Reading literature in/against the digital age: Shallow assumptions,deep problems,expectant pedagogies

Abstract

Following the popular theme of Nicholas Carr’s The Shallows: What the Internet Is Doing to Our Brains (2010), various commentators observe the erosion of what used to be called ‘reading’, but is now increasingly referred to as ‘deep’ reading. University English departments offer ‘close reading’ of literary texts as a corrective to the digital shallows, while worrying about a general decline in reading, and reading ‘for pleasure’, attributed to the digital. However this decline is created equally by the procedures of the discipline, including its core practice of ‘close reading’ which emerged similarly from entangled anxieties about technology, mass media and low/high culture. It makes sense that literary reading would be eroded in educational contexts (which overdetermine reading), and not just because of the erosion of deep reading and ‘deep attention’. Training in literature may lead, ironically, to a loss in the ability to read and understand literary texts and to draw on multimodal narrative literacies. This paper proposes a practice of expectant reading that, supported by social annotation, can re-centre reading and restore narrative and frameworks of expectation constitutive of literary meaning, while embodying the contract between reader(s) and text, and facilitate socially distributed reading.

Deep reading, or, reading

There is now no doubt among scholars that interaction with digital platforms is changing how we think. As the popular narrative has shifted from the hopes of the digital ‘information economy’ to fears about the ‘attention economy’ (Herbert, 1971), notably from an economy of abundance to an economy of scarcity, the cognitive effects wrought by the internet have registered particularly (in popular writing about the internet) as a transformation in reading (Carr, 2010; Hari, 2021; Jackson, 2008; Jacobs, 2011; Ulin, 2010; Wolf, 2018). Reading, the reading of literary texts, has become a focal point for a swirl of anxieties about the digital bordering on a moral panic. Following the popular – and evidently catchy – theme of Nicholas Carr’s The Shallows: What the Internet Is Doing to Our Brains (2010), various commentators have observed with alarm the erosion of what used to be called ‘reading’ but is now increasingly referred to as ‘deep’ reading.

The phrase ‘deep reading’ appeared as early as 1994, in Stephen Birkerts’ The Gutenberg Elegies: The Fate of Reading in an Electronic Age (1994: 19), which set the tone for nearly three decades of elegiac musings on the loss of attention and deep reading in a digital world (always the elegy never the eulogy). Deep reading has been discussed, increasingly, in terms of ‘deep attention’ (Carr, 2010; Hayles, 2007, 2010; Wolf, 2018). The term appears nowhere in Herbert Simon (1971), in Goldhaber (1997), nor in Davenport and Beck’s The Attention Economy (2001), but entered the lexicon in 2007 with Katherine Hayles’ ‘Hyper and Deep Attention: The Generational Divide in Cognitive Modes’. ‘Deep attention’ – which Hayles associates explicitly with thinking in the Humanities - has also been taken up as a term in psychology, and in corporate and self-help literature. Cal Newport in Deep Work: Rules for Focused Success in a Distracted World (2016) repackaged the notion of flow as ‘deep work’, contrasting this with ‘shallow work’. The goal of ‘deep learning’ became almost paradigmatic in education, following Marton and Saljo’s (1976) investigation of reading and recall, building on the theory of processing levels (Craik and Lockhardt, 1972). It is worth noting that as educationalists use these terms, they have more to do with worries about deep reading than with ‘deep learning’ and ‘deep attention’ in machine learning. The deep in machine learning is less qualitative (and metaphorical), more concrete and quantitative, ‘depth’ referring to the number of node layers in a neural network. ‘Deep attention’ as it circulates outside of these technical discussions, appears to have been born quite specifically out of the potent combination of rueful nostalgia and digital moral panic that has placed anxiety about reading at the centre of the discourse about (in)attention and the dangers of the digital. ‘Deep’ and ‘active’ reading thus emerge together as periapts of the digital moment.

‘Deep reading’ is difficult to define, largely because so many discussions, oriented to the shallows, focus on what deep reading is not. As English Professor Richard Nordquist summarises, shepherding a disparate set of quotes from disparate sources into a ‘Guide to Deep Reading’: ‘deep reading is the active process of thoughtful and deliberate reading carried out to enhance one’s comprehension and enjoyment of a text’ (2019). It is difficult to imagine the inactive process of thoughtless and accidental reading that is carried out neither for comprehension nor for enjoyment. Even more difficult to find this cognitive practice in actual readers. But Nordquist’s definition is rhetorical, finding coherence through a set of oppositions. Of course, as Brock Haussamen (1995) trenchantly observes in ‘The Passive Reading Fallacy’, there is no ‘passive reading’ – only a binary fiction, dreamed up by champions of ‘active reading’: ‘[All] reading is an integration of the active - the purposeful, the transforming - and the passive - the receptive and reactive’ (378–379).

Carr cites a study of MRI activity of subjects reading a short story: ‘[d]eep reading, says the study’s lead researcher Nicole Speer, ‘is by no means a passive exercise’ (2010). What Speer actually concludes is that ‘reading is by no means a passive exercise’, though Carr’s citation has circulated more widely than the study (Speer et al., 2009 cited in Carr, 2010). In ‘The Importance of Deep Reading’ (2009), Maryanne Wolf and Mirit Barzillai explain: ‘[b]y deep reading we mean the array of sophisticated processes that propel comprehension and that include inferential and deductive reasoning, analogical skills, critical analysis, reflection, and insight’ (32). Again, as with Nordquist’s definition, it is difficult to conceive of reading any text without activating this array of processes (unless one is merely sounding out the letters for another to interpret). Citing Wolf and Barzillai’s definition, David M. Durant silently replaces ‘deep reading’ with ‘linear reading’ (2017: 5). In this case, the modification seems justified; what they primarily experience as disrupted by the digital is the experience and practice of continuous linear reading.

‘Passive reading’ is clearly not a useful concept, despite its prevalence in educational practice and literature promoting ‘active reading’ via various classroom techniques. But it does have a clear usefulness in giving apparent shape and definition to notions of ‘active reading’. Though ‘active reading’ must logically suffer the same problem (in distinguishing itself from non-active or inactive reading), the fiction of ‘passive reading’ reinforces claims made for ‘active reading’ by a binary logic. This binary is supported by, and in its turn supports, other binaries like deep versus shallow. Thus, these ways of understanding and characterising reading come to appear obvious and natural insofar as they fit (and appear confirmed by) other binaries swirling around in the moral digital panic: if the internet is making us passive and superficial, deep and active reading will save us. But these terms are in fact often in tension.

Anxiety about passivity in the digital sphere requires ‘deep reading’ to be ‘active’ in a heightened way: resistant, critical and sceptical. In Reader Come Home (2009), Wolf worries about the ability to filter information and misinformation from ‘the daily deluge of eye-byte-sized information’ (22) and, in ‘The Importance of Deep Reading’, the urgent need to train ‘young readers to be purposeful, critical, and analytical about the information they encounter’ instilling ‘the kind of strategic thinking that is vital for online reading comprehension’ (Wolf and Barzillai, 36). Yet, what she experiences as lost and potentially under threat at the start of Reader Come Home is precisely an experience of surrender: ‘I still bought many books, but more and more I read in them, rather than being whisked away by them. At some time impossible to pinpoint, I had begun to read more to be informed than to be immersed, much less to be transported’ (145). This confusion emerges routinely in accounts of deep reading which call forth a champion against the shallows and will clepe ‘deep’ whatever might stand against the shallows so that vigilant critical filtering and immersion in a literary narrative come to be called by the same name.

‘Deep reading’ has also been identified with ‘slow reading’. Wolf and Barzillai perceive ‘slower, more time-consuming cognitive processes that are vital for contemplative life’ as being at the heart of deep reading (32). For Nordquist, deep reading is also slow reading, defined against skimming, scanning and other techniques of speed reading once considered specialised skills, technologies even, gained through active, conscious practice (as in Evelyn Wood’s Reading Skills, 1958), but now increasingly associated with passive digital habits. The slow/fast reading binary appears less muddy than either the active/passive or deep/surface reading binary. Fundamentally ‘slow reading’ denotes simply reading at an intentionally slowed pace. Without prescribing any specific type of reading, only a conscious decision to read more slowly, champions of the practice suggest that slow reading can improve comprehension, or allow for immersion, or a combination of the two. This seems to answer deep reading’s desire to have it both ways. However, in looking at the discussions of slow reading some of the same confusions start to creep in, with slow reading variously identified with ‘deep reading’, ‘critical reading’, ‘immersive reading’ and ‘close reading’. Moreover understanding ‘slow’ reading as a key component of slow education, within the broader ‘slow movement’, offers some resistance to such inclusion (discussed below). Essential to note here is that slow reading presupposes linear reading: the measured pace of slow reading is measured along a linear reading path.

Whether reading of literature is in fact in decline as a result of the digital, or in decline at all, is debatable. The practice of literary reading however plays a central role in accounts of digital strain, as the site where the disruption of continuous linear reading registers most strongly – where a tendency to shift reading mode, to shift the style or even the site of reading (hyper reading) is experienced as more disruptive, less easily disguised or integrated, more easily recognised than in non-literary reading practice that offers the continuity of reading to be informed. It is clear that, while immersive, literary reading is essential to these accounts, the binaries they depend on, binaries of deep versus shallow, deep versus surface, and active versus passive, taken separately, and especially taken together, proffer a weak and confused account of literary reading – or, continuous linear reading of a literary text.

Close reading and the disciplining of literature

English literary studies should yield a better understanding of literary reading, although the study of reading is surprisingly marginal in the discipline. Despite a general consensus that departments of English (literature) teach students how to read literary texts, understanding reading is peripheral to the intellectual labour of the majority of literary scholars and teachers. Recent critical efforts to unseat depth-seeking hermeneutics have coined a whole catalogue of types of ‘reading’ – ‘surface reading’ (Best and Marcus, 2009), ‘reparative reading’ (Sedgwick, 2002), ‘generous reading’ (Bewes, 2010), ‘thin description’ (Love, 2010), ‘allied reading’ (Squibb, 2015). But these are concerned with critical writing and the tension between interpretation and description, not with reading per se, and postcritique has had limited impact on classroom practice. ¹

While worrying (and complaining) about a general perceived decline in reading ‘for pleasure’ (immersive reading), widely attributed to generational habits that are also digital habits, university English Departments offer ‘close reading’ of literary texts (‘slow’, ‘difficult’ literary works), as a corrective to the digital shallows. In the main, however, English Departments do not study ‘deep’ literary reading, teach it, nor even necessarily effectively promote it. And this is not because they also increasingly offer the application of ‘close reading’ techniques to digital media and to texts that are not slow, difficult, nor literary. This article argues that, if literary immersion is being lost, crowded out in the digital shallows, English literary studies do not offer an obvious solution to this. The procedures of the discipline of English, including its core practice of ‘close reading’ contribute to, as much as they address, the erosion of ‘deep’ literary reading elegised in the literature of digital panic.

As Jane Gallop notes ‘English’ became a legitimate profession when its practitioners ‘stopped being armchair historians and became instead painstaking close readers’ (2007: 135). ² That happened in the 1920s under the influence of I.A. Richards and ‘practical criticism’ in Britain, and the ‘New Critics’ in America, who broadly promoted literary criticism as a practice of giving an account of what a text means or does by reading for how it means or does it. This idea of the English scholar as literary critic and close reader reached a high point of dominance in the 1950s and 60s, after which its appeal, like that of the larger Modernist programme it had served, necessarily declined. In the 1970s English Studies secured a professional (and elite) identity by embracing ‘high’ theory (and distancing itself from the popular implications of reader response theory). ³ But close reading remained entrenched in university and school classrooms.

Close reading’s rise can also be placed in the tradition of Gutenberg elegies traced at the beginning of this paper. In the early 1960s, Marshall McLuhan, who had studied at Cambridge under I.A. Richards, published The Gutenberg Galaxy: The Making of Typographic Man (1962), an obvious precursor to The Gutenberg Elegies, although Birkert’s book was written at the dawn of the World Wide Web. Even in the 1920s the burgeoning movements of Practical Criticism and New Criticism were, to a significant degree, a response to technology and mass culture. In the context of mass education, close reading offered a comparatively simple and accessible method for literary analysis, that proved ‘eminently teachable’ and open to ‘practitioners of varying sophistication’ (DuBois, 2003: 2), as Gerald Graff puts it ‘not just the student body but the new professors as well, who might often be only marginally ahead of the students’ (2007: 173). Its influence has even been described as ‘democratizing’. ⁴ But behind the impulse to spread literary education was a strong concern with protecting the study of literature against lower and common forms of culture and the effects of new mass media (Guillory, 2010). As Timothy Saunders notes, proponents of close reading have been ‘vexed by the threat they believe new media has posed to the public’s ability to read literature more generally, in the form of cheap paperback fiction, advertising, television, and cinema in the early 20th century, and of digital technology more recently’ (2020: 145).

Close reading has evolved strangely alongside the internet, which, as I will show, both supports and undermines its practices. English departments, again anxious about their place in scholarship (and under threat), have more or less accepted the status quo of an online environment that mocks the reading work of the classroom and makes practical criticism impossible. Yet, at the same time, the culture of classroom close reading has been hostile to the richness the internet can bringing to literary study - not only in distant and machine reading, ⁵ but in the circulation of popular media that proponents of close reading wished to close out and protect readers from, and the interpretive practices of vast online communities. ⁶

Slowing, digging, drifting

On the axis of shallow, passive reading and deep, active reading, close reading is supposed to be speed reading’s opposite. But both are extractive reading strategies. Although speed reading is fast, skimming or sifting for what can be quickly gleaned, and close reading is slow, plumbing the depths of a small area of text, in either case the aim is to maximise the successful extraction of information, whether horizontally or vertically. Speed reading is like strip mining and close reading like shaft mining. To maximise extraction, both approaches jettison the experience, here the flow, of reading. While the speed reader rushes forward, skimming over any detail in excess of her basic informational needs, and the close reader pauses, probing the text, both strategies resist the text’s temporal experience, in order to extract more from the text that a reading at ‘normal’ (roughly spoken) speed would allow.

When Elaine Showalter states that ‘close reading is slow reading’, she seems to describe reading that resists the flow of reading – both pace and immersion.

Close reading is slow reading, a deliberate attempt to detach ourselves from the magical power of story-telling and pay attention to language, imagery, allusion, intertextuality, syntax, and form. […] In a sense, close reading is a form of defamiliarization we use in order to break through our habitual and casual reading practices (2002: 28).

Arresting the flow of the text yields a laundry list of items that can be used to produce the text of the college exam-essay: ‘language, imagery, allusion, intertextuality, syntax, and form’. For Showalter, the search for these items ‘forces us to be active rather than passive consumers of the texts’ (28). While the cognitive work of reading literature (the sophisticated processes of deep reading that Wolf and Barzillai describe) is imagined as ‘passive,’ the extractive labour of close reading is figured as active. Like speed reading, close reading is selective. The close reader, equally, moves rapidly between extraction points, as is clear in the conclusion of Showalter’s paragraph: ‘[s]ince novels are very long texts, we don’t attempt a close reading of the whole book. Instead, we look at particular important sentences, sometimes even phrases; we may group them together to reach an interpretation or to illustrate an observation’ (98).

Showalter explicitly proposes resisting immersion to achieve ‘active’ extraction. But slow reading is understood differently especially by proponents of slow reading under the general umbrella and philosophy of the slow movement where the emphasis tends to be on surrender, not mastery, discovering what one is not actively looking for by surrendering fully to the reading experience. Here the dominant metaphor is reading as an attentive and patient act of listening to the voice of the other. One might say that in close reading, the reader slows down to dig deeper (having found their extraction point), in slow reading to drift deeper. Metaphors of ‘deep reading’ routinely confuse the depths of immersion (a pleasurable sinking into and surrender to the text, reading with the grain) with the depths of excavation (a masterful probing of the text, reading against the grain).

Interestingly Birkerts seems to attempt to bridge these poles – mastery and submission; control and immersion; digging and drifting - when he champions ‘deep reading’ as ‘the slow and meditative possession of a book’.

Reading, because we control it, is adaptable to our needs and rhythms. We are free to indulge our subjective associative impulse; the term I coin for this is deep reading: the slow and meditative possession of a book. We don’t just read the words, we dream our lives in their vicinity. The printed page becomes a kind of wrought-iron fence we crawl through, returning, once we have wandered, to the very place we started. Deep listening to words is rarely an option. Our ear, and with it our whole imaginative apparatus, marches in lockstep to the speaker’s baton’ (326).

For Birkerts, the reader possesses the book only by making space to be possessed. There is a degree of co-creation here. The reader is active, but not extractive. Reading allows a particularly attentive form of listening, not because reading is slow (or not exactly) but because the speed of reading is (always) variable, because we control the speed. For Birkerts then, when we read both slowly and deeply, we do it not to extract but to dream deeper.

Showalter’s Teaching Literature takes a predictably instrumentalist view of close reading as a technique that produces the discipline of English – in classrooms and in academic publishing. In contrast, for Michele Boulous Walker, author of Slow Philosophy: Reading Against the Institution (2016) and an exemplar of the ‘slow education’ movement, ‘slow reading’ opens the way to ‘a slow and transformative engagement with the strangeness of the world’ (2017). ⁷ But such a slow philosophy demands a fundamental reorganisation of her discipline and of an academy that has become organised around the values of productivity and efficiency. For Boulous Walker it is not enough to teach students in her discipline to read slowly. For real listening to occur, students and academics must, to some degree be liberated from the requirement to produce. ‘Teaching Literature’ wants no such revolution.

Extr(a)active reading and the second text

‘Teaching literature is teaching how to read’ in the words of Barbara Johnson, who joins Showalter, Hayles, Gallop and a majority who understand the discipline of English literary studies as concerned with and defined by its role in teaching reading, specifically in teaching skills of ‘close reading’ (2014: 347). However, as Robert Scholes explains, it is more often the case that emphasis is placed on teaching ‘writing’ since we can see and judge writing but we cannot see reading (2002: 166). Referencing Scholes, Poletti and Seaboyer et al. point out that, assuming that ‘literary studies does teach the skill of reading’, this is ‘largely accidental’ because we focus more on ‘whether a student can describe their reading in their writing, in the creation of a new text, not whether or not they have actually read the text they’re responding to’ (2014: 6). An obvious form is the college English essay. But the ‘second text’ is also regularly required and produced in the classroom – verbally rather than in writing – when students are called on to give an account of a text they have read.

It is worth also considering the other texts (built for purpose) that assist in producing the second text of the student’s ‘reading’/interpretation: ‘notes’, summaries and ‘guides’ predominantly sourced from outside of the university. As Fuller (2019) notes, ‘study guides’ became ubiquitous (and profitable) print business during the Cold War period, taking advantage of the cheap paperback market and offering readers a cheaper alternative to what passed as ‘mastery’ of the Great Books’ (335), which they proved did not require accessing the originals. Now distribution and proliferation via the internet makes auxiliaries even cheaper for the efficient student, who saves both time and money. The success of SparkNotes, created by Harvard graduates, speaks to both change (innovation) and continuity with the culture of Great Books that rose and reigned at Harvard along with the New Criticism.

Put simply, for the purpose of the class – if the purpose is producing a reading – it may be (perceived as) easier for a student to construct their account and with it their understanding of the literary text from these auxiliary texts – or at least with their help. The problem with such help is that, if it does not replace the text (rendering a reading of the primary text unnecessary) and if the student does still read the literary work, this is generally only after reading the supporting text. The efficient student will read the supporting text first, so that they know what to read for: the meaning and patterns they are expected to see – and will see them.

School and university English habituate students to reading works in English classes with extensive priming and guidance, and perhaps only ever reading literary works with such apparatus. As much as exposure to literary works might encourage reading, it might also make it less likely that students would read independently, without such assistance, or feel comfortable doing so. While study guides take away the uncertainty of the act of reading, they also take away a crucial aspect of textual meaning and meaning making. A student reading with the literary work and its ‘meaning’ already mapped out, loses the experience of being lost in the text – not the pleasure of immersion, but the experience of confusion, the narrative contingency, and the sophisticated processes of inference and prediction (as listed by Wolf and Barzillai) that are vital to reading. In sum, English literary studies may allow and even incentivise students to bypass reading, train dependency and perhaps most important for the discipline make it difficult for students to see the experience of reading as essential to its meaning.

Following Rosenblatt’s (1978) transactional theory of literature, which locates literary meaning in the transactional relationship between the reader and the text (12) and looks (less moralistically) to the context of reading and readers’ ‘stance’ along a spectrum from ‘aesthetic’ (process-oriented, immersive) to ‘efferent’ (product-oriented, extractive) (23), it follows that ‘aesthetic reading’ of literary texts would be steadily eroded in educational contexts which are product-oriented, and not just because of the much decried erosion of ‘deep reading’. Where ‘hyper active’ reading further exacerbates this, training students in literary departments leads, ironically but quite logically, to a loss in the ability to read and understand literary texts, whose meaning lies in their effects, and to draw on multimodal narrative literacies that they do possess to anticipate, calibrate, recalibrate and make sense of these effects. Cheryl Hogue Smith concludes that forcing an efferent stance to the literary text, undermining both efferent (extractive) and aesthetic reading, causes Basic Readers to become inactive, not reading (2021: 64). ⁸ But students of literature are more likely to become hyperactive, or (distinguishing from Hayles’ general symptom of the digital age based in the need for constant stimulation a specific mode of behaviour associated with the deferent stance and triggered by the deferent situation) extr(a)active, as they labour to construct the text’s meaning from outside sources, especially those offered by the digital environment.

Practical criticism in a Wifi zone

It is generally accepted that the multiculturalists won the canon wars of the 1980s and 1990s, in which defenders of the canon and the literary critic’s role as arbiter of the canon (established by the New Critics) suffered a decisive defeat. Subsequently the teaching of English literature became more varied and inclusive, open to a range of texts. However, classrooms continue to centre ‘close reading’ (broadly understood as close textual analysis) and close reading still most often involves demonstrating the complexity – and implicitly the value – of a text. That value is both cultural and pedagogical. Highly valued literary texts lend themselves to close reading (as it was designed). But texts that lend themselves to close reading are also highly valued. What ‘works’ in the university classroom also drives what is taught and thus considered to have value. This is particularly the case with English departments under pressure to teach fewer and shorter texts (Poletti et al., 2014: 2). Thus, a kind of canon continues to be produced in universities. Texts that are taught in schools and universities, typically canonical texts, are covered by online study materials. And when less canonical texts are taught in the university they are swiftly admitted into the canon (of the taught and teachable) by the appearance of such guides. ⁹

The student trained in English soon learns that a ‘good’ literary work (such as they encounter in literature classes) is one that can generate ‘close reading’, be that formalist, deconstructionist historicist, materialist, psychoanalytic, postcolonial, or other. The ‘good’ text, the classroom text, affords close reading and close reading is an affordance of the classroom text for both the teacher and the student. Whether the text is identified with high or low culture, the student knows that it must be deemed complex and valuable – something that can be shown through close reading – and that to make such a showing is the aim of the game. I.A. Richards, often credited as the founder of New Critical ‘close reading’, fought such bias in his famous practical criticism experiment in which he gave his students poems to analyse without disclosing the author. This was possible at Cambridge in the 1920s, in the absence of ubiquitous internet and wifi-enabled devices. ¹⁰

Close reading’s success in embedding itself in the academy means that when Barbara Johnson writes that ‘Teaching literature is teaching how to read’ (a sentence quoted widely and approvingly, though perhaps not quite comprehendingly, in the digital humanities literature) she means teaching how to close read:

‘... how to notice things in a text that a speed-reading culture is trained to disregard, overcome, edit out, or explain away; how to read what the language is doing, not guess what the author was thinking [...] to take in evidence from a page, not seek a reality to substitute for it.’ (2014: 347)

Johnson’s (deconstructive-leaning) characterisation of the typical literary classroom captures something of New Critical teaching in its heyday where students at Harvard in the 1950s were expected (as Paul de Man recounts): ‘not to say anything that was not derived from the text they were considering… in other words, to begin by reading texts closely as texts and not to move at once into the general context of human experience or history’ (1986: 23). De Man’s teachers were concerned about historicist reading and generalising about the human experience. But, by the 1980s among the realities to substitute for the page, Johnson could also include the large number of literary-isms that had emerged – with the notable exclusion of Deconstruction(ism), Johnson’s own interpretive method for which her account of close reading in Teaching Deconstructively particularly makes space. For Johnson close reading of the text ‘is the only teaching that can properly be called literary; anything else is history of ideas, biography, psychology, ethics, or bad philosophy’ (2014: 347).

‘Nothing outside the text’ is of course Foucault’s phrase, not Derrida’s – a rebuttal (and sly rebranding) of deconstruction as a ‘historically determined little pedagogy...which teaches the student that there is nothing outside the text…’. Critics debate whether this, also Spivak’s (1976) translation of ‘il n’y a pas de hors-texte’ (158), is a mistranslation. However, the phrase is treated as unproblematic in the context of close reading. David Banash directs his students at Western Illinois: ‘In a close reading you cannot write about anything outside of the text itself. (These issues are relevant in other areas of English studies, but not in close reading). Issues Banash means to exclude include… ‘the author’s life or beliefs (such as comparing biographical details to the text)’, ‘historical or contemporary contexts which seem related’ and ‘[your] personal experience of reading (feelings, difficulties, other reactions)’.

Although closely associated with close reading, Richards expressly recognised (as cognitivists do today) that readers always read with knowledge ‘outside of the text’. When he designed his experiment in Practical Criticism, Richards was perhaps more concerned with the authorial fallacy, historical context and authorial intention, ‘biography’, ‘the author’s life or beliefs’ (Banash) than with the fallacy of author-ity. However, suspending knowledge of the poem’s status was crucial to Richards’ project of introducing literary judgement to the English classroom and a reason his work is so closely associated with that of the New Critics. Removing authorial information would have a particular force in todays’ classrooms, where the challenge is no longer students’ trained dependence on history or biography (as it was in the 1920s), but, ironically, their trained familiarity with (and dependence on) the procedures of close reading as a form of classroom canonisation. Eliminating the author, as Richards did, might eliminate the status of the text so that students are forced to judge for themselves, not knowing whether the text ‘belongs’ in the classroom, whether it is supposed to be good. Under these conditions, ‘close reading’ becomes far more challenging, and more interesting. However, while the internet’s omnipresence makes this experiment in practical criticism practically impossible at the same time that it makes it vital (this paper is not a call for Wifi-disabled Practical Criticism), close reading’s institutionalisation in schools and universities may have made students less up to the challenge.

When De Man received his Ivy League training in close reading, students were directed:

to start out from the bafflement that such singular turns of tone, phrase, and figure were bound to produce in readers attentive enough to notice them and honest enough not to hide their non-understanding behind the screen of received ideas that often passes, in literary instruction, for humanistic knowledge (23).

There is a special irony here that close reading, detached from Richards’ practical experiment and entrenched in school and university literature classrooms would later afford students precisely a ‘screen’ behind which to ‘hide their non-understanding’. For a student who neither reads nor understands the text as a whole, ‘close analysis’ of (free association on) isolated words and images may function as a performative replacement for a base level comprehension of the text and the sentences of which it is comprised. De Man (1986: 23) describes close reading at its height and at its most undiluted in Harvard in the 1950s, with its understanding that close reading is only available to students ‘attentive’ enough to respond to ‘singular turns of tone, phrase, and figure’ and to make this their starting point. Compare with Richards who included in Practical Criticism a catalogue of the chief difficulties of critical readers and noted: ‘First must come the difficulty of making out the plain sense of poetry’ (1929: 36). Facing this same problem nearly a century later among students trained as close readers, Derek Attridge and Henry Staten have made their case for teaching ‘minimal reading’ (also ‘weak reading’ and ‘reading for the essentials’, developing a ‘basic poem literacy’ that allows students to see what a poem ‘literally says’ (2015: 2).

This is not only a problem for poetry. The institutionalisation of New Critical close reading made teaching literature synonymous with teaching poetry and, importantly, with reading (teaching) everything as though it were poetry (what Modernist critics understood as poetry). Current pedagogic interventions tend to focus on poetry. But prose and drama – or rather the more narrative-driven forms associated with immersive reading – perhaps suffer most from the distortions of close reading entrenched as universal method in English classrooms.

‘How to rig the AI’

Arguably an algorithm can write literature. It can, likewise, produce a literary interpretation. An AI cannot read literature. Yet, in English classrooms, greater value is still assigned to producing written interpretations according to the demands of the genre (well represented online), and discussion of reading is directed, intentionally or unintentionally, to this end. This is not surprising. Although reading literature is an enormously complex activity, and far from being fully understood, and although literature departments are held to teach students ‘how to read’ literary texts, what common readers do when they read, what they do with literary texts that allow for ordinary experiences of immersion and engagement, is of less interest to the discipline than the creation of specialist readings and especially of specialist readers.

After my analysis of deep/shallow, active/passive, slow/speed reading and a glance at the many types of ‘reading’ that name various critical approaches that ground scholarly production in the discipline, I do not intend here to propose any new type or subtype of reading. Reading is not the problem addressed in this paper; the problem is pedagogy. As I hope this discussion has shown, what is needed, more modestly, is an approach to teaching less oriented to reproducing the discipline (generating specialist readings) and more concerned with student’s practice of reading, that recognises the ways in which the literary classroom may, actively or passively, displace and devalue the ‘simple’, in fact deeply complex, activity of reading literary texts, and which works to make this more relevant to students and their interpretive activity. Were I to call the strategy I propose to achieve this ‘anticipatory reading’, it would have to be with the recognition that, just as all reading is active, so all reading is anticipatory (Trasmundi et al., 2021: 2). ¹¹ The name only aims to make that explicit, and central to interpretive accounts. Likewise, it should be clear that all reading is social, when I argue for the value of ‘social reading’, particularly digitally facilitated social annotation, in supporting this exploration of reader expectations and their relationship to textual meaning.

At the same time that immersive reading is fretted about on various public stages, it is also oddly assumed in literary departments, as it was when Cleanth Brooks and Robert Penn Warren suggested that the close reading process should begin with the student’s initial reading, ‘as full and innocent an immersion… as possible’ (1976: ix). It is important to recognise that this is not the default for students of literature (who are not innocent). Teachers cannot ask students to read ‘immersively’ or ‘just read’ and expect them to do so. Not because of reduced concentration spans or a shallowing of hearts and minds immersed in the digital, but primarily because reading is overdetermined by educational context. Students will be reading for (something). And whatever they are reading for, it will not be ‘pleasure’ or immersion – which may appear little help, even a hindrance, in accessing what they need to produce a ‘reading’ (the specialist text).

Calling for a practice of ‘critical immersion’ to replace close reading of poetry in English classrooms, Saunders (2000) notes that, while close reading appears to recognise and call for both comprehension and apprehension, its practices keep immersive reading separate and subjugated while the privileging of comprehension (a complete understanding of the whole) leaves students in a permanent state of apprehension, unable to experience their moment-by-moment apprehension of meaning in the text as pleasurable or meaningful. The contrastively more open-ended work of apprehending the literary text, ‘grasp[ing] it perceptually, but without an expectation of ever coming to perceive its scope or outer boundary in their totality, or to grasp and own them fully’ (144), it should be noted, is constitutive of immersive literary reading but is not understood or experienced as such by students trained to ‘close reading’. Saunders argues essentially for a perceptual adjustment (more for the discipline specialist than the student), rejecting the visual logic that privileges ‘distance over immersion or rational cognition over more experiential or emotional forms of understanding’ (153), arguing for the epistemological value of immersive experience, and calling for an emphasis on excluded (non-visual) sensorial aspects of reading to inform a multimodal and multifaceted ‘response’ he terms ‘critical immersion’.

There is no reason to assume that this sensorial experience is empty of epistemological content or that the real-time experience of apprehending, thinking, and responding it provides is negated by its lack of a concrete teleological goal. Immersive reading’s greater sensitivity toward the temporal and contingent experience of reading literature may well unfold different modes of understanding from those valued by close reading—such as incomprehension, hesitation, boredom, knowing when to embrace being lost or when to withhold evaluation—but these are not deficient forms of knowledge or in any way uncritical by nature. (153)

This leaves for students (as for teachers) the gap between reading and analysis, the ‘concrete… goal’, as Saunders discards the ‘more cerebral and disembodied enactment of reading as a knowledge-gathering process’ (153). But I would argue that it is possible to develop a reading practice, particularly for narrative texts, that privileges moment-to-moment apprehension, that orients the reader to apprehend the immersive reading experience with its ‘greater sensitivity toward the temporal and contingent experience’, with focused attention, and that, crucially, does not attempt to comprehend the text as an object detached from the reader’s lived environment and experience (148) – and, crucially, that this practice can meaningfully ground interpretation where these moment-by-moment apprehensions are shared. Saunders’ recalibration emphasises sensorial aspects that have been foregrounded in recent distributed language approaches to reading (Trasmundi et al., 2021), but, aside from attention to performance, related to the sensory quality and embodiment of voice), neglects its social aspect. Here, I will follow Gavriel Salomon’s proposition ‘that cognition is distributed among individuals, that knowledge is socially constructed through collaborative efforts to achieve shared objectives in cultural surroundings, and that information is processed between individuals and the tools and artifacts provided by culture’ (1993: 3).

Although English Studies dismisses reader response as a subjective free-for-all, Saunders seems closer than Rosenblatt to ‘anything goes’ individual interpretation ¹² as, rather than describing critical immersion in the classroom, he leaves it finally to Billy Collins’ poem ‘Invitation to Poetry’ to offer a vision of students reading multi-sensorially and, in a strong sense, creatively (155):

ask them to take a poem

and hold it up to the light

like a color slide

or press an ear against its hive.

I say drop a mouse into a poem

and watch him probe his way out,

or walk inside the poem’s room

and feel the walls for a light switch.

I want them to waterski

across the surface of a poem

waving at the author’s name on the shore.

But all they want to do

is tie the poem to a chair with rope

and torture a confession out of it.

They begin beating it with a hose

to find out what it really means

The complaint clearly addresses the English classroom: ‘… all they want to do/is tie the poem to a chair with rope/and torture a confession out of it…. to find out what it really means’. The poem’s speaker refuses this demand and instead the ‘Invitation’ demands of the student reader that they read poetry with a poet’s creativity and imagination. Here, the poet/poem models ‘best practice’ (in the banal language of education and training). But reading poems, all of which do some version of this, does not appear to endow students with that capacity to deny their training in anxious extraction and blithely ‘surf across the surface of a poem/waving at the author’s name on the shore’.

Less inspiring, but more compassionate to student readers, closer to their compromised, clumsy sometimes embarrassing experience, is Collins’ ‘Marginalia’, a poem which evokes readers’ sometimes awkward relationship with the technology of writing, in a set of self-conscious self-inscriptions ranging from the banal to the mysterious. Collins includes a predictable sampling of New Critical dogma: ‘Man versus nature’, ‘‘Metaphor’, ‘Irony’

One scrawls “Metaphor” next to a stanza of Eliot’s.

Another notes the presence of “Irony”

fifty times outside the paragraphs of A Modest Proposal

among more affective and eccentric inscriptions, like the poem’s memorable final line:

Pardon the egg salad stains, but I’m in love.

But all of these reading-writing acts are performative.

We have all seized the white perimeter as our own

and reached for a pen if only to show

we did not just laze in an armchair turning pages;

While ‘Marginalia’ is exhaustively cited (and annotated) by proponents of web-based annotation, Collins’ poem does not imagine that annotation has any analytical or interpretive power. And the gestures at literary interpretation that appear in the poem, emptily reproducing and amplifying tired New Critical pieties, appear especially rote and bloodless (Collins, 1996).

And if you have managed to graduate from college

without ever having written “Man vs. Nature”

in a margin, perhaps now

is the time to take one step forward.

The digital space however supports social reading and forms of social annotation distinct from the marginal performances and textual missed connections of ‘Marginalia’. While the decline of reading is a constant complaint, literary reading is, by some measures, booming on the internet (Pianzola et al., 2020) and it can be argued that the web is now producing the most viable media for the deep reading that it is understood to be eroding (Dowling, 2014). Web-based annotation appears to bridge the gap between literature in the classroom, where reading is moribund, and cultures of online social reading, where reading is thriving, and to bridge the gap between the discussion space of the classroom and the space of students’ reading (which despite accounting for the majority of nominal time on task, is typically done in isolation, unseen and unsupported). This bridge is not without its dangers. Digital annotation might bring reading into the classroom, into the centre of learning, or it might push it further into the margins. Textual annotation in collaborative, open-access editorial projects and social editions is productive and intuitive from a student perspective, though it facilitates a fragmentary approach, focused on writing rather than sustained reading, that students seem to assume by default when asked to read and annotate. In a more troubling example, on Reddit ‘thebeanintheback’ can be found contributing meaningfully to an AskReddit discussion of narrative stereotyping of LGBTQ+ characters but posts on a University of Michigan subreddit: ‘i literally write nonsense on perusall [a social annotation platform] and so do all my classmates. we reply to each others’ nonsense and upvote it and act like it makes sense. it sucks but once you figure out how to rig the AI, it’s not all that bad’.

Expecting notes v. noting expectations

Literary students cannot be made to read for immersion, or read for pleasure. But the emphasis in the classroom can be shifted to make these more relevant to textual analysis. The compensatory strategy I propose, drawing together reading and writing, is something like the classic drawing exercise popularised in Drawing with the Right Side of the Brain, where the trainee artist is enjoined, not to sketch an object, but to discover the object by sketching the negative space. Rather than requiring students to (re)produce the second text of a comprehensive interpretation, returning to Saunders’ distinction, students might instead be trained to inscribe precisely their ‘apprehensions’, specifically their expectations, as they move through the text.

The immersive literary text does what it does not through a set of formal ‘devices’ to be identified and enumerated (ambiguity, irony, paradox and tension in the New Critics; imagery, allusion, intertextuality, syntax, form and, with perplexing broadness ‘language’ in the lexicon of Showalter and English exam prompts). It does it through activating and interacting with expectations. Narrative engagement and pleasure are based on a sense (whether narrative, rhetorical, rhythmic/metrical or phonic) of where the text might go that makes what follows satisfying (or frustrating, or intriguing…). Noting and attending to these responses need not produce the subjective interpretation feared by the original close readers and by those who now reject the perceived subjective excesses of the ‘reader response’ tradition, particularly when these can be brought into dialogue, showing students the range of responses that a text may elicit, where interest in the individual reading shifts to interest in a larger pattern of responses.

Along with satisfaction, curiosity and interest (where there is a pleasurable and meaningful interaction with expectations), such a reading practice may also discover, following Saunders, ‘incomprehension, hesitation, boredom’. These ‘negative’ responses play a significant role in demonstrating to students how the text can fail to work successfully with a reader’s prior knowledge and expectations. Where students may find it difficult, initially, to identify and articulate moments in the text where an expectation is formed or worked, emotions may play a role, not as a barrier to productive analysis that can be declared irrelevant and excluded or turned from ‘negative’ to ‘positive’, but as a useful indicator. Asking how an emotional response might be related to expectation shifts emphasis from classroom affects to reading affects. Alongside purportedly ‘positive’ feelings like anticipation, excitement and satisfaction associated with the setting up, maintaining and resolving of expectations, ‘negative’ affects like frustration, disappointment confusion and boredom may equally signal a reader’s interaction with expectation. Not a bug but a feature of this practice, they may be a means to focus attention on what is happening in the reader’s transaction with the text.

While there are no ‘wrong’ expectations – students are required only to note their observations as accurately as they can – the notion of a ‘right’ answer may cause students difficulty. There is no such thing as a ‘correct’ expectation, only an expectation that is fulfilled – and the two are quite distinct. The aim of this practice would not be to correctly guess the end point as somehow signalled earlier in the text (pace the New Critical sense of the artwork as an organic whole) but rather to observe and note expectations formed contingently at various points. Producing a correct prediction has no special value or status here, particularly if expectations are based in (reversed engineered from) information found later in the text or gleaned from notes and summaries. This requires a different orientation to reading and a reading discipline that may feel, practically and theoretically, alien to students schooled to seek an account of the whole and not trained to focus on moment-by-moment development of the text.

Tracing apprehensions also inevitably uncovers misapprehensions, and this has value both for the teacher and the student. Richards’ exposure of student misreadings in his practical criticism classroom experiments raises misgivings today. Teachers are understandably apprehensive about the effect on students of exposing their errors, as well as Richards’ stated goal of leading students to ‘a successful interpretation, a correct understanding’ (1929: 336). However, significantly, Richards believed that scholars and teachers should regard misinterpretation not as an unfortunate aberration but as ‘the normal and probable event’ (322). This is a paradigm shift that entails far more than a move away from the notion of an incorrect reading (which is easily achieved and often proves banal and counterproductive). Viewing misinterpretation as ‘the normal and probable event’ within reading transactions entails positioning the text, or the textual transaction, not the student, as fallible. There is perhaps less of a distance between Richards’ practical critical and Rosenblatt’s transactional reading model than is commonly assumed (Graham and Wall, 1997: 12). Crucially, apprehensions and misapprehensions are a rich resource for understanding the text, a resource that, when shared with other readers, can be used to inform not a ‘correct’ interpretation but a layered understanding of what the text might try to do.

Digital tools, particularly social annotation tools, by allowing inscription in a shared online text, can enhance, and significantly extend this beyond the classroom and into an expanded space of social reading, though unstructured social annotation on any platform will be unlikely to have the same effect. Broadly, noting expectations as they read a digitised literary text shared via the internet, students can create a shared map (common property of a community of readers), highlighting transactions with the text and different assumptions informing reading, allowing the reader (teacher or student) to see where these converge and where they diverge. This would support contextualised peer-to-peer online discussion, the main focus of research into social annotation, where its effectiveness is well established (Morales et al., 2022) It might further enable students to make judgements about how the text works, not only based on their own interpretation, or on others’ that appear more persuasive, but also by viewing the pattern of responses that emerges from the activity of multiple readers. In this way, digital systems may support a reading practice based in apprehension where reading is, following Salomon (1993), socially distributed.

This draws a basic outline. There is more to say about more specific applications, for example how, in the context of teaching early modern drama, attending to horizons of expectations (Jauss and Benzinger, 1970) and the moment of reading accesses a multi-scalar view ‘emphasising the social-historical and cultural aspects of reading as well as the reader’s lived experience’ (Trasmundi et al., 2021: 2). I will not discuss here general accounts of the benefits and drawbacks of social annotation, factors informing the selection of specific social annotation platforms, or choices in how social annotation is deployed and framed, such as including or excluding lecturer annotations, synchronous versus asynchronous annotation, and unforeseen consequences of social annotation pedagogy – first, because existing evaluations of the effectiveness of social annotation consider its effectiveness for a range of purposes secondary to that purposed here and second, because the latter are complex and significant questions that require greater elaboration and particularisation. ¹³ Brown and Croft’s caution regarding the embedded risks that may make engaging in social annotation less effective and potentially more dangerous for students from historically marginalised backgrounds are relevant here, given the level of exposure of students, as well as cautions regarding ‘over-engineering’ of annotation assignments reducing positive outcomes. However, the strategy described meets the standards of Brown and Croft’s ‘critical social annotation’, leveraging the potential of social annotation for greater classroom equity (4). Crucially, and against the close reading tradition, a student reading in this way is not expected to be (or to construct herself as) an ‘ideal reader’ (Brooks, 1979: 600) who possesses all the background information and cultural assumptions to be able to successfully read a text without having to look outside of it. But a student can construct some idea of an ideal reader (and the ideal textual transaction) drawing on shared knowledge made available through collaborative reading.

A practice of social online annotation that calls students not to interpret text, but to note expectations, can re-centre reading and map narrative and affective frameworks of expectation, that are, as reader response critics have noted, constitutive of literary meaning, embodying the contract between reader(s) and text, and recognising the extent to which reading and literature are social. It is a question whether this practice would render readers pleasurably or painfully, aware of their reading. What is proposed here is not immersive reading or the critical immersion Saunders proposes. At base, it is still a reading and writing practice geared towards a form of descriptive explication, concerned (as the Richards and the New Critics were) with how the text works, not what it means but what it does (and how it does it). While it might be absorbing, clearly this practice of reading is not immersive in the sense of sweeping the reader along, or letting the reader forget herself. Rather, it constantly asks the reader to remember herself – and also other readers. While it emphasises continuous reading (following the flow of information as it is presented), such reading is necessarily marked, if not marred, by interruption: first as it trains the reader to pause to register narrative possibilities and second as it trains attention of the readings of others, making these available. The presence of collaborative annotation does allow for a kind of hyper reading, allowing the reader to skip between points of focus in the text (in which expectations are activated) and between their own reading experience and others’. The student still, in a basic sense, reads ‘on the prowl’ (Baron, 2017: 39), although searching for something that cannot be scanned for, but only found in continuous reading.

This practice, intended to challenge students and engage them more deeply in reading, could, like others before it, ossify into empty and repetitive gestures. Were it widely adopted, one might expect that a databank of pre-extracted expectation points for frequently taught texts would at some point become available. Given the efficiency of online providers in meeting the demand for content, sourcing essays and notes directly from students, those asked to note expectations might soon be able reach for the internet ‘to show/we did not just laze in an armchair turning pages…’ However, the collaborative maps drawn in those classes would have value (beyond what little they might be sold for on the internet). They would teach something – more, it must seem, than the interpretations being passed around online and passing for ‘knowledge’ of the discipline. The discipline might learn something from students reading together, discovering and shifting horizons of expectation.