Near final draft S.

Whiteley

A cognitive poetic approach to researching the reading experience

Dr Sara Whiteley, University of Sheffield ,UK

Near-final draft, submitted to:
K. Otterholm, K. Ivar Skjerdingstad, L.E.F.McKechnie & P. Rothbauer
(eds.) (forthcoming) Plotting the Reading Experience: Theory, Practice, Politics,
Wilfred Laurier Press.

1. Introduction
This essay investigates the experience of reading a particular poem: ‘I’ll Be There To Love
and Comfort You’ (Armitage 2010), by making connections between the poem’s language
and its experiential effects. My analysis uses the cognitive poetic Text World Theory
framework (Werth 1999; Gavins 2007) to consider the conceptual structures or mental
representations that are prompted by the language of the poem. I also draw on recordings of
readers discussing the poem to gain a sense of the salient experiential features associated with
the text. I argue that cognitive poetic analysis, which models the interaction between text and
mind, can offer some explanation for a number of the experiential effects discussed by
readers. Section 1introduces some of the central principles behind the cognitive poetic
perspective on the reading experience. It also provides some background information about
the reader response data I refer to, and introduces the poem under discussion. My analysis in
Section 2 centres on areas of similarity across the readers’ discussions and relates these
common responses to aspects of the language of the text.

1.1 Cognitive poetics and Text World Theory

Cognitive poetics is an approach to literary study which is informed by research in the fields
of cognitive linguistics, cognitive psychology and the cognitive sciences. It suggests that
readings of literary texts (and other texts, too) may be explained with reference to general
human principles of linguistic and cognitive processing (Gavins and Steen 2003: 2). There
are several variants of cognitive poetics: for instance, Stockwell (2005) distinguishes between
a predominantly North American tradition that emerges more from psychology and
linguistics departments, and a predominantly European tradition more associated with
stylistics or literary linguistics (see also Brone and Vandaele 2009). The approach in this
article is most closely aligned with the European, stylistically influenced strand of cognitive
poetics. Stylistic analysis is underpinned by the assumption that ‘the primary interpretative
procedures used in the reading of a literary text are linguistic procedures’ (Carter 1982: 4).
Leech and Short characterise the concerns of mainstream contemporary stylisticians to be:

1
Near final draft S.Whiteley

‘how, when we read, we get from the words on the page to the meanings in our heads and
effects in our hearts’ (2007: 287). These summaries reflect stylisticians’ interest in the
experience of reading, and also their belief that the language of the text or ‘words on the
page’ have a significant role to play in this experience. Because of its focus on the human
mind, cognitive poetics places extra emphasis on the role of the reader in literary
interpretation, representing a major evolution in stylistics (Carter and Stockwell 2008: 298) .

The central feature of cognitive poetic analysis is the relation of the structure of literary texts
to their presumed or observed psychological effects on a reader (Gavins and Steen 2001: 1).
A common misconception is that cognitive poetics is necessarily reductive and purely
objective in its approach to literature because of its focus on cognitive processes. But, as
Vandaele and Brone point out, it is the ‘felt qualities of mental life’ which cognitive poetics
seeks to address, rather than ‘mere “computation” or “processing”’ (2009: 2). Cognitive
poetics is strongly influenced by the experientialism of cognitive linguistics and is
fundamentally concerned with context: regarding human minds as embodied and embedded
in complex physical, social and cultural situations. Analysts are interested in exploring how
and why particular texts create particular interpretations or effects (Semino and Culpeper
2002: x) and in both the similarities and differences between different reading experiences
(e.g. Stockwell 2002: 1-12).

In cognitive poetics, reading experiences are studied through a variety of methods. Vandaele
and Brône (2009) make a useful distinction between the use of ‘first-person introspection’ on
the part of the analyst and ‘third-person observation’ through more empirical means (2009:
6). First-person introspection involves the analyst explicating their own interpretation or
experience of a text (for example see Semino and Culpeper 2002; Gavins and Steen 2003).
Vandaele and Brône describe first-person introspection as ‘indirectly empirical’ (2009: 7),
because analysts draw upon generalisable, empirically researched principles from linguistics
and cognitive science in the articulation of their analysis. ‘Third-person observation’ is more
directly empirical, meaning (in its widest sense) that it involves the investigation of the
responses of other readers in addition to or instead of the analyst themselves. Third-person
empirical methods range from the use of informal or anecdotal observations regarding other
readers through to more formal qualitative and quantitative studies of reader response. These
third-person approaches are becoming increasingly common as cognitive poeticists seek to
test and develop their claims about literary effect (for example, see Burke 2010; Gavins,

2
Near final draft S.Whiteley

2013; Stockwell 2009; van Peer 1986; Whiteley 2010). This essay draws on both first- and
third-person methods in order to investigate the experience of reading ‘I’ll Be There to Love
and Comfort You’, as I include both consideration of my own reading of the text and the
interpretations and experiences which readers shared during recorded discussions about the
poems.

The cognitive poetic framework which underpins my analysis in section 2.2 is Text World
Theory. Text World Theory is a cognitive linguistic discourse processing framework which
combines a detailed focus on the linguistic features of a text with consideration of the
conscious participation of a reader (for a comprehensive introduction, see Gavins 2007;
Werth 1999). Text World Theory conceives of discourses (meaning instances of linguistic
communication) as operating at two fundamental levels. First, they occur within a situational
context, which is called the ‘discourse-world’. And second, they involve a conceptual domain
of understanding which is jointly constructed by the producer and recipient(s), known as a
‘text-world’ (Werth 1999: 17). The discourse world involves two or more human participants
engaged in linguistic communication, and also incorporates all the perceptual, linguistic,
experiential and cultural knowledge which these participants draw upon during discourse
processing. Text-worlds are the mental representations which participants form in order to
comprehend the discourse. They are constructed through the interaction between linguistic
cues in the text and a discourse participant’s knowledge stores and inferences (Werth 1999:
7). These two levels and their interaction form the foundation of text-world analysis. Most
discourses require participants to imagine multiple text-worlds, which can be richly detailed
or fleeting and undeveloped, and switch between them as the discourse progresses. Text
World Theory is interested in the relationship between the text-worlds created during
discourse and the discourse’s experiential effect.

1.2 Reader discussion data

The reader discussion data I describe in section 2.1 was collected during the University of
Sheffield’s ‘Creative Writing in the Community’ project in 2010-11. My strand of the project
was informed by the cognitive poetic interest in context, and aimed to study how groups of
readers in different contexts in Sheffield responded to the same literary texts (e.g. see
Whiteley 2011b). I asked groups of students, academics and local reading groups to read

3
Near final draft S.Whiteley

three poems by contemporary British poet Simon Armitage. The six groups who took part in
the project are summarised in Table 1.

Table 1: Groups recorded discussing ‘I’ll Be There to Love and Comfort You’
Group No. of Gender Age range Highest Location of Length of
name participants qualification discussion discussion
in English
Academic 8 Mixed 20 – 45 PhD / MA Domestic 1 hour
Group (researcher
present)
Michael’s 8 All male 56 – 75 A-level/O- Domestic +/-30 min
Reading level
Group
Susan’s 8 All female 46 – 75 Degree / O- Domestic +/-30 min
Reading level
Group
Jane’s 5 All female 56 – 75 Degree/ A- Domestic +/-30 min
Reading level
Group
Student 4 Mixed 18 – 19 A-level University +/-30 min
Group 1 room
Student 5 All female 18 – 19 A-level University +/-30 min
Group 2 room

The top four groups are established reading groups who regularly meet in each others’
homes. They agreed to talk about poetry for this project, though they usually discuss
academic papers (in the case of the Academic Group), or novels (in the case of Michael,
Susan and Jane’s groups). The student groups were comprised of first-year undergraduates
studying English at the University of Sheffield who responded to an email asking for
volunteers. The table indicates the age range and gender of the groups, the length and
location of their discussions, and also their members’ highest formal educational qualification
in English (A-levels refer to college education at ages 16-18, O-levels to high school
education at ages 11-16).

When groups of people meet to talk about literary texts, their discussion typically involves
the negotiation of interpretations and accounts of the text’s effects (e.g. Peplow 2011; Swann
and Allington 2009). Such discussions provide an interesting window into literary reception
which cognitive poeticists can use to ‘broaden [their]conception of what responses need to be
explained by textual analysis’ (Myers 2009, see also Whiteley 2011a, 2011b, 2013). Verbal
interaction of this kind does not, of course, provide any privileged insight into the

4
Near final draft S.Whiteley

experiences or cognitive processes of readers as they read the poem, but no method of data
collection fully enables this (Steen 1991b). What it does provide is a ‘third-person’ insight
into the salient experiential features associated with the text in each particular interactional
context, for which cognitive poetic analysis can attempt to account.

I found that despite the different academic contexts, there were often clear parallels between
the way the poems were discussed and the types of responses which participants negotiated
(see below and Whiteley 2011b). Short (1996: xi) notes that, from a stylistic perspective, ‘the
fascinating thing that needs to be explained is that we often agree on our understanding of
poems, plays and novels in spite of the fact that we are all different’. In this article, my focus
will be on some of the broad similarities between the responses to the poem across the
different discussions.

1.3 The poem

‘I’ll Be There To Love And Comfort You’ by Simon Armitage is one of three poems which I
asked the groups to discuss, and is reproduced in Appendix 1. Armitage is a well-known
contemporary literary figure in the UK, whose work is both critically acclaimed and popular.
The poem comes from Armitage’s latest poetry collection Seeing Stars (2010), which is
widely regarded as an experimental departure from his established style. In terms of form, the
poems blur the boundaries between prose and poetry through their unconventional lineation.
In terms of content, reviewers often remark upon ‘how strange Armitage dares to be’ (Noel-
Tod 2010), describing the poems as ‘surreal’ and ‘absurd’ (Ruddock 2011; Noel-Tod 2010;
Beddow 2010). From a cognitive poetic perspective, these poems were interesting because
they seemed to present a particularly distinctive reading experience and I was interested to
see how readers engaged with their more challenging or surprising aspects.

In section 2.1 below I review some of the interpretations and effects which readers discussed.
In section 2.2 I use the Text World Theory framework to consider how the poem’s language
may have influenced the reading experiences which readers share and construct in their
discussions.

2. Analysis of ‘I’ll Be There to Love and Comfort You’

2.1 Reader responses

5
Near final draft S.Whiteley

Across the group discussions, participants were in broad agreement regarding a central
interpretation of the poem as about a couple who have lost a child, and in particular the
narrator’s emotions in relation to this loss. The negotiation of this interpretation in Michael’s
group is shown below:

A: ...the thing comes crashing through the wall and he’s back to something he’s
forgotten maybe and this is his child, maybe a dead child I think
B: Dead child
C: I think it s a dead child and its about loss and how loss is- how the trauma of
loss surfaces in strange ways off beat ways
A: In dreams maybe
(Michael’s Group)

However the discussion groups also negotiated a range of possible interpretative variations
regarding the age of the child and the nature of their death. Participants seemed more willing
to accept diversity in interpretation here, and in many cases referred to sections of the text in
order to support various deductions. Some of this range is illustrated in the excerpt from
Susan’s group:

A: I don’t know if the child was lost it was a miscarriage or a still born or died
very young
B: Or stolen or something [...]
K: I thought it was a slightly older child [...] I think with the fist it’s a sort of
violent thing - its tattooed and horrible and sinister [...] I think that might be
symbolic of some kind of act of violence that the child had died [...]
C: To me all that about ocean depths and the starfish of the child’s hand
swimming and swimming I think it’s a miscarriage
D: Unless it’s a child who drowned
(Susan’s Group)

In addition to this interpretative ambiguity, participants also noted the complexity of the
poem. It was described as ‘puzzling’ (Jane’s Group) and several participants across the
groups described reading it several times in order to make more sense of it.

A: I was confused, I was confused, I didn’t get it until I read it three times
B: Three times?
C: Yeah

6
Near final draft S.Whiteley

A: Yeah I read it two or three times

D: Me too
E: Mmm.
F: Its quite disorientating at first isn’t it? I think
(Academic Group)

Several groups made reference to the sense that the poem has two contrasting parts, and there
is a sense of disjunction between them: for instance a participant in Michael’s Group said:
‘Its in two parts isn’t it an everyday domestic going on next door and a surreal second half’,
whilst Susan’s group said: ‘You think the two things are related but actually they’re totally
unrelated’ (Susan’s Group). In the student and public reading groups, this contrast was
discussed in emotional and evaluative terms. For instance, Susan’s group describe feeling
surprised or unsettled by this disjunction:

 “[When] the poem starts off- you feel quite comfortable, telling a little story and then
all of a sudden something happens and you feel uncomfortable you know its very
odd” (Susan’s Group)
 “When I first read it I thought ‘oh!’, it really was a shock because I wasn’t expecting
that second paragraph, I was expecting a continuation of the first” (Susan’s Group)

Participants in other groups perceived a lack of connection between the two sections as an
unsuccessful aspect of the text:

 ‘it didn’t kind of work for me [...] it was just a sort of one way trip into this sci-fi
fantasy place’ (Student Group 1)
 ‘I couldn’t see what the relationship of the first half was to the second half, I think
each half in its own terms is good condensed descriptive stuff but I mean did anyone
else see a link between the first and second part?’ (Michael’s Group)

In section 2.2 I examine the conceptual structure of the poem and consider how it relates to
the experiences which readers describe.

2.2 The text-worlds of the poem

Like many poems, ‘I’ll Be There to Love and Comfort You’ constructs a narrative voice
which is distinct from the discourse-world author’s. The poem is written in the first-person
which establishes the existence of a narrator who is creating the text-worlds of the poem (see
Gavins 2007: 131-5). Text-worlds are created by two types of linguistic feature known as

7
Near final draft S.Whiteley

‘world-builders’ and ‘function-advancers’ (Gavins 2007: 36, 54). World-builders establish

the temporal and spatial parameters of the world and specify the entities and objects it
contains. Function advancers are propositions which propel the narrative or dynamic within
the text-world forward (Stockwell 2002: 137). The most significant text-world created in the
first nine lines of the poem relates to an event in the past, indicated by the use of past-tense
verbs such as ‘were’ and ‘said’. Temporal locatives lines 1-9 of the poem also indicate the
progression of time from the evening (‘Right then’) to the early hours of the morning (‘the
small hours’). The spatial location is indicated by noun phrases specifying place such as ‘the
house’, ‘the kitchen’ and ‘bed’ and locative prepositions such as ‘next door’. Entities (or
‘enactors’) in this world are nominated by pronouns and noun phrases: ‘I’, ‘Mimi’ and ‘the
couple next door’. Noun phrases also nominate objects present in the text-world, such as ‘a
pot of camomile tea’, ‘the tiled floor’, and so on. Function-advancers are often verb phrases
which indicate actions, events or descriptive relations (analogous to Halliday’s (1985; 1994)
notion of transitivity processes). In lines 1-9, the narrative of the poem is advanced with verb
phrases describing the actions of the couple next door, the verbalisation and action processes
of Mimi and the narrator, and events such as the clattering pans and pounding and
caterwauling.

In addition to establishing the parameters and events of the text-world, world-builders and
function-advancers also work to activate fields of knowledge which are relevant to the text’s
interpretation. Because each reader’s text-world will be fleshed out with aspects of their own
personal, experiential and cultural knowledge, Text World Theory recognises that our
individual mental representations will be somewhat unique (Gavins 2007: 18-33). For
instance, readers will need to activate their knowledge of marital relationships (‘Mimi my
wife’) and domestic locations (‘the house’, ‘the kitchen’, ‘bed’) to form a text-world
representation of the opening lines of the poem. My cultural and experiential knowledge
about marriage and English houses led me to imagine that the narrator is male, that the couple
live in a terraced or semi-detached house (which shares an internal wall with the
neighbouring property), and that when the characters ‘retired to bed’ they went upstairs to do
so. But readers with different cultural knowledge about marriage, or who imagine a bungalow
when they think of a house for example, may well enhance their text-worlds with different
inferential information.

Inferencing is a central process in text-world construction, and the overarching interpretation

of the poem as about child loss is also clearly a product of inferencing rather than explicit

8
Near final draft S.Whiteley

textual cues. A number of features of the text combine to activate our knowledge about the
loss of loved ones, in particular the behaviour of the central characters. For instance, both
characters immediately know who the pronoun ‘her’ refers to in line 26. Mimi has a strong
emotional and possessive relationship to this character (‘Ger her back’, lines 28-9), and the
narrator has specific knowledge about the person’s age (line 27). This young age and the final
image of a ‘child’s hand’ (line 38) provides further input to suggest that the lost person is a
missing child.

As noted in section 2.1, however, readers in the discussion groups seemed quite willing to
accept variation regarding the specific age of the child and the nature of their loss. Obviously,
one explanation for this variation in interpretation must be the different personal and
experiential discourse-world knowledge which readers bring to their understanding of the
discourse. But I would argue that other aspects of the text also work to permit and promote
such variation. Text World Theory enables deeper investigation of the other linguistic cues
which are important in creating this effect.

One such cue is the world-switching which is involved in conceptualising the poem. As noted
in section 1.1, it is common for discourses to create multiple text-worlds. Three worlds which
are central to the poem’s effect are shown in Figure 1 and will be discussed below. Linguistic
cues such as hypotheticals or modal expressions are one type of trigger for the construction of
specialised text-worlds known as ‘modal-worlds’ (Gavins 2007: 91-125). Text World Theory
posits that we conceptualise the relationship between different types of worlds spatially in a
manner which is intimately connected to our spatial conceptualisation of knowledge more
generally (Gavins 2007: 82). Modal-worlds are conceived as being more remote or distant
mental constructs existing at a different ontological level, because they represent unrealised
events or the (unverifiable) attitudes of a narrator. In line 9 of the poem, a modal-world is
created by the line ‘I was dreaming...’ (shown in Figure 1 as ‘modal-world 2’). This
expression indicates that the world-building and function-advancing information that follows
it (the asteroid on a collision course with planet Earth) is not part of the ontological reality of
the original text-world, so this world is formed at a greater conceptual distance.

Something particularly interesting happens in relation to the ontological structure of the

poem’s text-worlds with the use of the phrase ‘when unbelievably’ in line 11. The modal
adverb ‘unbelievably’ cues the construction of another modal-world (shown as ‘modal-world
3’ in Figure 1), in which a fist thumps through a wall. However, the ontological status of that

9
Near final draft S.Whiteley

world in relation to the existing text-worlds is ambiguous. The prefix ‘un-’ indicates the
contents of this world are less believable than the dream and should be held at greater

MODAL-WORLD 2
TEXT-WORLD 1 (‘I was dreaming...’)
time: unspecified past, location: space
evening into early morning objects: asteroid, Planet
location: terraced or semi- Earth
detached house, kitchen,
bedroom MODAL-WORLD 3
entities: narrator, Mimi, the (‘when unbelievably...’)
couple next door time: after dream of asteroid
objects: pot of camomile location: bedroom
tea, tiled floor, bed entities: narrator, Mimi
world objects: fist (knuckles, skin),
switch bedroom wall, headboard,
moon, etc.

Figure 1: A text-world diagram of the three worlds cued by lines 1-11 of ‘I’ll Be There to
Love and Comfort You’. World-builders are indicated. See Gavins (2007) for full diagram
conventions

conceptual distance. But this modal can also be used to express surprise over something
which did happen, for example in an expression like: ‘unbelievably, they gave me my money
back’. This interpretation would decrease the conceptual distance at which modal-world 3 is
constructed and make the narrator’s voice appear more reliable. The content of modal-world
3 also contributes to the ambiguity here. There is a similarity between the world-building
elements in worlds 1 and 3: as both contain the enactors Mimi and the narrator, and both are
set in the bedroom of the house. But there is also a similarity between the function-advancing
elements in worlds 2 and 3: as both involve a collision between a moving and a static object
(the asteroid and planet earth, the fist and the wall).

In my reading of the poem, the similarities between the world-building features of worlds 1
and 3 were the most compelling, and I inferred a connection between these worlds, assuming
that the text had switched from the dream in modal-world 2 to back the initial text-world. So
when the fist is described as punching through the wall in lines 11-12, I expected that it
would belong to one of ‘the couple next door’, previously mentioned. Later, in line 20, when
the narrator peers through the hole in the wall, one would expect the adjoining house to be
visible. However, the couple next door are never mentioned again, and by the end of the

10
Near final draft S.Whiteley

poem they fade from attention. Instead, from line 15 onwards, occurrences in modal-world 3
become increasingly unusual. Mimi isn’t woken by a loud crash, but a quite voice; the hole
doesn’t lead to a house but to a void seemingly filled with liquid and air. As the poem
progresses, these illogical world-building elements mark modal-world 3 as being more
ontologically remote than previous worlds. Readers who assumed modal-world 3 was a
version of the initial text-world are likely to have to reassess this connection.

In terms of conceptual structure, ‘I’ll Be There to Love and Comfort You’ establishes a
trajectory into text-worlds which are increasingly ontologically ambiguous and remote. This
kind of play with conceptual structure is common in absurd and surreal literary texts (Gavins
2000, 2013; Stockwell 2000), and is likely to be one of the reasons behind the confusion and
sense of disjunction which readers discussed in the data in section 2.1. Discourse participants
are usually motivated to construct a coherent representation of the discourse in hand, so when
readers become aware of inconsistencies or illogicalities which may have arisen in their
mental representations, action is normally taken to correct them (Gavin 2007: 142). In the
discussion group data, readers demonstrate a number of approaches to establishing
coherence. One option is to engage in ‘world repair’, in which an aspect of the existing text-
worlds is reconceptualised (Gavins 2007: 142). For instance, some participants
reconceptualise modal-world 3 as a continuation of the dream, or an account of the
supernatural, as the extract from Susan’s group shows:

A: you feel from a certain point that it became a dream [...]

B: It is isn’t it?
C: Depends
D: It depends whether you want to see it as a ghost story or it as a dream or
Many: Mmm
B: Oh I never thought of that, I never thought of it as a ghost story
F: Oh I didn’t think of it like that, no
C: I thought of it as a ghost story
B: //Did you? I thought it was a dream
D: //Did you I thought it was something supernatural
C: Something supernatural yes - but yes a dream is more convincing
(Susan’s Group)

A second, more drastic, option is ‘world replacement’, in which existing worlds are
abandoned and reconstructed anew (Gavins 2007: 142). This kind of strategy is demonstrated
in this comment from Student Group 1:

11
Near final draft S.Whiteley

I thought that the couple next door were this couple [Mimi and the narrator] in the
past, because when the hand comes punching through the wall its as if the past is
coming back, and then when he [the narrator] reaches back [its] as if he is reaching
back into the past for a child

World-replacement involves a significant reorganisation of the text-worlds, and here the

participant disregards linguistic cues in order to split the text into two spatio-temporal zones.
Aspects of the poem are re-interpreted metaphorically, so that the wall between the houses
becomes a division between past and present, and the couple next door become versions of
Mimi and the narrator. Although the text-world of the poem require at least some world-
repair in order to appear coherent, readers will respond differently to this requirement. Some
of the readers quoted in section 2.1 felt that the poem ‘didn’t work’ for them. Their evaluative
comments demonstrate a third option, which is to reject world-repair as being too arduous, or
symptomatic of bad rather than artful writing on the part of the author.

Processes of world-replacement are not the only forms of metaphorical processing prompted
by the poem. The language of the poem itself is highly metaphorical, which also contributes
to the complexity of its text-worlds and its experiential effect. In the discussion data, several
groups made explicit connections between the metaphors in the poem and different ideas
about the child’s age and loss (see for instance the extract from Susan’s group in section 2.1).
I would suggest that the rich knowledge domains activated by the metaphors in the poem are
another significant factor in the interpretative variety in the discussions.

A new understanding of the conceptual basis of metaphor was one of the most significant
developments in the rise of cognitive linguistics in the 1980s (e.g. Lakoff and Johnson 1980)
and more recently conceptual integration has added further sophistication to earlier models
(e.g. Faucconier and Turner 2002). Metaphor is now generally understood as a process of
blending two conceptual domains or ‘input spaces’, so that particular features of those
domains are combined to produce a new and unique mental representation. Gavins (2007:
146-64) describes a metaphor’s ‘input spaces’ as fleeting text-worlds which represent the
conceptual domains cued by the metaphor. These representations are then selectively
combined to form a ‘blended-world’. She argues that metaphors create a sense of ‘double

12
Near final draft S.Whiteley

vision’, as readers conceptualise both the text-world of the discourse and the blended-world
of the metaphor at the same time, making associations between the two.

In the opening line of the poem a metaphor is established when the couple next door are
described as using an abstract phenomenon (a ‘difference of opinion’) to test the physical
properties of a building (‘testing the structural fabric of the house’). One of the input spaces
created by this metaphor is the notion of testing the structure of a building, which to my mind
is something usually carried out by builders or engineers on bricks, mortar or steel, using
some kind of instrument, with the aim of establishing whether a construction has a secure and
safe structure. The second input space is the notion of arguing, which is something carried
out by two or more people, using gesture and language, with the purpose of reflecting and
perhaps settling their disagreements. In the process of interpreting this metaphor, readers
construct a blended-world in which certain aspects of these domains are combined and
reconciled. Blended-worlds have an ‘emergent structure’, meaning that new features which
do not appear in either of the original input spaces can be generated (Gavins 2007: 148). In
my blended-world, the couple, like builders or engineers, are somehow acting on the bricks
and mortar of the house. But their instrument is their argument – which by implication is
loud, emotional and physical enough to have a tangible effect on the bricks and mortar. In my
interpretation of this metaphor, the constructive aims of the builders are replaced by
destructive aims. So the couple seem to be trying to destroy the house with their loud
argument. In the ‘double vision’ created by this metaphor, the contrast between the
understated ‘difference of opinion’ in the text-world and the earth-shattering row in the
blended-world creates humour.

Text World Theory is interested in both the relatively isolated blending processes cued by
sentence-level ‘micrometaphors’ to the more extensive ‘megametaphors’ which are
maintained and developed across entire texts (Gavins 2007: 146-64). At the level of
individual metaphors, there is an interesting shift in their conceptual scope as the poem
progresses. The metaphors which form part of text-world 1 (lines 1-2 and 4-5) have domestic
input spaces which contain similar elements to the text-world they supplement: the walls of a
house and a kitchen. From line 15 onwards however, once modal-world 3 is established, the
metaphors begin to involve more remote input spaces which are temporally or spatially
distant from the target in the text world: a faraway country afflicted by an earthquake (lines
16-19), the ocean depths (lines 23-24), or a nineteenth century London street (lines 32-33).

13
Near final draft S.Whiteley

Everything the narrator sees, feels or hears through the hole in the wall in stanza 2 is
described metaphorically. As the text-worlds of the poem become more conceptually remote,
the metaphors incorporate increasingly novel input domains, which serves to amplify the
inferencing processes involved in reading the text.

Several metaphorical elements in the text also share broad semantic connections. These
semantic fields can act as input spaces for more extensive megametaphors which ‘feed
intricate inferential information into the processing of the entire text’ (Gavins 2007: 155). For
instance, the destruction present in the blended-world I describe above resurfaces
thoroughout the first stanza: in the pans falling to the floor (lines 4-5), and the rubble caused
by the fist / earthquake (lines 11-19). Another megametaphor has water-related input spaces,
such as the ‘life forms...in the ocean depths’ (line 23), the foggy river (lines 32-4) and the
‘pulsing starfish...swimming’ (lines 38-9). These water-related metaphors seemed particularly
influential in some readers perception of the child as neonatal, as indicated by a quotation
from Susan’s group in section 2.1. During the discussions I recorded, further connections
between these megametaphors were negotiated and developed by participants. For example,
Jane’s group linked the domain of destruction with the watery, neonatal imagery, with one
participant remarking: ‘Birth is quite violent, like something smashing through into
existence’. When interpreting the poem, megametaphors can be combined and weighted
differently to produce different effects, both at the moment of reading and during discussions
about the text. The multiple blended-worlds which are formed during the reading of this
poem exist alongside the text-worlds and contribute to variations in readers’ interpretative
responses.

3. Conclusions

Text World Theory analyses seek to account for the experience of reading through close
attention to the linguistic and conceptual structures involved in processing a particular
discourse. I have argued that the experience of reading ‘I’ll Be There to Love and Comfort
You’ is particularly influenced by the text’s trajectory into more conceptually remote text-
worlds, the way it disrupts connections between those worlds prompting world-repair, and the
rich, interconnecting metaphors which amplify the complexity of readers’ mental
representations. By relating my analysis of the text’s language to the interpretative accounts
which arose in readers’ discussions about the poem, I hope to have shown that Text World

14
Near final draft S.Whiteley

Theory has the explanatory power to consider both convergent and divergent responses to a
text. Whilst it has not been possible to do full justice to the complexity of the poem or
readers’ discussions here, I hope to have demonstrated some of the central principles of
cognitive poetics as an approach to studying the reading experience.

Note on transcription conventions

Discussion extracts have been transcribed as prose for ease of reading. Inverted commas are
used to show reported speech and question marks to indicate questions. Double slashes (//)
indicate overlapping utterances. Ellipses in square brackets ([...]) indicate sections omitted for
brevity. Comments in square brackets and underlining are my additions to aid clarity.

Acknowledgements
I am grateful to the Higher Education Innovation Fund at the University of Sheffield for
funding this research as part of the ‘Creative Writing in the Community’ project. Thanks also
to Joe Bray, Sam Browse, Alison Gibbons and David Peplow for their helpful input with
earlier versions of my analysis.

References
Armitage, S. (2010) Seeing Stars. London: Faber and Faber.
Beddow, A. (2010) ‘Seeing Stars by Simon Armitage’, The Literateur, available at:
http://literateur.com/seeing-stars-by-simon-armitage/
Brône, G. and Vandaele, J. (2009) Cognitive Poetics: Goals, Gains and Gaps, Berlin:
Mouton de Gruyter.
Burke, M. (2010) Literary reading, cognition and emotion: An exploration of the oceanic
mind. London: Routledge.
Carter, R. (1982) Language and Literature, London : George Allen Unwin.
Carter, R. and Stockwell, P. (2008) ‘Stylistics: retrospect and prospect’, in R. Carter and P.
Stockwell (eds) The Language and Literature Reader, London: Routledge, pp. 291-
302.
Fauconnier, G. and Turner, M. (2002) The Way We Think: Conceptual Blending and the
Mind’s Hidden Complexities, Plymouth: Basic Books.
Fauconnier, G. and Turner, M. (2002) The Way We Think: Conceptual Blending and the
Mind’s Hidden Complexities, Plymouth: Basic Books.
Gavins, J. (2000) ‘Absurd Tricks with bicycle frames in the text worlds of The Third
Policeman’, Nottingham Linguistic Circular 15: 17-33.
Gavins, J. (2007) Text World Theory: An Introduction, Edinburgh: Edinburgh University
Press.
Gavins, J. (2013) Reading the Absurd. Edinburgh: Edinburgh University Press.
Gavins, J. and Steen, G. (eds) (2003) Cognitive Poetics in Practice, London: Routledge.
Halliday, M.K. (1985) An Introduction to Functional Grammar, London: Arnold.

15
Near final draft S.Whiteley

Halliday, M.K. (1994) An Introduction to Functional Grammar, 2nd Edition, London: Arnold.
Lakoff, G. and Johnson, M. (1980) Metaphors We Live By, Chicago: University of Chicago
Press.
Leech, G.N. and Short, M. (2007) Style in Fiction, 2nd Edition, Harlow: Pearson Education
Ltd.
Myers, G. (2009) ‘Stylistics and “reading in talk”’, Language and Literature 18 (3): 338-44.
Noel-Tod, J. (2010) ‘Seeing Stars by Simon Armitage: review’, The Telegraph, available at:
http://www.telegraph.co.uk/culture/books/bookreviews/7851863/Seeing-Stars-by-
Simon-Armitage-review.html
Peplow, D. (2011) 'Oh, I've known a lot of Irish people'. Reading groups and the negotiation
of literary interpretation. Language and Literature 20(4): 295-315.
Ruddock, S. (2011) ‘Book Review: Seeing Stars by Simon Armitage’, available at:
http://www.writerscentrenorwich.org.uk/bookreviewseeingstarsbysimonarmitage.aspx
Semino, E. and Culpeper, J. (eds) (2002) Cognitive Stylistics: Language and Cognition in
Text Analysis, Amsterdam: John Benjamins.
Short, M. (1996) Exploring the Language of Poems, Plays and Prose, Harlow: Longman.
Steen, G. (1991b) ‘The empirical study of literary reading: methods of data collection’,
Poetics 20: 559-75.
Stockwell, P. (2002) Cognitive Poetics, London: Routledge.
Stockwell, P. (2009) Texture: A Cognitive Aesthetics of Reading, Edinburgh: Edinburgh
University Press.
Stockwell, P. (2000) ‘(Sur)real stylistics: from text to contextualizing’, in T. Bex, M. Burke
and P. Stockwell (eds), Contextualized Stylistics, Amsterdam: Rodopi.
Swann, J. and Allington, D. (2009) ‘Reading groups and the language of literary texts: a case
study in social reading’, Language and Literature 18 (3): 247-64.
van Peer, W. (1986) Stylistics and Psychology: Investigations of Foregrounding, London:
Croom Helm.
Werth, P. (1999) Text Worlds: Representing Conceptual Space in Discourse, Harlow:
Longman.
Whiteley, S. (2010) ‘Text World Theory and the Emotional Experience of Literary
Discourse’, unpublished PhD thesis, University of Sheffield, UK.
Whiteley, S. (2011a) ‘Text World Theory, Real Readers and Emotional Responses to The
Remains of the Day’, Language and Literature 20(1): 23-42.
Whiteley, S. (2011b) ‘Talking about ‘An Accommodation’: the implications of discussion
group data for community engagement and pedagogy’, Language and Literature 20 (3):
236-56.
Whiteley, S. (2013) ‘Ethics’, in P. Stockwell and S. Whiteley (eds) The Cambridge
Handbook of Stylistics, Cambridge: Cambridge University Press.

16
Near final draft S.Whiteley

Appendix 1

I’ll Be There to Love and Comfort You

1 The couple next door were testing the structural fabric

of the house with their difference of opinion. ‘I can’t
take much more of this,’ I said to Mimi my wife. Right
then there was another almighty crash, as if every pan
5 in the kitchen had clattered to the tiled floor. Mimi said,
‘Try to relax. Take one of your tablets.’ She brewed a
pot of camomile tea and we retired to bed. But the
pounding and caterwauling carried on right into the small
hours. I was dreaming that the mother of all asteroids
10 was locked on a collision course with planet Earth,
when unbelievably a fist came thumping through the
bedroom wall just above the headboard. In the metallic
light of the full moon I saw the bloody knuckles and a
cobweb tattoo on the flap of skin between finger and
15 thumb, before the fist withdrew. Mimi’s face was
powdered with dirt and dust, but she didn’t wake. She
looked like a corpse pulled from the rubble of an
earthquake after five days in a faraway country famous
only for its paper kites.

20 I peered through the hole in the wall. It was dark on the

other side, with just occasional flashes of purple or green
light, like those weird electrically-powered life forms
zipping around in the ocean depths. There was a rustling
noise, like something stirring in a nest of straw, then a
25 voice, a voice no bigger than a sixpence, crying for help.
Now Mimi was right next to me. ‘It’s her,’ she said. I
said, ‘Don’t be crazy, Mimi, she’d be twenty-four by
now.’ ‘It’s her I tell you. Get her back, do you hear me?
GET HER BACK.’ I rolled up my pyjama sleeve and
30 pushed my arm into the hole, first to my elbow, then as
far as my shoulder and neck. The air beyond was
clammy and damp, as if I’d reached into a nineteenth-
century London street in late November, fog rolling in up
the river, a cough in a doorway. Mimi was out of her
35 mind by now. My right cheek and my ear were flat to the
wall. Then slowly but slowly I opened my fist to the
unknown. And out of the void, slowly but slowly it
came: the pulsing starfish of a child’s hand, swimming
and swimming and coming to settle on my upturned
40 palm.

by Simon Armitage (2010)

17