PART I
Theorizing the Archive
Chapter 1
The Archaeology of the Manuscript:
Towards Modern Palaeography
Wim Van Mierlo
As students of poetry, we have conceptions of what a poem is and we have notions of what poetic creation is, but do we actually know how a poem comes into being? There certainly is a poetic tradition going back to the Romantics that sees creation as something mysterious, elevated, ungraspable, issuing almost out of nothing. Shelleyâs likening of inspiration to a âfading coalâ is probably the most famous expression of this. â[T]he mind in creationâ, he said, is fuelled to âtransitory brightnessâ by an âinconstant windâ, the moment of inspiration powering the brain to perceive a sublime glimpse of the uncreated poem. But inspiration is a fickle, uncontrollable power. To capture that potent but momentary glimpse in language is like chasing the wind. And so âwhen composition begins, inspiration is already on the declineâ (Shelley 1977: 503â4). Composition, for Shelley, is inspirationâs poorer cousin.
During composition, a poet is (I imagine) not always fully conscious of the mental, creative and cognitive processes that allow writing to happen. Yet he cannot be fully unconscious of these processes either. Shelleyâs idealized notion of inspiration may strike one as rather remote from the actual experience of writing. A modern poet who refined Shelleyâs theory is Robert Graves, who divided the creation of poetry into two distinct phases. In the first phase, the poem flickers into being almost subconsciously, as it is ârhythmically formed in the poetâs mind, during a trance-like suspension of his normal habits of thought, by the supra-logical reconciliation of conflicting emotional ideasâ. The second phase begins when the poet has emotionally âdissociated himself from the poemâ and begins âtesting and correctingâ it âon common-sense principlesâ, transforming the poem-in-the-rough into something that will âsatisfy public scrutinyâ, all the while making sure that ânothing of poetic value is lost or impairedâ (Graves 1995: 3â4). Graves thus significantly broadens Shelleyâs understanding of inspiration and spreads it across different levels of emotion and experience. What is significant, however, is that inspiration no longer excludes pen and paper, for, even in the first stage, invention manifests itself directly in writing.
Literary archives allow us to study that writing not only in its finished, but also in its inchoate, embryonic state. The work in progress, contained in drafts and manuscripts, offers fruitful insights in the physical processes that underpin its construction. Critique gĂ©nĂ©tique, a theory and practice devoted to studying drafts, analyses these processes. It emphasizes not the afterlife of the work, when the finished text is released to the public, but what comes before: the avant-texte, the text before it is âthe textâ. As a rich, dynamic network of emergent textual components whose development we can study, the avant-texte encompasses all the stages of literary creation (invention, conceptualization, planning and organization, drafting and revision) and the different modes of writing (note-taking, sketching, drafting, revising and correcting) that take place on a variety documents (notebooks, loose leaves, typescripts, page proofs).
As the manuscripts I will discuss in this essay show, the work of the poet involves a struggle not only with language, but also with paper and ink. Studying the growth of a poem through manuscripts may seem to demystify romantic notions of inspiration. But it cannot completely pass over these notions either. Revision is not only about mechanistic change, or about selecting the right word or expression; it is also about invention.1 Inspiration does not simply precede, but also happens during, composition. Manuscripts are, as Daniel Ferrer observes, the âdĂ©pĂŽts sĂ©dimentairesâ of invention (2011: 53).2
Even so, I shall not contend that by looking into the poetâs workshop one gains privileged access to the poetâs mind. The archive does not offer a way of reclaiming a process that remains unfathomable perhaps even to the poet himself. Yet if we accept that manuscripts can say something about creativity, we need to learn how to read their signs. Louis Hayâs classic adage â manuscripts have something to tell; it is time we made them speak â is still pertinent (Hay 1996: 207). The question that deserves our attention, therefore, is how do we distil the poetâs vision from his revisions? What the archaeology of the manuscript must investigate is the meaning behind the cancellations, insertions, substitutions and overwritings that are layered across the page.
While coming to terms with the creative origins of poetry (using as case studies manuscripts by Wordsworth, Keats and Wilfred Owen), my purpose with this archaeology is to expand the analysis of literary drafts to incorporate a more detailed palaeographical investigation. One of the main goals of critique gĂ©nĂ©tique is to disentangle the temporal aspects of writing from the âundifferentiatedâ manuscripts in the archive and, via a number of preparatory operations such as ordering, classifying and deciphering, distil from them the avant-texte (de Biasi 2004: 38).3 This process, however, depreciates the manuscriptâs spatial attributes, its look and appearance. Once the avant-texte is established, and the writing is âlifted offâ the page, the physical dimensions of the document are reduced to a one-dimensional text. Some recent work has moved from the mere deciphering of the words to analysing the graphic signs that are indicative of the processes that produced them.4 Even more so than the actual words and revisions, this palaeographical evidence provides information about the dynamics of composition. The way in which the hand moves across the page and the variations this produces (quick or slow, straight or slanted, spontaneous or contrived) is indicative of the creative intensity that drives the writing. The flow of the writing, the vigour of the pen, the boldness of the cancellations, the positioning of the writing on the page all inform us about not only the circumstances in which the writing took place, but also the characteristic habits (or usus scribendi) of the individual writer (Ferrer 1998: 256). As well as elucidating how the manuscript functions in its own right, this palaeographical information also highlights broader contexts of the biography of the text and the scribal culture that is at work at the time. While critique gĂ©nĂ©tique sees manuscripts as largely private and wholly idiosyncratic productions, writing nonetheless is a cultural phenomenon in which specific practices are shared at certain periods in time (Ferrer 1998: 259â60, 265).
What I mean by the archaeology of the manuscript is not something purely metaphorical; it offers a relevant conceptual and methodological perspective on what is essentially a book-historical matter. The challenges that archaeologists face when they interpret the past are similar to those encountered in the palaeographical analysis of modern manuscripts. The fragmentary writing found in drafts, notebooks, fair copies, typescripts and page proofs is similar to the shards of pottery or other remains of human activity uncovered in an archaeological dig. They are our only means of reclaiming the processes of creation from the past. Archaeology and manuscript studies essentially share the same basic hermeneutic problem: how does one begin to understand the evidence without already knowing what it means? Thus, archaeology as a discipline cannot function without the support of other specialities like history, art history, anthropology, geography, geophysics, archaeobotany and so on. Likewise, the study of literary drafts is not possible without the context of literary history, biography, genre studies, poetics, literary criticism, publishing history, reception history, paper history and the like. More to the point, archaeological research and manuscript studies involve a âmultistranded and multiscalar processâ â that is, it is not enough to create just one interpretation, but to make âsense of the data at different spatial and temporal scalesâ, particularly as the data is complexly diverse, illusive and frequently conflicting (Hodder 1999: 43, 78, 99). What defines archaeological research are the âdynamic, dialectical, unstable relations between objects, contexts and interpretationsâ (Hodder 1999: 84); with manuscripts, too, we need to recognize the variable relations between the physical documents, the contexts of their production and the interpretations of the textual and graphic signs. Only by connecting and comparing the evidence can we begin to make sense of what is in front of us.5
Beginning The Prelude
I want to begin elucidating these ideas by looking at the beginnings of William Wordsworthâs The Prelude in MS JJ (Figure 1.1), which is itself a part of Dove Cottage MS 19 (one of four notebooks that William and his sister Dorothy purchased at 1s each in preparation for their extended stay in Germany in 1798), and which contains the now-famous opening lines: âwas it for this / That one, the fairest of all rivers loved / To blend his murmurs with my nurseâs songâ (f.Zr) (1977: 115). These lines, however, pose a particular problem. They occur towards the back of the notebook from which the writing seems to proceed backwards, from the recto on to the corresponding verso and continuing from there on to the preceding pages. Most scholars consider âwas it for thisâ to be the âoriginâ of Wordsworthâs great poem on the growth of the poetâs mind, even though, when he wrote these lines in late autumn 1789, it cannot have been clear, even to him, what would come from them 50 years hence.
But are they really the beginning? The evidence is not entirely conclusive. MS DC MS19 already contained many other bits of writing composed after Wordsworth had arrived in Goslar. To start a new poem, he needed not only unused space, but also a place that he could easily return to without having to leaf through the whole notebook. The final 11 leaves of the notebook offered that space. Apart from that, the precise sequence remains elusive. Scholars have established that the order of inscription generally moves inwards away from the back in five different stages; within these stages the direction of writing moves mostly forwards, âin a zigzag fashionâ.6 Sometimes, Wordsworth turned the notebook sideways because the width of the page better accommodated the length of his lines; sometimes, too, he wrote down short sections out of sequence with the rest (1977: 3).7
Why did Wordsworth make work so difficult for himself? Why didnât he turn his notebook upside down and proceed forwards? This would have made theprocess simpler and more intuitive â moreover, he had done it elsewhere already.8 Could it not be that there is no order at all?
Figure 1.1 William Wordsworth, The Prelude, MS JJ, ff. Yv-Zr [ff. 90v-91r in DC MS 19]
While emphasis in recent scholarship has certainly shifted away from a âretrospective perspectiveâ on the âtwo-partâ Prelude towards a view that the poem grew from âa loosely connected sequence of fragmentary blocks of writingâ, the assumption remains that âwas it for thisâ was the beginning.9 The reasoning for this, however, was derived not from evidence in MS JJ, but from the order of the poem in two later fair copies: MS V, a copy that Dorothy produced with the assistance of her brother, and MS U, a duplicate copy by Mary Hutchinson, both of which open with âWas it for thisâ.10 The order of poem in these manuscripts is used to âreconstructâ the five-stage sequence in the earlier draft. The sequence in the fair copies, however, was the result of a further creative act by Wordsworth that intervened between MS JJ and the later manuscripts.
Linearity and Revision
Writing is inevitably a linear activity; it purposely moves forward towards the completion of the text. But the operative words âforwardâ, âpurposeâ and completion are not as self-evident as they appear. Just as origins are difficult to recover, the end the writer strives towards does not predetermine the course by which that end is reached. The composition process, and the trajectory it follows, is marked as much by deviation and indeterminacy as by straight, progressive development.
To take these ideas forward â of beginnings, linearity and process, and purpose â I want to look at the âCertainty Suspendedâ project of the Manchester artist Anne Charnock, a series of âwork artâ canvases inspired by the âtrack changesâ function in Microsoft Word (Figure 1.2). These works visualize the processes of revision that operate in artistic creation, showing writing as something that is layered and tentative, rather than settled and definite. For Charnock, the intertwining of contradictory sentences gives âsolidity to her own meandering thoughts and uncertaintiesâ (Charnock and Clark 2006: n.p.). The example I have chosen, âUncertainty Series No. 1â, specifically presents the idea of prevarication that underlies the process of revision. Between the first statement â âI always have a clear idea that drives my practiceâ â and the revised statement â âI donât often believe my end-result is driven by my original ideaâ â lies a movement that captures the transition from confidence to equivocalness through seemingly unpremeditated discovery. The truth about creation as a process is expressed as much by the revised statement as by the act of revision. And thus, perhaps paradoxically, the original statement, though placed sous rature, remains legible and does not entirely lose its value. This is the double existence of the cancellation, its âloss and gainâ, âemptiness and fullnessâ, âforgetfulness and recollectionâ (GrĂ©sillon 2008: 88).11 While the original statement has been negated, it has not lost meaning. Hence it remains (or becomes) possible to read the revised statement in relation to the original from whence it came. It is that connection between the two â that present absence â that gives âsolidityâ to the uncertainty.
Figure 1.2 Anne Charnock, âUncertainty Series No. 1â
Charnock thus consciously emulates what for any writer is an ordinary experience. Her aesthetics of revision evokes schematically what happens more intricately in a draft. Take, for example, the revisions that Keats made in the autograph manuscript of âOde to a Nightingaleâ (Figure 1.3; see also Hebron 2009: 138). The changes that Keats introduced are minor, but they add subtle depth to the poem, such as, for example, the change in the seventh stanza from âthe wide windows opening on the foamâ to âthe magic casementsâ. The windows, after all, are not real but imagined, recollected from an auditory vision of the nightingaleâs song in âfairy lands forlornâ (Keats 1978: 371). The change is clearly a reasoned change, but not all revisions are like this. Keatsâs rapid, almost desultory change of âkeellessâ into âperilousâ in the final line of the penultimate stanza is undoubtedly of a different order.12 This currente calamo revision happened almost without interruption, for âkeellessâ seems hardly a fitting adjective for the sea.13 âPerilousâ is very much le mot juste.
Figure 1.3 Detail of John Keats, autograph manuscript, âOde to the Nightingaleâ, Fitzwilliam Museum, MS 1-1933
What is interesting in Keatsâs dynamic of revision is not the choice of words, but the process of choosing, and the fact that we can practically observe Keatsâs mind tripping up. Apart from revealing the history of the textâs composition, manuscripts also tell us a good deal about method and craft, about aims, intentions and poetics, about the history and sociology of the text. Keats changing âkeellessâ to âperilo...