General

 iv.4.1.At first glance, the title of this article might be surprising insofar as fiction and referentiality belong to different conceptual categories: while referentiality is concerned with the relation between language and the world, fiction characterises a certain kind of referential relation. Furthermore, literary theorists more commonly oppose fiction to factuality (Hühn 2014), in line with studies in narratology (Cohn 2000; Fludernik 2018; Schaeffer 2014). However, the specificity of the lyric raises questions that are different from those in narratology, and using the notion of referentiality makes this shift explicit. The tension between fictionality and factuality that operates in narratology cannot be simply transposed into the lyric, as this transposition fails to acknowledge the difference between narrative and the lyric (Categories of lyric). Rather than considering the lyric as a domain in which we can apply the results of narratology, focusing on referentiality suggests that the lyric operates at a different level, one at which the question of reference is not solved with the appeal to fiction, but where the referential relation itself is brought into question.

 iv.4.2.This chapter investigates the ways in which fiction and referentiality are problematised in the lyric. Does the lyric move away from representation and towards expression of the poet’s inner feelings as Hegel and the Romantics suggest (Hegel 1986)? If, following the Hegelian line, the lyric is opposed to narrative epic, can the effects of the lyric be grasped under the traditional scope of make-believe that is at play in fictional worlds (Walton 1993; Pavel 1989)? Is fiction still a matter of make-believe or ‘as if’ as in traditional narrative fiction or is it a matter of a broader ‘seeing-as’ that outgrows the aims and scope of fiction qua narrative? The question of referentiality in the lyric can be approached from three different perspectives: 1. Objects; 2. Speaker; 3. Reader. While most scholars focus on the question of the speaker, as we will see in section III, it is also important to consider the presence (or absence) of objective references to space and time in poems, as they can be considered the most basic form of referentiality, and to acknowledge the role of the reader in conceiving the referential relation.

 iv.4.3.Are references to places or moments fictional in the lyric? As Dominique Rabaté argues while discussing Lamartine’s Le Lac, the poet ‘voluntarily removes the Lac du Bourget from its contextual frame, making it Le Lac (›the Lake‹). The despecifying of autobiographical circumstances allows the elegy to attain to allegory.’ (Rabaté 2017, 95) The removal of reference operates an abstraction that shifts from the particular autobiographical experience to a general allegory. Reference does not disappear completely, as words still retain their referential values, but it is relegated to a second degree of importance. Relegating reference leads to further explore what fiction is opposed to and recent theory of the lyric use the category of ‘nonfiction’ to describe this opposition to fiction (Culler 2015). The question of reference highlights an opposition between a conception of the lyric based on representation (modelled on painting, ‘ut pictura poesis’) and one based on expression (modelled on song, ‘ut musica poesis’): is lyric a matter of showing the world or showing oneself?

 iv.4.4.This question of showing oneself leads to a central concern for many lyric theorists, namely the category of the speaker. Is the lyric ‘I’ fictional? If so, what is its fictionality like? The referential status of the lyric ‘I’ is indeed problematic. If we are to distinguish the lyric from narrative, it seems that its fictionality should be of a different kind. This difference in kind might lead to a difference in the effects of the lyric that will require further characterisation. Lyric fictionality might not follow the same suspension of reference that operates within narrative fictionality. As we will see in section III, most critics have respectively argued that the speaker is either fictional, factual, or even metaphorical. Does a metaphorical logic substitute reference by figuration? (Why) should we single out reference in lyric poetry?

 iv.4.5.One might think that the question of the fictionality of the speaker can be solved by calling on resources from autobiography theory, especially Philippe Lejeune’s autobiographical pact (Lejeune 1989). However, there are no set conventions regarding the fictionality or the factuality of the lyric ‘I,’ and many lyric poems (and poets) play on this indeterminacy of reference. Antonio Rodriguez offers a more convincing reading of the idea of pact in lyric poetry, suggesting that there are different ‘pacts’ that characterise intentional frameworks for literature: lyric, fable-like (‘fabulant’), and critical (Rodriguez 2003, 94–96). However, the lyric pact does not involve a question of fictionality, factuality, or referentiality.

 iv.4.6.While the questions of referential objects and of the speaker concern the production and nature of the lyric, we might approach the problem from the perspective of the reader. What is the relation between the fictionality of the lyric and the reader’s response? Is this fictionality relevant from a reader’s point of view? When encountering a fictional narrative, readers often have in mind its fictionality and suspend their judgment. However, the experience with lyric poetry seems to be somewhat different as the lyric ‘I’ is sometimes confused with the poet. Is the experience of lyric poetry degraded if it is fictional? Or to the contrary does fiction add something to the experience of lyric reading? If it is the case, this ‘something’ needs to be characterised as more than mere added information. These questions reveal the difficulty in conceptualising the relations between fiction and the lyric, especially since these relations vary across geographical and historical contexts, as we will see in the next sections.

The Subject in Contexts

 iv.4.7.The problem of fiction and referentiality in lyric poetry finds its origin in the Greek concept of mimesis and Plato’s rejection of poetry and other mimetic arts from his ideal city. While Plato marks mimesis with a negative seal, Aristotle inverts this evaluation and makes it one of the central notions for most poetics to come. In his Poetics, he insists on the positive aspects of catharsis in regulating the bodily drives of the audience. As this article focuses on the tradition that originates in Greek thought, it will mainly approach the problem of fiction and referentiality from within the framework of Western philosophy. It can be argued that the question of fiction in respect to lyric poetry is less prominent in other traditions of thought and François Jullien for instance shows that Chinese wen has little to do with Aristotle’s mimesis. In this context, the allusive value in Chinese thought escapes the question of fiction and referentiality (Jullien 2003). This allusive value can be more fruitfully related to questions of metaphor and figuration that reconfigure the notion of fiction in lyric poetry, as we will see in section III.

 iv.4.8.The Greek problem of mimesis finds its contemporary counterpart in the notion of fiction and some scholars like Gérard Genette or Jean-Marie Schaeffer even suggest translating mimesis by fiction (Genette 2004;Schaeffer 1989). Is mimesis fictional and does the lyric follow the lines of mimesis? In considering the relations between lyric, fiction, and reference, we must bear in mind how these notions evolve and shape one another across historical, geographical, cultural, linguistic contexts. Is the same experience at play in Ancient Greece, in Romantic Britain, or in contemporary explorations of the self? If the conceptions of fiction and referentiality change through time, how do their relations to the lyric evolve? We will see in the next section how poets and theorists rethink the question of fiction throughout recent poetic history.

 iv.4.9.By placing poetry on the side of mimesis, Greek philosophers have marked the history of the relation between poetry and fiction. Even though they are not yet talking about lyric poetry as such, further interpretations of Plato and Aristotle bring lyric poetry in relation to problems of mimesis and fiction. However, not all lyric poetry can be considered fictional, as at least one specific subgenre contests this fictionality: the historical lyric. As Cathrynke Dijkstra and Martin Gosman argue, crusade lyric for instance challenges the fiction-referentiality distinction: ‘the fiction of the conventional reality derives its referentiality from the fiction of a conventionally presented surrounding historical world, and vice versa.’ (Dijkstra and Gosman 1995, 14) According to them, crusade lyric (and to a broader extent historical lyric) can neither be said to be fully fictional insofar as it attests events from the world nor to be fully referential as the historical world to which it refers is itself a poetic construct. Historical lyric thus raises the question: what is the content of the historical reference? Dieter Lamping suggests that historical lyric offers a different or ‘higher’ kind of truth than historical science (Detering and Trilcke 2013, 62–75). In the specific case of historical lyric, it seems that fictionality must be excluded in order to understand the reference to the historical moments and conventions that the poem conveys, even though this reference is a poetic construct.

 iv.4.10.Another aspect that we must bear in mind is the performative reading of the lyric, that has become prominent in contemporary debates, especially in Jonathan Culler’s Theory of the Lyric (Culler 2015). This performative dimension reveals that the problem of fiction only occurs within uses of language that value referentiality. If we place ourselves in a framework in which language is not a matter of representing the world but one of acting in the world, the question of fiction disappears, as the performative effects of language are always real and escape the truth-false dichotomy. Recent work on African philosophy of language shows that many uses of language are less related to reference than to performance, thus highlighting the ritualistic dimension of these uses of language (Wiredu 2004; Hamminga 2005). If we consider language in terms of force rather than representation, we come closer to a performative conception of language in which fiction is not a problem: the power of the lyric becomes like that of a ritual. However, as Culler argues, the lyric is not reducible to a ritual and it ‘involves a tension between ritualistic and fictional elements’ (Culler 2015, 7). These ritualistic elements are moments where language becomes action and the ritual a speech-act. In relation to poetry, the question remains to define what kind of speech-act is at play and what are the intended effects. With this performative dimension, language aims to affect the world rather than merely represent it and, in the case of literature, it aims to affect the audience, as Northop Frye argues (Frye 1957). Marc Dominicy also considers that speech-act theory can be adapted to lyric poetry and that poetic utterances are therefore to be considered like other speech-acts (Dominicy 2019).

State of the Art

 iv.4.11.Inasmuch as the relation between fiction and the lyric change according to the context in which the lyric is used, the experience of poetry also evolves throughout history and the effects on the audience vary accordingly (Attridge 2019). These variations affect the relation between fiction and referentiality in the lyric: for instance, Philip Sidney considers that the poet’s task is to create a fictive world to convey a certain truth. For him, truth is not linked to a historical fact, but to the imagination of the poet. In his Defence of Poesy, Sidney acknowledges this question of falsehood and fiction in respect to lyric poetry (Sidney 2004). The centrality of truth for Sidney also pervades the Romantic programme. Indeed, Romantic lyric aims at uncovering a hidden reality, at unveiling a ‘higher’ truth.

 iv.4.12.However, even with this aim in mind, Romantic poets adopt radically different views regarding the relation between fiction and referentiality in the lyric. Wordsworth and Coleridge famously oppose one another on this topic. Wordsworth defends a conception of the lyric as factual in the Preface to Lyrical Ballads:

The principal object, then, proposed in these Poems was to choose incidents and situations from common life, and to relate or describe them, throughout, as far as was possible in a selection of language really used by men, and, at the same time, to throw over them a certain colouring of imagination, whereby ordinary things should be presented to the mind in an unusual aspect; and, further, and above all, to make these incidents and situations interesting by tracing in them, truly though not ostentatiously, the primary laws of our nature: chiefly, as far as regards the manner in which we associate ideas in a state of excitement. (Wordsworth and Coleridge 2013)

 iv.4.13.The object of lyric poetry is common life and the poet’s task is to present it through unusual (poetic) aspects. Poetry is therefore a matter of presenting ordinary life in a way that is not visible in ordinary life. To the contrary, Coleridge defends the famous idea of a suspension of disbelief in lyric poetry: ‘it was agreed, that my endeavours should be directed to persons and characters supernatural, or at least romantic; yet so as to transfer from our inward nature a human interest and a semblance of truth sufficient to procure for these shadows of imagination that willing suspension of disbelief for the moment, which constitutes poetic faith.’ (Coleridge 1834, 174) Lyric poetry only holds a semblance of truth and leads to poetic faith. There is no matter of reference anymore, but only an action, namely the suspension of disbelief that leads to enter the fictional world of the lyric.

 iv.4.14.As we can see, in the context of a referential conception of language, the tension between fiction and nonfiction revolves around the question of truth as correspondence. Pushing Coleridge’s idea to an extreme, Edgar Allan Poe considers that poetry has nothing to do with truth: ‘To recapitulate, then: — I would define, in brief, the Poetry of words as The Rhythmical Creation of Beauty. Its sole arbiter is Taste. With the Intellect or with the Conscience, it has only collateral relations. Unless incidentally, it has no concern whatever either with Duty or with Truth.’ (Poe 1956, 470–71) The only thing that matters in poetry is taste, and truth has no say in the matter. Whether a poem is true or not is not of central significance: what matters is the creation of beauty. Poe thus evades the question of fiction and reference by placing the emphasis on the aesthetic. This shift coincides with the philosophical idea that truth can be problematic, as Friedrich Nietzsche suggests that the question of truth becomes a philosophical problem with a history that goes back to the Greeks. Is truth a matter of simply corresponding to the world or is it a matter of revealing something hidden in the Romantic sense? Nietzsche eludes the question by raising the more fundamental one of the value of truth (Nietzsche 2008). Why are truth and fiction important to lyric poetry? Does the question matter only because the Greeks have raised it?

 iv.4.15.In contemporary literary theory, the answers to the question of the fictionality of the lyric can be schematically organised in three groups: 1. Lyric is fictional, 2. Lyric is not fictional, 3. Lyric evades the fiction/nonfiction distinction. In this categorisation, the question of the speaker (sujet lyrique, lyrisches ich) is of central significance. Indeed, most of the debates around the fictionality of the lyric revolve around the status of the speaker. Is the speaker fictional? Is the speaker factual? Is the fictionality of the speaker important in our understanding of lyric poetry? One of the reasons for the centrality of the logical status of the ‘I’ originates in the idea that the confusion between speaker and empirical author is much more common in lyric poetry than in narrative forms of literature. The speaker thus becomes a problem in differentiating between the utterance and the act of uttering.

Lyric is fictional

 iv.4.16.In line with Sidney, Coleridge, and Poe, a few literary scholars consider lyric to be fictional. Monroe C. Beardsley defends such a view and considers that ‘The writing of a poem is not an illocutionary act; it is the creation of a fictional character performing a fictional illocutionary act.’ (Beardsley 1970, 58–59) According to Beardsley, the speaker of a poem must not be confused with the poet themselves and this speaker is fictional. Beardsley distinguishes poetic utterances from ordinary ones, thus following J. L. Austin’s rejection of poetic utterances from his speech-act theory (Austin 1975, 121). Barbara Herrnstein Smith adds to this distinction by separating natural utterances that are ‘verbal responses of an historically real person, occasioned and determined by an historically real universe’ (Smith 1971, 264–65) from fictive utterances to which poetry belongs. As she pursues: ‘What I would like to suggest, however, is that all poetry may be so regarded, that we could conceive of as mimetic discourse not only the representation of speech in drama, but also lyrics, epics, tales, and novels.’ (Smith 1971, 268) Smith argues that lyric poetry is fictional because it is not made of natural utterances. Both Beardsley and Smith attribute the fictionality of the lyric to the specificity of poetic language. Lyric poetry is fictional because its language is not ordinary. As we will see, conceptions of the performativity of the lyric contest this view.

 iv.4.17.Another way to argue for the fictionality of the lyric is to consider that it participates of an aesthetic illusion. Even if it is in a restrictive way, Werner Wolf considers lyric poetry to participate in such an illusion: ‘My thesis is that lyric poetry does indeed show a certain resistance to such an illusion, but that there are at the same time generic features, centred above all on the “speaker” of a poem and his or her experience in a simulated present situation, which nevertheless allow the emergence of illusion, if only in a modified or restricted form.’ (Wolf, Bernhart, and Mahler 2013, 185) According to Wolf, lyric poetry is not different from other forms of literature, and it takes part in aesthetic illusion. In other words, lyric poetry plays the game of make-believe that is central to fictional forms of art, as Kendall Walton argues (Walton 1993).

Lyric is not fictional

 iv.4.18.In Logik der Dichtung, Käte Hamburger defends the opposite position, namely that the lyric not fictional. She opposes the lyrical to the fictional genre and considers: ‘that the manner in which we experience a lyric poem as literature is quite different from the way we experience a fictional, i.e., a narrative or a dramatic work of literature. We experience it as the statement of a statement-subject. The much-disputed lyric I is a statement-subject.’ (Hamburger 1973, 234) In respect to the system of language, Hamburger argues, the lyric ‘I’ is not fictional: ‘we only experience a genuine lyric phenomenon where we experience a genuine lyric I: a real statement-subject, which guarantees the character of lyric statement as reality statement regardless of whether or not this I names itself in the first-person form.’ (Hamburger 1973, 291) However, the nonfictionality of the lyric ‘I’ does not entail its factuality for Hamburger. The subject is real but not necessarily factual.

 iv.4.19.In contrast, Gérard Genette goes a step further by considering the lyric as factual. To do so, he moves from the question of fiction to that of diction:

In lyric poetry, utterances of reality-thus authentic speech acts-do occur, but these are acts whose source remains indeterminate, for the lyric ‘I’ is inherently incapable of being positively identified either with the poet in person or with any other determined subject. The putative enunciator of a literary text is thus never a real person, but either (in fiction) a fictitious character or else (in lyric poetry) an indeterminate ‘I’—thus constituting, in a way, an attenuated form of fictionality. (Genette 1993, 12)

 iv.4.20.The indeterminacy of the lyric I implies that it cannot be considered fully fictional. This indeterminacy can be considered from various perspectives. Laurent Jenny suggests conceptualising the relation between the lyric ‘I’ and the poet through the notion of figuration (Jenny 1996). Following this idea, the figuration of the lyric can convey a form of metaphorical truth, in the sense of Paul Ricoeur’s theory of metaphor (Ricoeur 1975). In a different direction, Thomas Pavel’s notion of degrees of fiction brings to consider lyric poetry as a form of reduced fictionality (Pavel 1989). More recently, a move towards speech-acts in the theory of the lyric reveals how the lyric cannot be said to be fictional from a speech-act point of view, as Culler argues in his Theory of the Lyric: ‘it would therefore be wrong to embrace for lyric a notion of performativity correlated with fictionality.’ (Culler 2015, 128)

Lyric escapes the fictional/nonfictional question

 iv.4.21.These positions suggest that lyric poetry is either fictional and involves a suspension of disbelief or nonfictional with varying degrees of what this nonfictionality means. An interesting position is that of Walton who moves from considering lyric poetry a mimetic genre to which the suspension of disbelief applies to considering it a kind of ‘thoughtwriting’ that resembles speechwriting (Walton 2011). However, a third option is available, namely that the lyric is not affected by the question of fiction. In their response to Walton, John Gibson and Hannah Kim hint towards such an idea by emphasising the centrality of experience in lyric poetry: ‘Lyric poetry, however, tends to present selves in a radically and intentionally diminished manner, and rarely is the I presented in a biographic sense. In fact, the I of lyric poetry is often nothing more than a center of perceptual, cognitive, and affective attention: the subject of an experience. It is a self effectively reduced to a perspective.’ (Gibson and Kim 2021 2021) This move from a high to a low degree of fictionality might reveal, from an external perspective, that the question of the fictionality of the lyric is ill-framed. Indeed, perhaps that lyric poetry escapes the question of fictionality, that whether we consider it fictional or not has no impact on the reader’s experience. The effects of lyric poetry apply whether it is considered fictional or not. Rodriguez suggests that the question of fiction is only one perspective for reading lyric poetry and that it can be read in multiple ways: fictionality is not necessarily opposed to factuality in the framework of lyric poetry and we must consider the lyric reading as actualising the potentialities of the poem, with varying degrees of fiction, diction, and figuration (Rodriguez 2012, 157).

Contemporary practices, methods, and debates

 iv.4.22.One kind of contemporary poetic practice that highlights the question of reference is documentary poetry. Michael Leong has shown that there is a turn to document in contemporary North American poetry (Leong 2020), but a similar statement is valid for other geographical realms as documents pervade poetry in other languages. The turn to document is especially strong in France, with Franck Leibovici’s explorations of ‘poetic documents’ (Leibovici 2007; Leibovici 2020) or Christophe Hanna’s notion of ‘poetic dispositif’ (Hanna 2010). This documentary turn affects the question of fiction and referentiality in lyric poetry, insofar as the document becomes a referential object. While it might seem at first glance that poetic documents are opposed to lyric poetry, insofar as they would erase the subjectivity at play in producing poetry, such a picture is too schematic. Indeed, there is a lyrical transformation of documents through the decontextualisation and recontextualisation at play in documentary poetry. How do these documents affect fictionality? Does the presence of an ‘objective’ document create an actual reference? Or does the context of the poem necessarily break the referential link?

 iv.4.23.A combination of the lyric and the documentary can be found in Muriel Pic’s Elegies documentaires and L’argument du rêve. In these works, Pic produce lyric poetry from documents and archives. She describes Elegies documentaires as ‘a lyrical, atmospheric, and elementary experience of documents’ (Pic 2016, 81) that provides a different reading of archives. In this context, it seems that the connection of the poem to the document is central. But the value of the poem is not referential, as it refers to a document that is included in the poem. It further plays on the self-referentiality of poetry to produce lyric effects. Such a form of poetry comes closer to historical poetry that constructs a historical world to which the poem refers. Similarly, Pic’s documentary poetry constructs a reading of an archive to which the poems refer.

 iv.4.24.This documentary poetry opens to different aspects of the social dimension of poetry. If the lyric aims to affect the reader, the question of fictionality becomes secondary. From a pragmatic perspective, the effects are more important than the ontology. Forms of political, social, ecological lyric do not function on the mode of the fictional, but rather display references to concrete situations. The aim is not to produce a fictional discourse but to affect the reader in a certain way. This pragmatic aim escapes the questions of fiction and referentiality.

 iv.4.25.In line with autobiography and autofiction (Bode 2019), one of the contemporary forms such social and political poetry takes is autotheory. Combining documents, literary and theoretical references, and autobiographical elements, autotheory aims towards promoting a social and political message. For instance, Claudia Rankine’s Don’t Let Me Be Lonely and Citizen are subtitled An American Lyric. They combine documents and poems to produce a discourse on racial injustice in the US. In this context, the referentiality of the poems seems significant as it is a step towards action. To denounce injustices requires a form of truth or truthfulness. These examples of documentary lyric perhaps reveal that what is central is not the concept of truth, but that of truthfulness. This move from truth to truthfulness places lyric poetry as being somewhat outside the jurisdiction of the fiction/nonfiction debate. While there can be references in poetry (and with varying degrees), the reference is always an element of an experience and not evaluated in terms of truth and falsehood.

Topics for further investigation

 iv.4.26.The question of documents in poetry opens the field of multimedia and transmedia poetry. To what extent does the change of support affect questions of fictionality and referentiality? Does a transposition of a poem in a pictural or cinematographic format affect its relation to fictionality and referentiality? If we imagine the poem being staged as uttered by the poet, the confusion between the empirical author and the enunciative voice becomes stronger. And to the contrary if the enunciative voice is represented by a character other than the empirical author, the question of the identity of the two is cast aside. The image brings another layer of referentiality that answers the question of fictionality (either by going towards fiction or towards fact). But the image can also add to the confusion and play with it, oscillating between representation and non-representation, following the tradition of cinepoetry (Wall-romana 2013). How does the multimediality of the poem affects the question of fictionality?

 iv.4.27.A similar matter arises regarding performance and recent work on performance question the lyric voice (Novak 2011; Simecek 2023). Does the embodiment of the voice in a performance affect the audience’s understanding of the fictionality or referentiality of the poem? Does the performance affect the distinction between the performer and the lyric ‘I’? Rather than the use of images, the emphasis is placed on the use of the body and the voice. If the performer is the poet, the identity between the enunciative voice and the empirical author is strengthened. If the body enacts what the enunciative voice is saying, the figural dimension of the lyric is left aside and becomes a description of an actual act. The embodiment of the lyric voice could thus solve the question of fictionality by either embracing an identification between the speaker and the empirical author or by separating them. However, this embodiment does not eliminate the ambiguity of the relation between the lyric ‘I’ and the empirical author. The question of performance can thus be related to that of interpretation (as in music). What would a lyric interpretation be and what are its characteristics?

 iv.4.28.More than performance, how does interaction affect the question of fiction and referentiality in lyric poetry? If the experience of fiction is not fictional, does the reader’s or spectator’s interaction with lyric poetry affect the fictionality of the poem? The act of reading can indeed be altered if the reader believes in the fictionality or the non-fictionality of the poem. The experience of the poem can vary according to the fictionality of the poem. Digital poetry emphasises the interactive dimension of poetry (Bachleitner 2005). The mode of interaction can take various forms and some electronic poetry are even video games (Papa 2013; Magnuson 2023). How do these developments in multimediality and performance affect the experience of reading the lyric?