Actual Texts, Possible Meanings: The Uses of Poetry and the Subjunctification of Experience

Abstract Jerome Bruner’s experiment over 30 years ago suggested that imaginative literature had greater affordances for the ‘subjunctification’ of experience by those who heard it read aloud than did transactional prose such as a news article. By ‘subjunctification’, Bruner meant the capacity to use the resource (the short story, for example) to transform one’s experience of the world, to render understanding in more complex ways and to do more than get things done as they have always been done. This paper reports on a small-scale replication of the experiment that sought to measure differences in the affordances of poetry being read aloud compared to hearing a short story or a news article.


Introduction
In A. S. Byatt's novel The Virgin in the Garden, Stephanie Potter, a teacher of English, conducts a lesson on Keats's 'Ode on a Grecian Urn' . As she reflects upon the process she considers how, for her, effective contemplation of poetry is 'not the encouragement of self-expression, self-analysis, or what were to be called interpersonal relationships' but rather 'the induced, shared, contemplation of a work, an object, an artefact' . She goes on to consider the experiences of different readers as they approach poetry: some of whom require 'a key, mnemonics and an analytic blueprint' , those who acquire a 'visual memory of its shape on the page' , others who respond to its words and its 'run of grammatical and punctuational pointers' , and still others who need their minds to be clear of what she calls the 'curious clutter' of expectations of the poem (Byatt 1981, 76-77).
Individuals' responses to poetry vary widely. Many people in spite of (perhaps because of) their experiences at school are wary of poetry. T. S. Eliot recognises this in The Use of Poetry and the Use of Criticism: …difficulty may be caused by the reader having been told, or having suggested to himself that the poem is going to prove difficult. (Eliot 1933, 150) There is a perception that the language and the content of poetry are 'hard' , and this hardness becomes embodied in the very idea of poetry itself.
The exploratory study reported on in this paper forms part of a larger Arts and Humanities Research Council-funded project -'The Uses of Poetry' -concerned with identifying any benefits of experiencing, studying or producing poetry as a part of both formal education and life-long learning. Our study was modelled in part on a small experiment designed by Jerome Bruner (Chapter 2 of Actual Minds, Possible Worlds; Bruner 1986 ) in which a sample of university students was asked to re-tell a story from James Joyce's Dubliners. In our study, we asked a small sample of university students to re-tell a poem, a short story or a newspaper report. The data analysed in this paper emerge from their re-tellings as part of short semi-structured interviews. These were intended to explore whether significant differences emerge between participants' capacity to make meaning in response to different types of text, their willingness to explore alternative meanings and their awareness of how the language of different texts permits certain differentials of meaning. The study, in other words, was designed to explore how and in what ways interpretive responses to poetry might differ from responses to other kinds of written text. We were particularly interested to address a question left somewhat implicit in Bruner's study: does poetry differ as a 'mode of thought' from narrative and the 'paradigmatic'? Is there something distinctive about poetry and the poetic that is visible when the students produce what Bruner called a 'virtual reading' of the text in their re-tellings.

Poetry, narrative and the paradigmatic: different modes of thought and language?
The 'uses of poetry' in an age of narrative and accountability Narrative − drama, novels, screenplays, documentaries, docudramas, soap operas, video games, journalism, etc. -has dominated 'textual' representation in the twentieth and twenty-first centuries (Atherton et al. 2013). In all of these media, prose is dominant. The place of poetry and verse has declined (McAlister 2005;Snapper 2013) and the teaching of poetry seems to be impoverished (for a discussion of poetic culture, see Gioia et al. 2002). A 2012 report from schools' inspectors in England (Ofsted 2012, 15) concluded that in lessons ostensibly focusing on poetry, there was no exploration of language or ideas: 'the text could have been a railway timetable' . The same report speculated that government-initiated accountability pressures are contributing to this situation: 'national tests and examinations have too much impact' (15). Even amongst teachers of English, poetry is often low on their list of cultural preferences, and Cremin (2011) suggests that many English teachers rarely if ever read or listen to poetry unless they are preparing to teach it. The dominance of narrative in our culture combined with accountability pressures of an audit-framed educational system (Strathern 2000) has increasingly narrowed the place of poetry.
However, there is a long tradition of thought in philosophy and literature that argues for the distinctive personal, social and educational value of studying and producing poetry. Horace (following Aristotle) posited the educational function of poetry as a counter to Plato's view of poetry as morally corrupting. Philip Sidney ([1595] 2004) pursued this line with his defence of poetry as both pleasure and moral instruction, developing Chaucer's earlier distinction in The Parliament of Fowls ([1379] 1987, 383, l. 15) between 'lust' (pleasure) and 'lore' (learning). Shelley argues that poetry has a use-value: 'whatever strengthens and purifies the affections, enlarges the imagination, adds spirit to sense, is useful' (Shelley 1932, 97). Heidegger, in 'The Origin of Work of Art' ([1935] 2002), proposes poetry's capacity to 'reveal truth' through what he referred to as 'projective saying' , a form of language that offers the potential for the disclosure of truths by bringing forth articulations which are ordinarily hidden (44-47). More recently, Dylan Thomas asserted that a poem was able to 'change the shape and significance of the universe, helps to extend everyone's knowledge of himself and the world around him' (Thomas 1954, 169). Poetry, according to this tradition of thought, has distinctive properties or affordances for subjective self-realisation and capacity to act on and to shape the world.

'Subjunctivising reality': Bruner and the distinctive power of literary narrative
Jerome Bruner (1986, 11) distinguished between two 'modes of thought' − 'narrative' and 'paradigmatic' − signalling his continuing interest in literature and the power of literary narrative. For Bruner, these modes of thought or 'cognitive functioning' provided 'distinctive ways of ordering experience, of constructing reality' with 'radical' differences in their 'procedures for verification' (11). He did not consider these modes of thought hierarchically, but insisted on the potential of the 'paradigmatic imagination' (13) in scientific discovery. He argued for the meaning-making, truth-revealing potential of literary narrative and explored how the process of reading such narratives worked. In doing so, he drew on a characteristically eclectic range of intellectual resources including reader-response theory, American pragmatist philosophy, Russian formalist linguistics and French post-structuralism, combining these diverse resources with a psychologist's tendency to experiment. Literary narrative permits readers to perform 'virtual re-tellings' that transform texts in ways that he described as 'trafficking in human possibilities rather than in settled certainties' (26).
To attempt to illustrate this process at work he designed two small experiments with his research group. First, they analysed a short story ('Clay') from James Joyce's Dubliners and a chapter from an academic (anthropology) research text. Second, they asked an unspecified small sample of readers to re-tell 'Clay' and they analysed how readers' re-tellings transformed the original texts, using Tzvetan Todorov's (1977) concept of 'transformations' . Transformation, in Bruner's (1986) interpretation, was the way in which a 'simple, expository and non-subjunctive' statement was elaborated to become 'psychological' and 'contingent and subjunctive' (29). These elaborations focused in part on the action of the verb in a sentence. Bruner used the sentence 'x commits a crime' and showed how six simple transformations and six complex transformations (by modifying the verb phrase) could place the activity described in the sentence into 'a landscape of consciousness' (30). So while a simple transformation of mode could subjectify the action by inserting a modal auxiliary verb ('x might commit a crime'), providing an implicit context for the action, a complex transformation ('supposition' being one example) would begin to produce a more complex psychological reality (e.g. 'x foresees he will commit a crime'). This process of transformation -a process of permitting 'discourse to acquire a meaning without this meaning becoming pure information' (Todorov 1977;quoted in Bruner 1986, 31) -'thicken[ed] the connective web that holds a narrative together in its depiction of both action and consciousness' (30).
In their analysis of 'Clay' and the academic (anthropology) text, Bruner's team noted, on average, two transformations per sentence in the literary narrative and one every other sentence in the academic text. Their analysis of the re-telling of 'Clay' (told one day after the student had read the story), suggested that the multiple possibilities for meaning present in the original text were 'preserved' in the reading and re-telling with a greater number of simple transformations in the re-telling than the original and a similar number of complex transformations in both. For Bruner (1986), though, the most interesting feature of the re-telling was the reader's 'management of subjunctivity' (33). The literary narrative text, Bruner speculated, 'needs the subjunctivity that makes it possible for a reader to create a world of his [sic] own' (37). The contingent and subjunctive in the narrative mode does not lead to settled conclusions or 'certainties' , as it might in the paradigmatic mode, but permits 'the varying perspectives that can be constructed to make experience comprehensible' (37). Throughout Bruner's account of these two, small experiments and, indeed, in his general argument about narrative and paradigmatic modes of thought, poetry and the 'poetic mode' receives only scant attention. At times, metaphor and 'poetic translation' prove useful to him in arguing that the kind of transformations he identifies 'substitutes metaphors for both given and new' (24) and create powerful 'gaps' in sense. But Bruner deliberately shifts his focus in his theory about two modes of thought away from poetry. As part of our project, we were interested to explore Bruner's identification of two modes of thought with particular reference to the uses of poetry and to consider, on the basis of an equally small experiment, whether the kinds of resources Bruner drew on would help us to identify a separate 'poetic mode' .

Research design and method
Our study was undertaken on a university campus in the London area. Volunteer participants were sought through an open email call and through a direct approach by the researchers on the day of the activity. As Table 1 demonstrates, students from a range of subjects took part in the study. A sub-sample was then selected in order to provide some balance between male and female participants. Participants (n = 12) were read a short text -four heard a poem: 'Sentences' by Chris Jones; four a short story: ' A Clean, Well-lighted Place' by Ernest Hemingway; and four a newspaper report: 'The Tragedy of Guernica' by George Steer (The Times, 27 April 1937). After hearing the text once, they were asked to re-tell the story through a semi-structured interview. Participants were rewarded with a hot drink and a cake. Ethical approval for the study was given by the university's research ethics committee.
Our study differed from Bruner's in several respects. First, we did not complete an analysis of the three texts using Todorov's transformation concept prior to the research. We were not interested in the linguistic or stylistic properties of the texts but rather the sense people were able to make of them. Second, we completed both aspects of the research on the same day: the students were read and re-told the story in the same session. Third, we developed a short, semi-structured interview schedule and used this in post-reading questioning whereas Bruner just suggests the single reader he mentions was asked 'a good many questions' . Re-tellings − and the interviews in which they took place − were audio-recorded and fully transcribed. The transcripts were then analysed by the researchers in order to identify salient features of the re-tellings and for the participants' comments on the re-tellings and the texts more widely. Two analytic passes were made: first, to analyse the language of the re-tellings with particular attention to the verbs within sentences; second, to develop a broader and more thematic categorisation of the participants' comments on re-tellings and texts.

The role of narrative in creating meaning
'Narrative' -the re-telling of (elements of) the story or possible stories -was a way in which a number of the participants appeared to gain control over meaning and interpretation. When asked what the text they heard was about, the first instinct for many of the participants was to tell us the story, not to seek to establish any more abstract meanings of the texts they heard. This tendency we might refer to as 'narrativization' (White 2004) or 'storyising' .
It was interesting to note that the ways in which participants employed narrative was very different in response to the different text types. Narrative methods of exploring meaning were most evident in response to the poem (three of the four participants); two of the four participants who heard the short story also used narrative in discussing the text; no participants who heard the newspaper report used narrative as a method of making meaning. This interestingly suggests that participants' certainty or uncertainty regarding precise meaning was a factor in determining the extent to which they needed to use narrative to explore or 'play' with textual meaning. Participants who heard the newspaper report were quick and confident in their ascription of meaning. N1, for example, attaches a single meaning to the text: I think it is about a place which was bombed and … I don't think there is another … because it is more fact based or more details. Although there are some emotional verbs or words, it's like 90% objective information.
In spite of identifying 'emotional verbs or words' , which indicates a sense that the writer consciously employs language to colour readers' perception of factual events, the fundamental meaning of the text is perceived as unambiguous. Others hearing the newspaper report proved equally secure in ascribing meaning, albeit with varying degrees of precision: 'the bombing of Guernica' (N2), the bombing of 'a town in Spain' (N3) and most simply 'bombing' (N4).
Two of the participants who heard the short story explored meaning using narrative. S3 uses narrative detail -not all correct -at length: Basically this old man, this old character, is used to coming into this bar to have drinks, but it seems on a frequent basis he leaves there drunk. So a few occasions, about two occasions, he tried to commit suicide, but was rescued by his niece. And the older barman … the older barman [sic] … I don't know, probably like had some issues with this old man drinking so much and … I don't know if he wanted him to kill himself or something … kill himself and … well there's also a part in the story where the old man wanted to buy more drinks but because of the fact the older barman wanted to go home to his wife, wanted to close early, he decided not to sell the drinks to this old man.
It became evident that he was striving for narrative accuracy rather than meaning. It was not until prompted that he was prepared to propose a meaning: 'I think there was a message. I would say if I was to classify this story under one word, I would say … discrimination. ' S1 employed narrative methods in rather a different way, the act of recounting seeming to free alternative but closely related meanings in the text: …after telling it … I mean explaining what happened I can see that there is actually … can be like a different storyline that it's about the loneliness, about the feeling of loneliness, and that the older waiter, he is also kind of hiding in you know like crowded places, in well-lighted places because he thinks that it might help him to like handle his feelings of loneliness.
Here we see a more complex transformational act as envisaged by Bruner in his application of Todorov. The act of recounting the narrative triggers interpretation as the participant adds a layer of complexity to her initial view that the story was about confidence and youth by recognising the presence of its opposite.
In the space between Bruner's 'narrative' and 'paradigmatic' domains, it might be argued, reading and the construction of meaning are essentially acts of exploratory creative synthesis or recreation which relate to both but which sit fully with neither. Such interstitial creative processes recall Mikhail Bakhtin, who reflects on the nature of language as vehicle between addresser and addressee: every word is directed towards an answer and cannot escape the profound influence of the answering word that it anticipates. (Bakhtin 1981, 280) He posits here the possibility of a dialogic relationship between the reader, the author and the text they share, and in so doing focuses on the reader as co-creator. Texts that invite readers to intervene to create meaning he styles 'dialogic' . He also, however, identifies a body of what he calls 'monologic' texts that actively resist the reader's right to participate in meaning creation. Both types of text, of course, spring from the recognition that readers and writers 'share' text, but both seek to use this relationship in different ways. In a similar vein, Roland Barthes (1977) suggests that texts are either 'readerly' -allowing readers few if any rights of interpretation -or 'writerly' -allowing them the freedom to interpret widely. Such distinctions seem useful when considering the varied extent to which participants felt free to 'debate' meaning in relation to different text types.
Three of the participants who heard the poem used narrative as a method of engaging with meaning. P1 offered a straightforward narrative account of the poem, but the other two participants used narrative in rather different ways. P2's response explicitly explored a variety of narrative methods: the concept of imprisonment, the role of and delineation of character, point of view, shifting perspectives and narrative contexts. These were used in relation with one another as a means of (de/re)constructing meaning in the poem. This raises the interesting notion of how poets use narrative and the extent to which poetic narrative might be different to prose narrative. Participants' meaning-making processes, in other words, suggest that -perhaps because of sparsity of narrative detail -there are particular affordances appertaining to poetry. Or it may simply be that poetry is perceived by readers as a different mental and literary 'space' . As much is suggested by P2's observation that 'although it was in poetry format it felt more like a story' .
In a similar vein, P3 identifies a 'problem' in that the poem does not meet some of his expectations of poetry: you didn't think of it as like a poem with line breaks and stuff like that … cos I'm still getting out of the mentality of poetry having to rhyme all the time … so when you hear stuff like that and you think okay poetry can be realistic. This interestingly implies that he does not perceive poetry as being 'realistic' or mimetic, and that the literary 'space' of poetry relates to reality differently than other kinds of text. 'Poetry' in itself is simultaneously perceived as a method of representing and of making meaning: I think like the whole poetry metaphor (?) it's almost like … well it's almost like the state he's in, like what the drugs do to him. Because if you're like on weed and you're on like a real high state, it's like poetry being a high form of art so to speak, it's like a metaphor for that.
His perception of poetry as 'a high art form' suggests particular literary and cultural poetry 'baggage' that impacts upon the ways in which it is read. In referring to poetry as a metaphor, the participant suggests that poetry in and of itself becomes an integral part of the reading and meaning-making process, attributing it particular affordances.
Some participants used personal 'stories' and frames of reference as a way of creating meaning. Dialogic (Bakhtin) and 'writerly' (Barthes) texts encourage such interaction and in so doing automatically open themselves up to plurality. This was most in evidence amongst participants who heard the poem. In Bruner's terms, we may see this as part of the process of subjunctivisation, whereby the individual uses personal narratives as a means of 'building bridges' into the text. Such bridges, we might argue, allow meaning to flow both into and out from the text. As such methods were most in evidence in response to the poem, we postulate certain poetic affordances of language to affect personal frames of reference through the participants' use of personal narrative. P1 framed her response to the poem in empathic terms -'I just sort of pitied him' . Initially such a movement towards the situation of the persona is a means of creating meaning, but also reflexively influences the participant's personal frame of reference. Differently, P3 uses his previous reading and listening to shed light on meaning in the poem: I learnt about how you bring your own experiences or what you know to the poetry, and as we were reading that the things that I was thinking of was like Trainspotting by Irvine Welsh, I was thinking of punk bands like I don't know Remote (?) and Sex Pistols, Drawing the Vision (?) … like all containing people who some sort of internal conflict -that kind of thing.
He then went on to reflect upon the empowering dialogue between his previous knowledge and the poem: So whilst that not be the text exactly it was talking about … because that's what resonates with me -that's what I bring to it and also take back from it … like a give and take kind of thing.
The poem, in other words, becomes absorbed in the participant's cultural frame of reference for use both in 'reading' itself and other texts.

Willingness to engage in interpretation
These ideas in their turn relate interestingly to participants' willingness and ability to engage in interpretive reading. Bakhtin and Barthes recognise that processes of interpretation are affected by the extent to which the meaning of a text is perceived as 'fixed' -the extent, in Bruner's terms, to which a text encourages or discourages readers' use of subjunctivising 'space' . Participants who heard the poetry text were, in general, more likely (if not always happy or successful) to engage in such interpretive acts.
P1 recognises the poem's richness of potential meaning: 'There might be something to learn from it still…' and proceeds to discuss alternative constructions of the title -'Sentences' -looking to create new meaning on the basis of the ambiguity and 'space' afforded by the poet's language. P2 explores the same uncertainty: 'it was interesting cos the title "Sentences" makes you think that it's something to do with literature' . P4 picks up on the prison yard/ school yard dichotomy with which the poem begins and when asked about possible alternative meanings, chooses to 'write' an imaginative extension of the poem rather than looking for new ways of creating meaning within the confines of the text. The participant is working imaginatively with the narrative world of the poem and its characters and situations as a means of creating a transformational or subjunctivising 'space' .
The interpretive acts of those participants hearing the short story were rather different. This is, perhaps, related to the greater certainty they feel about the 'meaning' of the text. All participants hearing the story were happy to ascribe a meaning to it, and their interpretations centre less around negotiations of meaning. S3 explores the impact of opposites in the story, particularly Hemingway's delineation of youth and old age -'okay so old age and youth, they don't have to be so separate, you know' -moving to the interesting and mutually enriching conclusion that: 'You know bringing together two sometimes opposites together to create maybe a new kind of understanding. ' Interpretations of the newspaper report were different again. Whilst unwilling to challenge the basic premise that the text is about the bombing of Guernica, N2 observes: …yeah, there's always different ways of interpreting things, especially when you're looking at history and the way it's reported, because everyone will have a bias in the way they report it. They say history is written by the victor. This suggests that the meaning 'space' of this text might be shaped by alternative 'writings' rather than alternative 'readings' . So, whilst the participants who heard the poem and short story tended (to a greater or lesser extent) to create worlds of their own, as signalled by Bruner's study, here the direction is reversed. N2's analysis resides not with the 'events' of the text but rather with the troublesome issue of authorial intention, related representational choices and the impact of these choices upon readers' response: it felt a little bit like there was an opinion there that the author was maybe trying not to let it out but couldn't help it. So you couldn't tell whether it was supposed to be a piece that was promoting sympathy for the villagers or not.
N3 also approaches issues of meaning that are bigger than the text itself: it talked about a city being bombed and being bombed not for strategic reasons but to crush the spirit of a nation or people.
He then proceeds to the potential political impetus behind the construction of the text: the context in which it was published is what I'm saying, maybe it was to try and rally some more spirit if it was published at the time back in England.
As with N2, interpretation is around the implicit function of the writing rather than its explicit meaning. For N1 this was also an important idea. When asked for potential alternative interpretations she asked: 'do you mean a different way to describe it, do you mean the way he describes it or … [do] you mean the text could have been written in a different way?' The newspaper report, with its explicit meaning, proved more resistant to subjunctivising reader processes.

Impact of language
Different perspectives on the meaning of interpretation itself require us to consider the role of language in different ways. As the previous analysis of participants' interpretations and processes suggests, participants 'placed' themselves very differently in relation to each of the three text types. Varied approaches to text types predicate different readerly 'spaces' , and these in their turn have implications for readers' relation to the roles and affordances of language, their responses to language and their willingness/ability to interpret language. In other words, such perceived relations have implications for Bruner's subjunctification.
This was well exemplified when participants were asked to identify memorable aspects of the texts they heard. The language(s) of the texts functioned differently in each case.
P1 refers to the 'aggressive verbs' the poet employs and his use of 'descriptive language' . She locates the language of the poem clearly in the affective domain. P3 also identifies the affective capacity of language in poetry, picking up on the poet's use of the word 'terriers' and observing how this implies they are 'the best weapon they've got' . She proceeds to explore this interestingly in relation to personal frames of meaning making: if something's put in a context you're not familiar with it makes you ask why. Because for example if it said his words were … I don't know … were cats roaming around the house, I wouldn't really have thought twice about it. Cos (inaudible) a cat roaming around the house, I've never had one, but that's an image I'm familiar with. But when it talks about words being pitbull terriers it makes me think okay terriers are particularly vicious breed of dogs, so what is (inaudible) trying to get across with that.
Poetic language is perceived as more than a catalyst for meaning, it is in and of itself perceived as transformative.
Differently, S1 identifies Hemingway's use of the Spanish word 'nada' as particularly memorable. Although inaccurately defining the word ('necessary'), she understands that Hemingway is using 'some kind of strange pray [sic], with a very strange word "nada"' . Interestingly, the most memorable feature of the text is, therefore, an element that is not understood. This suggests that the uncomfortable place of 'unknowing' can prompt interesting and essential experimentation as a means of building personal meaning.
The same passage was identified by all of the participants who heard the short story. Whilst he did not understand the precise impact of the passage, S2 understood that Hemingway was parodying The Lord's Prayer and saw how this reflected the nihilism of the text. S4 commented interestingly on Hemingway's use of light for deliberate (poetic?) effect: …when he was talking about the electric light like showing the leaves. It's almost like for me I saw that dichotomy between like nature and like technological development or coming together. Similar poetic uses of language for effect were identified by two of the participants who heard the newspaper report. N2 was particularly struck by Steer's use of the phrase 'the rhythm of death': …it's quite catchy, and it really does portray that kind of idea that it's something done to a plan, carefully orchestrated and almost inevitable … yeah once a rhythm's set it continues… N3 observed: I was surprised … for a piece of journalism how creative some of the language was, I guess to paint a picture the word 'crumbling' stood out at the end … rather than being fact based.
He uses the word 'poetic' to describe some of Steer's language and is captured by its power: I found myself drifting in and out of the story and being more consumed by interesting words than the content … the language was quite figurative and descriptive and … so in that way it was sort of pleasurable to hear….
Note, however, that whilst the participants discussing the short story or the newspaper report comment on the writers' use of metaphor and image, the language of poetry seems to enter a new dimension whereby language is in and of itself metaphor. Whilst both the short story and the newspaper report included poetic uses of language, the particular affordances of language and its role in creating meaning in the poem are different.

Conclusion
Commencing from Bruner's exploration of Todorov's idea of transformations, we set out to explore how readers locate themselves in relation to texts of different types, including poetry. We were also alert to Bakhtinian and Barthesian conceptualisations of text, whereby literary texts are seen either as monologic/readerly or dialogic/writerly constructs and accordingly are either 'closed ' (or 'open') to plurality of meaning. Within the reading 'space' of our study, therefore, we sought to identify to what extent participants were faced with differing interpretive possibilities. On the basis of this small study, we would contend that the literary 'space' of poetry, whilst (and perhaps because) it makes particular language demands of readers, offers unique affordances. Readers, faced by a gap between the poem as language object and the poem as meaning are afforded a meaning-making 'space' in which they can deploy their own creative resources. To a demonstrably greater extent than participants hearing the short story and the newspaper report, those who heard the poem engaged in subjunctivising meaning-making processes as conceived by Bruner.
As we are considering the ways in which participants sought to interpret and create meaning from the texts they heard, the extent to which their own use of language was provisional and uncertain became a significant means of identifying this interesting subjunctivising 'space' in which readers doubt or struggle with their right to develop personal meaning from text. In some cases, such hiatuses may have reflected participants' sense that meaning lay somewhere beyond themselves and became somehow disempowering. In several cases, however, this led to creative personal processes through which participants inhabited the gap between what Bruner terms the 'narrative' and the 'paradigmatic' , using theories, imagination and the 'certainties' of personal experience as a springboard for hypothesising about meaning.
The language of the participants who heard the poem was particularly interesting. Words such as 'perhaps' , 'maybe' , 'possibly' and 'might' occur frequently. Their responses are also regularly punctuated by pauses and ellipsis in a way that those who heard the short story and newspaper report were not. This indicates their uncertainty and possibly their openness to plurality of meaning. In Bruner's terms, we were able to identify suggestive evidence of 'contingency' as the participants used different transformative possibilities (subjunctifying processes) as means of taking personal possession of textual meaning.
The small study we have reported upon here was intended to explore the further potentials of Jerome Bruner's original small study in Actual Minds, Possible Worlds. We expanded upon Bruner's original experiment by utilising a variety of text types in order to gather initial data surrounding an idea that Bruner only touches upon -the idea that the language of poetry might embody particular affordances not (or to a lesser extent) present in other kinds of text. On the basis of our findings with a small number of participants, we believe that the students hearing the poetry text, where narrative elements are sparser, were enabled (or obliged) to engage with language and meaning in ways distinctly different to those hearing either the short story or the newspaper report. We are reminded of T. S. Eliot's assertion in 'The Metaphysical Poets' (1951) that in engaging with the poets of the past, new writers are forced 'to force, to dislocate if necessary, language into meaning' . We believe this small study has demonstrated that readers, like the writers Eliot envisages, use language in similar ways through acts of subjunctification and that this is an issue warranting study on a larger scale.

Disclosure statement
No potential conflict of interest was reported by the authors.

Funding
This work was supported by the Arts and Humanities Research Council [grant number AH/ L006324/1].

Notes on contributors
Andrew Green teaches in the Department of Education at Brunel University London where he is Programme Director for the BA in Education. He has published many articles on the teaching of English in school and in Higher Education. Viv Ellis teaches in the Department of Education and Professional Studies at King's College London and as a visitor at Bergen University College in Norway.

Karen
Simecek is Senior Teaching Fellow and Director of Student Experience in the Department of Philosophy at the University of Warwick. She has published articles on the philosophy of poetry in the British Journal of Aesthetics and Estetika: The European Journal of Aesthetics.