The Listener and the Voice

Mildred: I like to hear you talk.

Wally Fay: Yeah? So do I. Something about the sound of my own voice that fascinates me.
Mildred Pierce (1945)

With the transition to prerecorded cylinders and discs, the pleasure of recorded sound became focused on the voice of the other, a condition of listening pleasure equally true of cinema, because of its incorporation of the recorded voice, and radio. In order to clarify the significance of this transition for the voices of women, I shall briefly review the ways two prominent theorists have attempted to describe the appeal of listening.

As one of the few early film theorists to address the topic of sound-without-image (Dziga Vertov's sound experiments being the other major example), Rudolf Arnheim attests to the pleasure "pure" sound could offer, describing it in terms eerily resonant of Guy Rosolato's "sonorous envelope," but on a massive, even national scale (Rosolato 1974, p. 81). In his book Radio, written in Germany in 1936, Arnheim conjures up the image of the radio audience as an unseeing, eagerly listening mass, enthralled as their imaginations thrill to the bidding of a (benevolent) disembodied voice. Arnheim pursues this impossible figure to its logical conclusion, defining an ideal radio announcer/personality as a "bodiless" voice.

It is very significant that certain expressive voices do not strike the naive listener as "the voice of somebody one doesn't see" and whose appearance can be speculated on, but rather transmit the experience of an absolutely complete personality.
(Arnheim 1936, p. 142)^[31]

In a chapter entitled "In Praise of Blindness: Emancipation from the Body," Arnheim argues that the recorded or transmitted voice has been

"purged of the materiality of its source" (ibid.). In broadcasting as in recording, "Resonance is eliminated, out of a very proper feeling that the existence of the studio is not essential to the transmission and therefore has no place in the listener's consciousness. . . . The listener rather restricts himself to the reception of pure sound, which comes to him through the loudspeaker" (ibid., pp. 143, 142). The notable lack of spatial signifiers described in Arnheim's ideal radio broadcast enables him to construct the disembodied voice as an infinite one (comparable, as will be argued, to the infinity defined and filled by the voice of the mother). Not only does Arnheim wish to eliminate our awareness of space, but also of the body—and, consequently, gender.

Roland Barthes contradicts Arnheim by insisting on the voice's relation to the body, the voice as physical signifier. For Barthes, rather than being "pure sound," the "grain" of the voice signifies, first, the body. "The 'grain' is the body in the voice as it sings. . . . I am determined to listen to my relation with the body of the man or woman singing or playing and that relation is erotic" (Barthes 1977, p. 188; italics added). It should be noted that Barthes is dealing exclusively with recordings of singers and not with live performance. The possibilities of ownership, repetition, and control increase the potential identification of the recorded voice as object and fetish.

Secondly, for Barthes, the voice exists in a constant, negotiated relationship to language. What he calls the "grain" is not merely a physical trace but "the very precise space . . . of the encounter between a language and a voice " (italics in original). Barthes aims to describe the effect of listening not to "the whole of music but simply . . . a part of vocal music." However, what he says regarding the space between the voice and language holds equally true for speech. Inasmuch as the speaking voice is involved in producing both language and sound, "the grain of the voice" is involved in "a dual production—of language and of music" (Barthes 1977, p. 181). When recorded, the voice, so often lost track of in the attempt to capture the meaning of the sounds articulated, reemerges, becomes a capturable object, a source of pleasure separable from its function within the symbolic field.

In Barthes' account, this awareness of the sound of the voice already creates a space or distance between the voice and language. Referring to Julia Kristeva's work, he differentiates between her concepts of the "pheno-song," which covers "all the features which belong to the structure of the language being sung. . . . In short everything in the performance which is in the service of communication," and "the geno-song," which

is the volume of the singing and speaking voice, the space where significations germinate "from within language and its very materiality"; it forms a signifying play having nothing to do with communication, representation (of feelings), expression; it is that apex (or that depth) of production where the melody really works at the language—not at what it says, but the voluptuousness of its

sounds-signifiers. . . . It is, in a very simple word but [one] which must be taken seriously, the diction of the language
(Barthes 1977, pp. 182–83)

—or, what might be called the sound of the sound.

In this elevation of diction (defined in the Random House College Dictionary as "the accent, inflection, intonation, and speech-sound quality manifested by an individual speaker"),^[32] Barthes goes so far as to differentiate between consonants and vowels. Consonants, in Barthes' reading, are on the side of the symbolic, setting limits, symbolizing restriction: "always prescribed as needing to be 'articulated', detached, emphasized in order to fulfill the clarity of meaning " (italics in original). Vowels, on the other hand, partake of the geno-song, encouraging the listener (as well as the singer or speaker) to dwell on a sound without limits: "There lay the 'truth' of language—not its functionality (clarity, expressivity, communication)," but instead its status as "pure" sound, as the place where one can discover the body in the grain of the voice. When sung properly, "the range of vowels received all the significance (which is meaning in its potential voluptuousness)." Barthes' preoccupation with voluptuousness reenforces the connection with the erotic—located in a specifically Oedipal place. He states, "Isn't it the truth of the voice to be hallucinated? isn't the entire space of the voice an infinite one?" (Barthes 1977, p. 184).^[33] "Infinite space," as we shall see, is frequently invoked as a primary (fantasized) characteristic of the mother/infant bond, symbolized by the all-encompassing maternal voice. By invoking the fantasy of a boundless space, Barthes reintroduces the figure of the mother.

Mary Ann Doane also describes the voice as being at once language and that which exceeds language and leads us back, again, to the mother.

The voice thus understood is an interface of imaginary and symbolic, pulling at once toward the signifying organization of language and its reduction of the range of vocal sounds to those it binds and codifies, and toward original and imaginary attachments, "representable in the fantasm by the body or by the corporeal mother, the child at her breast."
(Doane 1985b, p. 171 [italics in original]; quoting Rosolato 1974, p. 86)

What Barthes finally wants is a way to theorize the effect of the voice of the other on the listener, to give an "impossible account of an individual thrill that I constantly experience in listening to singing," and "to succeed in refining a certain 'aesthetics' of musical pleasure." Ultimately this new theory of listening "will certainly be individual . . . but in no way 'subjective' (it is not the psychological 'subject' in me who is listening; the climactic pleasure hoped

for is not going to reinforce—to express—that subject but, on the contrary, to lose it)" (Barthes 1977, pp. 181, 188).

And one way to "lose" the subject, to flee the symbolic, the letter, the Law, is in pursuit of the presymbolic. While it is not actually possible to regain access to a condition before language, it is a frequent characteristic of cultural production to appeal to its vestigial traces, the lingering nostalgia for that time. Barthes' difference with Arnheim rests on Barthes' insistence on (1) positing a male subject and (2) wanting it both ways—wanting an infinite space inside the voice of the mother and an erotic relation to the "grain" of the voice, which always points back to the body. This seeming contradiction grows out of the fact that our construction of a presymbolic condition is "superimposed upon infancy from a subsequent temporal and spatial vantage" (Silverman 1988, p. 74). The fantasy of the maternal voice is invented in a post-Oedipal present and imposed on the past. Barthes wants to recapture the prelinguistic state where difference is abolished while keeping the erotics—the desire for the mother as woman, as difference—possible only for a post-Oedipal male subject.

As psychoanalytic theory has been of great significance to feminist film theory and to work on sound, it might be helpful now to briefly outline the Lacanian scenario of subjectivity to which I shall be referring throughout this work.

According to Lacan, the mirror stage provides the child's first awareness of itself as separate from its mother, the world, and, most crucially, itself. The progression toward subjectivity requires a series of separations and losses. What becomes separate is objectified (the mother, the figure of the Other onto which the child projects imaginary attributes, whatever is external) as the child defines what is internal (its self) in terms of its separation from these objects. As Silverman succinctly puts it, "to the degree that the object has been lost, the subject has been found" (ibid., p. 7). With the accession to language, the subject is subsumed into the symbolic. However, as the subject has been constructed around this "splitting," being cut off from what it was, it is haunted by a sense of loss, absence, or lack. "The object thus acquires from the very beginning the value of that without which the subject can never be whole or complete, and for which it consequently yearns" (ibid.). As I shall attempt to show, the process of listening isolates, intensifies, and crystalizes the subject's relationship to the other (present through the medium of the voice). Thus, the listening situation can become a crucible where the subject works through its relationship with the other and, in doing so, momentarily, and pleasurably, reconstructs a sense of wholeness.

As an invention, the phonograph provided unique new ways of listening and dramatically different relationships to the voice. Given the theory cited above, it is possible to posit a psychological reason for the shift away from

home recording. The voice heard during playback is always the voice of the other—crucially, even when it is the listener's own. With home recording, the subject is forced to confront his/her voice wrenched from its internal echo of presubjective plenitude. Exposed, the subject's voice becomes comparable to the numerically limitless, but inherently less important , voices of others. It sounds too high (because heard for the first time outside the resonance of the cranium). It may sound monotonous and uninflected. It frequently sounds more like the voices of family members than oneself. The splitting of the subject is made unnervingly concrete when one is confronted with a playback of one's own voice. It is seldom a source of pleasure. Jean-Louis Comolli identifies the extreme care taken to preserve synchronization of the actor's voice and image in film as the sign of ideology at work, with dis-synchronization "scandalizing" one's sense of mastery and ownership (see Doane in Weis and Belton 1985, p. 58; Comolli 1972). How much more of an affront dis-synchronization must be when the technologically externalized, "objectified" voice is one's own.

With the sound reproduction industry's shift to marketing prerecorded phonograph records (a move that effectively replaced or eliminated home recording),^[34] the voice became openly and unproblematically that of the other, requiring a less drastic adjustment or fragmentation of the listening subject, and as such, could be substituted for other voices including, most crucially, that of the mother. The mother's actual voice is displaced to the realm of the "external" and "otherness" following the accession to language.^[35] Through displacement, it becomes possible to substitute the voice of the other for the powerfully specific voice of the mother. Listening thus becomes an intimate affair, infinite in promise because keyed to an ancient desire, without calling to mind a possibly overwhelming and ambivalent relation to an actual parent.

In discussing what Silverman calls the "fantasy of the maternal voice," it becomes clear that the nature of the subject's relationship to the maternal voice is spatial. Rosolato's famous description of the mother's voice as a "sonorous envelope" surrounding the child identifies the voice as a place, a place to be, a space that is at once everywhere and nowhere.^[36] The fantasmatic condition of existing within a place defined by the mother's voice can assume the form of either a terrifying miasma of non-meaning, from which the incipient subject struggles to escape through language, or a paradisiacal wholeness in which the child is blissfully united with the Mother. (Chion 1982 is Silverman's main example of the negative version of the fantasy.) The speaking subject recreates the mother's voice as "either cherished as an object (a) —as what can make good all lacks—or despised and jettisoned as what is most abject, most culturally intolerable" (Silverman 1988, p. 86).

Sound reproduction, as I shall attempt to illustrate, is particularly suited to the fantasy of submission to the imaginary because of its unique ability to create a position for the listener as isolated, surrounded by sound but alone,

and then to obliterate the subject's sense of alienation by re-merging subject and other. And it is "the fantasy of the maternal voice" that returns again and again as the model and locus of this imaginary operation.

Unlike in the mirror stage (where "here/there" and "internal/external" become clearly marked categories, because we can hear and speak at the same time), it is difficult in this phase, Silverman asserts, "to situate the voice, to know whether it is 'outside' or 'inside.' . . . Rosolato suggests [that the voice] can spill over from subject to object and object to subject, violating the bodily limits upon which classical subjectivity depends, and so smoothing the way for projection and introjection." In Silverman's view, "the child's economy is organized around incorporation" and "what is incorporated is the auditory field articulated by the maternal voice. The difficulty of establishing distance from the voice of the mother is suggested by its inclusion among Lacan's lost objects, or objets (a) , "those objects which are first to be distinguished from the subject's own self, and whose 'otherness' is never very strongly marked." The subject's difficulty in distinguishing between internal and external voices results from (1) the "double organization of the vocal/auditory system" (which has implications spanning the range of sound reproduction both in and outside cinema), and (2) attachment to the objet (a) , whose "loss assumes the proportions of an amputation," motivating fantasies of plenitude in regard to the voice, fantasies of recapturing and owning it as a fetish, since "once gone, the [lost object] comes to represent what can alone make good the subject's lack" (Silverman 1988, pp. 79, 80, 85).

The acoustic confusion of internal and external is not linked to the woman's voice per se. As Barthes illustrates, the voice we capture on recordings and choose to luxuriate in may be the most masculine basso profundo. If the "otherness" of the maternal voice, as objet (a) , is "never strongly marked" neither is its "difference." The "feminine" quality of the maternal voice, it could be argued, is one of the aspects projected onto it at a later stage, after the recognition of difference. The relation to the voice can be described in male/female terms, but the privileging of the female voice in the "fantasy of the maternal voice" is culturally conditioned. As Nancy Chodorow points out in The Reproduction of Mothering , "mothering" and "maternal" behavior are defined by a group of actions and are not inherent or gender-based.^[37] By trying to occupy a fantasmatic space where the subject merges with the "object" or the "other," where "the boundary separating exteriority from interiority is blurred by [an] aural undecidability" (Silverman 1988, p. 79), the subject, whether male or female, seeks to escape or forfeit gendered subjectivity in an attempt to merge with an undifferentiated other.

The desire for non-subjectivity precedes issues of gender (the "symbolic castration" of the accession to language preceding the awareness of sexual difference) and manifests itself clinically in hypnosis. In the mirror stage, vision abets language and the voice in inscribing separation of self and other.

Sound, especially as discussed here in reference to recordings and radio, can provide the illusion of repairing the split by reincorporating the other. The physical markers of hypnosis (the trance, insensitivity to pain, etc.) locate and reveal the symptoms of hysteria (the literal embodiment of psychic crises) in everyone as the body speaks an impossible desire. Hypnosis can be seen as clinical evidence of a persistent and unending confusion/confusability between inner and outer space, and as a sign of either gender's lingering, inherent desire to fuse with the voice of the other/mother, to dissolve what Lacan calls the "opposition between language and the phenomenal realm, . . . between meaning and life" (ibid., p. 8).

The confusion between subject and object in hypnosis is reiterated in clinical practice. Lawrence S. Kubie and Sydney Margolin have described the dissolution of the subject in response to the induction of a hypnotic state: "Once the subject is going 'under,' it is only in a purely geographical sense that the voice of the hypnotist is an influence from the outside. Subjectively it is experienced rather as an extension of the subject's own psychic process" (Kubie and Margolin 1944, p. 612). This condition is achieved through the classic methods of having the immobilized patient concentrate on repetitive visual stimuli (such as a metronome or swinging watch) as the hypnotist repeats various phrases.

This simultaneous restriction both on the motor and sensory side reduces to a minimum the variegated sensory contrasts upon which Ego boundaries depend.

It is the dissolution of Ego boundaries that gives the hypnotist his [sic ] apparent "power"; because his "commands" do not operate as something reaching the subject from the outside, demanding submissiveness. To the subject they are his own thoughts and goals, a part of himself.
(ibid., pp. 614, 612)

Such a condition is not restricted to classical hypnosis alone. Kubie and Margolin state that "such an obscuring of Ego boundaries, so dramatic in its manifestations when it is total, is frequently encountered in normal psychology as a partial phenomenon." In fact, their description of the efficacy of the techniques of the early hypnotists suggests the fundamental underlying appeal of listening to recorded sound through headphones. "The melodramatic maneuvers of the old-fashioned hypnotist" helped "concentrate [the subject's] attention on one field of sensation and to withdraw attention from all others " (ibid., pp. 613, 617; italics in original).

Hypnosis illustrates how the isolation of sound (and in cinema, sound and image) can return the immobilized spectator to the infinite space before the internal/external, self/other distinctions achieved with subjectivity. "This dissolution of the Ego boundaries creates a psychological state which is analogous to that brief period in early infancy in which the mother's breast in the

mouth of the infant is psychologically a part of that infant far more than his own toes and hands, as much a part of the infant's Ego as is his own mouth" (ibid., p. 612).

The relinquishment of subjectivity, of the "I" as distinctly separate from the other, allows the individual to recover or recreate a pre-linguistic condition where s/he existed in an infinite space, united there with the mother and the body, objets (a) that the recorded voice momentarily returns to us.