Among the several dead horses on display in this book, the canon has perhaps been beaten up the most. As it was the centrepiece of multiculturalism’s critique of the traditional structure and emphases of the field of literary studies, the controversies over its status over the last half century or so have provoked an unmatched volume of scholarly angst and ire, and very few of the field’s topics have elicited concern of similar scale outside the academy. Perhaps its most provoking problem over these years has been the one that Henry Louis Gates voices in the first epigraph above (putting aside, for the moment, the implications of his phrase ‘medieval never-never land’). Gates uttered this cri de coeur over three decades ago, yet even then it was already something of a retrospective lament, as the institutional and attitudinal changes were well afoot that have at this point loosened – and in some instances broken altogether – the curricular grip, in university English departments at least, of ‘the high canon of Western masterpieces’. Nonetheless, the problem that Gates’s remark identifies continues to be a provoking one, suggesting some unfinished business. For example, a more generally formulated version of it serves as a point of departure for Ankhi Mukherjee’s recent What Is a Classic?: Postcolonial Rewriting and Invention of the Canon. As Mukherjee observes near the outset of this book, ‘The canon has historically been a nexus of power and knowledge that reinforces hierarchies and the vested interests of select institutions, excluding the interests and accomplishments of minorities, popular and demotic culture, or non-European civilisations.’² To recycle my metaphor from Chapter 1, the canon has functioned as an aesthetic Trojan Horse of ideology and domination, one that has helped to perpetuate the naturalisation of specific norms and to enforce principles of social and political exclusion. Yet recognition of this fact is, for Mukherjee and Gates, not itself a solution but a starting point, one that leads them to the more difficult question of what sort of response is called for.

With Gates, Mukherjee and many others, most of us without hesitation will lament the damage the canon has caused and condemn any continued operation in this regard. Yet, also with Gates and Mukherjee, we will discover that we have not thereby solved the problem. This problem, this chapter argues, resides in the fraught relation between the readiness with which we may affirm the sentiment of the chapter’s first epigraph and the recognition, expressed in the second, that despite ‘judgment’ falling out of use as a description of the literary critical enterprise, we have never ‘actually stopped judging’. The continued haunting of the ghost of judgement, as I called it in Chapter 1, bespeaks the persistence of the category of literary value in the enterprises of literary criticism and teaching, however much it may be acknowledged or suppressed. Indeed, with this latter remark, Gates aligns himself with those critics who, as I mentioned in this book’s introduction, began in that period to voice concern about the neglect of literary value or literary distinctiveness in the scholarship being pursued under the banners of historicism and cultural studies. Mukherjee, a quarter of a century later, voices similar sentiments.³ In my terms, both critics at some level recognise that inasmuch as the adjective literary remains a meaningful qualifier to the projects that they pursue, the category of value is part of those projects’ most basic apprehension of the object of study. Hence, not surprisingly, while both Gates and Mukherjee move quickly from a singular canon to plural canons (plural in both synchronic and diachronic dimensions), neither then seeks to do away with the latter. Instead (to oversimplify, massively, their complex and quite differently focused arguments) they attempt to understand the role canons may have in a more just society.⁴

How successful they are in this effort I will leave their readers to assess. Certainly, extending from before Gates through Mukherjee and beyond, there have been many literary theoretical calories burned in the effort to defend, or at least to analyse, the endurance of canons.⁵ Yet the very fact that this work continues to be pursued suggests that a palpable if sometimes rather vague discomfort with the idea of canons persists within the field of literary studies. To be sure, the practicalities of syllabi, anthologies, graduate reading lists, and so on ensure that something akin to canons in effect still shape teaching and scholarship in English departments generally. But my sense is that they do so typically at a sort of arm’s reach – as, say, porously delimited sets of provisionally selected texts, sets that may have a particular focus and do not present themselves as necessarily excluding the claims of others.⁶ Unquestionably (to me at least), the flexibility and broadening accomplished by this pluralisation are to be celebrated. For example, assuming that one may not retain both, one may feel that the gain outweighs the loss if, for a second-year course on the bildungsroman, one makes room for Tsitsi Dangarembga’s Nervous Conditions by dropping Great Expectations. Nonetheless, despite this pluralisation’s apparent success, the notion of canonicity that it still necessarily involves remains, at some conscious or subconscious level, troubling to many of us, and the reason for this (or at least one of the principal reasons) is, again, the relation between the ideas underlying the two epigraphs from Gates.

Drawing on the formulations of the preceding chapter, we may, on the one hand, understand the problem of canonicity identified in the first epigraph as an affordance of the subterranean working of literary value’s characteristic loose binding. As we saw illustrated with Boccaccio’s meditations upon Dante’s value vis-à-vis that of his pagan predecessors, ascriptions of literary value tend to seek strengthening through interlinkage with networks of other kinds of value, with literary value and those other kinds of value thereby constituting a mutually affirming circuit. Because of loose binding, that is, literary value, in practice, is almost never just literary, and as Gates’s remark identifies, one of the networks of value with which the value of the canon has been traditionally interlinked has been that of white supremacy. One of the affordances of loose binding, in short, is a reciprocal linkage between the canon and white supremacy. Obviously, once we have unearthed this particular interlinkage, we may reject and seek to undo it. On the other hand, however, as the second epigraph insists, and as I have argued throughout this book, insofar as our projects are conceived as literary study, we cannot escape the category of literary value and the activity of value ascription – the acts of judgement – that in practice constitute that study. Because of literary value’s characteristic loose binding, then, we may hardly evade some interlinkage with networks of other kinds of value, at greater or lesser levels of indirection, of which we are more-or-less conscious. And those interlinkages, in affirming some sets of values, necessarily do so in some differential relation with others sets of values. Hence, in the privileging, however provisionally and porously, of some texts over others that the activity of value ascription necessitates, we also privilege some other-than-literary values over others.

Facing this situation, we may be tempted to begin by identifying those other-than-literary values that we believe are right and good and then exercising literary judgement in explicit correlation with those values. This sort of response to the problem is a topic that I take up in the next chapter, in relation to literary interpretation. Here, we may simply note that the fact that both the Right caricaturises the Left, and the Left the Right, as performing this bully pulpit exercise should give us pause. More fundamentally, that the literary value of, say, Othello may easily be interlinked with any number of mutually hostile sets of other-than-literary values indicates that loose binding (as we saw with the Boccaccio example) cuts both ways. As much as it affords connection to networks of other kinds of value, its very looseness entails that those connections are never sure. In linking Othello’s literary brilliance to the value of a critique of white supremacy, we may thus inadvertently elevate a text that may continue to reinforce white supremacy, or something just as bad.

To translate this back into Gates’s terms, any act of canon recognition, no matter how politically and socially nuanced, grants a voice to some at the expense, potentially, of rendering others voiceless, and therefore incurs the risk of committing the kind of injustice that Gates identifies in the first epigraph. Not surprisingly, therefore, many of us hesitate to provide, if not simply avoid altogether, the rationale, aim or basis (at least a fully articulated one) for the literary selection – that is, the literary valuation – that we have not ceased to perform, since such formulations seem inevitably to tend toward canon affirmation. The result is the same sort of inchoateness or lacuna in respect to literary value that I have noticed throughout this book: a gulf within the everyday practice of criticism and teaching between what we are willing to claim that we are doing (e.g., assembling reading lists) and what we actually are doing (canon-making). Despite several decades of arguing about, for and against canons, we still find ourselves bumping up against canonicity, even if – perhaps especially if – we would rather just put that category aside. It is this particular gulf between thinking and doing, as made evident in these everyday practices, that this chapter explores.

In the pages that follow, I turn first to a recent, baldly defensive moment in the history of Chaucer studies to provide an extended example of this gulf within the practical, institutional realities that govern our work and to delineate some of the gulf’s characteristics. Next, to underscore how the gulf is not merely an idiosyncrasy within Chaucer studies, I consider in this respect a couple examples of similarly practical but more general defences of literary study, as they appear in departmental administrative documents and opinion pieces in professional periodicals. As we will see, in all these instances literary value – and more specifically a defence of the value of the canonical text – becomes the proverbial elephant in the room. As one might expect, then, this uneasy refusal to acknowledge the obvious has provoked in other forums a rather voluminous stream of attempts to account for canonicity and to advocate for canons in some form. Hence I next turn to an illustrative trio of such attempts, one pertaining to Chaucer studies, one to medieval literary study more broadly and one to literary study in general. Holding this trio up to the account of literary valuing that I developed in the preceding two chapters, I suggest that inasmuch as these three defences of canonicity are representative, they disclose how the very problem with which this chapter begins – the problem of canonicity marked out by its two epigraphs – in some fashion ineluctably reemerges. In this chapter’s conclusion, then, I offer some suggestions for how to respond to this dilemma between wishing, for good reasons, not to acknowledge the elephant in the room, and acknowledging that elephant but then bearing out those very reasons that had prompted our avoidance. By no means claiming to have found a solution to this dilemma, I instead seek to reframe it in terms of this book’s account of literary valuing and to offer ways to think about the impasse that aim to make it scholarly and pedagogically generative rather than a sort of spot on the carpet upon which we have set our feet.

Defending the MLA Chaucer Division

Medievalists will readily recognise the ironies in Gates’s offhand reference to a ‘medieval never-never land’. The medieval functions here as the familiar paradoxical bogeyman, simultaneously an eclipsed period of ignorance and barbarity (think Vikings in horned helmets) and a point of origin for idealised (rightly or wrongly, depending on one’s point of view) social, cultural or political values (think Camelot, or, more starkly wrongly idealised, Ku Klux Klan ‘knights’). The Peter Pan allusion plainly signals that we are to understand this instance of the trope as politically retrograde nostalgia for an era in which the ‘high canon of Western masterpieces’ rested untroubled (except, presumably, for border skirmishes involving this or that minor poet). One irony, of course, is that, at least for literatures in English, in the Middle Ages there were only the trace beginnings of anything like a canon, and these appeared only at the very end of the period and possessed a cultural and institutional footprint that was miniscule – that is, profoundly marginal – in comparison to the total artistic and intellectual output of the time.⁷ Another irony, therefore, is that in so misprojecting a contemporary injustice onto the premodern past, Gates’s othering of the medieval follows the same general logic as the racial othering that he sees performed by the ‘high canon’. This irony does not, obviously, excuse that racial othering. Rather, it suggests that the medieval, and more specifically the marginal position of medieval literature within the field of English literary studies, has special purchase on the ongoing problem of canonicity.

Most of the English literary medievalists whom I know have, however grudgingly, grown to accept their marginal position in university curricula and in scholarship. (Indeed, that such a seemingly large proportion become department chairs perhaps owes something to how unthreatening they seem in their acceptance of their marginality.) Their collective outcry in 2013 was hence all the more remarkable. The triggering event was the decision by the MLA Executive Council to reconsider the division and discussion group structure into which the MLA had long parsed the many and ever-increasing literatures that fall under its umbrella (a structure last revisited in 1974). One of their proposals was to collapse the three divisions devoted to medieval British literature – Old English, Chaucer, and Middle English Language and Literature Excluding Chaucer – into one. In ‘imagining a structure that works for the MLA in the twenty-first century, that reflects members’ current field affiliations, and that makes space for areas that are currently underrepresented or absent’, they asked the following of the members of the Chaucer Division Executive Committee:

The basic rationale for the proposal, that is, was to make room for other, formerly silenced or newly emergent voices, Gates’s ‘invisible … unrepresented … unrepresentable’ – a rationale that, as a general principle, I suspect most medievalists today would affirm and even incorporate into their own practice, precisely in the fashion of my dropping Dickens in favour of Dangarembga. Nevertheless, every medievalist whom I spoke to about the proposal was outraged by it.

Unavoidably, part of what provoked this response was merely self-interest. Feeling as though our bowl of gruel was already small enough, we balked at the demand that we accept one-third of our usual portion. The then-Executive Director of the New Chaucer Society (NCS), Ruth Evans, made no attempt to disguise this motive in the email notifying society members of the proposal. I quote in full her entire second paragraph:

The threat, Evans indicates, is to our very viability as a subfield: fewer sessions at the MLA means fewer medievalists giving papers, fewer medievalists in attendance, fewer job interviews for medievalists and hence, presumably, fewer of those jobs themselves. To be sure, Evans does briefly mention a more conceptual objection – the ‘absurdity’ of ‘representing … a millennium in only two sessions’ – but she does not elaborate. Her readers in this case were not likely to pause to observe that the six division sessions have never been the only ones at the convention representing the Middle Ages, nor to weigh the relative injustice of the scarcity of sessions pertaining to a whole millennium against that of the scarcity of those pertaining to a whole continent, such as Africa. (I certainly did not so pause.) And nor would we wonder whether our smaller footprint at the MLA would in fact simply be the inevitable effect, rather than a further cause, of our smaller footprint in the field of literary studies as a whole: that is, whether the MLA would simply be accommodating itself to the real loss of medievalist tenure lines in English departments, and the cascading effects of that, rather than furthering such loss. Instead, facing the threat of losing ‘our places at a very important table’, we were determined not to go gently.

But of course this threat to our existence does not by itself argue against the proposal, as anyone knows who has sought to explain why the extinction of an obscure animal species matters to someone not prone to worry about such things. The question our outrage begs is why medieval English literature – and in particular Chaucer studies – continues to deserve its place at that important table. It was naturally this question, then, that the formal response sent to the MLA from Evans and then-NCS president Alastair Minnis strove to answer. Acknowledging the anomaly of ‘Chaucer and Shakespeare’ as ‘the only two named authors with separate Divisions’ (and so obliquely acknowledging the inheritance of those authors’ canonical status), they locate the origin of that anomaly for Chaucer in the institutionalisation of Chaucer studies in the nineteenth century separate from the study of Middle English language and literature.¹⁰ Quite aware, however, that the original establishment of Chaucer studies as a distinct subfield is not in itself a reason to preserve it, Evans and Minnis proceed to make their case on other bases:

The initial move in this passage is shrewdly savvy. Placing Chaucer in scare quotes at once obliquely references the prior legitimating function of the poet’s canonical status and signals acceptance, even approval, of the now post-canonical sensibility of the field. It is hence no longer Chaucer the poetic genius, but rather ‘Chaucer’ as the name of a sort of scholarly neighbourhood that continues to prove an attractive place to reside, that necessitates its continued privileged status. Attesting to this attractiveness is the sheer number of scholars who have taken up residence there and the corresponding volume of their scholarly activity: two journals, thousands of downloads, hundreds of conference attendees, and so on. Moreover, far from a gated community, this neighbourhood is one with frequent and mutually enriching exchanges with the cosmopolitan downtown of, for example, sexuality studies and PMLA itself.

I do not know how the numbers that Evans and Minnis cite compare to their parallels in, say, Milton studies. My hunch is that the Chaucer neighbourhood may not be as conspicuously distinctive in these respects as their arguments suggest, but of course achieving some quantitative threshold is not really the letter’s rhetorical intent. Rather, the point that the paragraph as a whole makes is that the reason to keep the Chaucer Division is not Chaucer but the scholarship that ‘Chaucer’ continues to engender (even and especially among ‘increasingly younger’ researchers). A moment’s reflection, however, may prompt the suspicion that this is rather akin to claiming that it is not pizza that one likes but the feeling of contentment that one has when one eats it. If that feeling of contentment may be had other ways, then there is no basis for privileging pizza, so the question circles back to whether there is something special about pizza itself. In other words, the question circles back to whether the ‘vitally important’ scholarship that resides under the heading ‘Chaucer’ actually depends upon something distinctive about Chaucer’s works. If it does not, then the rationale for having both a Chaucer Division and an other-than-Chaucer Middle English division becomes more difficult to fathom.

Perhaps sensing this, in the next paragraph of their letter, Evans and Minnis argue,

That Middle English literary scholarship is ‘highly distinctive’ in respect to scholarship on Old English literatures is perhaps a claim that both groups of scholars would at least to some degree be willing to accept, if for the linguistic differences alone – although this is far from a settled or uncontested view (e.g., while the Conquest demarcates at least somewhat ‘different histories’, to what degree do the subfields’ ‘intellectual approaches’ differ today?). But for those of us who have devoted much of our research to other-than-Chaucer Middle English texts, the claim that at present we have ‘very different institutional and intellectual approaches, different histories … different agendas, different constituencies’ may seem simply nonsensical. (And for some Langlandians and Gowerians, I would guess that it may provoke no small amount of irritation.) I am quite sure that Evans and Minnis were aware of this when writing the above sentences. Both well know that the prepositional phrase ‘in the age of Chaucer’ in the title of the society’s flagship journal performs the important work of, among other things, recognising the shared perspectives and interests of those working on post-Conquest English literatures. But their audience in this letter was not those scholars but instead those who were perceived as threatening the viability of medieval literary studies as a whole. And because their argument for the preservation of three divisions is based not on canonicity but on the distinctiveness of scholarly activity, they inevitably found themselves overstating that distinctiveness.

Nowhere in the letter do Evans and Minnis attempt to argue that Chaucer’s works have any special value in themselves, as of course to do so would be to invoke the idea of canonicity, which, as they were no doubt acutely aware, no longer has suasive efficacy in the field at large. Nonetheless, as I have suggested, that idea remains stubbornly half-submerged, in the form of an aporia or inchoateness, in their defence of a Chaucer Division. Likewise, the formal letter sent by the Executive Committee of the MLA Chaucer Division (which at the time included Holly Crocker, Kathy Lavezzo, Jessica Rosenfeld, Mark Miller and Kellie Robertson), while broader in scope and more detailed, runs into much the same problem.¹¹ The authors of the letter eloquently defend the study of medieval British literatures generally as at once distinctive and in mutually productive conversation with the rest of the field. With regard to the Old English Division, the authors emphasise ‘fundamental’ linguistic differences, ‘cultural and political factors’ distinguishing the pre-Conquest period and the ‘different methodologies and different habits of thought’ of Old English scholars (no doubt aware, as we may assume for Evans and Minnis, of the eyebrows, or hackles, the latter characterisation may raise in some quarters). For the Chaucer Division, then, the letter begins by recognising ‘the poet’s unique place in the history of English’. Yet, lest their addressees suspect a canonical argument, the authors immediately clarify that they mean the history of English as a language and thus Chaucer’s ‘incorporation of French words and his sensitivity to dialectical diversity’. Notwithstanding the fact that Christopher Cannon has taught us to be sceptical of Chaucer’s actual impact on the language,¹² the authors surely understood that such old philological arguments would carry little weight, and hence they quickly move on to an assertion that closely echoes the primary justification offered by Evans and Minnis:

The resonance here with the remarks of Ralph Hanna that we saw in Chapter 1 – that, despite his wish not to write about Chaucer, ‘the canonically central medieval poet demands the attention of anyone involved with Middle English textual dissemination’ – is striking.¹³ Similar to how Hanna draws upon and further promulgates Chaucer’s literary value, despite his explicit desire to do otherwise, the arguments of the Chaucer Division letter rest upon a literary value that they not only push aside as that division’s raison d’être but also avoid even acknowledging. The reason to preserve the division, the letter claims, is not Chaucer but ‘Chaucer as a focal point’; it is not anything special about Chaucer’s works but the ‘critical conversation’ those works engender. The implication that remains (strategically) unstated is that, say, Piers Plowman does not provoke ‘critical conversation’ of the same scope and value, and hence combining the Chaucer and the other-than-Chaucer divisions would impoverish the subfield of late medieval literary study generally (a scenario that the authors phrase diplomatically as ‘crowd[ing] out other field interests’). The implication, that is, is that Chaucer’s works are special.

For both letters, therefore, the necessarily unstated reason to preserve a separate Chaucer Division remains the reason why that division was established in the first place, which is essentially the same reason for the equally anomalous separate Shakespeare Division: the author’s longstanding and repeatedly reaffirmed canonical status. To be sure, as Evans and Minnis note, Chaucer is no Shakespeare in this respect. Each functions as a literary value touchstone that at once measures and legitimates the literary quality (including poorly judged quality) of the works of others that fall within its orbit. But for Chaucer the latter include primarily just late medieval English texts, while for Shakespeare they potentially include virtually all literatures – and dramatic productions – in English, not to mention in other languages. As I reviewed in Chapter 1, from the able research of David Matthews, Stephanie Trigg, Tim William Machan, Thomas Prendergast and others, we have a satisfyingly historicised and conceptualised account of Chaucer’s function in this respect.¹⁴ As these scholars have shown, it was the combination of Chaucer’s self-evident, apparently timeless literary excellence, his uncanny ability to seem always current, with the manifest historical distance of his language, manuscripts and culture that enabled the birth of modern Chaucer studies in the nineteenth century. For the founder of the original Chaucer Society, Frederick J. Furnivall, Chaucer’s poetry was self-evidently worth recovering, reconstructing, preserving and transmitting. And also for Furnivall, this presumed value, and the necessary scholarly activity it engendered, in turn anchored Middle English studies more generally. As the logic went, if one medieval poet was so valuable, then others might be so too, once the scholarly work of recovering their achievements has been performed; or, at the very least, these others can help us see just how valuable that one is.

This logic – of Chaucer’s paradoxically simultaneous exceptionality and representativeness – has proven quite enduring. To highlight just one institutionally impactful repetition of it, I offer the editors’ introduction to the 1966 inaugural issue of ChR:

Chaucer’s canonical status is here, from the first sentence on, explicitly assumed. It needs no defence in itself, but rather in its very stable certainty it may serve as a means to recognise the literary value of ‘many writings [that] have not in the past been properly understood and appreciated’. In the terminology of Chapter 2, inasmuch as recognition as ‘literature’ is one and the same as ascribing literary value (however much or little), the manifest value of Chaucer provides the axiological anchor point from which one may recognise ‘in [other] medieval literature an artistry … perhaps even more worthy of study’ than the ‘contemporary literature’ that is axiologically more proximate to ‘teachers and students’. Chaucer’s self-evident value, amid the controversies and excitements of 1960s literary criticism (and with the canon wars still around the corner), once again led the way, thenceforth to be institutionalised in one of the still-thriving journals that bears the poet’s name. (And given this rationale for Chaucer-centrism, it is all the more striking that back in 1926, as Patricia Clare Ingham reports, the forty members of the MLA ‘Chaucer Group’ in attendance at the association meeting voted against merging with the ‘Middle English Language Group’, as the thirty-seven members of the latter proposed, with the decision resting on a single vote.¹⁶ Apparently, Chaucerian noblesse oblige only extends so far.)

That this self-evident literary value has in fact never ceased to function in this manner, despite the pluralisation of canons and widespread suspicion towards canonicity as a principle of privileging some texts over others, was undoubtedly recognised on some level by the authors of both letters to the MLA, as well as by other leaders within Chaucer and Middle English Studies. For example, the questions that the MLA proposal exhumed in this respect spurred the other-than-Chaucer division to devote one of its 2015 MLA convention sessions to the topic ‘Rethinking the Place of the Author’, a roundtable in which Evans participated, contributing a talk entitled ‘What Is a Chaucer?’¹⁷ Indeed, Evans in particular has repeatedly returned her penetrating gaze to the issue of Chaucer’s institutional centrality and value, with her research and reflections in this regard culminating in her 2021 NCS Presidential Lecture. Entitled ‘On Not Being Chaucer’, this talk reconnoitred the rocky critical terrain to provide an explanation for why Evans ‘still want[s]‌ to read Chaucer’.¹⁸ Yet, at the time of the earlier controversy and for the missives directed to MLA leadership, Evans and the others just as certainly recognised that – for the purpose of resisting a proposal whose rationale plainly shares the general sentiment of this chapter’s first epigraph – they could scarcely mount any defence that even remotely suggested, as Gates puts it, the ‘return of “the” canon’. As a consequence, the defences they did mount beg the question of whether the undoubtedly impressive scholarship that ‘Chaucer’ has engendered actually needed Chaucer at its centre, other than for the scholarly findings pertaining to Chaucer specifically; or, if it did, whether it continues to so need Chaucer.

Let me be clear, however, that in pointing this out, I am in no way questioning the wisdom or prudence of the authors of these letters. Quite the contrary: I am profoundly grateful to them and relieved that they achieved their aim. In their shoes, I would have attempted the same arguments but much less eloquently. My purpose here, instead, is to underscore how the category of canonicity persists in our basic practices and institutional manoeuvres, despite our laudable desire to remedy the social injustices of which it has been one instrument among many. Inevitably, the conflict between this persistence and this desire appears, within the practical occasions that these letters represent, as a degree of conceptual incoherence – one facet of the general problem of literary value.

Defending literary studies

Chaucer studies, as it is centred around a poet who is simultaneously canonical and marginal, likely exhibits this facet of the problem of literary value more baldly than other areas under the literary studies umbrella. Nonetheless, a similar incoherence is not hard to spot even in the most general defences of the field of literary studies, especially in those relatively unguarded ones with practical aims directed toward external audiences who do not already assume that the field has value. In particular, as the graduate students who absorbed the critiques of the canon in the 1980s and 1990s have grown into senior faculty in the 2000s and 2010s, the departmental self-definitions over which they preside have come to reflect their unease with canonicity as a defining justification. Recognising the conceptual tautology, ideological and socioeconomic instrumentality and historical relativity of literary value, they are no longer likely to advertise notions of literary greatness and genius in, say, their departmental promotional material. They have accepted that these notions were often vague and never subject to demonstration that (as we will witness again below) was not either logically circular or pendant on external authority, with the latter always potentially in service, at some level, to the sort of retrograde ideology targeted in this chapter’s first epigraph.

For example, about eight years ago from the time of this writing, as part of the assessment plan for the English major at Indiana University South Bend (a regional branch of the Indiana University system offering BA and MA degrees in English), the department was required to develop a mission statement for the undergraduate major, ‘a clear and concise description of the ultimate principles that guide the work of the program’, as the university assessment experts communicated to us. This is what we agreed upon:

Despite the not insignificant effort it took to compose and agree upon this statement, my hunch is that in this post-canon age, its tenor and even details are fairly typical of English departments’ self-characterisations composed for similar purposes. In several ways, for example, it is merely a more measured, blander version of the claims that Azade Seyhan makes for the value of the English major that I quoted in the preceding chapter.²⁰ It foregrounds, especially, skills (more so than Seyhan in this respect), lays claim to broad, self-evidently valuable categories of knowledge (‘cultural and historical’), is careful to include a nod to its creative writing constituencies (‘creative expression’) and even finds a way to preserve something of a rationale (‘many traditions of literatures in English’) for past investments in literary specialisations in periods and genres that still at that point characterised the department’s organisation of faculty lines and, hence, curriculum.

In these regards, it may be usefully set alongside Paul Jay and Gerald Graff’s 2012 exhortation, in an Inside Higher Ed piece, for humanities departments to reorganise themselves around ‘critical vocationalism’, giving voice to what has become a prominent trend in twenty-first-century English department self-refashioning.²¹ Although the IU South Bend mission statement was drafted in ignorance of the article, it was motivated in no small part by the same anxieties about the status of the humanities and accordingly echoes many of the article’s prescriptions. Jay and Graff urge humanities departments to abandon their resistance to ‘our culture’s increasing fixation on a practical, utilitarian education’ and instead embrace the fact that ‘many heads of philanthropic foundations, nonprofits, and corporate CEOs … have lately been extolling the professional value of workplace skills grounded in the humanities’. The critical vocationalism that they advocate

In particular, humanities departments should emphasise both to their internal and external constituencies ‘the range of useful professional competencies with which a humanities education equips 21st-century students’. Such students learn ‘to read carefully and to write concisely’; they learn to ‘analyze and make arguments in imaginative ways, to confront ambiguity, and to reflect skeptically about received truths’; and, in encountering ‘texts of diverse cultures’, they are able ‘to put themselves in the shoes of people who see and experience the world very differently from their own accustomed perspectives’. All these abilities are ‘skills that are increasingly sought for in upper management positions in today’s information-based economy’ and ‘transnational marketplace’. While Jay and Graff continue to believe ‘that studying philosophy, literature, and the fine arts … have a value in and of themselves apart from the skills they teach’, for them in the end it is those skills, not that value, that justifies the continued existence of humanities departments: ‘there is no defense of the humanities’, they declare, ‘that is not ultimately based on the useful skills it teaches’.

Faced with declining numbers of majors, no few English departments have, I would guess, adopted at least some of the language and emphases advocated by Jay and Graff (among many other such advocates), especially in self-justifying documents like the IU South Bend English major mission statement. Notably in the latter, entirely absent is any recognition of literary value, just as in Jay and Graff’s article the notion of ‘value in and of themselves’ yields place to ‘the useful skills it teaches’. Taking note of this absence, a sceptical reader of the mission statement – that is, the very reader to whom it is directed – might reasonably hesitate over its inclusion of ‘the many traditions of literatures in English’: why must literary texts be the vehicle – or even part of the vehicle – for the development of the listed skills? Aside from ‘creative expression’ – which this sceptical reader might find merely tautological with ‘traditions of literatures’ – what special claim do literary texts have on ‘critical thinking … cultural and historical knowledge, skills at and methods of textual analysis and research’?

Given that the anthropologist sitting next to me when first I typed the preceding sentence vigorously affirmed that all those skills are ones fostered by his field – as I suspect they are by many others – the answer would seem to be: no special claim at all. To be sure, as we have seen in the preceding chapters, no small number of attempts have been made within what I have been calling ontological accounts of literary value to substantiate this sort of special claim. I consider a very small subset of similarly oriented attempts below. At this point we may simply observe, as I did in Chapter 3 in respect to Seyhan’s claims, that the inclusion of ‘traditions of literatures’ in the mission statement must rest on the implication that the study of literary texts facilitates the development of the listed skills in a uniquely efficacious manner (or, at least, that literary study lends the skills unique qualities); or on the unstated assumption that literature is worth engaging for its own sake, even if just as a side-effect of developing the skills; or on the banal fact that in literature departments literature is what is read. But since the first two of these justifications depend in turn on an unstated claim about literary value – literature’s distinctive efficacy at fostering the listed skills, or its intrinsic value – and the third is no justification at all, the absence of any recognition of literary value produces the same sort of incoherence that we saw with the letters protesting the elimination of the MLA Chaucer Division.

Obviously, if there is little-to-no justification for the inclusion of literary texts, then there is even less justification for the inclusion of any particular literary text – that is, justification for any canonical selection of texts. Conversely, however, if we do admit one of the above justifications – say, the most modest one, that the study of literature lends unique qualities to the skills of ‘critical thinking … cultural and historical knowledge, [and] skills at and methods of textual analysis and research’ – then the question arises as to whether some literary texts facilitate this better than others. Most literary scholars, even those sympathetic to the position of Jay and Graff, would, I believe, answer this question in the affirmative. For example, Robert Scholes, in his book English after the Fall urges English departments to put aside literature as their disciplinary centre in favour of what he terms ‘textuality’, arguing that ‘the business of English departments is to help students improve as readers and writers, to become better producers and consumers of verbal texts … It is a humble business, but it is the only justification for the existence of these departments.’ Nevertheless, towards this ‘humble’ end, Scholes retains the category of literariness as a scalar quality. He just would neither limit the category to a predefined set of text types nor insist upon a hard line between literary and non-literary:

As I pointed out in Chapter 2, even a poor judgement of literary quality (‘very low on the literary scale’) registers the literary by means of a minimal ascription of literary value, and here for Scholes that value, offhandedly given within parentheses, is the more-or-less Horatian one of ‘combination of textual pleasure and power’. Putting aside the (no small) question of what this ‘pleasure and power’ consist of and how they can be measured comparatively, we may therefore recognise that even in this proposal for shifting the central concern of English departments away from literature, literary value still has a place (if a less prominent one). Because certain texts have more of this value than others, they presumably ought to be privileged in some fashion, even if just within that diminished place. If for Scholes the net cast by the notion of literariness is wider than that of English departments past (and in fact Scholes’s net is not much wider than the one medievalists have long used, as it includes, for example, biblical texts), there are nonetheless still better and worse fish.²³

Hence, while the ‘Sample Program in Textuality’ that concludes Scholes’s book may only fitfully resemble Gates’s spectre of ‘the high canon of Western masterpieces’, the canon-making impulse – that is, canonicity – persists. Indeed, it makes an unmistakable appearance, if an indirect and brief one, even in Jay and Graff’s exhortation of humanities departments to focus on skills. As evidence for the ‘range of expertise’ and ‘concrete value’ provided by a humanities education, the authors cite the example of Damon Horowitz, ‘a leading figure in artificial intelligence and the head of a number of tech startups’, who ‘took a break from his lucrative career to enroll in Stanford’s Ph.D. program in philosophy’. They report that Horowitz discovered that his sabbatical from the technology world actually increased his value in it. As Horowitz himself concludes in the Chronicle essay cited by Jay and Graff, ‘You go into the humanities to pursue your intellectual passion; and it just so happens, as a by-product, that you emerge as a desired commodity for industry.’²⁴ Horowitz’s initial work in artificial intelligence focused on natural language processing, and he writes about how, upon confronting limits to what he was able to have machines accomplish, he went to graduate school with questions about ‘the nature of thought, the structure of language, the grounds of meaning’. He discovered to his happy surprise a long and rich history of seeking answers to those very questions, not just in ‘analytic and continental philosophy’, but also, among other disciplines, in ‘literary theory’, one of the fields in which ‘thinkers explore different aspects of how we create meaning and make sense of our world’. Revealingly, after mentioning this inclusion of literary study among Horowitz’s panoply of humanities pursuits, Jay and Graff conclude their summary of Horowitz’s experience by reporting how he realised that his previously merely ‘computational’ understanding ‘of cognition failed to account for whole expanses of cognitive experience (including, say, most of Shakespeare)’. Horowitz realised, that is, that without the humanities, computer science could not account for the texts of Shakespeare. To the unstated question of why it should need to account for those texts, the assumed answer is that they are a key source for ‘whole expanses of cognitive experience’: presumably, they are uniquely capacious representations of human consciousness – which view, albeit cryptically implied, seems at base not so different from Harold Bloom’s uber-canonical view of Shakespeare.²⁵ In Jay and Graff’s report, therefore, the value to the technology industry of Horowitz’s humanities training is corroborated by the canonical literary value of Shakespeare.²⁶ It is hence a deepened appreciation for Shakespeare (alongside whatever higher salary Horowitz could demand) that helps to justify a humanities PhD – the very appreciation, of course, that used to justify a PhD in English, because Shakespeare, as the most canonical of ‘the high canon of Western masterpieces’, was self-evidently worth appreciating in ever deeper ways.

Defending canonicity

The canon-making impulse, then, or the reemergence of some principle of canonicity – the ‘expanses of cognitive experience’ encompassed by Shakespeare or Scholes’s Horatian ‘combination of textual pleasure and power’ – persists even in those very forums that seem intent on providing an alternative to literary value as a disciplinary rationale. It appears there as a species of the return of the repressed, as that which the adjective ‘literary’ requires but which, for all the reasons summarised above, we would just as soon not dwell upon, perhaps not even acknowledge. It is precisely that which we have seen was repressed, in a much more focused manner and rather more consciously, in the letters of protest to the MLA. In other forums, however, literary scholars have not been so coy. Provoked in part by this very impetus towards repression over the last several decades, they have been increasingly interested in unearthing, accounting for and defending this repressed. They have endeavoured to return to the problem of literary value, typically in some ontological fashion, in order to provide firmer ground for a discipline that still has ‘literary’ as part of its name.

For this chapter’s purposes, a particularly revealing instance of this effort is Mark Miller’s NCS blog post entitled ‘Why Do We Care about Chaucer?’²⁷ As mentioned earlier, Miller was one of the signatories of the Chaucer Division letter to the MLA, and since his post appeared just over a year later, it seems not unlikely that it was at least partially inspired by his recognition of that letter’s axiological lacuna. Indeed, Miller seems to call attention to this very gap when at the outset he declares that ‘we must have an answer’ to the ‘fundamental question’ voiced by the blog’s title, ‘since we enact answers to it all the time whether we think about it or not’; and in what follows he proceeds to contrast Chaucer with two Ricardian contenders for canonicity, those stars of the other-than-Chaucer MLA division, Langland and Gower. Seemingly recognising both the continued dependence of Chaucer studies on some claim for the special value of Chaucer’s texts and our reluctance to articulate that claim, Miller sets out to provide such an articulation. And with the admirable succinctness and directness encouraged by the blog form, he proposes two explanations for that value (‘embarrassingly retro’ ones, he worries), the first baldly ontological, the second apparently more genealogical: ‘1) Chaucer’s a genius; 2) Chaucer’s poetry, particularly the Canterbury Tales, is exceptionally well suited to the material and ideological conditions of higher education’.

As the latter explanation is more readily elaborated, Miller begins there. He argues, ‘Unlike the equally brilliant Piers Plowman, the Canterbury Tales lends itself wonderfully to the extraction of a short stretch of text for sustained examination.’ The key phrase ‘equally brilliant’ signals that this particular special value of extractability, while ascribed to an aspect of textual manner (to use my terminology from Chapter 2) that distinguishes Piers from the Tales, does not distinguish the two in respect to literary value more broadly considered, but rather in respect to the more utilitarian value of alignment with our typical pedagogical practices. Simply put, the Tales helps us appear to be better teachers, because in contrast with Piers its ‘difficulties … often chunk themselves into bite-sized morsels’. Miller, in genealogical fashion, is well aware of the historical contingency of this value, or, in my terms, its position within the historically persistent axiological constellations that have determined and sustained these typical pedagogical practices – those constellations that, as no few scholars have discussed, from the second half of the nineteenth century led to the emergence of English as a discipline.²⁸ It was, Miller writes, a ‘sheer historical accident that Chaucer, unlike Langland, wrote in a way that the institutional structures of 20th-century higher education ended up finding convenient’. This special value ascribed to the manner of Chaucer’s text, therefore, is not an ontological value-in-itself but rather functionally serves (that is, is linked to and hence strengthened by) the value of those ‘institutional structures’.

Yet, even if we put aside the question of why we should in turn care about those structures (the question that might be asked, not by employees within those structures, but by, say, the sceptical legislators funding them), we may recall, genealogically, what those structures have depended upon. In particular, in the US at least, the regimen of close reading and the accompanying pedagogical and professional privileging of short, nuanced passages of text were, of course, the results of the academic institutionalisation of New Criticism. And, as I reviewed in Chapter 1, at the heart of New Criticism lay judgement, the comparative assessment of literary value, with something like Gates’s ‘high canon of Western masterpieces’ serving as both anchor and outcome of that assessment – a ‘high canon’ that had, moreover, always privileged Chaucer. Thus, as it turns out, if we should ‘care about Chaucer’ because his texts lend themselves to the way that we teach, then we are caring about Chaucer because he has been and remains canonical. The ‘sheer historical accident’ of the privileging of Chaucer is no accident at all, but rather a historical chain of contingent ascriptions of value across axiological constellations. As recalled above, an already canonical Chaucer played a key role in the institutionalisation of English as an academic subject, and so not surprisingly English pedagogy has been influenced by the characteristic features of his texts. Chaucer fits our teaching because our teaching, at least to some degree, has been fitted to Chaucer. As I have suggested in the preceding chapters, when we follow the interlinkages of the axiological constellations of current ‘institutional structures’ with constellations of the institutional past, genealogical analyses almost inevitably lead back to ontological claims.

Characteristically perspicacious, Miller is well aware of this relation between genealogy and ontology (although not in those terms, obviously), and hence he goes on to consider his first answer, Chaucer’s ‘genius’, in relation to the second, Chaucer’s functional suitability to the classroom. One ‘sign of genius’, he suggests, is that Chaucer’s text ‘so persistently interrogates its own grounds that it leaves its best readers in a state of suspension’, but this quality, he notes, it shares with Langland’s.²⁹ What distinguishes it, again, is its accessibility in ‘bite-sized morsels’, ‘10–20 line chunks of poetry’. But now, beyond their pedagogical utility, Miller emphasises how those morsels remain fresh: they remain present to us in ways that ‘reward continual reexamination, that always seem capable of producing fresh insights, that always seem to be there ahead of us as we learn to think in new and different ways’, so that ‘we are tempted to say that Chaucer somehow anticipated the insights of feminism, or psychoanalysis, or whatever combination of discourses we happen to have learned from’. The feature of textual manner underlying this freshness, Miller then suggests, is the ‘very condensed condensation’ of Chaucer’s writing, by which he means how frequently Chaucer packs complex, multifaceted signification into single lines, phrases and words. Hence, even after six hundred years of being read, Chaucer’s texts still prompt new understandings of their intricate webs of meaning. This feature sets Chaucer’s oeuvre apart, Miller argues, from that of Gower, which, while similarly accessible in chunks, ‘has not proven nearly as receptive’ to the broad sort of interpretive freshness that Chaucer’s has.

Not many scholars of Middle English literature, I think, would contest Miller’s characterisation of these two features of textual manner in Chaucer’s works – its extractability and condensation. A few, perhaps, might query how much they truly distinguish those works, pointing to the extractability and semiotic density of, say, the first twenty-one lines of Piers Plowman. And these few might further wonder whether Chaucer’s texts just seem to possess these features of manner more so than others because considerably more energy has been invested into the scholarly and pedagogical recognition of them. But this nitpicking would not be fair to the aphoristic spirit of the blog format in which Miller, to his credit, was willing to put forth answers to a question that indeed needs answering. Nor would it be fair to voice reservations about the predictability of the features that Miller highlights (otherwise phrased, perhaps, as resonance, textual polysemy and self-referentiality), which have featured in many prior attempts to define the nature of canonical texts. Miller himself labels his explanations ‘retro’, and, as his longer-format work ably attests, he would be entirely capable of providing more richly theorised versions of them.³⁰ And, in any event, these reservations are mere border skirmishes in comparison with the more crucial question that the identification of the features seems irresistibly to beckon, a question that Miller (again to his credit) voices explicitly in his concluding paragraph: if we grant that Chaucer matters because of the extractability and condensation of his texts – if we follow Miller and ascribe value to those features of manner, and if we assume that those features are more available in Chaucer’s texts than elsewhere – then, as Miller puts it, ‘why do we care about that’? What, in other words, is the value of this set of specific literary values?

To answer, ‘we care about extractability and condensation because this is what makes texts valuable’, would be, of course, to respond tautologically, to declare in effect that we value what we value. And in a tightly bound situation, this is often enough (e.g., I like to eat pizza because it’s yummy, and I like to eat yummy food because, well, it’s yummy). But the very voicing of the question here indicates the obvious – that, as typical with literary value, the situation is one of loose binding, in which such tautologies are usually not enough. When we granted the values of extractability and condensation to Chaucer’s texts, we did so on the implied promise that some other, greater value would confirm or prop those more strictly literary ones. To move beyond the latter, we necessarily enter into the network of valuing, in which value ascriptions arise as mediations of other value ascriptions. The answer, then, to the blog’s titular question, ‘why do we care about Chaucer’, will ultimately involve the axiological ballast that I described in the preceding chapter. We will care about Chaucer because doing so means giving attention to something that we care even more about. Understandably – but also strategically – Miller makes only vague gestures toward this something in the conclusion of his post. He points to a pair of value systems ethical in nature and conversely related to one another: on the one hand, ‘a kind of Arnoldian humanism, newly revitalised, among other places, in the return to formalism, the recent critical emphasis on ethical self-cultivation, and some of the directions taken in affect theory’, and, on the other, ‘the cultivation of something very different: not the self, but critical habits of mind that interrupt the circuits of identification that make for Arnoldian horticulture’. Again, it would not be in the spirit of the blog format to ask for more explanation and defence of these value systems or justification for why Chaucer, or even the literary as a category, is especially well suited to them. It suffices to observe that the formulation of such an explanation, a defence and a justification would be a difficult, highly contestable task, and that the justification in particular – as suggested by Miller’s invocation of Arnold and the latter’s famous formula for culture as ‘the best that has been thought and said in the world’ – would almost certainly circle back, logically, to the fact of Chaucer’s persistent canonicity.³¹

What Miller’s blog thus illustrates with such admirably succinct clarity is how, when we set out to defend canonicity, we may offer it as its own defence, tautologically, as consisting of those features of manner that most appeal to us in an already canonical text, and/or we may wish it to consist of something more than that sort of academically cultured taste – and because of loose binding, the latter move is difficult to forestall. But once we move into the network of valuing in search of that something more, we encounter both the circularity and indefinite extension of that network. The circularity entails that, say, ‘Arnoldian humanism’ and Chaucer’s canonical value become mutually defining of one another. The indefinite extension means that however counter to our intentions, an ‘Arnoldian humanism’ may possibly wind up underwriting something like Gates’s ‘high canon of Western masterpieces’, that is, it may underwrite social and political values that we find anathematic.

The possibility of the latter is provocatively made evident in a blog post by Lynn Arner. Wryly entitled ‘Why Do We Care More about Chaucer than Gower?’, it diplomatically advertises itself as an affirmation of Miller’s post but in fact constitutes a neat counterpoint, venturing into the network of valuing and settling in a rather different place.³² Arner accepts Miller’s claims for Chaucer’s suitability to ‘the material and ideological conditions of higher education’ and for the semiotic density of his texts. She relates the former, however, to the twenty-first-century value most North American and British universities ascribe to diversity, a value to which the ‘motley crew’ of the Canterbury Tales pilgrims and the work’s ‘wide range of genres and poetic forms’ seems well-tuned. Yet for Arner the Tales merely creates ‘the illusion of an inclusive world’. Although it is not the aim of her post to substantiate this view, she persuasively points both to the fantasy status of the Tales’ depicted inclusivity and to the actual social narrowness of even that fantasy. Arner then understands the higher education institutional ascription of value to diversity as ‘a neoliberal rhetoric of inclusion’, suggesting that this ascription itself serves in practice not social justice but a system of global capitalism. Whereas for Miller the notion of the ‘ideological conditions’ of higher education leads him to consider the privileging of short passages of texts, for Arner it leads to the disciplinary formation of subjects suited to the social order.

Similarly, Arner follows the ascribed value of resonant, semiotic density – the ‘endless proliferation of meaning’ within Chaucer’s texts – to the ‘ideology of individualism’ that higher education seeks to promulgate. In typical Chaucer pedagogy, ‘readers are encouraged to invent their own innovative readings, a multiplicity that acts as proof of each student’s own unique talents and capacities’. At the same time, Chaucer’s semiotic density ‘thwart[s]‌ a stabilization of politics’, resulting in the ambiguous, mobile, self-contradictory politics of his texts. For Arner, this is part of Chaucer’s strategy for insulating art from social instrumentality, evidence of his belief – in stark contrast with Gower – ‘that authors who explicitly attempt to produce socially responsible poetry generate dull, sterile art’. For these reasons, Chaucer becomes ‘an ideal figure to celebrate in humanist approaches to authorship. Chaucer seems to speak from no position in particular but seems to stand above the fray; he appears to occupy a position outside social conflict in his day, providing a neutral, enlightened vision of society’. Arner points out, however, that Chaucer actually speaks from the position of ‘the proto-bourgeoisie’. Chaucer’s easy irony and poetic sprezzatura thus make him ‘a model to emulate for those who aspire to the types of bourgeois characteristics and aesthetics rewarded in our educational system’. In contrast, Gower is more honest, forthright and even visionary about the politics of literature. While Arner fully acknowledges the more unpalatable aspects of Gower’s social position and politics, she nonetheless understands Chaucer’s contemporary as striving for ‘a more politicised version of literature with readers ultimately acting in more socially conscious ways as a result’. And this is why ‘we’ – by which Arner means the neoliberal institution of higher education – care more about Chaucer than Gower.

It is not my purpose here to assess the merits of Arner’s response to Miller. My point, rather, is the simple one that both scholars begin with the same or similar features of manner in Chaucer’s texts, and both follow the values that they ascribe to those features into the network of valuing towards other values that they perceive as most proximate, which is to say, all the way to those values that matter to them. Hence canonicity – whether Chaucer’s or the competing possibility of Gower’s – cannot be defended or critiqued without an appeal to values that either are ultimately not especially relevant to literature (e.g., neoliberalism) or, to the extent that they are relevant, are ultimately not wholly distinguishable from the claim for canonicity that started the venture (e.g., Arnoldian humanism). As we have seen, it is this very axiological slipperiness and circularity that in many quarters have encouraged canonicity’s exile, especially in practical rhetorical situations, such as the letters to the MLA and English major mission statements. Yet, also as we have seen, exiling canonicity (as in the letters to the MLA) has not vanquished it but has rather had the effect of encouraging its return in some more-or-less reformulated, newly justified fashion (as in Miller’s blog).

It would not be difficult to show that the general characteristics of Miller’s defence of canonicity, just as with the general characteristics of defences of Chaucer that do not mention his canonicity, are not merely idiosyncratic of Chaucer studies but are in various permutations rather typical of other recent defences marshalled with more extensive theorisation and at a wider scope of application. The ontological approaches to literary value mentioned in earlier chapters, for example, could be recanvased in this light. Here, however, I will consider just two other such defences, which, given their authors’ well-established critical accomplishments, I hope will be sufficiently illustrative: very briefly, a short essay by Thomas Prendergast directed toward Middle English scholars, and, at rather more length, a mini-monograph by Frank Kermode addressed more broadly to the field of literary studies. Both of these explicitly take as their point of departure approaches to canonicity of a genealogical nature, find those approaches lacking in some respects (albeit to rather different degrees) and offer an ontological account of literary value as a defence of canonicity.

In his chapter on ‘Canon Formation’ in A Handbook of Middle English Studies, Prendergast acknowledges the explanatory power of historicising or genealogical accounts of canon formation but argues that they are not by themselves sufficient – that ‘the larger forces of history are inadequate to explain the canon’. He sets out, then, to identify ‘a quality that is necessary (if not always sufficient) to make a text canonical’, or, in my terms, he sets out to identify those features of textual manner that necessarily possess literary value and hence justify canonicity. Or, more precisely, he ascribes value to those features in the process of recognising them and proposes that value as the basis for canonicity. Although Prendergast devotes a large section of the essay to Chaucer, he takes as his exemplary text Sir Gawain and the Green Knight and focuses on the way in which it both calls attention to its own mystifications and yet remains, in the end, mysterious. It signals its repressions, but ‘what truly sets the poem apart is the extent to which the poem’s repressions are irrecuperable. And it is this lack of epistemological certainty that seems to underwrite the canonicity of the poem.’ The traditional label for this intractable epistemological uncertainty, Prendergast tells us, is ‘wonder’, which the poem both evokes and ‘meditates on’ in a way that teaches us that ‘demystification is never quite complete’.³³

Having then, like Miller, isolated a canonical textual quality (which, as a kind of inscrutability in which meanings proliferate because they are always just out of reach, shares a family resemblance to the semiotic density that Miller identifies in Chaucer’s text), at the end of his essay Prendergast seems to acknowledge – again like Miller but more obliquely – that this identification is somehow insufficient. By itself, it amounts to the claim that canonical texts are those that possess the quality that make them canonical. Hence the pressure of loose binding is felt, impelling the nagging question of why this quality is a valuable one – why, if we are indeed to accept this as a quality attached to the manner of Gawain, we should care about it so much that we privilege texts that possess it. In response, Prendergast informs us that ‘what recent work on wonder reveals is that wonder is that which we experience when confronted with novelty, potentially leading to revolutionary or utopian thinking’, although, like Miller – and for the same reasons – he does not elaborate. Instead, he concludes by asserting that Gawain and other Middle English texts considered canonical (even if not always considered so) ‘all share this quality’.³⁴ As in Miller’s blog, therefore, canonicity first seeks to be its own justification but then, urged forward by loose binding, also seeks justification in other-than-literary values (‘revolutionary or utopian thinking’) that stretch far beyond the horizon of the traces of their mediations.

Kermode, in a pair of essays based on his 2001 Tanner Lectures at Berkeley that were published as his penultimate monograph, similarly seeks to define the qualities of the canonical text in the face of genealogical critiques that understand the canon in terms akin to those of Gates’s in this chapter’s first epigraph – as Kermode puts it, for example, ‘a wicked myth, designed to justify the oppression of minorities – a political propaganda weapon now at last revealed as such and, as the word goes, “demystified”’. But, much more intensively than Prendergast or Miller, Kermode seeks to insulate literary value from the taint of utility that may accompany any recourse to other-than-literary values. He seeks to define the specifically literary value that ought to lie at the heart of literary criticism instead of those values associated with critical work on ‘for instance, gender and colonialism’.³⁵ In my terms, he seeks to define a tightly bound literary value, one that is nonetheless explicable in non-tautological terms even while being autonomous, in the sense of not mediated by other ascriptions of value.

In his first essay, drawing upon Jan Mukařovský and Roland Barthes, Kermode offers a reading of Wordsworth’s ‘Resolution and Independence’ to illustrate his contention that ‘[a]‌gain and again we find in the best of the poetry a curious blend of delight and dismay’. It is this specifically literary value of ‘juxtaposition or collision of pleasure and dismay’, he claims, that marks the canonical text, not such other-than-literary values as ‘collusion with the discourses of power’. In his second essay, Kermode seeks then to account for the relativity of value by addressing the manifest historical determinations of canonicity, initially by acknowledging the ‘element of chance’, which of course for medievalists – aware of the sheer luck that has given us, say, Sir Gawain and the Green Knight or the Book of Margery Kempe – goes without saying. More important for Kermode, though, is the way in which canons are tested and modulated according to the manner in which readers at any given historical juncture experience (or fail to experience) the requisite ‘juxtaposition or collision of pleasure and dismay’. Drawing upon Gadamer, Kermode observes that the ‘canonical text … must be made to answer to our prejudices, and they are necessarily related to the prejudices of our community, even if in reaction to them’. ‘So a canon changes’ according to these prejudices (or, in my terms, the axiological environments of a particular time and place), and ‘the changes renew the supply of both pleasure and its potent derivative, dismay’.³⁶ Hence, while the specific set of texts that constitute the canon is subject to historical determination through axiological network pathways, the specific literary value that canonical texts possess, the combination of pleasure and dismay, remains historically constant and, so tightly bound, may therefore serve as a transhistorical principle of canonicity. Literary value per se escapes mediation, even while any given experience of literary value, Kermode in effect acknowledges, is necessarily mediated through the axiological constellations that determine that experience. Kermode’s tightly bound pleasure principle, in turn, may ground the discipline of literary criticism as an intrinsically literary endeavour, one that is distinct from, say, the more-than-literary ideology critique that exposes ‘collusion with the discourses of power’.

To work through the implications of Kermode’s argument, we have the benefit of the responses by Geoffrey Hartman and John Guillory, which were published in the same volume. Since Hartman, over his long, storied career, had of course himself proffered many a searching consideration of literary value, we should not be surprised to find that he expresses reservations with Kermode’s attempt to provide literary value the particular tight binding that he specifies. Hartman queries Kermode’s choice in a way that suggests the pressure of loose binding, posing a question that is in effect a more elaborated version of Miller’s ‘why do we care about that?’: ‘what general cogency, beyond being a promesse de bonheur, a reward for a more complex understanding of tradition or acculturation, does the criterion of “pleasure” have, revived by Kermode?’ While registering the pleasure that he, like Kermode, receives from canonical texts, Hartman ultimately finds that the particular pleasure that Kermode identifies is not self-evident in a way that enables it to carry value-in-itself weight. Drawing upon Lionel Trilling, he sees Kermode’s literary value as in fact anchored by a politics – ‘the eudaemonic nihilism of a liberal, progressive politics’ – that stands in (blind) opposition to ‘an anticonsumerist force calling itself spiritual, and often in total contempt of pleasure, indeed of worldly society as such’.³⁷ Hartman in effect suggests that Kermode achieves tight binding simply by ignoring the other-than-literary values that actually prop up his pleasure principle.

To build his case, Hartman offers a counter-reading of ‘Resolution and Independence’ that understands the poem’s primary dynamic not as one of pleasure but of power, claiming that the ‘powerlessness’ of the leech gatherer, ‘who seems scarcely alive, has to become a source of power for the poet’. Recalling Wordsworth’s ambivalence even toward The Terror, Hartman argues, ‘Unpower/power, not the pleasure/unpleasure complex, is the [poem’s] problematic subject.’ Hartman concludes that Kermode’s defence of canonicity, by insisting on the tight binding of the value of pleasure, therefore ‘skirts the political impasse that presently makes literary criticism, not only literature, a troubled mirror of our culture’.³⁸ In my terms, Hartman points out that not only are the ascriptions of the literary value of pleasure/dismay to specific texts necessarily mediated through ‘prejudices’ – axiological constellations of political and other values – but so is the very conception of that value itself. (By the same token, Kermode’s other species of change, the force of chance, emerges out of axiological constellations that determine what sorts of happenstances are possible to begin with.) As I discussed in Chapter 3, any one kind of value is only conceivable through reciprocally distinguishing relations with other kinds of value, whether those relations are supportive or hostile. We care about pleasure/dismay (or Hartman’s alternative, unpower/power) because, whether we acknowledge it or not, we care, or do not care, about the other values that enable us to conceive of that very sort of pleasure, and, in a loose binding situation, those other values cannot be left unacknowledged without incurring some incoherence. Thus, despite Kermode’s evasions, literary judgement (pleasure/dismay) still resides on the slippery slope of, say, Gates’s ‘medieval never-never land’ (politics).

In contrast, Guillory – whose Cultural Capital remains among the most influential of genealogical examinations of literary value – is wholly willing to grant pleasure the status of a value that needs no further justification. Given a world in which literary value is, as Guillory has argued, linked in practice to the other-than-literary value of cultural capital, one may seek to resist the larger value system encompassing the latter by identifying and then refusing that linkage. In some ways mirroring, therefore, the final movement of Cultural Capital, Guillory embraces the idea that ‘the pleasure of the literary work is … its chief reason for being, and … the communication of that pleasure to the readers of criticism is at least one of the purposes of criticism’.³⁹ But in a loose binding situation, to refuse a linkage without putting another in its place is also to refuse to answer Miller’s question of why we should ‘care about that’. With loose binding, that is, such a refusal foregrounds the arbitrariness of the choice of value. Guillory – unlike Kermode – accepts this arbitrariness, and thus what he objects to in Kermode’s argument is the privileging of the specific sort of pleasure that Kermode identifies as the principle of canonicity, since without the axiological ballast of an other-than-literary value, that privileging no longer possesses any rationale. Guillory does not question, in other words, why we should care about pleasure (he is happy to answer, in effect, ‘because it is pleasurable’), but instead rejects the claim that we ought to care about one particular sort of pleasure so much more than others, aesthetic or otherwise, as to require the very business of canon-making and the discipline that performs that work. To so privilege one such pleasure, Guillory suggests, is to give up the refusal of the linkage that in fact enables one to embrace pleasure as the end of the literary experience.

Since Guillory is unwilling to grant ‘a higher status’ to the pleasure that Kermode identifies, he also does not extend special privilege to ‘the domain of culture the higher authority upon which cultural criticism was and continues to be based’, as that would simply be broadening the linkages that he wishes to refuse. Rather, to sever pleasure-finding from canon-making, Guillory locates the former in a host of other value constellations: ‘Our speech, our manners, our bearing, our dress, our houses, our furnishings, our public spaces and private entertainments should all be beautiful, should deliver their measure of aesthetic pleasure.’ In effect, as long as ‘aesthetic value’ is everywhere, its inevitable linkages do not as much matter, and so in its very ubiquity it gains a kind of quasi-independent status. In contrast, to identify one species of aesthetic pleasure as more important than another necessarily involves – as we have seen with Hartman’s critique – the ‘prejudices’ that are something more than pleasure, and so Guillory declares that we cannot ‘generalize any principle from the experience of aesthetic pleasure that would ground a principle of evaluation or canonicity’. Since the ‘complex pleasures’ of so-called high art must then constitute just one kind of pleasure among others, no more or no less valuable than any of the rest, we must ‘retreat from attempting to make the connection between the quality of pleasure and the judgement of canonicity’, not the least because the latter judgement, in necessarily moving beyond pleasure per se, undoes pleasure’s quasi-independent status and thus threatens the very pleasure that was so elevated.⁴⁰ In short, in a loose binding situation, to relish in good faith pleasure as of value in and for itself, as Guillory is willing to do, requires us to recognise the arbitrariness of its selection over other values and hence its inability to serve any larger axiologically anchoring function.

Since it is possible (though, I believe, inaccurate) to understand Hartman as keeping canonicity in place but arguing for a different grounding principle, Kermode, in his response to the critiques that concludes the volume, directs the most heat toward Guillory and in a manner especially revealing for present purposes. Summarising Guillory’s argument, he states that the latter ‘attacks the notion that some things give more pleasure than others, holding that it is unfair to claim privilege for the “higher”, since what is normally thought to deserve that label can be regarded as a “very minor subculture in a vast domain of cultural production”’.⁴¹ Yet, as we have seen, while Guillory does indeed reject the privilege granted ‘the “higher”’, he does not make any claims about relative amounts of pleasure, and this slippage between quality and quantity enables Kermode to side-step Guillory’s actual critique, which targets the privileging of one kind of aesthetic pleasure over another.⁴² By assuming instead a single kind of aesthetic pleasure that is provided in different quantities (not unlike, interestingly, Scholes’s ‘literary scale’), Kermode in effect merely repeats, without offering any defence of, his privileging of ‘the “higher”’, wondering aloud whether Guillory ‘does not have experience of the difference between serious fiction and rubbish’ and expressing regret that those who know better ‘say … that they have wholly comparable experiences from a television soap and Dante’.⁴³ Because of Kermode’s singular tightly bound value, all cultural products are measurable by the same scale, and thus a canon necessarily follows. And in practice the measuring stick turns out to be, despite the forces of chance and change, uber-canonical texts such as Wordsworth’s and Dante’s. Just as with Miller’s and Prendergast’s defences, then, we arrive at a position in which canonicity is its own justification. For Kermode, it is a self-evidently ‘higher’ pleasure that bestows canonicity upon a text across the flux of historical change, and yet it is already canonical texts that define what constitutes this pleasure in respect to all other texts. The autonomy of Kermode’s literary value turns out to rest upon the familiar tautology that canonical texts are so because they possess the quality of canonicity – as indeed is inevitable, because such a tautology is ultimately one and the same as a claim for tight binding.

As we have seen, both Hartman and Guillory take Kermode to task for his occulted incursion of this tautology, each in effect recognising the demands of loose binding, with Hartman accepting those demands by pointing to possibly defining other-than-literary values, and Guillory seeking to evade those demands by embracing the anticanonical implications of the tautology. Kermode, in simply reasserting the autonomous, privileged status of his chosen literary value in his more informal concluding remarks, cannot help but to betray the lacuna in his reasoning. For example, in the rather tart remarks that frame his response to his interlocutors, he states,

The business of the critic, Kermode insists, is the discerning and communicating of specifically literary value (the sort of pleasure that he has identified), not ‘other topics that some work of literature happens to present’, whose ‘value would pertain to another subject’. Inasmuch as a critic focuses on those ‘other topics’, Kermode avers, she is simply not doing her job. But in respect to keeping those other topics at bay, and hence literary value autonomous, the phrase ‘no nose you can trust’ is (to mix his metaphor with my own) the card falling from Kermode’s sleeve. Although in its specific context this phrase refers back to Kermode’s invocation of William Empson’s remarks on the use of theory in criticism, in the larger context of what has preceded and what will follow, it also cannot help but to invoke, through an irresistible synaesthesia, the category of taste (and, indeed, when initially referring to Empson’s remarks, Kermode acknowledges an ‘enological analogy’). Kermode, not surprisingly, does not elaborate on what having a ‘nose you can trust’ entails nor addresses whether or how one may obtain such a nose. Tellingly, when expressing disbelief at Guillory’s supposed flattening of ‘the difference between serious fiction and rubbish’, he simply remarks that the experience of this difference ‘is a fact of life, however difficult it may be to philosophize it’.⁴⁵

Although I cannot here embark upon any adequate discussion of the category of taste, we may simply observe that by invoking that category, Kermode inadvertently discloses how firmly he has backed himself into a corner. On the one hand, Kermode invokes the category precisely so that he does not need to explain ‘the difference between serious fiction and rubbish’, since taste, as Lucien Karpik (drawing on Hannah Arendt) notes, ‘is idiosyncratic, and therefore no argument can prove any overall superiority’. But on the other hand, by implying that true literary critics would share his taste (and that those who do not ought to find other employment), he suggests that taste is emphatically not idiosyncratic but rather a shared judgement that definitively separates, for anyone actually paying attention, the ‘serious’ from ‘rubbish’. Yet as Karpik further notes, in contrast to taste, judgement ‘is totally in the world. Because it embodies a norm, it is inseparable from all other judgments’.⁴⁶ In my terms, judgement is a value ascription performed by an actor as a mediation of the value ascriptions of other actors, within an infinitely receding axiological network. Judgement expressed as taste, then, is a species of value ascription that presents the mediation of other value ascriptions as not such mediation, and thereby gains authority through that stratagem. In Pierre Bourdieu’s famous formulation, taste is ‘a class culture turned into nature, that is, embodied’.⁴⁷ It is an immensely powerful mechanism by which aesthetic value may command all sorts of other values (or vice-versa).

Kermode’s terse acknowledgement of how difficult it is to ‘philosophize’ the distinction between ‘serious’ and ‘rubbish’ hence serves as an abrupt erasure of the axiological constellations within which any such distinction must necessarily be made. Indeed, in naming this distinction a ‘fact of life’, he renames as fact what are quite plainly values (the judgements of ‘serious’ and ‘rubbish’) – and facts, of course, are the one thing that values are not. (Or, more precisely, they are always already saturated with values from the moment that they are experienced as facts.) Kermode’s attempt to secure his literary value of pleasure/dismay from any sort of axiological mediation thus eventually bumps into a sort of conceptual incoherence kindred to what we have seen in other defences of canonicity, only Kermode seems wistfully to imagine judgement-as-taste as also becoming the ground of and gatekeeper to the profession. By refusing to acknowledge the other values that would prop up his insistence that the ‘collision of pleasure and dismay’ is present in canonical texts in the fashion he has described and worth caring about so much as to serve as the basis for an entire profession, he can only say, in effect, either you get it or you don’t, and if you don’t, you should stop talking about literature.⁴⁸ Faced with the prospect in which literary value is only experienced as such within an infinitely receding axiological network, he would like to transform the portion of the network comprising the institution of literary criticism so that all its actors ascribed value in the same way – the way that he does (if not necessarily to the same text). Unfortunately, for Kermode, there will always be those who favour the cheap wine.

Canonicity is dead: long live canonicity

In the wake of this perhaps ungenerous reading of Kermode’s defence of canonicity, let me reiterate that I am not arguing that his or any of the other defences that I have considered is inadequate – in the sense that categorically better defences might be marshalled – or that this general endeavour is not a worthwhile one. Rather, as I mentioned in the preceding chapters, defences of canonicity, as ontological defences of literary value, are among the crucial tasks of the field of literary studies. Complementary with rather than antagonists to genealogical analyses of literary value, they help to chart the expanse of the network of literary valuing. A further consideration of why this is so, what its implications are for how we go about teaching and studying literature, and how we might then navigate the problem of canonicity, is the task of this final section.

What the prior sections of this chapter have sought to trace is the double-bind of canonicity, the basic idea of which I introduced with the chapter’s pair of epigraphs. If we avoid defending canonicity, being aware of its ideological and conceptual pitfalls, but retain the category of the literary (in however limited a fashion), canonicity inevitably reemerges in some unformulated but logically mandated way whether we acknowledge it or not. But if we do acknowledge this reemergence and accordingly seek to provide some formulation for canonicity, we find it always slipping from our grasp into the endless relays of interlinked networks of value. To refuse to recognise that slipping is to chase our own tails, arguing some untenable form of the tautological claim that canonicity is the quality of canonical texts. Yet to recognise those networks, even vaguely, is to discover that the values that would finally give canonicity axiological ballast both keep circling us back to where we started and are always just around the corner (of the next mediation), never quite in reach, and that we cannot prevent those cascading mediations from potentially involving values that we may find anathema.

One seemingly ready solution to this double-bind is that which Guillory proposes in his response to Kermode: why not just discard canonicity but keep literary value? Once we have recognised that there is no single scale upon which the value of all texts may be weighed, should not our canonical burden be lifted – should not we be free to enjoy literature, and the study of literature, without needing literature also to serve some larger purpose that is ultimately not literary and perhaps ideologically problematic? The ethically and politically charged nature of critical discourse at present suggests that we cannot so easily be let off the hook, and the formulations of the preceding chapters help explain why. If ascriptions of value to features of textual manner occur differentially by way of mediations of other ascriptions of value across the axiological network, then in most if not all cases, among the most proximate mediating ascriptions will be ones involving like features of perceived manner in other texts. Hence, our ascriptions of value to textual manner emerge differentially not just in kind but also in degree. In short, within the network of literary valuing, judgements of literary value inevitably include judgements of relative literary value.

I am of course just stating the obvious, in a rather abstract, convoluted fashion, which the simplest example illustrates. If I find, say, the stress pattern of two lines of text musical, I am able to do so in part because ‘musical’ is an established literary value that other pairs of lines have in greater or lesser amounts, as made available to me through mediated value ascriptions, whether mine or other actors’. Thus, while we may readily follow Guillory in refusing to enthrone a single quality as the principle of canonicity, we may not so easily follow his assumption that we may have literary value without hierarchy. Although there is not a single scale of literary value as Kermode seems to wish there were, as long as texts seem to possess similar features of manner, the value ascribed to them, because it is differential, is in practice unavoidably scalar. Indeed, as the sheer ubiquity of ‘top ten’ lists for this or that attest, one the greatest pleasures that we seem to get from literature, or any aesthetic experience, is that of relative judgement and the construction of hierarchies of value.

Such hierarchies of value are necessarily idealising within the vector of the scale that they delineate, and they therefore entail a version of the present absence dynamic that we saw with extensions to totalising values in Chapter 3. Once value becomes scalar, every judgement involving that value – however provisional, spontaneous or minute – is haunted by the shadow of that value imaginable as fully realised, perfectly achieved, available in the greatest plenitude possible. Integral to each act of scalar judgement is an invocation of the phantasmagorical perfection required for that judgement to occur – an invocation that may be more-or-less conscious, more-or-less conceptually definite and more-or-less temporally stable. In an axiological network constellation of any persistence, these invocations may crystallise into a palpable principle of hierarchy, which is to say more simply that what may emerge out of any top ten list is a principle of top-ness, however implicit or inchoate that principle may be.

Canonicity is this principle of top-ness writ large. It is the notional ideal that any act of scalar judgement, which is to say virtually any recognition of the literary, necessarily invokes. It is the simulacrum of the ideal literary work, according to whatever specific ascription of value is being performed, that that ascription requires and invents.⁴⁹ It is, for example, the imagined perfect realisation of Scholes’s ‘combination of textual pleasure and power’, Miller’s ‘very condensed condensation’, Prendergast’s ‘wonder’ and Kermode’s ‘collision of pleasure and dismay’. But it is also the more inchoate or even wholly unacknowledged principle at work when we select texts for syllabi, choose an anthology or decide which conference sessions to attend. It is that which any defence of the value of literature or of literary study – on the basis of, say, literature’s facilitation of critical thinking, empathy, negotiation of ethical complexity and cross-cultural understanding – necessarily invokes, as the imagined vessel conveying the maximum possible amount of all those virtues. In short, we cannot study something that we call literary without positing the category of canonicity, whether or not that category explicitly enters into our research or classrooms. Nor can we avoid canonicity within a cultural studies framework that features literary texts merely as one cultural artefact among others. For, as I argued in Chapter 1, insofar as we recognise that someone has registered those texts as literary, someone else’s canonicity – through a series of transmutations through the network of literary valuing – becomes also our canonicity.

While canons themselves are wholly mutable, therefore, we cannot eliminate canonicity without also eliminating the category of the literary. And this returns us, one final time, to the dilemma that has been this chapter’s overarching purpose to describe. On the one hand, many of us hesitate to formulate principles of canonicity because of what we have come to learn about canonicity’s conceptual pitfalls and ideological liabilities. And then, with the resulting aporia in the centre of our object of study, we find that in some important respects we have nothing to counter scepticism towards the value of our subfield or discipline as a whole, since these institutional formations cannot escape resting upon the value of that object. On the other hand, those of us who do venture to formulate principles of canonicity produce accounts that cannot withstand the very conceptual pitfalls and ideological liabilities that keep the rest of us at bay. When we attempt to fill the aporia left by our reluctance to judge, we cannot help but reproduce the problems that led to that very reluctance. The question, then, is where do we go from here?

Although any answer must necessarily be provisional, if we approach the problem in the simple terms of bad faith (relying on canonicity even while not believing in it) versus blind faith (committing to canonicity, regardless of the consequences), one possible response readily emerges: to have faith in the necessity of having faith. That is, we may explicitly embrace the central role of canonicity and simultaneously make reflection on the intractable challenges of formulating canonicity’s principles itself one of the field’s defining tasks, both in scholarship and in the classroom. As I suggested in the preceding chapter, some degree of an explicit adoption of an axiological framework may be beneficial in this regard. By simply acknowledging something like that framework as the field’s big tent, we would bring to the surface the axiological negotiations that we are already performing, and the field as a whole might thereby be better oriented towards understanding the intractable challenges of canonicity as one of its ongoing tasks. If we were then to draw on this framework as way to delineate our own positionality within the axiological network, as I also suggested, we might better recognise the everyday work that we do – say, putting together a syllabus, writing an annotated bibliography, producing a critique of a scholarly tradition – as scalar ascriptions of value that are mediating others’ scalar ascriptions of value and, in this way, invoking canonicity by creating hierarchies of value. We could then turn some of our attention to those invocations, investigating in some form or another the manner and extent of the traversal of the axiological network that our work necessarily involves. In other words, in performing the work that we are already doing, we would also be putting our axiological cards on the table, so to speak, so that both we and our scholarly and pedagogical interlocutors may consider what it means to have a winning or losing hand. This disclosure, moreover, would have the benefit of possessing a ready answer to the question of ‘why should we care about that’, for it is precisely an explanation of why we care.

As a pedagogical example of this sort of response to the problem of canonicity, I offer an instance of my own recent experience – or, less arrogantly and more accurately, an example of an awkward pedagogical collision with the problem of canonicity that baldly exhibits the very inchoateness that I have described for others, in response to which I sought to cobble together a more coherent approach. Not long after I arrived at Agnes Scott College in 2015, I recognised that my new institutional home afforded an opportunity to rework the Canterbury Tales course that I referred to at the beginning of Chapter 1, to transform the course formally in ways that for years it had been informally tending. As its mission statement reads, ‘Agnes Scott College educates women to think deeply, live honorably and engage the intellectual and social challenges of their times’, and students embrace this mission passionately. Exceptionally diverse racially, socioeconomically and in sexual orientation, the students as a group are nonetheless generally united in their progressive attitudes toward gender and LGTBQ+ issues.⁵⁰ Given this institutional context, the fact that over the years gender and sexuality had increasingly been taking centre stage in my pedagogical approach to the Canterbury Tales and the lack of any specific mandate for teaching Chaucer at all, I decided to refashion the course to make it one of the rotating topics for the ‘Studies in Gender and Sexuality’ course that is cross-listed with the Women’s, Gender and Sexuality Studies department, rechristening it ‘Gender, Sexuality and Chaucer’.

The first time that I taught the course, I naturally focused on the importance of the cross-listing. In the terminology of this book, I presented as the course’s axiological network anchor point progressive values regarding gender and sexuality, assuming that what we would care most about would be feminist theory and criticism, gender and sexuality theory, and the injustices that these focuses would bring into view. Chaucer and his works were, in this respect, one or more mediations away in the network, nudged from the centre in favour of an initial set of values that I believed I shared with the students. Hence the aim of the course was to discover what light the study of a premodern literary text such as the Canterbury Tales could cast on our more central concerns, most often by way of comparisons and contrasts between medieval and modern.

As the semester progressed, however, I increasingly felt the pressure that this approach put on the rationale for focusing on Chaucer rather than any other premodern author or just any text whatsoever. When students inquired into what the study of Chaucer could tell them about modern feminist theory, for example, Chaucer’s massive shortcomings as a feminist quickly and often became the most imposing topic in the room. We continually bumped into the question of why, in order to study gender and sexuality, we were studying texts by a relatively prosperous white fourteenth-century Englishman, one whose literary representations of gender and sexuality – no matter how ironic, ambiguous and self-reflexive – plainly reproduce elements of Western misogyny.⁵¹ We bumped into the very question, that is, that has recently been cogently and urgently reiterated by a number of feminist medieval literary scholars – the question of whether, in respect to progressive values of gender and sexuality, Chaucer’s works are not just beside the point but in fact detrimental. As Suzanne Edwards puts it in one of the two recent special issues of ChR that, broadly speaking, consider this topic,

Or, as Sarah Baechle and Carissa Harris query in their introduction to the other special issue, ‘How can we, as scholars committed to ethics and social justice, write about or teach Chaucer’s work without upholding the patriarchal and white supremacist institutions the poet and his oeuvre have advanced?’⁵³

To be sure, I began the semester equipped with what I believed was a set of student-friendly prompts for considering this question, but I soon had to admit (to myself) that none of them was wholly adequate. The axiological orientation of the course asked students to find ways to link Chaucer to the more urgent concerns of gender and sexuality, and most students achieved this by taking the ready path of focusing their contributions to discussion and especially their papers on Chaucer’s shortcomings or just plain toxicity in those respects. Such a focus as a scholarly project has performed and continues to perform the salutary work of disenchantment, of inspecting the House of Chaucer and discovering the holes in the roof and the skeletons in the basement that overeager prospective purchasers wilfully overlook. Inasmuch as other scholarly projects (my own not excepted) continue to perpetuate Chaucer’s enchantment, the former project bears repetition and elaboration.⁵⁴ In the context of my course, however – in which most students had at best only a very vague sense of a House of Chaucer to begin with – I found myself in the awkward position of needing to build that house for a sceptical audience just so that we had a reason to tear it down.

One solution to this awkwardness would have been simply to walk away from the falling-down house altogether, as Edwards in particular seems to recommend. But in the middle of the semester that remedy obviously would have incurred an even more disorienting awkwardness, and so I stuck with the basic plan, tweaking it as best I knew how. Over and over, the students expressed informed, perceptive views about Chaucer’s shortcomings as a feminist, but the sheer repetition of this line of argument, applied to this text and that, began to feel like shooting fish in a barrel. While on the whole the students did gain a more historically detailed, nuanced understanding of the longevity and intricacy of Western misogyny and patriarchy, this served mostly to fill in the gaps of the picture of the past that they had already largely drawn. Pedagogically, I felt that I had let them down, as they deserved an educational experience that went beyond such a repetitive confirmation of expectations.

Very likely, a more skilled, knowledgeable and creative teacher would have had greater success with the same course framework. And there is no doubt, too, that my identities as a white-passing Asian straight cis man, occupying the authoritative position of professor, affected classroom dynamics in respect to the course’s thematic focus. Nonetheless, I am vain enough – and had close enough relationships with some of the students – to believe that no small portion of the awkwardness lay in the course’s framework and not just in its delivery. Working on this present book during this time, I was prompted to realise that I had incurred a blind spot very similar to the one that I have described for, say, the letters to the MLA Executive Committee, only in my case it scarcely achieved those letters’ strategic success. There was a yawning conceptual gap between the values ascribed to issues of gender and sexuality and the pedagogical requirement to consider these issues in relation to Chaucer. I had sought to provide a rationale for what was manifestly still a practical centring of Chaucer (despite his relegation to the third term in the course title) that did not rely on the proposition that there was something special about Chaucer’s works – the very same sort of gap, or inchoateness, that we have seen in the attempts to defend literature without recourse to canonicity.

The actual most basic rationale for the continued centring of Chaucer was, obviously, my own investment in Chaucer’s literary value, the inheritance of Chaucer’s longstanding canonical status that pervades my experiences of his writing. By leaving that rationale to the side, I founded the course’s framework, in effect, on a non sequitur. In response, taking cues from my work on this book, the next time that I taught the course I reversed the axiological polarity of my approach. Laying my axiological cards on the table, in place of the basic question of what we can learn about gender and sexuality by studying Chaucer, I framed the course as an inquiry into what we can learn about Chaucer by way of feminist, gender and sexuality theory. The value of Chaucer’s texts, that is, I established as one of the course’s premises, just as it had been positioned for countless other courses back in the era before the so-called culture wars – Gates’s ‘medieval never-never land’. But, in contrast to what I suppose to have been the case in that never-never land, I made this premise explicit and described it as what it is: contingent. Chaucer’s literary value would not be an unquestionable, mysterious aura of greatness, but an institutionally established and perpetuated privileging of one author over others, a privileging that has without question been sometimes complicit with other kinds of privileging through the various networks of other-than-literary values with which his literary value has been interlinked.

Within this framework, the course evolved into an exploration of why anyone might value Chaucer’s writing enough to find rewarding the discovery of the complex ways that it represents and meditates upon gender and sexuality, no few of which ways, notwithstanding their complexity, we might find discomforting or simply reprehensible. While students once again readily perceived that Chaucer was no feminist, they more often took that perception as point of departure rather than a conclusion. That point of departure led them to see how, among other things, Chaucer’s artistry so often depends upon the contradictions and repressions of a patriarchal social order that shaped him and in which he was complicit. As a result, we gained more insight into both Chaucer’s artistry and those contradictions and repressions and their long histories. In effect, we more successfully traced the complex relays among value ascriptions – my own, but more importantly those of the students, those within Chaucer’s texts and intertexts and those within Chaucer scholarship – following them from literary values (Chaucer’s and others’) to the values of various other kinds that attend diverse understandings of gender and sexuality. In presenting canonicity explicitly as an institutional fact and also a problem to be explored, I found that even anticanonical impulses led to rich critical discussions. The problem of canonicity proved to be generative rather than limiting.

I do not wish to generalise this experience beyond what its circumstances may bear. Agnes Scott’s distinctive student population, my identities and teaching style, the particular classroom dynamic, the agency I had in respect to the curriculum, and so on, were all inextricable contributors to the outcome. Under different conditions – say, with a classroom including many cis white men, under an institutional mandate to teach Chaucer every year – something more in line with my first approach could be a better fit. I do not mean, therefore, to suggest how Chaucer ought to be taught or even to insist that he be taught at all. Additionally, I do not mean to claim that my revised approach represents any special innovation. Rather, I suspect that wiser teachers than I have long taught Chaucer in a similar fashion.⁵⁵ In fact, this suspicion significantly motivated my choice of this example, as it suggests that my recommendation of having faith in the necessity of having faith does not necessarily require major adjustments to current practice. In my case, the adjustments consisted of just foregrounding a previously occluded belief in Chaucer’s literary value and simultaneously encouraging an inquiry into the conditions that sustain that belief.

In a rudimentary way, the example of my pedagogical having faith in the necessity of having faith points to a more general, far-reaching affordance of an explicit, reflexive inquiry into the problem of canonicity, namely, a route to the effect that I described in the previous chapter’s discussion of defamiliarisation. Canonicity, as the idealised simulacrum of literary value that any scalar act of literary valuing invokes, lends itself rather directly to extension from features of textual manner to totalising kinds of other-than-literary value – as we saw exhibited in Boccaccio’s extension of features of textual manner to the duelling totalising values of God and power, by way of the canonical value of Dante’s literary achievement. Canonicity, as itself a present absence that may be manifested in any specific act of literary valuing, easily draws into itself more totalising present absences, functioning in this way rather like an axiological black hole. Because canonicity is quite catholic in the range of these totalising kinds of value that it attracts, the more energy that we put into a reflexive negotiation of the problem of canonicity in any particular situation, the more that negotiation may disclose of our axiological environment. The more thoroughly, precisely and reflexively that we explore why, say, Chaucer’s works deserve canonisation, the more profoundly and exhaustively we may plumb our own axiologies. By this I do not only mean that we confront our own value ascriptions in an unusually broad and deep manner. I mean also that we recognise both the contingency of those value ascriptions and, simultaneously, the necessity that they function in any given moment as if they were not merely contingent. In respect to this chapter’s epigraphs, we recognise that in judging, we risk aligning ourselves at some level with values we find anathema, but we also recognise that we nonetheless may not cease judging: once again, we have faith in the necessity of having faith. And so at the very least we are on guard, and we empower others to hold us to account for the judgements that we cannot cease making but which we can alter.

This kind of converse, dynamic and indeterminate pair of axiological recognitions, I argue, is one that literary study in both its scholarly and pedagogical forms is particularly well suited to foster. Although these recognitions are potentially available in virtually any field, within literary studies, because of the very problem with its object of study – the problem of literary value, the problem of canonicity – they may powerfully facilitate a richly varied and extensive engagement with the category of value per se. They may shine a spotlight on our being as actors in an axiological universe, on how value ascriptions are begotten by and beget other value ascriptions, and on the relations between values and the actors ascribing those values, which include the so-called objects to which values are ascribed. Whether in a monograph, syllabus or class discussion, if we bring to the surface the relays from ‘I like this book’ to ‘you ought to read this book’ to ‘this book is good for you’ – even while still seeking, in good faith, to construct those relays – literary studies can illuminate the mediations that link ‘like’ to ‘good’ across a wide range axiological domains.⁵⁶ In short, that we cannot avoid canonicity, even if we would like to, may itself be worth making one of our central concerns – as it is something that we all share within the experience of literary study. And we may then use to our advantage the fact that canonicity also happens to be that which much of the extra-academic world has already long understood to be our area of expertise.