In our everyday reading practice we rarely think about space; our eyes fixate on the printed words, following the direction of the text from left to right and from top to bottom. Despite the fact that all printed literature, be it prose or poetry, is defined by gaps – between the words and paragraphs, margins, headings and subheadings – we hardly notice the actual space of the page.

By contrast, visual poetry (and, more specifically, concrete poetry) breaks free from these traditional linear structures and treats the page like a canvas. As much a picture as a text, these poems use space creatively and invite the reader/viewer to interact with it. In fact, since linear reading is disrupted, the reader must use the space to navigate their way through the poem. The French Symbolist poet Stéphane Mallarmé was the first to fully exploit the creative possibilities of the page in his poem Un Coup de dés jamais n’abolira le hasard.

Some visual poems even make space an integral part of the meaning, creating clearly distinguishable gaps and interruptions. Eugen Gomringer’s concrete poem Silencio from 1954 is the most famous example of this tendency.

In the ‘Poetry Beyond Text’ project, we have been interested in how readers of visual poetry interact with, and derive meaning from, space. Is it possible to ‘read’ space and if so, how does this affect our understanding and experience of a poem? Does space render meaning ambiguous? Does it open up or close down interpretive possibilities? Our research has focused particularly on the disruptive function of space, in which we attempt to understand at what point, and to what extent, spatial layout thwarts the reading process.

In order to explore these questions, team members at the University of Kent conducted a series of experiments that would provide us with some information about how readers of visual poetry perceive, understand and process space. We used the same fourteen co-researchers for all four experiments detailed below:

Experiment 1: Explorative Approach

Participants were shown a selection of visual poems in an eye tracking laboratory, which they viewed on a computer screen, moving through each one at their leisure. Their eye movements and pupil dilation were recorded in real time. They were then asked to fill in a questionnaire which asked them about their enjoyment of the poems and how they perceived the function of space. They were presented with a series of statements about the role of space and were asked to indicate which of the statements best applied to each poem.

Eye Fixation Heat Map

Experiment 2: Reading Space in Un Coup de dés jamais n’abolira le hasard

In this experiment, we attempted to test some of the claims made by Virginia La Charité in her book The Dynamics of Space: Mallarmé’s Un Coup de des jamais n’abolira le hasard (Lexington, KY: French Forum, 1987) about the way the space of the page directs and determines the reading process. The experiments were carried out in groups of three or four. We gave each of the participants a copy of the translated version of the poem and asked them to fill in an evaluative questionnaire, as well as a series of statement derived from La Charité’s book, to which they were asked to respond with true or false answers. These activities were then followed by an open discussion, which was recorded.

Experiment 3: The Disruptive Function of Space

Cognitive theorist Reuven Tsur in Toward a Theory of Cognitive Poetics (Brighton; Portland: Sussex Academic Press, 2008) claims that poetry is ‘organised violence against cognitive processes’. He states that the reading of poetry involves the modification (or, sometimes, the deformation) of cognitive processes, and their adaptation for purposes for which they were not originally “devised”. We were interested in whether these claims could be tested empirically in relation to spatially manipulated texts.

In this experiment, we took a piece of prose and three poems (one in traditional verse layout, and two visual poems with one more radically spaced out than the other) and manipulated each one to look like each of these four layouts. Each participant was presented with one poem from each set on the eye tracker. They were then asked to fill in an evaluative questionnaire. At the end of the experiment they were asked which of the texts they had seen had been spatially manipulated.

Experiment 4: Creative Exercise – Cut-ups

The spatial position of words functions as an additional means of expression in visual poetry, which complements, complicates and/or enhances the semantic message of the text. This experiment was designed to explore whether spatial information can be retrieved just as semantic information. It was carried out in groups of three or four. Participants were shown a visual poem – ‘here are the lovers’ by Augusto de Campos – and were asked to rate their enjoyment of the poem and to comment on its meaning. They were then given a cut-up version of the poem and asked to recreate the original from memory using these individual components. Seven participants received the original words, whilst the other seven were given black lines approximating the length of the words. These two control groups were used to find out the extent to which semantic clues affect the memory of shape and layout.

What did we find?

In Experiment 1 we found that:

  1. The more linearly the words of a visual poem are arranged, the more likely it is that the reading pattern follows a top to bottom, left to right pattern.
  2. When the sign systems are mixed together in a more radical fashion, and when readers are perplexed and uncertain about which of them dominates (i.e. the verbal or the visual elements), the confusion as to which perceptual mode they should adopt is visible in their eye movements.
  3. In those poems where conceptual and material spaces overlap, where an empty space enclosed within a poem signifies, such as in ‘Silencio’, most participants actually fixate on the empty space whilst they ponder the meaning of the poem.

Experiment 2 revealed little evidence to suggest that Mallarme’s use of space and fragmented textual layout was perceived negatively by the co-researchers. La Charité’s claim that the poem creates a crisis for the reader was difficult to quantify, since, whilst the poem was seen as challenging, it was not really considered to be frustrating or confusing. In fact, one issue that came up in the discussions was that being challenged is most often a positive experience that forces one into an intellectual engagement with the work.

One of the main observations to emerge from Experiment 3 was that the co-researchers seemed to have an instinctive sense of how the poems should be laid out. Their own reading of the content of the poem and the themes expressed by it appeared to have influenced the way they reacted to the spatial layout. Semantic meaning was related to the visual layout, and where there was a strong correspondence between the two the response was more positive. One of the more surprising findings was that a lack of space can create as much cognitive rupture as an excess of space, and semantically difficult poems require more space for intellectual activity to take place.

It can be asserted from the results of Experiment 4 that the perception of space and spatial layout plays a key role in the reading/viewing of visual poetry. Whilst readers inevitably pay close attention to the words, searching for meaning on a semantic level, the formal properties of the work play an integral role in this process. As with the previous experiments, CRs judge the poems on the basis of whether or not they perceive the pictorial qualities as supporting the semantic meaning, and vice versa.