Semiotika ISSN 1392-0219 | eISSN 2424-547X
2024, vol. 19, p. 8–32 DOI: https://doi.org/10.15388/Semiotika.2024.1
Jimmie Svensson
University of Gothenburg, Sweden
E-mail: jimmie.svensson@lir.gu.se
https://orcid.org/0000-0001-6155-9228
https://ror.org/01tm6cn81
Summary. This article examines the process of transformation from a written poem by Louise Glück to the poet’s oral reading of it, from a semiotic and intermedial perspective. Glück uses Poet Voice, a widespread reading style characterized by a highly stylized prosodic evenness and predictable use of pitch. Whereas an essential trait of traditional metrical verse is its interplay between the artificial, regular meter and a prosodically more natural reading, non-metrical verse such as Glück’s depends on its visual segmentation into lines, often in conflict with syntax. By using Poet Voice, which conflicts with conventional prosody, these essential characteristics of poetry can be transmediated to the oral reading. While this does suppress expressiveness, it helps produce additional meaning in a way specific to poetry, by means of iconicity. The breakthrough of non-metrical verse around 1900 was accompanied by a visual turn; continental theory further contributed to an increased focus on the written text and to skepticism about a traditional lyrical “I.” The second half of the 20th century, however, saw an oral turn, with the poet emerging as the reader or performer of his or her own texts. In this historical context, this oral reading style balances conflicting tendencies by imitating characteristics of the written text: undecidedness, neutrality, and openness to interpretation.
Keywords: iconicity, media representation, poetry performance, Poet Voice, verse
Received: 11/03/2024. Accepted: 09/09/2024.
Copyright © 2024 Jimmie Svensson. Published by Vilnius University Press. This is an Open Access article distributed under the terms of the Creative Commons Attribution Licence, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Louise Glück (1943–2023) was awarded the 2020 Nobel Prize in literature “for her unmistakable poetic voice that with austere beauty makes individual existence universal” (The Swedish Academy 2020). At first glance, this citation might be mistaken to refer to how she used to perform her poems, as Glück indeed practiced a highly distinctive, formal reading style, being a prominent representative of what is now commonly known as Poet Voice. Recognizing the actual, sounding voice rather than “voice” in a merely figurative sense, could have been an attempt to address an ongoing turn towards aurality and orality in contemporary culture and literature (after all, the 2016 laureate was none other than Bob Dylan). Digital media technologies are rapidly changing the conditions for the production, distribution, and consumption of literature; at least in the Nordic countries, the streaming audiobook format, which appears to favor popular fiction, is now dominating the book market (Berglund 2024: 29, 57–58). To what extent more challenging, literary genres will be affected by the audiobook boom remains to be seen. So far, regarding poetry, another oral turn, with its roots in the pre-digital era, seems to be more significant. Since the 1950s, the poet has emerged as the principal reader of his or her own work at new types of reading events (Middleton 2005: 63–68; Wheeler 2008: 130–136), and the live reading now seems to be the most common way of encountering poetry, at least in the United States, Denmark, and Sweden (Serup 2017: 45).
Notwithstanding this situation, the printed format reigns almost supreme when poetry is being written about, in research as well as in literary journals and daily newspapers. Recordings and live readings may have the potential to make poetry more accessible and less exclusive, but for this to happen, substantial critical and scholarly attention must shift to the auditory text and, for that matter, to performative aspects such as body language, gestures, and audience interaction. Here, I confine myself to the former, to what Charles Bernstein (1998) calls the audiotext, i.e., the voiced written text (for the latter aspects, see Novak 2011 and Bäckström, Führer, and Schirrmacher 2022), and more precisely to the widespread, dominant, and highly stylized reading style of Poet Voice, with Glück as the example. Poet Voice is characterized by its monotony, the repetition of a single vocal cadence, so prominent that after a while one might hear only this pattern and thus be distracted from what the poem is about, its content. It is often perceived as annoying, boring, emotionless, insincere, and so forth, and this frustration with the style is the starting point of Marit J. MacArthur’s pioneering work on poetry readings (MacArthur 2016; MacArthur, Miller, and Zellou 2018; MacArthur et al. 2022), to which I will return in some detail presently.
In this article, I bring a semiotic and intermedial perspective to the table, with the aim of exploring what and how Poet Voice and its use might mean, and how the meaning of the written poem is transformed when read in this style. With a better idea of how to listen to it, we may find it less annoying, or at least more comprehensible. The key notion is iconicity, one of Charles Sanders Peirce’s three types of signification, introduced about 1890 (see, for instance, his Collected Papers 2.297–302, in Peirce 1932), which has attracted much cross-disciplinary attention in the last few decades. The icon is to some degree similar to what it stands for; following Lars Elleström (2010, 2015), I understand iconicity in literature as a continuum ranging from stronger, material similarity within the same sense category (an “o” representing the visual shape of the moon), to the weaker, cross-sensorial and cognitively more complex and abstract similarity often showcased in verbal metaphors and similes (“love is like a sweet melody”). However, I will refrain from using Peirce’s trichotomy of iconic signs, as this would require explication beyond the scope of this article. After the icon, the two other main sign types or functions are the index, which signifies based on an existential or physical connection, and the symbol, based on conventions or, as Peirce prefers to put it, “habits.” Signs in use often have a mixed character, consisting of iconic, indexical, and symbolic traits, although one type of representation often dominates, depending on the perspective. For instance, a portrait can stand for a person (icon), a wisp of smoke for fire (index), and a letter for a sound (symbol).
Due to his focus on logics and mathematics, Peirce downplays the material aspects of signification, and this, in addition to the fragmentary status of his writings, poses a substantial challenge. When the cognitive result of signification is the important thing, materiality is only of marginal interest; it is then essential that, for instance, the words “wild iris” signify (by symbols) a certain plant that sometimes bears blue flowers, not whether these words are perceived visually on a page or heard uttered. This is clearly not the case in the arts: a Picasso painting or a Glück poem is remarkable not only for what it represents, but also, and perhaps mainly, for how it represents. Elleström’s (2014a) adaptation of Peirce to the study of media and the arts, to which my approach is greatly indebted, is further integrated into a model of intermedial and multimodal analysis (Elleström 2014b, 2020) that helps us understand communication and interaction with various media. In this article, using Elleström’s terminology, media transformation, i.e., the transfer of meaning/content/form between media, will be studied in detail. More specifically, the source medium or media product is the printed title poem of Glück’s 1992 collection The Wild Iris, and the target medium is her oral reading of the poem from the same year, available on Poets.org, the Academy of American Poets website (link).
When media transformation involves media products belonging to distinct media types, of different semiotic modes, characteristics from the source medium must be more thoroughly transformed or reconstructed to still be recognizable when represented by the target medium: for instance, when visual, iconic configurations seen on a painting are transferred to a poem, using verbal language, which is predominantly symbolic (ekphrasis). In contrast, as the written poem and the recited one both use verbal language, it is comparatively easy to transfer essential characteristics between them. The audiotext might even be perceived as nothing more than a remediation, a repetition of the same media product with the same symbolic signs. The text and the audiotext differ, however, when it comes to material interface (text on paper or screen versus soundwaves), and, consequently, there are differences in how we perceive them with our senses and in how they relate to time and space. This, as I will demonstrate, makes the process anything but seamless and means that some of the characteristics must be significantly transformed (cf. Have and Pedersen 2020, on the differences between reading a novel and listening to an audiobook).
It is evident, according to MacArthur et al. (2022), that Poet Voice as well as the reading style typical of spoken word poetry, at the other end of the spectrum, were shaped by their different sociotechnical contexts in the United States. Poet Voice is strongly associated with the “mainly white rooms” of the so-called academic poetry reading (which might not have any direct counterpart in Europe, where creative writing programs at university level emerged later). The subdued reading style is a consequence of this academic setting: poets are given extended time, reading from books, with a quiet audience withholding its applause to the end. Conversely, the dynamic, energetic vocal performance style typically practiced in spoken word venues reflects the participatory norms of the Black church and Black music: poets there often perform with time limits, competing for the favor and approval of a boisterous audience (MacArthur et al. 2022). In semiotic terms, the different reading styles serve as indices (Peirce’s second type of sign) of their respective venues of origin.
To go deeper into the iconic facets of Poet Voice, it is necessary to explore its prosodic features in detail. According to MacArthur, in her first article on the subject, based on and exemplified by readings by Glück, Michael Ryan, Juliana Spahr, and Natasha Tretheway, Poet Voice is characterized by three qualities:
(1) the repetition of a falling cadence within a narrow range of pitch; (2) a flattened affect that suppresses idiosyncratic expression of subject matter in favor of a restrained, earnest tone; (3) the subordination of conventional intonation patterns dictated by particular syntax, and of the poetic effects of line length and line breaks, to the prevailing cadence and slow, steady pace. (MacArthur 2016: 44)
These characteristics partly overlap, with the fundamental frequency or vocal fold vibration, which is perceived as the pitch of the voice, as the highlighted acoustic quality. For the purposes of the present article, they need some further explication.
(1) One and the same intonation pattern is used throughout the reading. The same cadence, or what might colloquially be called melody, is repeated again and again. The cadence starts at the upper end of the poet’s pitch range, gradually going down to a lower pitch, and then the next cadence follows with a predictable return to the higher pitch. The pitch range, the span between the highest and the lowest tone produced in the reading, is comparatively narrow, approaching the monotone. (2) In contrast, a speaker or reader perceived as expressive tends to be more varied and unpredictable, and to use a relatively wide range of pitch, and while intonation and semantic content normally agree, in Poet Voice the reading is consistent and does not change with changing emotions or content. Consequently, all poems sound more or less the same or, as MacArthur puts it with reference to the Glück reading she is using, of “Witchgrass”: “We feel that we are listening to any poem, not this particular one” (MacArthur 2016: 48).
(3) Not only is Poet Voice indifferent to or at odds with content, but it might also seem to override what would be the more conventional intonation or way of putting stress on words. In everyday communication as well as in an audiobook novel, intonation and stress are normally used to guide the listener, to signal what is new or important information, but in Poet Voice, this function is largely displaced. In Glück’s slow reading, stresses are too even and too evenly and tightly placed. In the very first sentence of “The Wild Iris,” “At the end of my suffering / there was a door.,” “end” and “suffering” seem about equally stressed, while a more conventional reading might have instead emphasized “end.” In “suffering” the last syllable seems prolonged and oddly prominent. While a more expressive reading might rise to “a door,” as this could signal surprise or new hope, Glück follows her cadence, going down. This is how the first half of the printed poem appears on the page, with the plant, the wild iris, as the personified speaker:
At the end of my suffering
there was a door.
Hear me out: that which you call death
I remember.
Overhead, noises, branches of the pine shifting.
Then nothing. The weak sun
flickered over the dry surface.
It is terrible to survive
as consciousness
buried in the dark earth.
(Glück 1992, 1)
When reading, Glück presents this section in five cadences, each starting high and ending low, with long pauses in between to make the units easily perceptible:
1. At the end of my suffering there was a door.
2. Hear me out: that which you call death I remember.
3. Overhead, noises, branches of the pine shifting. Then nothing.
4. The weak sun flickered over the dry surface.
5. It is terrible to survive as consciousness buried in the dark earth.
The iconic functions of prosody are highly complex (see also El Zarka 2013 and Perlman, Clark, and Falck 2014), and cannot be discussed in further depth here, but it is safe to say, as suggested, that Poet Voice entails severe reduction of the iconicity present in everyday speech and conventional reading styles. The very basic iconic links that normally exist between language form and meaning are here largely abandoned: a word is said in a certain way because of its position in the repeated cadence, not because of its semantic weight.
MacArthur has continued her work on poetry readings by developing an open-source toolkit producing a set of prosodic measures. Based on the results of a quantitative study of 100 American poets, the previous understanding of Poet Voice is refined, while at the same time made less strict, allowing latitude for the highly individual and psychological nature of speech perception (MacArthur, Miller, and Zellou 2018). This means that readings exhibiting only a few of the Poet Voice qualities, or to a lesser degree than Glück’s, for instance, might also be perceived as belonging to the style. Some adjustments might also have to be made when discussing Poet Voice in poets of languages other than English. In my own application of MacArthur’s methods and tools to a Swedish corpus, in which the presence of a more formal, less expressive style with traits associated with Poet Voice turns out to be substantial, especially in female poets, I suggest that Swedish poets using Poet Voice tend to adopt an even narrower pitch range, compensating for the fact that intonation in Swedish goes up and down more than in English (Svensson 2022: 196). If and to what extent Poet Voice is practiced in languages other than the ones mentioned, I must leave for native listeners to discover and investigate further. To my knowledge, the term has so far otherwise only been used in a Mexican Spanish context (Meza 2023).
A reading like Glück’s, with its repeated cadence and even stress pattern, is reminiscent of a meter-focused reading of a traditionally metrical poem (which “The Wild Iris” is certainly not), only distinguishing between the strong and weak syllables of the metrical feet or units. In both cases, the meaning of the words seems sidestepped and might even be completely lost on the listener, drenched in the predictable, monotonous singsong. This similarity might even have contributed to earning this reading style its popular, pejorative name, Poet Voice, if perceived as a ridiculous attempt to recreate the elevated reading style of bygone centuries.
This calls for the introduction of Elleström’s (2014b: 11–20) two interrelated and theoretically, if not always practically, discernable aspects of media transformation: transmediation refers to the transfer of ideas or narratives across media, and media representation to how one medium represents the characteristics of another. In a film adaptation, the two aspects might be fairly easy to separate: the film transmediates the plot of the novel, although using different means, mainly moving images and speech, and perhaps also music, to represent the plot first represented by written text. The sensory configurations of film and novel are different, but equivalent in that they can successfully trigger corresponding representations of the same characteristic, the sequence of events. In this process, the film can draw attention to its source medium, the novel, by also representing the novel’s media specific or typical way of representing. For instance, passages from the literary text could be read in voice-over, or the scenes could be broken up by inserted chapter titles, as if the film were a novel.
Going back to Glück’s reading, the two aspects seem more entangled and complicated. A reading emphasizing a metrical pattern signifies a performance, a speech act far from the mundane, indicating that this is in fact poetry. This characteristic, which rather belongs to the genre of poetry than to the materiality of Glück’s written poem, as it is not metrical, is successfully transmediated, that is, reconstructed by different means, by the Poet Voice reading, the target medium, if it is interpreted in the same way. For this to happen, though, it is probably necessary for the listener to interpret the Poet Voice reading as an imitation of the meter-focused reading style, that is, as an iconic sign in which this source medium (again, rather the genre it belongs to) is represented. There is, however, more to the picture, and it is necessary to further investigate what is transferred, and how it is done.
In her very first collection, Firstborn, Glück is experimenting with meter as well as rhyme. A modernist writing in 1968, she frequently bent the rules in a way the likes of Keats or even Yeats would certainly not have done. In these lines, the iambic pentameter is clearly used as a template:
Requiring something lovely on his arm
Took me to Stamford, Connecticut, a quasi-farm,
His family’s; later picking up the mammoth
Girlfriend of Charlie, meanwhile trying to pawn me off
On some third guy also up for the weekend.
But Saturday we still were paired; spent
It sprawled across that sprawling acreage
(“Labor Day,” Glück 1995: 13)
The use of meter is an instance of what Roman Jakobson (1960) refers to as the poetic function of language. Equivalence in language form, mainly auditory form, is allowed to dictate (or, differently put, is projected on) the sequence of words. In this case, the equivalent units are the iambs (an iamb consists of one weak/unstressed syllable followed by a strong/stressed syllable), and they would have made it impossible to substitute “Requiring” and “lovely” in the first line with shorter or longer synonyms such as “Needing” or “beautiful.” Equivalence at other levels is in play too: “arm” is irreplaceable because it rhymes with “quasi-farm” in the preceding line. Jørgen Dines Johansen, building on Jakobson’s concept of the poetic function, which might well appear in non-poetic language, explains how poeticity works in the literary text:
Semiotically speaking, poeticity is first and foremost concerned with surplus coding, i.e. with adding rules in addition to language rules, for the production of literary texts, and for their reception too, in the sense that the reader is supposed, at least unconsciously, to understand and respond to what is characteristic of the individual poetic texts.
[…]
The difference between everyday and poetic speech is that in the latter the rhetorical devices are made more complex and refined, and this refinement allows poetic texts to represent states of affairs and relationships that are not stereotypical; indeed, sometimes it allows them to represent what has not yet been put into speech. (Johansen 2007: 121)
The added rule of meter is such a rhetorical device. Lines following the same meter inevitably vary in rhythm, and there is always some strain or conflict between the meter and a prosodically conventional or more natural way of reading the line. Even in the comparatively harmonious first line of the quotation (which scans, with the stressed syllables underlined: “Requiring something lovely on his arm”), it would sound odd to stress “on,” which happens to be in a strong position according to the meter. In a famous passage from An Essay on Criticism (1711), Alexander Pope both explains and demonstrates how the tension between meter and prosody can be used (Pope 1993: 2224, from line 362): in “A hoarse, rough verse should like the ocean roar,” for instance, he illustrates said roughness by placing “rough,” which should be emphasized in a natural, prosodic reading, in a metrically weak position. But this violation of the meter, perceptible because the pattern has been allowed to establish itself in the preceding lines, is meaningful and gives rise to iconic representation, resulting in the interpretation that the auditory form of the lines is illustrating the meaning of the symbolic words.
Apart from adding an extra iamb, Glück does something similar to Pope in her second line, by awkwardly placing “Connecticut” so that its first vowel either breaks the alternating pattern by means of a surplus weak syllable or must be omitted in the reading. Further down, a weak position is in fact omitted: “we spent” would have been the obvious, metrically correct wording, but Glück only writes “spent.” The faulty meter created by this word can be interpreted as an iconic representation of the relationship in question now being exhausted or drained of (sexual) energy, reinforcing the perception of the word as the adjective (as opposed to being the past participle of the verb, which of course is the syntactically correct option). As the most common meter in serious English poetry, a tradition dominated by men, it has been argued (see Finch 1993: 135–139) that iambic pentameter has been perceived as patriarchal and is sometimes used as a negative symbol of male authority, especially by female poets. With this interpretative context in mind, these and other violations of the pentameter in Glück’s poem seem to represent the awkwardness or ugliness of the situation depicted.
This interplay or conflict between meter and prosody in metrical verse, commonly described using the musical metaphor of counterpoint (e.g., Hartman 1980: 25), is highly psychological: we are supposed to perceive and respond to the two parts or voices at the same time (see also Tsur 2008: 171–173). Something similar is at work in Glück’s reading of “The Wild Iris.” Her Poet Voice cadence, like the iambic meter in her early poem, is superimposed on the text, to some extent in agreement with the syntax, as each cadence ends at the end of a sentence, or at least at a punctuation mark, and as the falling intonation normally used in declarative sentences is maintained, but often deviating from a potential, more conventional way of reading. While the pentameter line always has ten or eleven syllables, the cadence is more flexible with regard to duration and syllables, and while the meter is intersubjectively determinable from the text, even by silent reading, Glück’s cadence is only added in her oral reading. It is possible that she had the cadence in mind when writing, using it like a template in the same way as she undoubtedly used the iambs and the pentameter in her early poem, but, if so, this cannot be ascertained from her text.
Nevertheless, there is a similarity in structure between the two reading styles that goes beyond the mere similarity in sound. The Poet Voice reading can be understood as an iconic representation of how poeticity comes about through sound in metrical verse, and in the process, the media characteristic of poeticity is transmediated from the source to the target medium. Once the similarity is noted, it might trigger further semiosis, in which the context and discrepancy, the fact that Glück is clearly not writing metrical verse, is considered. Glück’s intertextual—or rather interaudiotextual—recollection of a form of poetry now nearly extinct, if associated with dead poets of old, might then be interpreted as an iconic representation of the speaking flower’s experience of being in the ground during winter, recollecting death, which the listener is encouraged to “hear”: “Hear me out: that which you call death / I remember.”
Notwithstanding the experiments with meter in her early work, the mature Glück is firmly grounded in the modernist free verse tradition, as evident from the printed poem “The Wild Iris.” As touched on briefly in the previous section, this aspect of the source medium belongs to its genre, a category or subcategory of media, what Elleström (2020: 54–60) calls a qualified medium, defined by convention and history, and by our experience of individual media products of this category (such as Pope’s An Essay on Criticism). Of course, metrical verse and free verse have many qualifying aspects in common. For instance, as both belong to the main category of poetry, they might well be open to interpretation in a way that is not desirable in non-literary texts, and their representation of objective reality cannot be expected to be the same as that in a news article. Regarding form and poeticity, however, free verse is qualified in a different way.
The breakthrough of free, non-metrical verse around 1900 marks a visual turn: it is only verse to the eye, as the only feature that distinguishes it from prose is its graphic division into lines (Lorentz 1980). It has even been suggested that the tedious, meter-focused recitation practice of the 19th century was a reason underlying the venture into free verse in early modernism (Finch 1993: 88). Primarily, the change of poetic paradigms is connected with technical novelties of that time period: “What good are the poetic mnemonic techniques of rhyme and meter when wax rolls can store not only substance and tone but real sounds?” (Kittler 1990: 236). Furthermore, the typewriter made it possible to write poems that looked more like printed text and to individualize the shape of the poem, as a graphic signature specific to the poet (Nyberg 2009).
The main unit in free verse is the graphic line, projected onto syntax, or conventional sentence structure, and this verse–syntax interplay corresponds to the previous, mainly auditory one between meter and prosody (Kjørup 2008). It manifests itself most conspicuously in enjambment, that is, when the syntactic unit is left incomplete by the end of the line and has to continue over to the next one. Poets’ use of enjambment has been duly investigated in terms of iconicity (e.g., Halter 1999; Müller ٢٠١٠), and examples are abundant in Glück’s early poetry, for instance in these opening lines, again from a poem in her debut collection (1968):
Lived to see you throwing
Me aside. That fought
Like netted fish inside me. […]
(“Hesitate to Call,” Glück 1995: 10)
By separating “throwing” from its object, “Me,” the movement of the reading eye can be perceived as bodily movement, as a spatial tossing aside of the word “Me.” This is an example of the surplus coding mentioned earlier: through the iconic sign, the literal meaning of this phrasal verb is highlighted, while the more conventional understanding of “throwing me aside” might not imply any physical movement at all. Glück’s mature style is more subtle (see also Dodd 1992: 152–155), but two comparable, iconic enjambments are nevertheless to be found in “The Wild Iris.” In the first half, there is “The weak sun / flickered over the dry surface,” and in the second half, which it is now time to quote, “the stiff earth / bending a little”:
Then it was over: that which you fear, being
a soul and unable
to speak, ending abruptly, the stiff earth
bending a little. And what I took to be
birds darting in low shrubs.
You who do not remember
passage from the other world
I tell you I could speak again: whatever
returns from oblivion returns
to find a voice:
from the center of my life came
a great fountain, deep blue
shadows on azure seawater.
(Glück 1992: 1)
Another manifestation of verse–syntax counterpoint is the pseudosentence that the verse line creates, or rather invites the reader to create, as the line’s perceptual wholeness creates a spatiotemporal framing effect (Kjørup 2008: 87–90). Pseudosentences are often ill-formed according to conventional linguistic criteria, adding syntactic ambiguity that should not be dismissed as mere misreading, as it would if this happened in a piece of prose. Here, if the syntactical unit is first interpreted as if ending with the line unit, “being” in “that which you fear, being” seems to be a noun (as in “you have a fear of being”), and “unable” in “a soul and unable” seems to be an attribute of “soul” (as in “an unable soul”).
In other cases, the function of verse–syntax counterpoint is harder to explain in semantic terms. Verse with frequent divergence between the two, in what Reuven Tsur calls divergent style, also including meter–prosody counterpoint, tends to be perceived as more emotional due to its weak, undifferentiated shapes; this means more information for the reader to handle, generating cognitive energy, similar to active emotional processes (Tsur 2008: 56–61, 100–104). This might be the case here, and, in the first part of the poem, it is worth pointing out that the frequent enjambments, suggesting movement, are followed by full stops, marked by interpunctuation and stanza breaks, suggesting halted movement or stillness. In the second half (which is otherwise not graphically distinguished), the stops are less frequent, the sentences flowing without halting. Thus, the verse throughout the poem can be interpreted as an iconic representation of the process that the wild iris goes through, first only stirring in the ground, half-dead, gradually reaching upward and bursting into bloom.
It is now time to recall how, from the mid 20th century, the poetry reading became increasingly important. Free verse poets, recognizing the visual as an essential qualifying aspect of their art, face a dilemma when their poems are to be performed. Undoubtedly, some effects of visual verse–syntax counterpoint can be effectively rendered in sound. The end of the line can be distinguished, as Glück does with “At the end of my suffering,” by pausing, holding back “there was a door” in a way corresponding to how it is held back on the page by spatially being placed in the following line. As temporal perception takes priority in auditory signs, a listener might perceive the withholding as an iconic representation of a process unfolding over a period of time. Of course, everything takes place in time and space, but the pausing might not work as well when the temporal aspect is less pronounced, as in “flickered / over the dry surface.”
Spatial perception takes priority in visual signs, but it is important to remember that differences between the senses are often bridged by our cross-modal capacities (Guttman, Gilroy, and Blake 2005; in relation to poetry, see also Ma 2011). Silent reading triggers the auditory imagery of the sounds of words, in what is known as subvocalization or silent speech (Smith, Reisberg, and Wilson 1992). Particularly when it comes to poetry, readers have been trained to read with a heightened attention to sound, as this is routinely put forward as a qualifying aspect of the genre. The experienced reader can tell what words rhyme and identify the meter or other prosodic patterns without hearing or voicing the poem. Conversely, most readers are hardly in the habit of visualizing the corresponding written words when they hear someone reading—the visualization they trigger would instead concern the content. “Subtitling” while listening, if audiences could even be persuaded to do this, would most likely not be worthwhile, due to our limited short-term memory capacity. The audiotext is temporal also in that its material interface disappears as soon as a sound is uttered, while the reader of a written text is in full control of its temporal sequentiality and can go back and forth and overview the poem as a stable, spatial object.
In addition, while pausing might help the listener hear the pseudosentence created by the verse line, the ambiguity will be lost. Pausing after “I can speak again: whatever” will reinforce the interpretation that “whatever” is the object of the clause (as in “I can say whatever”), but then the correct phrasing, that it starts another clause, “whatever / returns from oblivion,” would be obscured. In a more conventional reading or in everyday speech, the syntactic ambiguity would be even harder to convey: if “whatever” were first clearly indicated as ending the sentence (by falling intonation), it would be permanently disconnected from the rest of the clause. In a vocal performance, the oral reader is forced to choose between the alternatives, while they can both be kept active in the mental performance of a silent reading (cf. Tsur 2008: 185–187).
In conclusion, due to the different media possibilities and limitations of visual text and auditory text, it is not to be expected that the effects of verse–syntax counterpoint in the source media product, the written poem, can be recreated in the corresponding passages of the performed one. What is transmediated in the reading is instead the counterpoint effect as such, a characteristic of the qualified medium (sub)type of free verse. However, the projection of the Poet Voice cadence adds surplus coding of its own, creating new interpretative options and ambiguities, as well as iconic meanings, that are not transmediated from the written poem. A counterpoint effect is present every time the listener feels that Glück reads a phrase or a sentence differently from how it would normally sound, and this defamiliarizes the expressions and highlights language form. For instance, “Hear me out” of the second stanza might look like an instance of the everyday phrasal verb, unnecessarily asking us not to interrupt, but it certainly does not sound like it. In combination with the presence of the strange reading voice, and with the repeated references to voice later in the poem, the listener is invited to understand the expression as a metaphor for the process described in the poem: we are supposed to hear the nearly dead plant come out of the ground and burst into flower.
Next in Glück’s reading, “that which you call death” lacks the prosodic cues that would make it unambiguous: she would probably have read an apostrophe to personified death—“that which you call, Death”—in much the same way. The “you” could alternatively be understood as addressed to the listener/reader, or, bearing in mind that the speaker is a plant, to all humans or to the clearly human subject present in other poems of the book. The “You” later in the poem suggests that the pronoun indeed refers to another, while the way Glück reads this first “you” could suggest that it instead means one, and that it thus includes the speaker. The unclear identity of this “you,” if we allow it to be so, could then be made meaningful, by iconicity, if we notice the parallel to the question of identity that should have been raised as soon as we realized that the “I” of the poem is a plant: is the poem to be understood as an allegory, and, in that case, of what or whom?
Such effects—and these are just a few examples—mirror the effects created by meter–prosody and verse–syntax counterpoint, but by deploying different means. They require that the listener be open to them and be ready to perform them mentally, make an interpretative effort, or at least make poetic sense of them.
Arguably, the most prominent and important function of the Poet Voice cadence in “The Wild Iris” is that it helps Glück to subsume the reading into one overarching iconic sign, representing the process that the “I”, the wild iris, is undergoing. As noted, earlier in the written poem, the punctuation in combination with the enjambments also iconically represent this process, and as there is some detailed structural correspondence here between the text and the audiotext, this can be viewed as an instance of more detailed transmediation between them.
The cadence is the constant, creating the strong and easily perceptible shape by which change is measured and contrasted. In what has here been referred to and quoted as the second section, starting with “Then it was over,” Glück’s reading rate is faster than in the preceding section, in which the plant is still lingering in the ground (117 versus 86 words per minute), or, more accurately, the pauses between the cadences are much shorter in the second section. The words are not read faster, and the reading speed within the individual cadences seems to be nearly the same: the ten words of the very first cadence, for instance, are even read slightly slower than the ten words of the penultimate cadence. In the first section, all cadences end where the text has full stops, while this is only the case with half of the cadences in the second section, in which the full stops come less frequently and there are three colons. This, which in the written text could be read as a mere encouragement to read faster or with shorter pauses, is materially realized in Glück’s reading.
The Poet Voice cadence does not change, but is the basic pattern in relation to which the very subtle change is made perceptible, contrasting the first half or so, where she carefully respects the punctuation marks, marking them with long pauses, with the second half, where she shortens the pauses and even at times seems to override the punctuated boundaries. It is as if the waves of repeated cadence units become more frequent and intense. She is thus representing the death and stillness in the winter earth versus the life and bloom of spring, conveying how the flower “returns to find a voice.”
With Poet Voice, the poet is able to transmediate one of the most important characteristics of the printed free verse poem into an oral reading. Consequently, the reading appears severely deprived of expressiveness or empathy with the lyrical “I.” But far from being only an unfortunate but unavoidable side-effect, this might even be the main point. As far as I know, Louise Glück never talked about or explained her choice of reading style, and the only helpful statement that I have come across from someone reading in a similar way is this one from Athena Farrokhzad, one of the most prominent Swedish poets of the last decade:
I want the listeners to become flooded by language, and to be able to pick out what is important by themselves, without having it spelled out in an overly dramatic reading. My poems tend to be rather saturated with affect, and I think that a restrained reading style makes them emotionally stronger, while a more expressive one would reduce the emotional import.
(Svensson 2020: 108, my translation)
In this account, Poet Voice serves as a rhetorical device, inducing the listener to mentally add an interpretation, like a second voice to the sounding one, a perspective that supplements what has already been demonstrated here about the counterpoint effect in different forms of poetry. In addition, what Farrokhzad says here about the author taking a step back, leaving the interpretation to the reader, might form associations with the turn to textuality in French theory from the 1960s onwards. Thinkers like Barthes, Derrida and Foucault have been highly influential in their attempts to destabilize and reject the view that writing is subordinated to or only echoing speech, and they instead assert textual autonomy (see, for instance, Wesling and Sławek 1995). In the text as text, no authorial voice is present to guarantee its meaning, and the “I” is a textual function rather than the expression of a coherent subject that precedes and produces the text as speech: “writing is the destruction of every voice, of every point of origin. Writing is that neutral, composite, oblique space where our subject slips away, the negative where all identity is lost, starting with the very identity of the body writing” (Barthes 1977: 142).
These and related ideas have set their marks on poetry: poets abandoning the “I” altogether, or switching to a more fragmented subject position different from the lyrical “I” of romanticism and high modernism, and replacing conventional free verse, which tracks the speaking voice of a subject, with non-linear, typographical experiments (Perloff 1998; see also Patterson 2013 and Perloff 1999). Louise Glück as well, in her own moderate, sophisticated manner, questions and destabilizes traditional lyric form. According to Daniel Morris (2006: 22, 213–216), the book The Wild Iris, including a range of subject positions as the self is divided into three parts—God/the voice of nature, the author-gardener, and the flowers—is her most ambitious attempt in this direction. In her earlier work, by merging the autobiographical with different personas, most prominently from classical mythology, she avoids “confessionalism,” a derogatory, dismissive label often imposed on female poets, and distances herself from her speakers (Dodd 1992: ch. 5). This is an important aspect of how she makes “individual existence universal,” as the Nobel citation puts it. When it comes to the reading style, it might be that she and other female poets adopt a less expressive one to avoid the emotionality routinely associated with women and their voices, instead imitating the authority associated with the (stereo)typically male, less expressive voice (cf. MacArthur, Miller, and Zellou 2018: 46).
Parallel to this situation and the new status of the lyrical “I,” over the last half decade or so, the oral reading has become close to mandatory, at least for poets anxious to promote themselves. For a poet careful about the precarious relations between body, voice, and the “I” of the text, the live performance in particular poses a dilemma: on stage, Peter Middleton (2005: 33) points out, alluding to Barthes, the “dead author” rises again, occupies the subject position, and performs the text as if it originated in the present, in front of the audience. In the audio recording, the presence of a voice tends to do the same, but to read a poem like Glück’s “The Wild Iris” in a conventional or expressive way, as an actor would, to let us imagine the speaker as a character, the lyrical “I” behind it, would go against the grain of her project. Instead, by adopting Poet Voice, she can perform the text without having conventional prosody, by highlighting various moments, phrases, and words, interpret its content or emotional status. In addition, the stylization rendered her voice truly unmistakable, and made it a trademark (an index) of the poet, as opposed to merely representing the “I” of the poem.
As mentioned earlier, the two aspects of media transformation, transmediation and media representation, are often impossible to distinguish from one another in practice, and this is certainly the case here. A text remains voiceless, without a subject, until someone reads it, be it aloud or in silence. By imitating this voicelessness of the text, this characteristic of the source medium is transmediated to and represented by the reading. In other words, a Poet Voice reading will to some extent give an iconic representation of the text’s neutrality, undecidedness, and openness to interpretation—it will be voice representing text.
Berglund, K. 2024. Reading Audio Readers. Book Consumption in the Streaming Age. London: Bloomsbury Publishing Plc.
Bäckström, P., Führer, H. & Schirrmacher, B. 2022. The Intermediality of Performance. Intermedial Studies. An introduction. Ed. by Jørgen Bruhn and Beate Schirrmacher. London and New York: Routledge, pp. 198–224.
Barthes, R. 1977. Image Music Text. Transl. from the French by Stephen Heath. London: Fontana Press.
Bernstein, C. 1998. Introduction. Close Listening. Poetry and the Performed Word. Ed. by Charles Bernstein. New York: Oxford University Press, pp. 3–26.
Dodd, E. 1992. The Veiled Mirror and the Woman Poet. H. D., Louise Bogan, Elizabeth Bishop, and Louise Glück. Columbia: University of Missouri Press.
Elleström, L. 2010. Iconicity as Meaning Miming Meaning and Meaning Miming Form. Signergy. Ed. by Jac Conradie et al. Amsterdam and Philadelphia: John Benjamins, pp. 73–100.
Elleström, L. 2014a. Material and Mental Representation. Peirce Adapted to the Study of Media and Arts. The American Journal of Semiotics 30, pp. 83–138.
Elleström, L. 2014b. Media Transformation. The Transfer of Media Characteristics Among Media. Basingstoke: Palgrave Macmillan.
Elleström, L. 2015. Visual, Auditory, and Cognitive Iconicity in Written Literature. The Example of Emily Dickinson’s “Because I could not stop for Death.” Iconicity. East Meets West. Ed. by Masako K. Hiraga et al. Amsterdam and New York: John Benjamins, pp. 207–218.
Elleström, L. 2020. The Modalities of Media II. An Expanded Model for Understanding Intermedial Relations. Beyond Media Borders. Intermedial Relations among Multimodal Media, Volume 1. Ed. by Lars Elleström. Basingstoke: Palgrave Macmillan, pp. 3–91.
El Zarka, D. 2013. Pragmatic Functions and the Biological Codes. Evidence from the Prosody of Sentence Topic. Prosody and Iconicity. Ed. by Sylvie Hancil and Daniel Hirst. Amsterdam and Philadelphia: John Benjamins, pp. 109–126.
Finch, A. 1993. The Ghost of Meter. Culture and Prosody in American Free Verse. Ann Arbor: University of Michigan Press.
Glück, L. 1992. Audio recording of “The Wild Iris.” Available at:
https://poets.org/poem/wild-iris-audio-only [Accessed: 3 March 2022].
Glück, L. 1992. The Wild Iris. New York: Ecco Press.
Glück, L. 1995. The First Four Books of Poems. New York: Ecco Press.
Guttman, S. E., Gilroy, L. A. and Blake, R. 2005. Hearing What the Eyes See. Auditory Encoding of Visual Temporal Sequences. Psychological Science 16 (3), pp. 175–179.
Halter, P. 1999. Iconic Rendering of Motion and Process in the Poetry of William Carlos Williams. Form miming Meaning. Ed. by Max Nänny and Olga Fischer. Amsterdam and Philadelphia: John Benjamins, pp. 235–249.
Hartman, C. O. 1980. Free Verse. An Essay on Prosody. Princeton: Princeton University Press.
Have, I. and Pedersen, B. S. 2020. Reading Audiobooks. Beyond Media Borders. Intermedial Relations among Multimodal Media, Volume 1. Ed. by Lars Elleström. Basingstoke: Palgrave Macmillan, pp. 197–216.
Jakobson, R. 1960. Linguistics and Poetics. Style in Language. Ed. by Thomas Sebeok. Cambridge: Cambridge University Press, pp. 350–377.
Johansen, J. D. 2007. A Semiotic Definition of Literary Discourse. Semiotica 165, pp. 107–131.
Kittler, F. 1990. Discourse Networks, 1800/1900. Transl. from the German by Michael Metteer, with Chris Cullens. Stanford: Stanford University Press.
Kjørup, F. 2008. Grammetrics and Cognitive Semantics. Metaphorical and Force Dynamic Aspects of Verse-Syntax Counterpoint. Cognitive Semiotics 2, pp. 83–101.
Lorentz, O. 1980. Poesie fürs Auge. Kritik des freien Verses. Heidelberg: Verlag Lambert Schneider, pp. 83–124.
Ma, M.-Q. 2011. The Sound Shape of the Visual. Toward a Phenomenology of an Interface. The Sound of Poetry / The Poetry of Sound. Ed. by Marjorie Perloff and Craig Dworkin. Chicago and London: The University of Chicago Press, pp. 249–269.
MacArthur, M. J. 2016. Monotony, the Churches of Poetry Reading, and Sound Studies. PMLA 131 (1), pp. 38–63.
MacArthur, M. J., Zellou, G. and Miller, L. M. 2018. Beyond Poet Voice. Sampling the (Non-) Performance Styles of 100 American Poets. Journal of Cultural Analytics 3 (1), pp. 1–72.
MacArthur, M. J., Rambsy, H., Wu, X., Ding, Q. and Miller, L. M. 2022. 101 Black Women Poets in Mainly White and Mainly Black Rooms. Los Angeles Review of Books. Available at: https://lareviewofbooks.org/article/101-black-women-poets-in-mainly-white-and-mainly-black-rooms/ [Accessed: 3 August 2023].
Meza, A. 2023. Entre la voz del poeta y la voz poética: Una escucha asistida por computadora al audio literario de Eduardo Lizalde. PoéticaSonora MX. Available at: https://poeticasonora.unam.mx/entre-la-voz-del-poeta-y-la-voz-poetica-una-escucha-asistida-por-computadora-al-audio-literario-de-eduardo-lizalde/ [Accessed: 8 November 2023].
Middleton, P. 2005. Distant Reading: Performance, Readership, and Consumption in Contemporary Poetry. Tuscaloosa: University of Alabama Press.
Morris, D. 2006. The Poetry of Louise Glück. A Thematic Introduction. Columbia: Missouri University Press.
Müller, W. G. 2010. Metrical Inversion and Enjambment in the Context of Syntactic and Morphological Structures. Towards a Poetics of Verse. Signergy. Ed. by Jac Conradie et al. Amsterdam and Philadelphia: John Benjamins, pp. 347–366.
Novak, J. 2011. Live Poetry. An Integrated Approach to Poetry in Performance. Amsterdam and New York: Rodopi.
Nyberg, L. 2009. Bodies of Poems. Graphic Poetics in a Historical Perspective. Oxford and New York: Peter Lang.
Patterson, I. 2013. No man is an I. Recent Developments in the Lyric. The Lyric Poem. Formations and Transformations. Ed. by Marion Thain. Cambridge: Cambridge University Press, pp. 217–236.
Peirce, C. S. 1932. Collected Papers of Charles Sanders Peirce. Volume 2, Elements of Logic. Ed. by Charles Hartshorne and Paul Weiss. Cambridge: Redknap Press.
Perlman, M., Clark, N. and Falck, M. J. 2015. Iconic Prosody in Story Reading. Cognitive Science 39 (6), pp. 1348–1368.
Perloff, M. 1998. After Free Verse. Close Listening. Poetry and the Performed Word. Ed. by Charles Bernstein. New York: Oxford University Press, pp. 86–110.
Perloff, M. 1999. Language Poetry and the Lyric Subject. Ron Silliman’s Albany, Susan Howe’s Buffalo. Critical Inquiry 25 (3), pp. 405–438.
Pope, A. 1993. An Essay on Criticism. The Norton Anthology of English Literature, Volume 1. 6th edition. Ed. by M. H. Abrams et al. New York and London: WW Norton, pp. 2217–2233.
Serup, M. G. 2017. The Poetry Reading. SoundEffects 7 (1), pp. 44–62.
Smith, J. D., Reisberg, D. and Wilson, M. 1992. Subvocalization and Auditory Imagery. Interactions Between Inner Voice and Inner Ear. Auditory Imagery. Ed. by Daniel Raisberg. Hillsdale: Lawrence Erlbaum, pp. 95–119.
Svensson, J. 2020. Poet Voice på svenska. Poetisk tradition i uppläsningar av Ida Börjel, Athena Farrokhzad och Jenny Wrangborg. Nordisk poesi 5 (2), pp. 92–110.
Svensson, J. 2022. Fjärrlyssning. En kvantitativ studie av uppläst svensk samtidspoesi. Edda 109 (3), pp. 184–205.
The Swedish Academy. 2020. The Nobel Prize in Literature 2020 [Press release]. Available at: https://www.nobelprize.org/prizes/literature/2020/press-release/ [Accessed: 23 September 2023].
Tsur, R. 2008. Toward a Theory of Cognitive Poetics. 2nd edition. Brighton: Sussex Academic Press.
Wheeler, L. 2008. Voicing American Poetry. Sound and Performance from the 1920s to the Present. Ithaca: Cornell University Press.
Wesling, D. & Sławek, T. ١٩٩٥. Literary Voice. The Calling of Jonah. New York: State University of New York Press.