Thinking with Literature
I must first thank the Balzan Foundation and the Swiss Academies of Arts and Sciences for organising this event. I’m extremely happy to be here, and not only because I have been fortunate enough to win this amazing Prize, but also because this is the kind of intellectual interdisciplinary discussion I most enjoy. A conversation, I hope, as it will eventually become in the later part of the afternoon. Professor Milner is a very hard act to follow, especially as I am left-handed. I hope it won’t be held against me, it probably will confuse some of my operations.
My research activities cover a period of nearly half a century: I began working on my doctoral dissertation in 1960. Rather than providing a linear narrative of that period, I think it will be more interesting and productive to focus on my central preoccupations and working methods, which, as it now seems to me when I look back over my career, have remained surprisingly constant. I say “surprisingly” because, in the course of that half-century, the field of literary studies has undergone a transformation which a doctoral student of the early 1960s could not possibly have foreseen: the rise of “structuralism”, which borrowed from both linguistics and anthropology, and the fierce debates that it provoked between traditionalists and the avant-garde around the symbolic date of May 1968; the mutation of structuralism into a sometimes arcane post-structuralism, which drew on the philosophy of Jacques Derrida and the post-Freudian psychoanalytic theory of Jacques Lacan; the advent of an exorbitantly relativistic postmodernism; and out of that again the broadly “ethical” agendas that flourished above all in the 1990s – feminism, gay studies, postcolonialism, the new historicism – and still make their presence felt today amid the current preference for a cultural history in which literary texts are always read for their social and historical import alongside other kinds of document.
The first of my “constants” arises directly from that brief external history of fifty years of debate – often acrimonious debate – in the field of literary studies. Fascinated in the earlier stages of my career by this ever-changing intellectual landscape, I did my best to assimilate the various new approaches as they appeared over my horizon: they were after all the product of some remarkably brilliant critical thinking, and they were fired by the energy of a whole new post-war generation. I always believed that, at the very least, they had the merit of stretching the arguments of literary criticism to their utmost limit, often through paradoxical expressions such as the celebrated “death of the author” announced by Roland Barthes. They prevented us falling back on reassuringly comfortable habits disguised under the name of “common sense” or “real scholarship”. However, I also believe that there is such a thing as good scholarship, which must be based on a scrupulously empirical foundation, and which is as difficult as the most excruciatingly complex arguments of the theorists. No single theory or method is guaranteed to deliver the goods. Each has its insights, each may be appropriate to a particular critical or historical task.
Although structuralism now appears to be as dead as Barthes said the author was (biographical criticism is flourishing again in a new guise), it taught me things that remain to this day part of my outillage mental, my intellectual tool-kit. Indeed, what I have attempted to do as successive waves of literary theory washed like a tsunami over the field of literary studies is to read for the insight, to look for the decisive inflection of thought that drove each major contribution to that unrelenting series of reappraisals. The mistake, in my view, is to espouse a literary theory and assume that it can be applied systematically to a whole range of different texts and literary periods; in general, those applications rapidly become mere repetition, the theory reducing the text to a caricatural image of itself. The primary interest of literary study, I believe, lies in the fact that the modes of thought inherent in literature itself are infinitely diverse and flexible. Our task as academic specialists of literature is, delicately and painstakingly, to uncover those modes of thought, to make them accessible, rather than overlaying and obscuring them with a programmatic theory or logic of our own making.
So my first constant is “reading for the insight”, and it applies to the reading not only of critical theory and methodology, but also of literary texts themselves, and of whole configurations of texts: for every potential literary corpus, there will be a core problem or cluster of problems that one seeks to resolve. My second constant is a working method that remains uncertain of its outcome until the project or study is complete (or even later). When I started my doctoral work, my supervisor suggested that I should have a look at late sixteenth-century religious poetry. I spent a lot of time reading dozens of minor poets, but had no idea where it was all leading to until two things happened: I began to explore devotional treatises of that period in the hope that they might throw light on the poetry; and a friend who was working on English literature suggested I might look at Louis Martz’s classical study The Poetry of Meditation, which demonstrated systematic links between English poetry of that period and the methods of religious meditation that flourished in the Counter-Reformation. Suddenly the whole thing began to make sense; a shape emerged out of a random heap of materials. Ten years later, I was planning a book on the myth of Bacchus in the French Renaissance; at the same time, I was reading humanist treatises and handbooks on how to write, and especially on how to imitate ancient authors. Those two tracks were already connected in my mind, but in what one might call a weak way: the connection didn’t tell one anything new. Then I saw that, while Erasmus and other humanists were promoting the concept of copia or stylistic abundance, French vernacular writers seemed fascinated with the image of the cornucopia, which they used as a metaphor for several different domains of experience. That convergence or collision produced an explosion: my book became a study of copia and cornucopia, using a figurative device to link what would otherwise have been mere rhetorical theory to essential human values and problems. My next book Recognitions had a similar history, which I shall not recount here; I would simply say that I now understand it, twenty years after the event, as a contribution to a cognitive view of literature. I would not have put it in that form at the time: cognition has only emerged gradually as a central theme in my work, but that there is a connecting thread is now quite evident to me.
A last example to bring the record up to date. The book I am completing at present, on the afterlives of the character Mignon from Goethe’s novel Wilhelm Meisters Lehrjahre, began with a potentially enormous corpus comprising novels, poems, plays, songs and operas in German, French and English over some two hundred years. When I gave lectures on it, my audiences often looked puzzled or frustrated: they kept asking, “But where’s it all going? what’s your central argument?” I had a lot of local and partial arguments, but no all-embracing rational. I’ll come back later to how I have resolved this problem: for the moment, I’ll simply leave it hanging in the air.
One thing is clear: this way of doing research has some practical disadvantages. In 1984, I applied for a British Academy Readership which would have given me three years of research leave in order to write Recognitions. Unfortunately, at the time when I applied, my sense of what the book was about was still somewhat embryonic; the interviewing committee, who asked me the same kinds of questions as those lecture audiences, understandably decided to select another candidate, and I was obliged to write the book in the interstices of a busy teaching career. I have had other similar experiences since then: no institutional authority has ever liked my projects in advance.
As I have always told my doctoral students, the open-ended heuristic approach to research means living with uncertainty, which can be extremely uncomfortable, but in my personal experience it is far more productive. The Balzan Foundation has in fact proved that point, for me at least: the very studies that failed to achieve major grant support while they were being written are the ones now cited as the reasons why I have been awarded the Prize.
The third of my constants is what I would call the principle of connectivity. The most promising subjects for research are often ones that lie at the point of intersection between two apparently quite different domains. What I said a moment ago about the bringing together of copia and cornucopia is one example; another is the interplay, in Recognitions, between the history of poetics and the form that recognition scenes take in works of imaginative literature. Why are these intersections or fusions so productive in literary studies? Because literary works are not composed of linear rational discourse: they themselves are full of interconnectivity. They connect things that our minds connect in everyday experience, things that logic separates in order to achieve clarity. One of the consequences of this is that they have a value as historical evidence, a value that is particular to literature: they show how the different perceptions and modes of perception that were habitual at the time when they were written might have been connected in that world. My work on the French Renaissance in the later part of my career focused on the writings of Rabelais and Montaigne as virtual encyclopaedias of the ways in which it was possible to think and feel in that period: as large and complex cultural objects inscribed with the signs of their provenance. Of course adjustments have to be made for fiction, because it is not, and is not meant to be, an exact representation of the world. One needs to use control samples. But at the same time, it is important to avoid what I would call the contextual fallacy. Nothing can be understood without context, but if we assume that externally constructed contextual materials can tell us everything we need to know about the meaning of a work, we only get a mirror image of what is already known. Context in that sense is only the beginning. It has to be used to filter out false perspectives, illuminate obscure references, and so on, but in the end, the process should work the other way round: the literary work itself will tell us, in its unique way, how to perceive the context as someone in that period might have perceived it. To take a straightforward example, the doctor figures in Madame Bovary and Middlemarch do not merely confirm what we know about the status of provincial doctors and the state of everyday medical practices in those cultures; they provide us with a complex, holistic, joined-up picture of medical practice and its interface with the life of the community.
Clusters of texts, literary and other, are clearly much better instruments for reading the past than single works. Such clusters multiply the possible number of strands and of points of connection between them. They also serve as checks and balances for one another. We thus come here to a further constant which is in fact an extension of the last one: the power of the corpus. Literary history has always been concerned with aggregates of texts, either to designate what are usually called literary “movements” or to uncover the continuities and discontinuities of literary tradition (for example, the reception in modern times of the works of classical antiquity). My conception of a corpus, however, is rather different from that. In the first place, I would insist, once again, on the principle of connectivity, of finding points of intersection between different clusters of texts. If one constructs a uniform corpus that is meant to demonstrate a single theme, or represent a single category, one may well be excluding from the outset the very things that will prove to be interesting. For an example, I return here to my study of Mignon’s afterlives. Mignon reappears after Wilhelm Meister in the German Romantic novel, in Walter Scott’s romances, in the Victorian novel, in novels by Balzac and Zola and innumerable minor writers in France, in the opera Mignon by the French composer Ambroise Thomas, first performed in 1866, which became the smash hit of the later nineteenth century opera scene across Europe and beyond, in Frank Wedekind’s Lulu plays and Alban Berg’s opera, in a late novella by Gerhard Hauptmann, in Angela Carter’s Nights at the Circus, and in English, German and Italian films, not to mention the hundreds of settings by nineteenth- and twentieth-century composers of the songs she sings in the original novel. The Mignon corpus is thus not only large but also heterogeneous. It moves across several different genres, across three major European cultures (with offshoots elsewhere), and between different levels in the cultural hierarchy: Mignon appears in the work of canonic writers like Goethe and George Eliot, but also in popular stories, novels, dramas and musical dramas. In short, she is a crossover figure, a character who virtually personifies the impossibility of constraining literature within conveniently separate, rationally constructed categories. The question “What is the object of study here?” could be answered by saying, for example, that it is a certain kind of connection between narrative prose and lyric poetry, or between literature and music; that it is the shifting representation in the last two hundred years of female adolescence and its vulnerabilities; that it is the representation of the exploited street-child or child performer; and so on. All of those themes and issues are central to Mignon’s afterlives. But instead of choosing one of them and relegating the others to the background, I decided in the end that the primary object of study was the corpus as such, in all its heterogeneity, all its dendritic outgrowths.
Of course, I still had to structure the materials and present them to the reader in an assimilable form. And by “assimilable form”, I mean a 300-page book, not an encyclopaedia or a catalogue. The solution I have come up with in the last few months is to present the Mignon corpus as an exhibition, a designed collection of materials, in a series of four chapters that are the equivalent of different rooms in a museum. But I have also provided an extensive first chapter that sets out the methodological principles on which the exhibition is based, and a final chapter offering a synthetic overview and an answer, or a set of converging answers, to the question those lecture-audiences put to me five years or so ago.
This is not the place to go into the detail of what the answers look like. I shall simply pick out two central points. The first is that the corpus is a sample and an example of how cultural history (including literary history) itself operates. The Mignon story is a story told over and over again in different ways, like a myth – except that this myth was invented in modern times and is distinctively a myth of modern times. It is also a story set to music and experienced by thousands and thousands of people collectively. It generated a whole series of popular images in France and Germany, and those images themselves circulated throughout Europe, for example on a series of late nineteenth-century German postcards carrying photographs of Mignon lookalikes in sentimental poses. The story remains for an astonishingly long time in the collective memory: Angela Carter’s powerful late twentieth-century version shows that it is still capable of generating new imaginative forms in our own day. The book will thus be structured primarily as an investigation into the methodology of cultural history and of the workings of cultural memory. The second point brings me to a further constant. I said earlier that I now think of my book Recognitions as a contribution to a cognitive view of literature. The word “recognition” (and the Greek word anagnorisis on which it is based) carries in its etymology the theme of knowledge, and fictional works in which recognition scenes play an important role may be regarded as having in some sense a cognitive structure: they feature characters who are groping towards some truth that eludes them, often until it is too late. In such cases, knowledge, when it arrives, creates an explosion, a sudden transformation of all the known reference points, a drastic reinterpretation of the situation: one only has to think of Oedipus to get the point, although there are of course joyful and healing recognitions as well, for example in Shakespeare’s romances. Mignon’s story, too, is a recognition story, and I would argue that it provides nineteenth- and twentieth-century writers and their readers with a template that enables them to reimagine and rethink (for example) what it is like to be a vulnerable female on the threshold of sexual maturity, how difficult it is for her to speak about the traumas and the abuse that she has suffered, and what that state means in relation to the ethical assumptions of the particular culture shared by those writers and readers. Of course it is true that philosophers, psychologists and social commentators of the day wrote about these things in a rationally ordered and explicit way, but I would argue that the literary corpus represents a process of collective rethinking that exceeds any such reflection or commentary.
Let me offer you a metaphor for what I have in mind. There is a famous drawing known as the “duck-rabbit” picture which has provoked interesting reflections from, among others, the philosopher Ludwig Wittgenstein and the great art critic Ernst Gombrich. The drawing looks like the head of a rabbit from one angle, the head of a duck from another. It is impossible to see both together: one either has to see the duck or the rabbit. Yet the picture itself does not change. In literary study, it often happens that a single work produces diametrically opposite interpretations; there are also fierce debates between those who read literary texts as historical documents and those who prefer to treat them as aesthetic objects. It seems to me more helpful to regard the work itself as a single object which can hold in suspension as it were many different strands and perspectives, some of which may seem incompatible with one another when seen from a rational perspective. In addition to ducks and rabbits, they may be readable as foxes or swans or fish or even human beings. Our job as specialists of literature is, first, to keep that multiplicity in play in our readings, and then to understand it as a unique alternative way of thinking about the world.
I ask you to bear all that in mind as I come now to the last of my constants, which is a belief in the value of methodological enquiry. The word “method” is derived from the Greek hodos, a road or pathway, with the prefix meta connoting pursuit: the overall sense might be rendered by the phrase “the way forward”. In his Seventh Letter, Plato refers to philosophy itself as the “marvellous pathway” (hodon thaumastên), and Descartes chose the title A Discourse on the Method for the little treatise of 1637 that radically transformed philosophy, not so much in its content as in its working principles. I would like to insist here on the distinction between methodology and the kind of literary theory that I evoked in my rapid historical sketch at the outset. I have always been interested in literary theory, but I could never have been a theorist: abstract thought eludes my grasp. I like working with the untidy, unpredictable but endlessly fertile materials we refer to loosely as literature. On the other hand, I have always regarded it as essential to be aware of one’s own working methods and to make them explicit as part of any literary study. For over a quarter of a century in Oxford, I ran methodological seminars for doctoral students; since my retirement, I have conducted short-term seminars and workshops at the universities of London and Oslo on interdisciplinary methodology. What I like most of all is to get people who are working on quite different things to sit round a table and enter into a dialogue with one another, using samples of their own work as a point of departure (just as we are doing today, thanks to the Balzan Foundation).
The basic rule is that they should explain their work to the non-specialists in the audience precisely by focusing on their working method, on questions such as how they constructed their corpus or archive, the relation between empirical and theoretical perspectives, the logic and rhetoric that are particular to their discipline, the way in which their work overlaps or intersects with other disciplines, and where the limits of that interdisciplinary relation lie. I have found that in nearly all cases, this approach results in a lively and productive debate, so I have decided that it should represent a central thread in my proposed project for the Balzan Foundation. I shall use the resources the project puts at my disposal for discussion groups and methodological workshops rather than for large formal conferences. The dialogue I want to set up will be open-ended, heuristic, Socratic in character, although of course I expect it also to result in excellent specialist publications by the individuals involved. And the subject I have chosen for the project is “Literature as an object of knowledge”, both in its historical aspect and in its contemporary interdisciplinary embodiment: what are we, who profess to study literature, actually studying, and how can we justify that study in relation to our sister disciplines? What (to put it in another way) is the cognitive value of literature and literary study? That question has remained with me throughout my career, and it is the one I want to leave with you here today.