woven cloth of λέξις, ἁρμονία, and ῥυθμός showing its figure — melos, the word Vicente Chuliá gave back to our time, the incarnation of necessity in voice and breath and gesture and address, from which structure and signification and narrative and style and every cultural meaning afterward crystallize, and which none of them can generate, explain, or exhaust.

    So I will end where I began, with the two formulations that carry the whole weight, and I offer them not as conclusions but as thresholds:

    Meaning is not what music says. Meaning is what becomes necessary while music lives.

    And, in its fuller form:

    Meaning is the lived experience of necessity — the moment when melos incarnates itself so completely that life recognizes itself in sound.

    That last phrase is the one I most want to leave standing, because it says why music matters at all, and it says it against the entire tradition that would make music matter by making it carry information about the world. Music does not matter because it talks about life. It matters because it exists like life — because a human being is meaningful not by carrying information but by becoming irreplaceable to us, their meaning identical with their irreplaceability, a child, a friend, a teacher, a beloved, none of them reducible to any description, each of them a necessity that did not have to be. Music reaches that same ontological level. It does not describe the necessary. It is the necessary, appearing. And in the moment it appears — in the moment the phrase turns toward us with the face of destiny and the flesh of contingency, and we are met — life recognizes itself in sound. That recognition is meaning. Everything else is sediment, and the science of the sign is the geology of the sediment, and it has taken the sediment for the world.

    I obviously do not propose to have the last word on all of this. What I call an ontology of comparecencia cannot have a last word, because the event it studies happens again each time, always for the first time, never exhausted by what was said of it before. And the word I have leaned on hardest — melos — I did not find alone: I received it from Vicente Chuliá, as one receives a filiation and not a doctrine, and I return it here changed but not disowned. I propose only to have found the right threshold, and to have named, against a century that named everything else, the one thing the theory of meaning left out: that before the sign there was the song, and in the song, meaning had already, and wholly, and necessarily, appeared.

ADDENDUM

on the idea of "Phenomenological Neccesity", which I refer to in this essay

    The phrase phenomenological necessity may appear suspiciously vague, especially to readers trained in the analytic tradition, for whom necessity already possesses an established family of meanings, that is, things like logical necessity, metaphysical necessity, nomological necessity, conceptual necessity, deontic necessity. It may therefore seem that I have merely introduced yet another species into an already crowded taxonomy. However, I do not believe I have. Indeed, one purpose of this essay has been to suggest that what I call necessity has almost disappeared from philosophical view precisely because philosophy has gradually identified necessity with what can be demonstrated, formalized, generalized, or shown to hold independently of experience. My use of the word points in the opposite direction. It names not a property of propositions or worlds but a mode of appearing.

    Phenomenological necessity is not the necessity that is. It is the necessity that appears. That distinction is decisive. A phrase by Gabriel Fauré, George Enescu, or Franz Schubert is not metaphysically necessary. Nothing prevented another phrase from being written. Every note could have been otherwise. Every silence could have occurred elsewhere. Every tempi could have unfolded differently. Yet something remarkable happens in the living event of music. Contingency begins to appear under the aspect of inevitability. The phrase gives itself—not to logic but to experience—as though it could no longer have unfolded differently. This appearance is not an illusion to be corrected. It is the very phenomenon requiring description.

    Phenomenological necessity is therefore neither objective nor subjective. It is not objective, because it is not a property inhering in the score or in the sequence of notes. It is not subjective, because it is not merely a private feeling projected by the listener. It belongs to the encounter itself—to the event in which a work, a performer, and a listener mutually constitute a world in which contingency comes to appear as destiny.

    This is why I have repeatedly spoken of comparecencia rather than perception. Perception receives an already constituted object. Comparecencia names the event in which the object itself becomes what it is by appearing before another. Necessity belongs to that appearing.

    One might say that phenomenological necessity is the experience of retrospective inevitability. Once the event has genuinely happened, the alternatives disappear—not because they were never possible, but because they no longer belong to the world that has now come into being. Freedom crystallizes into fate without ceasing to have been free.

    This distinguishes phenomenological necessity equally from psychological compulsion. It is not simply the feeling that I cannot imagine another continuation. One often cannot imagine another continuation merely because of familiarity or convention. Countless clichés possess predictability without possessing necessity. A perfect cadence may be expected without being inevitable. Habit is not destiny.

    Necessity appears only where a living event generates its own law from within itself. For this reason phenomenological necessity cannot be reduced to probability, expectation, convention, information, prediction, coherence, or causal determination. It belongs to another order entirely: the order in which meaning first becomes possible.

    If I were forced into a single definition, it would perhaps be this: "Phenomenological necessity is the mode of appearing in which an irreducibly contingent event gives itself, in lived experience, as retrospectively incapable of having been otherwise." Everything else in this essay follows from that proposition.

And if this concept ultimately requires an entire essay of its own, I would regard that not as a weakness of the present argument but as evidence that the deepest questions rarely permit themselves to be answered only once. Indeed, if this ontology is correct, phenomenological necessity is not simply one concept among others. It is the very form under which meaning first enters the world.