Poem as song: the role of the lyric audience.

The article examines the phenomenon of lyric formalism--the view that poems wholly contain their meaning--from cultural and cross-cultural perspectives. It argues that the view presenting lyrics as pure self-contained expressions, not addressed to anyone, is part of a long cultural history that began in Romanticism and that led to the New Critics' formalism. It is culturally specific and must be studied as such. Through a reading of some key Romantic-era statements on the lyric by Wordsworth, Shelley, and Hegel, this article shows the increasingly problematic status of the lyric addressee as a cultural notion. On one hand the addressee was important as the beneficiary of the poet's genius, but on the other hand s/he was neglected as non-essential to the truest form of self-expression. Ultimately the lyric addressee was repressed, though never entirely. Since poems were not regarded as addressing anyone, they were not meant to directly communicate meaning from speaker to listener; meaning was rather generated somehow within the listener. What the listener received, then, was only the form or music of the poem which triggered his own inward responses. Thus thought and music were split off from each other in a way that did not happen in other poetic traditions, like that of Arab poetics. In modern Western culture, poems were divorced from songs in both the popular mind and in high literary theory. Song became regarded as opposed to communication, and the poem as pure thought or text without a performative framework. This segregation of song from poem, music from text, must be acknowledged as culturally specific and belongs to a certain literary period. A glance at poetry within Arabic culture offers other alternatives, where the musical dimension is not contrasted to the textual, but is joined to it.

**********

Lyrical formalism--the view that poems are complete aesthetic units that wholly contain their meaning, as a vase contains flowers--is sometimes viewed as if it was imposed on poetry by the Russian Formalists and the American New Critics. There is a common impression that critics like Cleanth Brooks ripped poetry arbitrarily out of its personal, cultural, and historical context and stuffed it into their "well-wrought urns," detached and self-complete. (1) But it may be that lyrical formalism in criticism came to reflect an already developing formalist tendency in literature generally, a tendency towards textual self-containment originating with the Romantics and pushed farther by the Modernists. (2) Reader-response theorist Jane P. Tompkins argues that the modern emphasis on the literary meaning of a text (its self-contained "message"), unlike the Classical or Renaissance emphasis on its social effects, implies that the inter-personal relations of author to audience became less important in the modern age. (3)

Similarly, orality theorist Walter J. Ong sees the formalist tendency to regard texts as containing their meanings, rather than delivering the meanings of a writer to a reader, as a result of the turn towards mass literacy and thus to private reading. (4) Ong says that written texts, both literary and not, were increasingly regarded as what he calls independent "closed fields," cut off from an immediate awareness of authors, means of distribution, modes of performance, and audiences. While poetry was the literary genre in which the closed field was most emphasized by New Criticism (perhaps because its short span could be isolated more completely), the closed and decontextualized text according to Ong was the norm across all genres, literary and non-literary. Accordingly, neither Tompkins nor Ong focuses on the lyric genre individually.

But formalism is tied so closely to the lyric genre in both critical and pedagogical contexts that we must look beyond Ong's and Tompkins' theses. Literature professors are more likely to assign close readings of a poem rather than an excerpt of equal length from a novel, and most modern western readers--even those who have never heard of formalist criticism--feel lyric to be more closed-off from its readers. Thus there must be special reasons why this particular genre, the lyric, was singled out as the literary closed field par excellence.

The aim of this essay is to follow Ong and Tompkins, but with a special emphasis on lyric formalism from the perspective of its structure of address, its audience relations. The notion of lyric address, as we will see in an examination of key Romantic manifestoes on lyric poetry, was increasingly denied in modern theory, even if it remained institutionalized in practice. Much more than novels or plays, poems became imaginary emblems of purely subjective experience outside of discursive performance contexts, outside address networks. Rather, they became viewed as pieces of thought without a performative effect in public life, held tight in the urn of the self or of the printed page until they are spiritually infused into the reader. The argument here will be that, along with poetic address, the whole dimension of poetic performance was repressed, so that the outwardly performed song was divorced from the inwardly conceived poem as it had never been before. Correspondingly, the song as performance became downgraded to something that affects the hearer, but does not communicate anything to him. How did this happen'? And what are the effects today of this divorce of audience-addressing song from the non-addressed poem? This essay will examine the western lyric genre as a modern cultural creation, and the cultural processes that split off song from poem. For contrast it will look at the place of the lyric in one very different non-western tradition in which song is regarded as a close sibling of poetry rather than its unworthy poor cousin, in order to start regarding lyric not only as closed formalist text, but as performance too.

Studies of the history of poetry tend to agree that, before the Romantics, it was highly unusual to think of a poem as unaddressed, without addressee. Poetry then still had a strong affiliation with oral forms that presupposed a "you" listening to the poetic "I." The influential studies of oral poetry by Albert Lord show how early poetic recitation was above all a means of preserving communal memory and convention for "us," rather than of advancing individual vision for "me." (5) Far from showing the epic singer alone on a mountaintop, Lord's research demonstrates the radical dependence of the bard on audience reaction, arguing that this was the model for the epic and folk poet from ancient times up to the rise of print. Similarly the medieval troubadour and minnesinger were not expressing themselves only for the inner joy of it, but were also taking part in ceremonious contests and courts of love subject to audience evaluation. The addressed audience was essential to the lyric event. As late as the Renaissance, according to David Lindley, the lyric was not regarded as a "voice talking to itself," as T.S. Eliot described lyric in the modern age, but was rather almost always what Lindley calls a "directed performance": words addressed to a listener with a certain purpose in mind. In Philip Sidney's An Apology for Poetry, the love poem is not just a well-wrought set of images evoking a poet's erotic impulses, but is spoken to the lover with the aim of "persuading" her. (6) Such a poem could thus be conceived as a rhetorical event, a directed performance hitting the audience with a performative force less overt than the older bard performances, but just as audience-dependent. If rhetoric is the art of persuasion through words, and if affecting the audience is a form of persuasion, then we could point to a whole dimension of lyrical performative rhetoric that is only imaginable when the possibility of poetic address is acknowledged. The pre-modern, pre-Romantic poet was not just speaking something that the audience may or may not hear, not just singing to himself, but was performing a lyric address of the audience, talking to them as much as any storyteller, in a poet/audience collaboration.

But as Ong shows, poetic address was rejected by the Romantics and the Modernists after them who saw the lyric as the genre of subjectivity freed from social determinations. They projected back onto literary history their own growing sense that poetry should be not a speaking to others, but an internal process of self-exploration by and for the self. (7) In a sense this was the ideal of Romantic literature generally. As older models of literature as the passive mirror of reality were replaced by new Romantic models of literature as lamp shining its own light or dreamlike vision out onto the world (to borrow M. H. Abrams' famous metaphor), (8) the status of the literary addressee became questionable. If literature is basically dreamlike, then how can one truly address a literary work to anyone else, any more than one can address a dream to an audience? Lyric in particular, as the least context-bound genre, became the emblem of pure vision beyond social and historical pressures, the genre of self-reflection and self-discovery in which the addressee is essentially the speaker talking to himself. Yet in fact the addressee did not entirely disappear either in theory or in practice. Rather, the Romantic lyric remained a paradoxical form of address that, one could say, repressed its own addressing. In much Romantic political verse, for example, the poet transcendentally "speaks to" a new age, a new society, a utopia, or an idea, without pragmatically speaking to any actual addressee from the old society who might hinder or question his vision. He speaks to the entire world, and also speaks to no one in particular. Similarly the common Romantic apostrophe ("O ...!") could be seen as a symptom of this repression of lyric address. Apostrophe maintains the formal importance of a structure of direct address, but while emptying out the addressee, which is often inanimate or otherwise unable to answer back (as when Shelley addresses a skylark). Again, this is somewhere between address and non-address: Romantic apostrophe affirms the convention of poetic speaking-to-someone-else, even as it empties that same convention.

Similarly ambiguous as address is the Romantic genre of the song or ballad, which was reaffirmed as a vestige of older pre-industrial communal life, with the poet as bard speaking to his community, but which in fact became something meditative and private, very unlike traditional ballads. Romantic narrative poems with balladic content imitated older ballads not as a performative oral event (poet addressing song to audience in a performance), but rather as a set of formal qualities (rhythmic regularity, repetitive refrains, fluidity, enchanting or lulling qualities). Keat's "Eve of St. Agnes," for example, with its balladic theme of star-crossed lovers and midnight escape, is unimaginable as a performed event: it is not meant to be addressed orally and theatrically, but to be read privately and reflectively. Thus removed from its oral discursive context, and placed in a context of self-reflexiveness and subjectivity, the form of song could be figured as a song to oneself: an occasion for thought. Its song-like qualities would not be used musically to enchant a gathered audience, but rather internally to enhance the sense of private fantasy. The privately read, unsung song became an emblem of the self's power to transcend its environmental constraints through literary experience--yet this was precisely the genre that retained all the marks of its earlier public oral performance status. Thus, like Romantic apostrophe, Romantic song both ejected and indirectly reinstated the addressee. More generally, the paradox of the absent-yet-present addressee was part of the larger Romantic paradox about the self-centered genre of the lyric being used for political and social-visionary ends. How could a genre that was becoming less and less interested in its audience as addressee also claim that it had the power to change the world of that same audience? Such questions about the absent-yet-present lyrical addressee point to a very deep crisis in not just Romantic poetry but modern poetry too, which has inherited it, as we will see. Modern lyric has been basically formulated as a paradoxical address that does not address, oriented to an audience that must not really be there to hear, but that also cannot be fully denied.

One can examine this uncertain status of the lyrical addressee in detail in one of the earliest and most influential Romantic statements about the lyric, the Preface that Wordsworth wrote in 1800-02 to the Lyrical Ballads, published in 1798. This essay has much to say about the relation of poet to reader, yet in other ways cancels the reader out entirely; thus it reveals the radically confused status of the modern lyric as non-addressed address. On one hand, Wordsworth's Preface advances interpersonal address as a poetic ideal. Indeed at first glance one might think that Wordsworth was deliberately intending to carry over into Romanticism the earlier importance of audience address and communal orientation in oral-performative traditions of poetry as song. He says that he was convinced by "several of my Friends" that if readers better understood the philosophy behind his poems, those poems might he better able "to interest mankind permanently." (9) Thus the Preface itself is introduced not as an interpretive commentary on the poems, but as a social thing--an interpersonal bond between poet, friends, and readers. And Wordsworth makes these social relations central not just to this Preface or this volume of poems, but to poetry generally. "[W]hat is meant by the word Poet? What is a Poet?," he adds in the 1802 revision of the Preface--and the way he continues his questioning helps answer his question: "To whom does he address himself? ... He is a man speaking to men." The poet here is thus defined not in terms of his material or his inspiration or his style, but rather exclusively in terms of whom he addresses. It is not the poetry that makes a poet, but his audience. This is not to say that subject matter or style is unimportant in this Preface. But the ordinariness of the subject matter and the simplicity of style that made this work so revolutionary are explicitly and repeatedly justified as a certain form of address, in terms of the effect on the readers. Poetic simplicity is not an aesthetic virtue here, but a social one: "I have wished to keep the Reader in the company of flesh and blood, persuaded that by so doing I shall interest him" (244). The poem itself would seem nothing but the verbal instrument used to maintain the bond of interest between the reader and the poet. Far from Brooks' idea of the unaddressed poetic object, a well-wrought urn alone outside channels of discourse, Wordsworth's ideal poem seems a verbally performed social relationship, a connection between people--a legacy of the older audience-oriented sung-poetry tradition.

But in fact this is questionable, and the contrast between Wordsworth and the New Critics may not be as sharp as it first seems. The language Wordsworth uses to describe the interpersonal poetic bond becomes peculiarly formal at times in the Preface, and it is here that doubts about this bond arise. He says for example that "by the act of writing in verse an Author makes a formal engagement that he will gratify certain known habits of association: that he not only thus apprises the Reader that certain classes of ideas and expressions will be found in his book, but that others will be carefully excluded" (237). As before, poetry here appears as a social transaction: the poem is a sort of contract between poet and reader. Indeed soon after this he uses the verb "contract" explicitly. Expressing regret that many readers will feel he has broken this poetry-contract by writing such odd poems, he says that "I will not take upon me to determine the exact import of the promise which, by the act of writing in verse, an Author in the present day makes to his reader: but it will undoubtedly appear to many persons that I have not fulfilled the terms of an engagement thus voluntarily contracted." This acknowledgement of a contractual bond between poet and reader might seem to be strongly affirming that a poem is a form of address in which the receiver is essential. Just as there is no contract without two signers, so too may there be no poem without a reader as well as a speaker. But in fact it could be said that Wordsworth's contract language is suggesting the possibility of exactly the opposite view of poetry. It makes the audience a presence that is obligatory rather than assumed. The audience is becoming something that must be spelled out and contracted, rather than taken for granted. Earlier notions of poetry as address rarely considered the audience contractually bound, since it was obvious that a good singer would attract an audience. By contrast, Wordsworth's legalistic affirmations of the necessity of the audience raise questions about the status of poetry without such an audience. And when Wordsworth expresses fears that his poems might be perceived as a broken contract with the reader, he implicitly raises thoughts about a kind of poetry that does break that contract (as so much modern verse would proudly do after him). Expressing the oral idea of interpersonal poetic address in the terms of formal engagements and contracts, Wordsworth makes the poet-reader bond of address seem not more natural and essential, but less so.

Indeed it may be relevant that Wordsworth's most famous definition of poetry in this Preface ignores the audience entirely, confirming his modern role as pioneer of private poetic introspection. Wordsworth's poet may be a man speaking to men, but when he defines poetry as "the spontaneous overflow of powerful feelings" which are "recollected in tranquillity," he makes no mention of anyone hearing or receiving the poet's overflow or recollection. The word overflow, invoking an image of feelings like a liquid spilling over the edge of a fountain's basin--or of a New Critic's well-wrought urn--suggests that the dominant metaphor here for poetic composition is not the addressed or sent word, but the filled container. And a container does not have an addressee. The poem is not what is transferred or communicated from poet to reader; the poem is what flows out of the poet alone, This does not mean that the reader is absolutely denied, since there is nothing to stop the reader from soaking up the poetic overflow. The point is not that poetry is impossible to receive, but that receiving it is not central to this Wordsworthian view of poetry as overflow, as it was central to his earlier view of poetry as address. And when the audience does receive the overflowing poem, it is not by means of a direct transaction between poet and reader, but by a more indirect and passive process whereby something happens to the poet to make his feelings overflow, and the reader happens to be affected by them. Indeed, Wordsworth's phrasing throughout this explanation is a labyrinth of passive-voice verbs. He says readers' feelings "are modified" by certain thoughts when reading a poem, and certain habits of mind "will be produced," so that the readers "must necessarily be ... enlightened." It is as if he is doing all he can grammatically to avoid saying that the poet affects the reader: rather the poet is affected, and the poem is produced, and the reader is affected. This ambiguous phrasing evades the conflict of attitudes towards the audience in the Preface. Wordsworth, pioneer of literary modernity with its emphasis on the self communing with itself, seems also to be trying to preserve the older oral address of bards and minstrels in which the audience is taken for granted. Of course, that older bardic audience cannot be resurrected wholly: Wordsworth for example never refers to the poet singing for his neighbors, but only speaking to them. He takes pains to show that the best poetry, illustrated by an example from Gray, is always basically like prose. Thus he is trying to move away from song and from the older oral audience that expected musical performance as part of poetry, revising that audience while preserving its essential role. Yet at the same time he is flirting with the modern idea that poetry might be a forum for the pure expressive mind without thought of audience. This resulting tug-of-war between two notions of poetic audience is apparent, indeed, in the very title of the volume of verse. The older oral view of poetry as sung for an audience (ballads) is juxtaposed with another term from the newer German-idealist view of poetry as the expression of an overflowing selfhood (lyric). The accent is on performance and audience address in the first, and on subjectivity and self-expression in the second. David Lindley has noted the tension between "narrative" (balladic storytelling) impulses and "personal" (meditative) impulses in Wordsworth's and Coleridge's title for this volume, but the tension between self-expression and audience-address in the two words of the title is no less important. (l0)

This same tension is felt in Wordsworth's own poems too, especially the poems about poetry itself. While one cannot confirm an entire theory from a single poem, it is nevertheless suggestive to examine one of Wordsworth's best known poems representing a singer in Ong's "closed field" who is definitely not a man speaking to men, and who is the counter-model to his own poetic role. "The Solitary Reaper" (1805), depicts a girl working in the fields and singing not in address, but precisely to no one but herself, yet overheard by the speaker of the poem. He hears only the accidental effects of her singing--the melody--but not the content, so that no communicative message is sent. This may be because he is too far from her, or because, as he explained in a note, she is singing in Gaelic:

   Will no one tell me what she sings?--
   Perhaps the plaintive numbers flow
   For old, unhappy, far-off things,
   And battles long ago:
   Or is it some more humble lay,
   Familiar matter of to-day?

The mention here of epic or bardic oral material ("battles long ago"), and of ballad or "lay" forms, emphasizes how not at all bardic or ballad-like this poem is, since the girl has not been heard and her epic content not communicated. If Wordsworth's speaker had heard not just the melody or "numbers" of the reaper's song, but its communicative content as well, then it would be possible to say that an address had occurred. But as it is, there has been song without addressed meaning. Song has been detached from communication; the only communication that occurs here is what the poet communicates to himself upon not hearing the words of the song. Thus the failure to hear the singing girl is a poetic victory. If the poet had heard and recorded the girl's content, then Wordsworth's poem--and by extension much of modern poetry--would lose its characteristic self-reflexiveness and introspection.

The singer/poet rivalry in this poem reflects the same tension around the notion of audience that the title of Lyrical Ballads did, since here too the audience is both dismissed and secretly required. As audience of the girl, the poet is paradoxically both present and absent (after all, the poem is called "The Solitary Reaper," even though she was clearly not solitary at all, if the poet was there to hear her). He has it both ways: he both hears and does not hear the Reaper's song, hearing the sound but not the sense. This detachment of song from communicated meaning could be said to herald modern verse, in which the modern poet sets himself far away from communication, too far to hear it, and thus forcing each listener to generate thought from within. If in oral cultures the singer and the poetic communicator were one, now they were split into two separate and distanced figures, which modern poets were forced to choose between. Some poets pursued cognitive complexity---communication-- by increasingly avoiding "singing" and song forms (balladic refrains, heavy rhyme schemes, etc.), moving closer to the philosophical or witty vehicle of the epigram, as David Lindley shows occurring from Donne onwards. (11) Others chose to sing, but using the song form as an opportunity for verbal virtuosity that renounced communicative ease (the wordplay of Gerard Manley Hopkins, or the short but intellectually difficult songs of Blake and Yeats). The symbolist idealization of music as the aim of all the arts was a glorification of the non-signifying grandeur of music, which does not use signs, and thus does not communicate or address any signified meanings. Insofar as art was musical, it was self-enclosed: grandly non-communicative and non-addressing.

The eclipse of poetic address is perceptible only twenty years after Wordsworth's Preface, in the next major nineteenth-century manifesto of the lyric, Percy Bysshe Shelley's Defence of Poetry (1821), which makes almost no mention of the possibility of direct poetic address between people. Largely a Platonic argument that poetry has a divine capacity for illuminating human potential, the Defence seems based on the assumption that anything spoken by one human to another would be merely earthly and profane, outside the sanctity of the lyric. True poetry, while it may be felt, is too transcendent to be addressed to anyone--or, since poetry is like music for Shelley, too transcendent to be performed for anyone. As with Wordsworth's ballads that must not be sung, Shelley's poetry is music that must not be played for an audience. Thus Shelley describes the poet as a music-maker unaware of any audience he may have: "A poet is a nightingale, who sits in darkness and sings to cheer its own solitude with sweet sounds." (12) A poet writes above all for himself, not to end his solitude by addressing others, but to cheer the solitude he preserves. Even the audience who happens to hear this song is for Shelley unaware of the music-maker: "... his auditors are as men entranced by the melody of an unseen musician, who feel that they are moved and softened, yet know not whence or why" (16). Thus the audience does not know where the art is coming from, and the artist does not know where it is going. Neither sender nor receiver is aware of any transaction or sharing having taken place. And since birds have no verbal meaning to communicate, song here becomes, as in Wordsworth's "Solitary Reaper," primarily non-communicative. One cannot address a meaning to anyone in the language of birds. In his "To a Skylark," the song of the bird is inspiring not because it communicates thoughts to the listening poet, but precisely because it does not communicate thoughts--song, being void of thought, thus opens a space for free imaginative conjecture within the poet:

   What objects are the fountains
   Of thy happy strain?
   What fields, or waves, or mountains?
   What shapes of sky or plain?

These are questions, not statements and thus they merely point to the empty space where the communicated content of song could be. Thus the music of the song has reached the audience's mind (the audience being the poet hearing the bird), but the meaning has not reached it, and so is self-invented from within the urn of the mind, rather than communicated from beyond it.

Shelley's other musical metaphor for the poet in his Defence implies this split between poetic musicality and poetic communication even more strikingly. "Man is an instrument," he says, "over which a series of external and internal impressions are driven, like the alternations of an ever-changing wind over an Aeolian lyre, which move it by their motion to ever-changing melody" (4). Here the lyre is no longer the symbol of civic performance by a poet before an audience, as it was for the ancient Greeks. Shelley's lyre is not civic because its audience is not mentioned. This human lyre, played by the wind, is imagined without player or hearer. It is a strange musical performance indeed, with no performers on stage, and no audience in the hall. There is only the instrument, playing itself. This wild metaphor is consistent with Shelley's regular tendency to show poetry outside the bond of address between poet and hearer. He admits that poetry is received by listeners or readers, and he admits (though less often) that poetry is written by poets, but he seems curiously reluctant to admit that poetry is an address by a poet to a reader. In those places where he dwells upon the creative act, he makes the poem seem a testimony to itself rather than a message to others, as when he calls poetry "the record of the best and happiest moments of the happiest and best minds" (50-51). The word record, unlike display or transmission, does not necessarily imply an audience: the mind is recording itself, not necessarily delivering that record to someone who wants to hear it. An audience may happen to read it, just as police records may be read, but, as in Wordsworth, audience participation is made incidental to the poetic act, not essential.

Yet oddly this does not imply any neglect of audience reception, or of the power of poetic influence. On the contrary, Shelley's essay paradoxically emphasizes the power of poetic influence even as it neglects the notion of poetic address that seems the first step to any influencing. Even when Shelley is asserting that the audience does not know "whence or why" their pleasure comes from, he is also asserting that the transmission of that pleasure is morally essential:

   [T]hose who read [Homer's] verses were awakened to an
   ambition of becoming like to Achilles, Hector, and
   Ulysses... the sentiments of the auditors must have been
   refined and enlarged by a sympathy with such great and
   lovely impersonations, until from admiring they imitated.
   (16)

Thus the hearer may be strongly affected by a poem without needing to know, or care, where the poem came from. For it is not the poet who influences the audience here, but the poem that does. As with Wordsworth's passive verbs, Shelley repeatedly makes poetry, rather than the poet, the active poetic influence. Poetry "makes immortal all that is best and most beautiful in the world," poetry "acts in a divine and unapprehended manner," poetry "spreads a figured curtain," poetry "transmutes all that it touches." That is, it is not the poet who spreads a curtain or transmutes or acts, but poetry.

While the human poet seemed dehumanized into a lyre or object for Shelley, then the poetic object (the text) seems very much humanized in its force and moral effectiveness. Indeed more than humanized; poetry is given more moral force than even humans have. Shelley links poetry not just with human but with divine powers throughout the essay, as when he says that poetry "is indeed something divine" (48), that it preserves "the interpenetration of a diviner nature through our own" (51). Whose diviner nature is he indicating: the poet's, or our own, or some nature outside both? It is far from clear. Shelley similarly says that "we are aware of visitations of thought and feeling" (51), but does not specify whether these are our own feelings visiting us with a divine force, or whether the poet has visited us. The poet must clearly have something to do with it, since Shelley is speaking of poetry rather than private meditation. Yet the effect of the poet, in his broken model of poetic address, is reduced to a ghostly presence, as the spooky word visitation suggests. Perhaps Shelley is wary of calling the poet himself divine. In any case the force of the poem is made divine, and the poet who, despite having made the poem, recedes into a feeble faraway memory.

The poem for Shelley is clearly not Wordsworth's "man speaking to men;" for Shelley the poem speaks to men. This image of poetry's divine force, larger and higher than the man who wrote it, may be taken as an early sign of the Victorian cult of poetry as an autonomous spiritual force. Divinity can more easily be attributed by isolating the message from its earthly transmission. And yet the Victorian spiritualizing of poetry to a place beyond human address was just as paradoxical as it seemed in Wordsworth. As M. H. Abrams has shown, this spiritualizing "de-addressing" of poetry was brought about by the very poets who most believed that their new poetry could effect revolutionary changes in the human sphere. (13) The reception that was made essential to poetry in the big picture was downplayed, or even forgotten, in the smaller picture of men talking to men. One might say that address was sublimated. Both Shelley and Wordsworth aspired to write a prophetic poetry heralding a new age for humanity, yet in stressing the spiritual autonomy of this poetry they tended to neglect the notion of address that prophecy requires.

The Romantic split between music and communication through signs persists in modern poetry, often visible in the Modernist love of "difficulty" or communicative obscurity, accompanying the technical brilliance that seems a legacy of audience-directed performance. Poetry is often envisioned by the Modernists as a song whose meaning, however grand and impressive, is not definitely addressed or communicated, or even meant to be understood--as with Wordsworth's singer singing too far away and in Gaelic.

To take a canonical modernist example, Wallace Stevens' "The Idea of Order at Key West" shows art as a powerful force, ending with the "maker's rage to order words." But this power is not enacted through communication: the poem's representative artist is an unidentified singing woman who, like the Solitary Reaper, neither addresses her song to the hearer nor communicates any identifiable meaning or thought. Her song is purely sound and form and event; it is not at all communicative address. The only address in the poem occurs, significantly, at the end of the poem when the poet asks a certain Ramon Fernandez (identified by scholars as a French literary critic) to "tell me, if you know," why the singing ended and what the ending entailed. Address seems thus conceivable only in poetic commentary and interpretation, in conversation with a critic. It does not seem to take place in poetic creation itself, which is unaddressed and pushed out of interpersonal relations, even though the singer is present on the scene. The singer in Stevens' vision is, to return to T. S. Eliot's phrase, a "voice talking to itself," with music as the sanction of that self-address. Insofar as we take Stevens' poem as a modernization of Wordsworth's manifesto-like poem about how poetry works, it seems arguable that Wordsworth's portrayal of the alienation between song and communication has had a Modernist rebirth in Stevens' poem. Moreover, the recognition that communication can only occur among critics in the audience, and not between the singer and the audience, seems emblematic of Modernist difficulty generally, as if the Modernist poet became often the obscure singer herself.

This modern rendering of lyric unaddressed and uncommunicative became not just apparent in poetry itself, but institutionalized in Western criticism. By the middle of the nineteenth century John Stuart Mill could write, as if with Wordsworth's and Stevens' singers in mind, that "eloquence is heard, poetry is overheard." (14) Mills' poetic audience is at best an eavesdropper, an unessential kind of listener, even though clearly necessary. The same sort of paradox apparent in Wordsworth's and Shelley's manifestoes about the lyric audience were thus lodged in criticism. They had been lodged in philosophical aesthetics as well, by Hegel and his successors who had originally deeply influenced Coleridge and English Romanticism.

Hegel, mapping the literary genres largely in terms of metaphysical categories of self and other, put forth the influential doctrine that the lyric poet "internalizes" all outer being (audience as well as worldly subject matter) into inward expression. Hegel's doctrine that lyric is essentially linked to selfhood was deeply influential, and is still so, even despite attempts like Rend Wellek's to disprove once and for all that lyric is any more self-centered than the other genres. (15) And as lyric was tied to self, its link to the outer world and the audience was reduced. Unlike Hegel's epic poet who "disappears" before the "objectivity of his creation," thus keeping a bridge between self and object, the lyric poet for Hegel makes everything into a pretext for selfhood, "for self-expression and for the apprehension of the mind in its own self-expression." (16) Indeed the outward content matters so little that even trivial topics can be lyrical for him (1114). What about cases in which something external provokes a lyric, though, as when a poet is invited to commemorate a public event? Hegel says even these lyrics are inner rather than outward-relating: if such poemes d'occasion are successful lyrics, then they avoid falling "into dependence on the external stimulus and the purposes implicit in it," so that the poet "shall entirely assimilate and make his own the objective subject-matter. For the truly lyrical poet lives in himself" (1118-9). He thus ignores the possibility that the poet could continue to "live in himself" while addressing something outside him. Even when the poem is about a local hero whom the poet is praising, as in Pindar's in ancient Greece, Hegel still insists that the poet is not really about that hero, but about the poet himself:

   [W]hen Pindar was invited to sing the praises of a victor
   in the Games, or did so of his own volition, he still so
   mastered his topic that his work was not a poem about a
   victor at all, but was sung out of the depths of his own
   heart. (1130)

Again, the logic is revealingly odd: why is Hegel assuming that a poet singing "about something" could not also at the same time be singing "out of the depths of his own heart," as if outer reference could not coexist with inner subjectivity in a poem? Similarly the civic officials who commissioned Pindar's poetic act are an outward presence that Hegel must call inner in his theory. The point here is not to critique Hegel's appreciation of poetic subjectivity, but only to question why his idea of inner subjectivity is incompatible with the idea of audience and the whole external world, which he says the good poet must subordinate. For Hegel the lyric must only be about inwardness, so all the outward aspects of outside reference and address are, as with Wordsworth and Shelley, downplayed or dropped.

This means that song, for Hegel, can no longer mean what it used to mean. When for example he states that singing springs from the same source as lyric poetry, he does not mention performance-oriented singing before an audience, but rather what sounds more like singing in the shower. He says that "senseless gibberish, tra-la-la, singing purely for the sake of singing, prove to be a genuinely lyric satisfaction of the heart for which words are more or less a purely arbitrary vehicle ..." (1122). Why give this peculiar illustration of singing? Singing nonsense to oneself may of course be authentic singing, but why assume that this sort of singing is more "genuinely lyric" than singing for an audience? The question of why Hegel mentions this example of singing nonsense to nobody, rather than of singing comprehensible words to an audience, underscores the same detachment of song from communication, thought from address, that we saw in Wordsworth's and Shelley's treatment of poetic musicality. Like them, Hegel never says lyric cannot have an audience or be communicative. He says that communication between self and audience may occur: "self-expression may be a deeper urge to communicate the heart's deepest feelings ..." (1129). But this "may be" shows that communication is only a possibility here, not an assumption. As with Shelley and Wordsworth, address is optional for Hegel's lyric, but not essential; what is essential is self-expression alone.

There is a cultural dimension to this as well. Hegel's doctrine of lyrical subjectivity gives him a criterion for judging and ranking not only poems, but also entire cultural traditions of poetry. Having described lyric as self-expressive rather than other-addressing, Hegel criticizes certain lyrics, and lyric traditions, that fail to dwell sufficiently in the inner subjective realm, that fail to be sufficiently urn-like or self-contained. Among such failures Hegel includes what he calls, in gross orientalist categorizing, "the oriental lyric," which he associates mainly with Persian, Hebrew, and Arabic poetry (1148ff). The oriental lyric is weak, he says, because the oriental poet is "unable to find any firm support in himself," and thus the poem lacks a firm and central subjective inwardness. Instead of affirming the superiority of the inner over the outer, the oriental poet "cancels himself in face of ... substance and power ... he strives, ever unsuccessfully, to attain an association with it in his feelings and ideas" (1149). He is unsuccessful because he lacks sufficient inward ideas and feelings to associate the outer substance and power; he is a frail inwardness facing a too-powerful outside, and fails to attain "independence and freedom" from it. Unlike the European lyric's supposedly strong selfhood, the oriental lyric reveals to us "not the poet in his inner life ... but only his self-cancellation in face of external objets and situations." This self-cancellation, or "liberation from himself and from anything and everything single and particular," may become a kind of empty parody of selfhood, "a naive expansion of the inner life which easily loses itself in the limitless" (1150). Hegel says this lost-in-the-infinite self may remind us of human nothingness before the divine, which explains why oriental lyrics often sound like religious songs of devotion, with "the character of a hymn-like exaltation." But for Hegel this oriental exaltation is not appropriate to the highest glories of the lyric genre, since the lyric displays the self as central, while the hymn points to something higher than the self and beyond it. This is precisely the crux of Hegel's criticism of the oriental lyric overall: it does not recognize the poet's subjectivity strongly enough, and lets elements outside that subjectivity have too much power. What is really at stake in his cross-cultural survey of lyric, then, is the distinction between a self-contained voice and an outward-addressing voice, one that addresses either an external object or an external infinity.

While rejecting Hegel's orientalist judgment that Arabic poetic selfhood is weak, one might nevertheless take seriously his feeling that there is a major difference between the notion of selfhood in the Romantic European lyric and the "eastern" one. Without privileging as he does the supposedly strong detached self of European lyrics, one might understand him to be pointing out that certain non-European lyrics--like Arabic ones--acknowledge some external presence beyond their speakers. Hegel might be pointing, however crudely, to ways in which Arab lyrics present the sort of audience-addressing selfhood found in other literary genres and in common discourse. If the tendency of the European-style lyric after the Romantics was to encourage the tree play of isolated subjectivity, the Arabic lyric may have retained a more traditional pre-Romantic notion of "directed performance," an exchange between inward and outer. Such externally directed song may not be failing to achieve the supposed European autonomy of self, as Hegel judged, but rather may be not even aiming for such an ideal.

Indeed, there seems to be little evidence in Arabic literary history of any "de-addressing" impulse comparable to that of Wordsworth, Shelley, or the poets that followed them. Throughout the Arab poetic canon of criticism there seems to be little impulse to consider song--even intricately crafted song--as anything other than a communicative address between people. Song in this tradition, generally speaking, seems to have remained an idea of something "sung-to-an-audience," received and understood by an auditor whose role was always essential. If music was linked by Romantics like Wordsworth, Keats, Shelley, and Hegel to an exalted state of transcendent non-address where meaning arises from within rather than being communicated between people, musicality in Arabic criticism seems viewed very differently. While one cannot sum up the entire tradition of Arab poetics here, it does seem possible to say that, on the whole this tradition has tended to regard musicality as intensifying communication rather than impeding it, as with Wordsworth's or Stevens' singers. Music in this poetics is rarely, if ever, seen as opposed to meaning; usually it complements or fulfills meaning.

The song-like character of Arabic poetry, predominantly an oral performance, is demonstrated by the ease with which poems became songs in medieval Arab-Islamic society. "Virtually all major poets provided-intentionally or otherwise--lyrics to musicians," writes the historian of Arab literature Wen-Chin Ouyang, mentioning al-Isfahani's medieval compilation of tunes and poems, Kitab al-aghani. (17) This view seems to date from the very origins of Arabic verse. Adonis in his history of Arab poetics goes so far as to assert that the "pre-Islamic poem was like a song." (18) Adonis emphasizes the importance of tarab or "a state of musical delight or ecstasy" in this tradition, which he sees as based upon an "aesthetics of listening." (19) Yet this is not the Solitary Reaper's non-communicative song, listened to without understanding, but rather a song whose musicality enhances "its effectiveness as a means of communication." (20) Al-Farabi in his tenth-century Great Book of Music (Kitab al-musiqa al-kabir) places poetry and music in the same artistic genre of works governed by meter, pause, and rhythm. Here again, this assertion of the musical orientation of poetry in no way implied an evasion of clear poetic content or audience-communication, since Al-Farabi repeatedly links musicality not just with poetic effects, but with poetic speech. Even when a critic like Al-Jahiz could argue that pre-Islamic poetry was separate from thought, he still insisted that poetry must be understood, in grasp beyond thought, and thus communicated. (21) This link between communication and musicality survived the advent of Islam.

The transition to the new norms of the Qur'an brought of course a huge shift in worldview in which the older pre-Islamic verse was often denounced as sensual. But what is striking is how poetic musicality was not itself denounced, and how seldom verbal music (rhythm, meter, fluency) was rejected or criticized in the Islamic era. Certainly it was never attacked by any Muslim thinker in the fierce way that Plato attacked sensual music as interfering with clear thinking in Book II of The Republic. Al-Farra' in his meticulous exegesis The Meanings of the Qur'an (Ma'ani al-Qur'an, c. 800) discusses the musicality of the holy text by comparing it to the meters of pre-Islamic orality. (22) Rather, he takes it lot granted, like many other Arab Muslim theorists after him, that music enhances meaning. Later in the ninth century, Ibn Qutayba in The Problematic of the Qur'an (Mushkil al-Qur'an) followed him by similarly discussing the musicality of the Qur'an not as a merely formal aspect of verbal rhythm-making, nor as a distraction from meaning, but rather as reinforcing meaning. Thus he mentions rhythms and rhymes as among the features with which God chose to protect "the repository for Arab sciences" which is the Qur'an. (23) Music, in other words, protects knowledge and communication. Wen-Chin Ouyang details the ways in which poetry in medieval Arab culture was designated as a branch of the sciences, albeit a minor one. (24)

Indeed one could risk the generalization that a mutual support between musical meter and communicated knowledge is foundational in Arab criticism. Through the variations of Qur'anic commentary, one could trace a basic agreement in such different critics as al-Rummani, al-Khattabi, Al-Baqillani, Ibn Qutayba, and Al-Farra' on the presumed compatibility of music and communicative meaning.

Those theorists and poets who reject poetic musicality are scarcely exceptions to this rule, since those same thinkers tend often to reject communicative (rather than mystically interior) meaning as well. Adonis, modernist poet and defender of textual free play, tends to see the vestiges of oral performance in Arab verse to be stifling and anti-intellectual, and defends what he calls a less performance-based, and more Qur'an-inspired approach to verse. He finds this approach exemplified by medieval "modernist" poets like Abu Tammam and Abu Nuwas, and by the critic Al-Jurjani, who affirmed the free play of the mind over the musical tyranny of the ear, thus rejecting meter, which in his eleventh-century Indications of Inimitability (Dala'il al-I'Jaz) he describes as "of no concern to us." (25) But these turns away from musicality are also turns away from communicability according to Adonis, who calls them rather mystic self-reflections on knowing the unknown, records of the mind talking to itself. This sort of modernist introspection clearly has little need either for addressing others communicatively. The key is that Adonis does not embrace music as non-communicative mode, as Wordsworth and Stevens did. Downgrading communication does not lead him to an idolization of music as form, as in the modern West. Adonis's Arabic modernism thus offers no real counterpart to the pervasive modern western poetic separation of communication from musicality.

The maintaining of friendly relations in Arab cultures between music and communicated meaning may explain why Arab audiences do not share the modern Western prejudice that a song addressed to an audience cannot really be a profound poetic experience, even if musically impressive. Taking into account the extraordinary prestige of singers as poets in the Arab world, one can say that for the average Arab audience there seems to be no feeling that the audience-directed performance of song is in any way intellectually inferior to a poem. Indeed the notion of audience-directed, communication-intensifying poetic enchantment or tarab mentioned by Adonis, is, according to musicologist Virginia Danielson, still a common measure of praise for song among Egyptians. She reports the mainstream view that tarab is not a counterpart to communication but a necessary part of the communicative act, since "the words ... do not contain all the meaning," and thus only the voiced or sung words can be mobilized into full meaningfulness. (26) The greatest Arab singer of the modern age, Umm Kulthum, was quoted as affirming the necessity of communicating meaning to an audience in song: "The singer who does not articulate accurately cannot reach the heart of the listener and is not capable of perfect artistic rendition." (27) In this respect Umm Kulthum as singer may be usefully imagined as the direct opposite of Wordsworth's singing Solitary Reaper, whose poetic value as singer lay in the fact that her articulated words did not reach the heart of the listener (since only her melody did). Wordsworth's singer was an overflowing self oblivious to audience, a "formalist" singer, while Umm Kulthum was more like Wordsworth's inter-personal poet, a "man singing to men." As in oral poetic performances, her awareness of her audience altered the form of her text, when she repeated lines or sections of songs that elicited applause, or even stopped the song entirely when the shouts and whistles were particularly overwhelming. (28) The fact that the audience response is not considered external to her performance, but a part of it, may be signaled by the frequency with which that audience response is audible on virtually all her recordings. (By contrast, inclusions of audience response in Western recordings tend to be rare, limited to special "live" performances and thus marked as exceptional.) This audience response, again, was a communicative phenomenon, involving words articulated and understood. Thus the line "Give me my freedom, set free my hands" from "al-Atlal," which provoked huge outbursts of applause that "might bring the performance to a complete halt at these points," was associated with Palestinian and Arab struggles, or Egyptian oppression, depending on the time and context of performance. (30)

Considering Umm Kulthum as lyric performer addressing her audience may be a startling, but appropriate, attempt to bring the genre of lyric out of formalism, back into analyses of discourse networks that include speaker and addressee. The very study of song as lyric event, or of a lyric poem as performative event in a network including speaker and audience, forces the critic to look beyond the poem as a self-contained text. Remembering the song connotations of the very word lyric, which still in English refers to the words of songs sung to audiences, helps

remind us that poems are events--not just texts, or containers of meaning--as much as songs are. And this step would be profitable for the contemporary study of lyric, which has seemed resistant to anti-formalist approaches. It is well known that much literary theory of the last decades has been anti-formalist, in the sense of aiming to discredit the notion of the text as having a self-contained or immanent meaning.

Various schools of criticism from Structuralism through Deconstruction and afterwards differ mainly in how they propose such an opening-up. The tendency most recently, in the heyday of New Historicism and Cultural Studies, has been to open closed fields by expanding knowledge of context, of the practical discourse network of speakers, listeners, writers, publishers, and readerships in which the text is sent and received. The aim is to do so not merely to provide "background," as older literary studies did, but to understand the circumstances in which the texts had effects as social acts or performances. Literature on this view is always a speech act, doing things in a performative sense; the assumption is that words do things as well as mean things, in a culture as in a conversation. All this is fairly well known even to beginners in literary theory. But what is less acknowledged is how lyric poetry, as a specific genre, has been neglected by these developments in cultural and historical criticism. H. Aram Veeser has noted the way in which New Historicism has paid disproportionately great attention to the genres of the novel (especially the nienteenth-century novel) and drama (especially Renaissance drama), and has had little to say about lyric poetry. (29) Beyond a few scattered exceptions there has been virtually no counterpart in the study of poetry to the interesting New Historical studies of Shakespearean drama or the Victorian novel as social acts performed with social effects.

Poetry thus seems still stuck in the formalist urn, despite many critics' feelings that urns are out of fashion, and this may be due to the long European cultural history of seeing poems without addressees. Seeing the poem as outside direct discourse may imply that it is outside social context as well, thus hard to study as a New Historicist.

The more socially coded genres of novel and drama obviously profit from a simple supplying of historical or sociological context in order to understand their effects as social acts or performances. Those genres were addressing audiences openly anyway, so the New Historicist's task is simply to broaden and deepen one's understanding of this audience reception. With the generally less socially referential genre of lyric, this need is just as strong, but less obvious and different. The first step here is to fully acknowledge that the lyric is a form of audience address at all. In order to do this, one must in turn acknowledge that some of the very features focused on most carefully by the New Critic formalists--like meter, and patterned repetition--are clear vestiges of its addressed, performed, sung past that must be regarded as such. Thus in the lyric, more so than in the other genres, formal analysis may in this sense be inseparable from a study of performance. Emphasizing that meter and repetition (rhyme, for instance) are characteristics of song helps us recall that lyric non-address is only an effect performed through address, as we saw with Wordsworth's "Solitary Reaper" and Stevens' "Idea of Order at Key West." Denying the audience is not the natural stance of the lyric poet, but is rather one audience-addressing speech-act effect among others, to be studied as such. Denying performance is part of the performance, as much as acknowledging it is part of the performance in other lyric traditions. Just as Umm Kulthum's audience's response to "set free my hands" became part of her performance itself, as well as part of its textual meaning (giving a national political significance to the bondage theme, and so on), so too is any lyric poem's meaning inseparable from its own audience-addressing performance.

Remembering the musical basis of the lyric, the links to song not as marginal but as central to this genre, would help us remember that poems are always addressing someone, even when they pretend to be hiding in urns.

Notes

(1) The reference here is to Brooks' famous study of poetry as iconic or self-contained structures of meaning, The Well Wrought Urn: Studies in the Structure of Poetry (New York: Harcourt, Brace & World, 1947).

(2) Walter J. Ong has argued, for instance, that the beginnings of Brooks' New Criticism can be traced back to Samuel Taylor Coleridge. See Ong, "The Poem as Closed Field," Interfaces of the Word (Ithaca, N.Y.: Cornell U P, 1977) 213.

(3) See Jane P. Tompkins, "The Reader in History," Reader-Response Criticism: From Formalism to Post-Structuralism, Jane P. Tompkins, ed. (Baltimore: Johns Hopkins U P, 1980).

(4) See Ong, 213ff.

(5) Albert B. Lord, The Singer of Tales, Harvard Studies in Comparative Literature 24 (Cambridge: Harvard U P, 1960) 20ft.

(6) David Lindley, Lyric (London: Methuen, 1985) 63-64.

(7) Ong, 223 ft.

(8) See M. H. Abrams, The Mirror and the Lamp: Romantic Theory and the Critical Tradition (New York: Oxford U P, 1953).

(9) William Wordsworth and Samuel Taylor Coleridge, Lyrical Ballads: The Text of the 1798 Edition with the Additional 1800 Poems, R. L. Brett and A. R. Jones, eds. (New York: Barnes and Noble, 1963) 236. All references to Wordsworth henceforth will be to this edition.

(10) Lyrical Ballads, 71.

(11) Lindley, 9 ft.

(12) Percy Bysshe Shelley, "A Defence of Poetry," Shelley's Defence of Poetry: Browning's Essay on Shelley, L. Winstanley, ed. (Boston; London: D. C. Heath & Co., 1911) 16. All references to Shelley henceforth are to this edition.

(13) M. H. Abrams, "The Spirit of the Age," in Romanticism and Consciousness: Essays in Criticism, Harold Bloom, ed. (New York: Norton, 1970) 91-119.

(14) John Stuart Mill, "Thoughts on Poetry and Its Varieties," Disxertations and Discussions: Political, Philosphical, and Historical (New York: Henry Holt, 1874-82) I, 97. Quoted by Ong, 223.

(15) Wellek refers to the fallacy of a "supposed inwardness" of lyric experience that "never be demonstrated." Rend Wellek, "Genre Theory. the Lyric, and Erlebnis,'" in his Discriminationsv Further Concepts of Criticism (New Haven: Yale U P, 1970) 251-52.

(16) G. W. F. Hegel, Aesthetics: Lectures on Fine Art, Bernard Knox, trans. (Oxford: Clarendon, 1975), vol. II, 1113. All references to Hegel will henceforth be to this edition.

(17) Wen-Chin Ouyang, Literacy Criticism in Medieval Arabic-Islamic Culture (Edinburgh: Edinburgh U P, 1997) 78.

(18) Adonis, An Introduction to Arab Poetics, Catherine Cobham, trans. (London: Saqi Books, 1990)31.

(19) Adonis, 27.

(20) Adonis. 31.

(21) Al-Jahiz, Kitab al-bavan wa'l-tabyin (Cairo, 1960), vol. I, l l 8. Quoted in Adonis, 27.

(22) Adonis, 38.

(23) Quoted in Adonis, 39.

(24) Ouyang, 61 ff.

(25) Al-Jurjani, Dala'il al-i jaz (Cairo, 1969) 144. Quoted in Adonis, 45.

(26) Virginia Danielson, The Voice of Egypt: Umm Kulthum, Arabic Song, and Egyptian Society in the Twentieth Century (Cairo: The American U in Cairo P, 1997) 12.

(27) Al-Naqqash, Raja', "Liqa' ma'a Umm Kulthum," in Lughz Umm Kulthum (Cairo: Dar al-Hilal, 1978) 42. Quoted by Danielson, 139.

(28) Danielson, 147.

(29) Danielson, 147, 168.

(30) H. Aram Veeser, "Introduction" to The New Historicism, Veeser, ed. New York: Routledge, 1989).