10 Sep 2010

Research example: The plasticity of the manuscript; the flattened text

By Wim Van Mierlo, University of London

This is a rejoinder (or conjoinder, rather) to Carrie Smith’s fine post on the about Spelling and strikethrough, typing manuscripts and silent editing. The kind of “silent editing” that she describes in Christopher Reid’s edition of The Letters of Ted Hughes is all too a common a practice. To smooth over minor slips of the pen seems for most a sensible thing to do; a minor intervention to help readability can only be a minor offence. Yet it begs the question, as Smith rightly asks, how much of the editor is there in the text. What annoys me most about it, I guess, is that it is always done in the name of the reader: it is the reader (editors or publishers claim) that do not want such troubling detail. These are usually also the kind of editors that find footnotes or annotations disturbing.

I don’t find slips of the pen disturbing. The Oxford Collected Letters of W.B. Yeats (under the general editorship of John Kelly) is in a fine example of an edition that retains detail of the hand that wrote the letters and thus gives a sense of the man behind the letters.

There are other features, too, that don’t find their way into printed editions. Among Ted Hughes’s letters to Leonard Baskin in the British Library (some of which Reid included in his volume) are written on “aerogram” paper, a type of blue, lightweight, gummed paper that could be folded to form an envelope which was used for airmail. Inevitably the “letter” sometimes overran the space on the paper, and Hughes would squeeze in a few extra lines, an ending, a greeting or a PS, to the sides of the page. None of this is ever found in a print edition of letters. Nor are any of the other features that one customarily finds in autograph letters—the space left between date, header, address; the inward-tapering left margin, and so on.

So there is a larger issue—an issue that has to do with the differences between manuscript and print. Print flattens text. A manuscript has three dimensions; the printed page has only two. The manuscript page, like the printed page, has width and length; the text moves—generally—from the top left corner to the bottom right. Generally, I say, because the writing on the manuscript can move in any direction in a way that print (barring some exceptional cases) does not. Print is limited by the sequentiality of the text; words appear in a strictly linear order. The text on the manuscript, by contrast, can move in any direction; the process of writing creates an order that is relational. Furthermore, as the text in the manuscript is not subjected to a strictly linear form, the temporal element of the manuscript is highlighted: the time of inscription becomes more apparent through the apparent simultaneity of textual elements—through co-existence of different moments of inscription—within one space.

The printed page of course has its own spatial possibilities; from the use of marginal glosses and footnotes in learned books to the dancing words of Apollinaire’s Calligrammes, examples of the versatility of mise-en-page abound. But it is simply that the manuscript naturally lends itself to an exploitation of all the dimensions of the page. Every manuscript is, in a sense, a palimpsest, even when the words are not actually written on top of one another. This is the plasticity of the manuscript.

The printed page, moreover, is not well suited to capture the elasticity of hand-writing: words written hastily, words written with great care; words crammed in a corner; words written on top of other words; words that are malformed. Print cannot quite adequately handle these. And yet they are what makes a manuscript a manuscript. They tell you something about the moment of inscription, the psychology of the writer, the circumstances in which the words were written.

An example of this can be found in an annotated copy of W.B. Yeats Poems (1901), now at Senate House Library, which contains some very sloppy pencil marginalia in an almost childish hand among otherwise reasonably clear and legible notes. The conclusion that the annotator, the poet and artist T. Sturge Moore, had taken his book with him and was writing on an unstable surface immediately presented itself. Or perhaps the surface was stable, but the writing hand was being rocked about by motion? It seems likely that Sturge Moore was reading his book on the train.

Besides handwriting, print can still at best only approximate the idiosyncratic features of a manuscript. With proper page-making skills, and a powerful DTP software package, one can go a long way towards capturing the irregular elements of a manuscript. It wouldn’t be totally impossible, for example, to produce a diplomatic transcription of one of Swinburne’s drafts, with his beautiful drooping, ever-expanding, never-ending lines. But the more detail is added, the less clear the printed text actually becomes. With a manuscript like this, the fact that one has practically no leading—the spacing between the lines—on sections of verse that curve downwards one above the other, could make the printed text actually less legible than the manuscript.

So print is less flexible. But approaching the point differently, one can accept that this is precisely the function of print. The printing press may have been invented to increase the rate at which texts can be duplicated and disseminated, but one of its effects was that it has standardized text. A book produced in the sixteenth century does not look essentially different from a book published in the twenty-first century. The coming and going of blackletter printing in Germany is certainly the best indication of how this standardization works. Even the world wide web still in many respects emulates the printed page, despite claims about a digital revolution. Print flattens text—indeed. But it is supposed to do so. And this has important implications for the way we perceive manuscripts.

Just a few months into the writing of his new book, Finnegans Wake, James Joyce wrote to Harriet Shaw Weaver, on October 1923, sending him some new drafts: “Today I send you the rough sheets with a plan of the verse and a forgotten page of H.C.E. But please don't read them yet—in fact, they are illegible”. He was not exaggerating. Joyce’s much-reworked early drafts, though sometimes written in a large, clear hand, are challenging for anyone to read, because of their inchoate nature and elaborate overlay. Soon after Joyce would send Weaver a fair copy or typescript which would be much easier to read. But illegibility seems to be part and partial of his practice. About a month earlier he had written to her about another piece he had completed: “I had it typed at once in order to read it”. This time for his own benefit. There’s a point, in other words, at which the manuscript reaches saturation. The page, literally, has its limits; as the striations become so intricate that the page is bursting at the seams, the need to have the text flattened arises. The writer, after all, must also be a reader. Writing cannot forever dwell in the realm of the possible; if composition wants to move forward, the text needs to be fixed at some point, even if for a brief moment.

To return to the issue of silent editing—what position, then, should the editor or transcriber take? When editing documents, one should not interfere with the text of the document. That should be the first principle, and so I personally wouldn’t regularize occasional, but obvious errors and slips of the pen. But as an editor one has a duty towards the printed text as well. In their edition of Finnegans Wake (The Houyhnhnm Press, 2010), Danis Rose and John O’Hanlon have rendered the opening line thus: “riverrun, past Eve and Adam’s, from swerve of shore to bend of bay, brings us by a commodious vicus of recirculation back to Howth Castle & Environs”. In the phrase “to Howth Castle and Environs” from the 1939 text, they emended “and” to “&”. They did so on the grounds that that was what Joyce first wrote in his manuscript (in a draft from 1926), and that (they argue) Joyce preferred the ampersand over the “distancing effect of the word ‘and’ in this context”. Editorially, critically, this is not a wholly straightforward decision, however. The “&” was indeed changed to “and” without Joyce’s express authorization on the galley proofs for the 1939 Faber edition— though remarkably after having been left to stand by several typists and typesetters as the text was transmitted in typescript and page proof through several iterations, including serialization in transition magazine. Now the ampersand is a ordinary feature of a manuscript—a symbol used for brevity, economy—that would in normal cases be normalized in print. The fact that it did not happen for so long has created an editorial problem.

Joyce’s typists and typesetters have, in other words, inadvertently acted as good documentary editors, until someone finally did erase the manuscript feature. This example shows just how plentiful the issues are when confronted by the idiosyncrasies of the minutiae in a manuscript. Editors (including editors like Reid) are not always conscious of these features—and the “manuscript culture” they encounter—in the manuscripts in front of them. Documentary editors, by contrast, are faced with a plethora of issues which they have to solve practically and with consistency. What, for example, if the slip of the pen is not really a slip? What when it poses a problem not unlike the ampersand in Joyce? What, in other words, if the misformed word is neither an abbreviation, nor a slip, but something not fully formed like this:

This detail from a manuscript by Lady Augusta Gregory for The Unicorn from the Stars (1908), a play she wrote in collaboration with W.B. Yeats, has a phrase which reads “that work in the business”. The word “the”, however, does not look like the; it rather looks like “te”. Yet to transcribe it that way would be pedantic, if not nonsensical. “Te” is not a standard abbreviation, the way it was customary up until the eighteenth century to shorten “the” to “ye” or “which” to “wh” or “wch”. The use of these abbreviations had something to do with the physical processes of writing by hand and using a quill and ink. (The disappearance of many of these abbreviations can, I guess, be attributed to the introduction of the steel pen nib.) Similarly, Lady Gregory “te” is a kind of elision effected for ease of writing and for speed, so that the pace of the hand can better keep up with the flow of the brain.

The straight slip of the pen is of course similar. What causes the writer to write one word for another is a momentary cognitive lapse; the intended word takes a different shape as the hand is forced to jump, like a faulty record or CD to keep up, with the flow of inspiration. The difference between the slip of the pen and “te” is that the slip is not a mistake that occurs consistently.

The case might therefore be made in favour for diplomatic transcription of “te”, but then the problem really has no end. What is it that the transcriber actually sees: is it “te”? Or is it “the”, with its letters rung together? Or even “th”? Take this example from the same folio, which reads: “but the weight of the business falls on Henry & myself – I wouldn’t”

The words “falls”, “myself” and “wouldn’t” show the same signs of being swiftly formed as “the”, but they are clearly no abbreviations or contractions, yet one needs some imagination (and a practiced eye) to decipher the individual letters of the double “ll” in “falls” and “my” in “myself”; even the standard contraction in “wouldn’t” is interesting because it has the apostrophe, but no separation of the letters “n” and “t”. Looking again at the first example, one can ponder whether “that” is that or this, two words that in Lady Gregory’s hands are not easy to distinguish. In this case it is “that” for the word fits syntactically with the rest of sentence, but the untrained would first take this word as “this”.

Such detail is intriguing and cumbersome at the same time. Pragmatically one sometimes has to flatten the text and opt for what is intended, not for what one sees, or thinks one sees, on the page. Ultimately, the plasticity of the manuscript is not something that can rendered in a perfect way, but as long the business of editors is to edit, they need to make decisions. However, what has to be kept in mind is that the text that results from the editorial process—whether through silent editing or diplomatic transcription—is a representation, not a reclamation of the manuscript.


  1. This comment has been removed by the author.

  2. I find this an extremely interesting concept - viewing the manuscripts as texts not flattened by print. It makes me think of manuscript drafts in a completely different way.

    In reference to the "would'nt" abbreviation in Lady Augusta Gregory's manuscripts, I coincidentally picked up on the similar way of abbreviating "not" in Emily Dickinson's manuscript dated "about August 1881" and reproduced in print as the poem 1559 in R.W. Franklin's The Poems of Emily Dickinson Variourum Edition Vol. III. The poem contains the following lines:

    "The Butterfly upon the Sky
    That does'nt know it's Name
    And has'nt any Tax to pay
    And has'nt any Home"

    It made me wonder whether this kind of abbreviation was a common variation within the standard in the 19th century.