Challenges in Editing Modern Literary Correspondence: Transcription

1. Foreward
2. Subject Matter
3. Classifications
4. Sources of Information
5. Audience
6. Sensible and Sacred
7. Politics of Transcription
8. Spelling
9. Punctuation
10. Eccentricities
11. Print Ain’t Handwriting (duh)
12. A Model Edition?
13. Follow or Invent Rules?
14. Different Source Materials?
15. The Value of Junk?
16. Different Media?
17. Transcription Is Translation
18. Against Pedantry
19. Transcription Is Characterization
20. My Modest Proposal
21. Conclusion

Foreward

The first draft of this lecture appeared in TEXT (Transactions of the Society for Textual Scholarship), ed. D.C. Greetham and W. Speed Hill (New York: AMS Press, 1984), pages 257-270.

Back then, the letters of a literary legacy existed in three static forms:

1. Holograph — handwritten or typed sheets of paper
2. Facsimile — photoduplication on paper or film
3. Print — transcription on typeset pages

An intangible digital form has since emerged and is poised to supersede the others — because of its utility (which is mind-boggling). That’s why I chose it for kickstarting George Moore’s literary legacy. This revised draft of my 1984 lecture takes the digital form into account.

At around 6,500 words, the lecture is longer than what I would write, were I starting over from scratch. I haven’t abridged because it’s all as pertinent now as it was years ago, though admittedly a few redundancies could drop to the cutting room floor. But not today.

Quotations inserted without footnotes are from personal correspondence before there was email. I keep those documents in hanging file folders — an old-fashioned alternative to scanning and OCR. Can you imagine a world where people wrote letters to share their ideas? That’s the lost world of a literary legacy.

Subject Matter

Since I975 when I began preparing a “collected” edition of the letters of George Moore, it has seemed that few if any rules govern the editorial treatment of literary correspondence. By “literary correspondence” I mean the day-to-day letters of major authors which were not written for publication, but still became part of a legacy.

In the absence of rules, we editors are like a party of weekend hunters, stalking and shooting our prey according to whim and habit. It may be assumed that diverse editorial techniques yield a more interesting if unpredictable bag.

That said, I’m here to argue for better control over our sport, for I believe that agreement on common standards and methods will make editors more efficient for their publishers, more helpful to their readers, and more worthy of their authors. Moreover tighter control will optimize legacies for our new digital age in which common standards and methods matter more than in the cottage-like industry of traditional publishing.

I have limited the present inquiry to the task of transcription: the discipline of transforming holograph into letterpress. At a future date my range may broaden to annotation of content and composition of prefatory matter. Decisions about transcription affect those tasks and ought to precede them. 

Some editors, including Leon Edel and Nigel Nicolson, recommend that we discover a more rational purpose than we have now for the publication of innumerable private letters. Yet most scholars agree that all the writing of an important author — even ephemera — is the stuff of legacy. The question is not if, but how the letters should be published.

Classifications

My inquiry is also limited to larger or comprehensive editions of letters. In that regard, I want to settle my definitions of three bandied-about words. In order to help readers identify the kind of work they have opened, one of these monikers ought to appear on the title page:

• Complete Letters — presenting all the letters known to survive
• Collected Letters— presenting just a representative sample
• Selected Letters — presenting a subset for a special purpose

Thus the Sanders and Fielding edition of Thomas Carlyle’s letters is Complete, though the editors called it Collected. Joseph Blotner’s edition of William Faulkner’s letters is probably Collected, though he called it Selected. Henry Nash Smith’s correspondence of Mark Twain and William Dean Howells is Selected, since it is a subset of those writers’ correspondence. 

By the same token, my edition of George Moore’s letters is indubitably Complete, though I called it Collected in 1980. Who knew? I think we can all agree that making title pages consistently characterize what follows is fairer to readers. We should drop the waywardness and get in line.

Sources of Information

My reflections on editorial practice in this lecture are based on three separate sources of information. 

The first source is my survey of nineteenth- and twentieth-century letters. I examined forty-seven editions published after 1900. The start date reflects my agreement with Bradford Booth, editor of Anthony Trollope’s letters, that “uncritical and dishonest tampering with texts on the part of nineteenth-century editors has made it necessary to repeat much of their work and to reevaluate the character of many prominent men of letters.” ¹

The second source is my personal correspondence and conversations in 1980-1981 with thirty-two editors of literary letters. I sought their guidance on how to edit George Moore’s letters for publication (a project that is only now coming to fruition in the digital age). I haven’t contacted more editors for this revised lecture, though I did broach the subject with one who didn’t engage.

The third source of information is my own experience of editing George Moore’s letters. The first thousand (of six) landed in my 1980 PhD dissertation at the University of Reading. Moore’s holographs seemed to beg for editorial sympathy — rather assistance — but I was wary of tampering with his handwriting and lessening the fidelity of my transcriptions.

Audience

Finally bearing on my inquiry is the question of readership. It may now be appropriate for me to contradict Walter Harding, editor of Henry David Thoreau’s letters, who warned that “any editor of manuscript materials is immediately faced with the problem whether he is to aim his text at the general reader or the scholar.”² Dan Laurence, editor of George Bernard Shaw’s letters, writes, “there no longer is a general reader. With rare exceptions, editions of letters just don’t sell.”

Thus according to Dan and me in the 1980s, the rules of editing letters should be devised by scholars for scholars, for those were the only readers. I no longer feel this is true. Editions of letters, like most scholarly publications, don’t sell for two addressable reasons.

First they don’t sell because they’re extravagantly expensive, as though the books are branded luxury merchandise. Editions are evidently priced for libraries rather than readers (analogous to the way healthcare is priced for corporations rather than patients). Individuals can’t afford to buy editions of letters or for that matter any scholarly publications, even if they want to. Why is that? It makes no sense to me.

I contend there are plenty of general readers for editions of literary letters, but the price must be right; and the right price for this type of content is free.

The second reason scholarly publications don’t sell is because they’re cluttered with technical paraphernalia — what George Moore called “the imposture of the expert” — that interferes with literary texts in a vain attempt to “interpret” them. Thus readers are faced not just with letters as an art form, but also with “protective” exegesis laid on, like linen wrapping a mummy. Ordinary readers — and even many specialized readers — don’t need or want that wrapping, but can’t simply shake it off. It makes any scholarly publication not just way overpriced, but extremely tedious.

I contend there are plenty of general readers for editions of literary letters, but the text must be right; and the right text for this type of content is uncluttered.

Sensible and Sacred

For editors of literary correspondence, holographs have the highest rank in the hierarchy of source materials. We search for them tirelessly, like knights in quest of the Holy Grail. We study them intensively. Our ambition to explain and disseminate them in printed form to colleagues and students — though somehow ordinary readers are lost in the mix. By ordinary readers I mean people who are neither paid nor obligated to read books; they do it for pleasure.

My expanded ambition is to present the letters of George Moore to ordinary readers, not only to professors and students; moreover to make the letters freely available online, in active texts that behave intelligently on devices, rather than in volumes that gather dust in stacks or languish in file directories that don’t behave nowise.

It’s useful to think of holograph as makeshift divinity. Homage has been practiced in two distinct ways: I call them The Sensible Text and The Sacred Text. Each uses an approach to transcription that barely overlaps with the other. This was true in 1984; I’m not sure about today because one camp may have foolishly declared Mission Accomplished while I was working on other things. I hope note.

Editors of The Sensible Text optimize transcription for readers. Like their authors (letter-writers), they present letters as narrative art, the purpose of which is to tell stories: what’s happening, when and where, to whom, and wherefore. They want literary letters to inform and edify and please readers the ways, more or less, they informed and edified and pleased recipients. To accomplish this, they normalize text so that it’s easy to read and understand — just as intended by the letter writer. 

Thus editors of The Sensible Text correct messy spelling and punctuation that is easily ignored in handwriting, but distracting or confusing in print. They wrap text, alter line and paragraph breaks, use the correct cases of letterforms — all for the purpose of making a letter easier to read in print than it is in holograph.

In contrast to that, editors of The Sacred Text replicate holographs in the medium of print; they would presumably choose to replicate holographs in the digital medium as well. They preserve both the literal meaning of the letter-writer’s words and a more subtle meaning suggested by the way the words are written. 

That subtle meaning may be contained in errors of spelling and punctuation and in myriad eccentricities including capitalization, paragraphing, marks of emphasis, cross-outs, repetitions, and abbreviations. More intricate holographic features, including elisions, fusions, and expansions, were carefully identified by Merrell R. Davis and William H. Gilman, editors of Herman Melville’s letters.

Politics of Transcription

Some editors of The Sacred Text believed that their methods of transcription predominated in textual scholarship of the twentieth century. Ralph Aderman, Herbert Kleinfield, and Jenifer Banks thus assert that their diplomatic transcription of Washington Irving’s letters conformed entirely with “modern textual principles.” Though Irving himself said that his errors should be corrected before his letters were published, they were not.

My survey of modern editions indicates that Aderman, Kleinfield, and Banks were somewhat arbitrary in their identification of textual principles. Only twenty-seven percent of the editors claimed to make diplomatic transcriptions.

This figure would probably be lowered after closer examination than I had time to make. For example, Bradford Booth wrote, “I have in no instance altered the text” of Trollope’s letters, but he did alter punctuation. Gordon Ray, a model editor of The Sacred Text, inserted paragraph breaks into William Makepeace Thackeray’s letters. 

Spelling

Where spelling is concerned, only forty-three percent of the editors neither normalized text silently nor corrected it in brackets. “Modern textual principles” do not require the faithful transcription of every spelling error. When Allan Wade corrected William Butler Yeats’s spelling, he explained that not to correct it would “in the long run appear merely tediously pedantic.”³ David Krause, editor of Sean O’Casey’s letters, believed that were Wade to have performed diplomatic transcription of Yeats, “the result would be chaos.” 

Wade, Krause, and Sir Rupert Hart-Davis all expressed the opinion that Yeats requires and deserves help with his spelling. John Kelly, editor of a newer Complete letters of Yeats, writes: “In general, I am all in favour of tidying up texts as long as this does not interfere with the primary meaning. I think that something written in manuscript looks quite different when in print, not merely in the obvious sense, but that it gives out different signs and involves different expectations of accuracy etc.” 

Rather than having observed “modern textual principles,” Dr. Kelly “shilly-shallied over whether to correct Yeats’s texts and in the end decided, despite what I have just said, not to.” Wade and Kelly decided the same question differently, individually, regarding the particular problems created by Yeats’s handwriting. Say what you will, this is not a systems approach.

Punctuation

Regarding punctuation, sixty-one percent of surveyed editors altered or supplied it for the sake of clarity. Among those who didn’t, Leslie Marchand is perhaps the most articulate defender of faulty punctuation. He explained that Byron’s errors of punctuation were caused by ignorance rather than intent, but to alter or supply it “detracts from the impression of Byronic spontaneity and the onrush of ideas in his letters, without a compensating gain in clarity.”⁴ 

I read this to mean that there would be a gain in clarity, but that it would not compensate for other losses. Marchand continued: “In fact, it may often impose a meaning or emphasis not intended by the writer. I feel that there is less danger of distortion if the reader may see exactly how he punctuated and then to determine whether a phrase between commas or dashes belongs to one sentence or another.” 

It is reasonable to wonder, as Sir Rupert Hart-Davis did, that since Byron intended a specific meaning when he wrote letters (despite his ignorance of punctuation), is it not better that the highest scholarly authority on Byron identify that meaning, or should it be left to the imagination of casual readers and less knowing scholars? 

Moreover, ”Any half-wit can copy,” writes Leon Edel. The great editor, such as Professor Marchand, can and should do more. Marchand himself recommended Gordon Haight’s George Eliot Letters as a model for future editors. “Punctuation I treated rather more freely,” Haight wrote to me, “regarding the reader’s convenience as more important than the mechanical reproduction of text.” Eliot’s punctuation was altered or supplied for clarity.

Eccentricities

Regarding other holographic eccentricities, it is difficult to reach consensus since most editors are reticent about them. Betty Bennett, editor of Mary Shelley’s letters, writes: “I believe that holographic eccentricities do reveal as much about an author as the words themselves.” 

Yet Davis and Gilman’s sixty-four pages of dense textual notes tell me little more than that Herman Melville’s handwriting was hurried and careless. Melville’s editors warned that diplomatic transcription would lead to “a typographical monstrosity.“⁵ Walter Harding, editor of Thoreau, has seen the monster and feels that academically sanctioned aspects of diplomatic transcription are work “carried to an idiotic extreme.” 

For perspective, I am guided by James F. Beard, editor of James Fenimore Cooper’s letters, who writes: “I consider it absolutely essential to acknowledge a fundamental difference” between the media of holograph and print. “Failure to make and acknowledge the distinction can be absolutely disastrous and incredibly expensive.”

Through examination of conflicting differences between media, it should be possible to estimate the proper scope of printed and digital media and to arrive at a sympathetic view of The Sensible Text.

Print Ain’t Handwriting (duh)

The problem of George Moore’s handwriting showcases many of the difficulties of printing holograph. Moore struggled in secondary school with his inability to master the basic skills of reading and writing. Most of what he called his education took place in racing stables, artist’s studios, and cafés. 

At first sight, his letters thrilled me. Disregard of orthography and punctuation, peculiar flourishes and messy blotches — all exuded character. Language unharnessed in the “onrush of ideas”!

My earliest transcriptions at the Berg Collection of the New York Public Library (I still have them) preserved all of the original spacing and line breaks. The awful consumption of time this caused forced me to abandon some scruples, but I continued to transcribe thousands of letters paying heed to every elision, fusion, and expansion and every accidental capitalization. I say accidental because in my judgment Moore’s decisions to write what looks like a capital A or C were often made entirely in his wrist.

Having made diplomatic transcriptions, I suffered a shock. The robust and engaging George Moore of the holograph was not in my typescript. Vividness had become sloppiness, music was noise. Michael Millgate writes that his goal in editing Thomas Hardy’s letters has been to render the “flavour” and the “bouquet” of the holograph letters. I was dismayed because I tasted peppercorns and smelled vinegar. 

The cause of my problem eventually became clear: I had pushed Moore’s letters from one medium into another, where different textual values prevailed. Leon Edel calls this process “translation.” He advises that “putting holograph into print is itself a sea-change.” 

With greater awareness of inevitable and crucial alteration in the character of handwriting when it is typed, I became convinced that my desire to replicate Moore’s holographs with print (The Sacred Text) was mistaken and would produce results that were less truthful than absurd. The same holds for the passage to digital, though digital modes of presentation can ameliorate the problems.

In order to put my emerging ideas in the context of wider experience, I sent a questionnaire to several eminent editors. My questions, in various wordings, were these: 

1. Do you regard a particular edition of letters as a model for future editors?
2. Did you invent your own rules of transcription or follow established rules?
3. Should autograph, typed, and printed source texts be governed by different rules of transcription?
4. What exactly might the retention of slips and careless errors of spelling and punctuation contribute to the printed version of a holograph?
5. Do you acknowledge a fundamental difference between the media of handwriting and print? How should a difference influence decisions concerning the transcription of holographic eccentricities?

A Model Edition?

Fifty percent of respondents did not recommend any edition of letters as a model. In general, they felt that unique problems created by an author’s correspondence are too great to be solved or even managed with a formulaic approach. Only three editions received more than one vote. 

1. James Hepburn, editor of Arnold Bennett’s letters, and David Krause, editor of Sean O’Casey’s letters, both nominated Sir Rupert Hart-Davis’s Letters of Oscar Wilde. That’s my choice too. Richard Ellmann, who was preparing a new biography of Oscar Wilde, wrote that “Rupert’s edition of Wilde is so magnificent that I feel tempted to endorse anything he has told you” in regard to editorial practice. 
2. Gordon Haight, editor of George Eliot’s letters, and Michael Millgate, editor of Thomas Hardy’s letters, both nominated Kathleen Tillotson’s Letters of Charles Dickens as a model. 
3. C. Richard Sanders, editor of the Carlyles’s letters, and Leslie Marchand, editor of Byron’s letters, nominated Gordon Haight’s George Eliot Letters as a model. 

None of these three exemplary editions supplies The Sacred Text. Moreover editors who avoided formulae by not nominating a model also avoided endorsing, among other things, a diplomatic transcription imposed in the name of “modern textual principles” on Washington Irving.

It seems clear from the survey that our profession doesn’t actually have a coherent body of “modern textual principles.”

Follow or Invent Rules?

All of the respondents synthesized editorial practices from their own instincts and from critical examination of other editions. Some including Gordon Ray and Leon Edel were more original than others, but none suggested that their editorial practice resulted from imitation rather than imagination. 

Even editions issues by the same press display considerable diversity of technique. In the view of one publisher’s senior editor, Mrs. Arthur Sherwood of Princeton University Press, a body of ground rules for editing letters would save money, time, and grief spent in harnessing the average editor’s imagination to the dogcart of practicality.

Different Source Materials?

The replies to question three, as to whether handwritten, typed, and copied letters ought to be transcribed differently, produced no consensus. 

A “copied” letter is the most problematic, since it is written by someone other than the letter-writer. Many of George Moore’s letters were typed by an amanuensis or copied by scholars, collectors, booksellers, or editors. It seems certain that errors in these letters are not attributable to Moore. Should I allow such inauthentic errors to remain in the text or should I correct them? Fifty percent of editors advocate the same methods of transcription regardless of the nature of the source text. David Krause asks: “What else can or should an editor do?” 

A creative approach to copied source material would be to alter it so that it conforms with the letter-writer’s known holographic character. This seems risky, since most editors are less than fully articulate when describing holographic character. 

Thomas Lewis, editor of Hart Crane’s letters, advises that “misspellings in printed sources should not be transcribed.” Robert Elias “rectified, as though they were typographic errors, all mistakes, inconsistencies, and carelessness attributable to Theodore Dreiser’s secretaries.”

Though Professor Elias does not voice the majority viewpoint, his method makes sense. Whether or not eccentric handwriting belongs somehow in print, editors should be able to agree that the typographical blunders of a secretary or journalist, like the careless markings of a bookseller or scholar, have only an accidental relation to a literary legacy and can be excluded.

The Value of Junk?

The replies to this question support the view that slips of the pen and careless errors of spelling and punctuation provide a holograph with some of its vivifying powers. George Moore died almost twenty years before I was born, yet the lively look and feel of his handwriting let me imagine that I can hear his voice and see his hands in their pantomime of consciousness. Just what are those qualities of handwriting that inspire what Richard Ellmann calls “pietism” among editors? 

Michael Millgate writes that Thomas Hardy’s eccentricities “may tell us something about his mood that day, or about his attitude towards a particular correspondent.” Nigel Nicolson offers an example: “The wrong spelling of a name can be a clue to how well, or how little, the author knew about that person. For instance, Virginia Woolf, in her first letter to her future husband Leonard, spelled his name Wolf. A year later she married him. The correct, or incorrect, spelling of the names of casual acquaintances (and she was usually correct) does indicate, I think, the trouble she took to identify people, and hence her social attitude.” Mr. Nicolson also suggests that common words could be significant “as an indication of how even the most literate people can sometimes fail to observe what they have often seen correctly spelled. I don’t think it has any psychological significance, as some would say.” Karl Beckson, editor of the mentally unstable Arthur Symons’s letters, decided that “orthographic peculiarities should be preserved to reflect the writer’s state of mind.” 

That brings to my mind a particular letter in which George Moore tried four different spellings of “eunuch” before opening a dictionary, in a state of irritation, and carving the correct spelling at the top of his letter, above the salutation. 

When editing the letters that passed between John Synge, W. B. Yeats and Lady Gregory, Anne Saddlemyer “wished to convey through the edition the basic trust each had in the other’s comprehension — even when they were all annoyed with each other.” By preserving their holographic eccentricities, Professor Saddlemyer extended her trust to the comprehension of readers.

Aside from trust, eccentricities “give a sense of ‘flow’” in Professor Saddlemyer’s edition of the letters of Synge. Similarly, J.T. Boulton writes that holographic eccentricity “gives us a more personal ‘feel’ — demonstrates that the writer is human like his reader!” Professor Boulton believes that ambiguous punctuation and slips of the pen put a reader of his edition “(so far as text goes) in the same position as the original recipient” of D. H. Lawrence’s letters. 

Norman Kelvin, editor of William Morris’s letters, similarly desires to recreate past moments of confusion “in the belief that the reader will prefer to work things out or ponder the ambiguities as Morris’s actual correspondents may have had to do.” Betty Bennett finds that oddities of Mary Shelley’s handwriting produce “tone.” Dan Laurence, on a more literal level, writes: “If I were dealing with an author who was a chronic misspeller, I would retain all the errors, as I think it adds verisimilitude to the text, which should be a reasonable reflection of the personality and education and experience of the writer of the letter.” 

On a less solemn note, Richard Garnett remembers that “Lady Diana Cooper’s manuscript of her memoirs spoke of someone who had a cold in which nose and throat reinfected one another in ‘a viscous circle’ — a brilliant joke, unintentionally brought about by dotty spelling.” 

Slips of the pen and careless errors can make us laugh; they create flow, feel, tone. C.L. Cline removed them from George Meredith’s letters, where they seemed to signify nothing, but he writes that they “flavor” the letters of Thackeray; Professor Millgate writes that they achieve “bouquet”; when Richard Ellmann mentions the feeling of “haste” he seems to echo J.T. Boulton’s “dash.”

These terms describe the intangible spirit of the holograph. This spirit can have a pleasing, magical effect on the reader of a handwritten letter. But can that effect make it into printed pages?

Different Media?

The replies to question five help to clarify differences between the media of holograph and print. Leon Edel, having deeply appreciated the superior quality of holograph and questioned whether it can be replicated in print, formulated editorial practices which he knew would worry his pietistic critics. Perhaps as a gesture of good faith, he has asked scholars to note that we live in “the age of photoduplication,” in which The Sacred Text can be disseminated on paper or film. 

This is not always true. Experience taught me that some curators of important manuscripts forbid photoduplication (what we now know as digital scanning). For example, I was unable to order photocopies of Moore’s letters at the Bibliothèque Littéraire Jacques Doucet in Paris and the Tolstoy Museum in Moscow. 

Some literary correspondence is owned by individuals or autograph dealers who forbid copying in anticipation of commercial transactions. Some curators who share facsimiles do so only at prohibitive rates. Finally, the products of certain forms of photoduplication — microfilm, rapid copy, low-resolution scan — frequently impair rather than clear the vision of an editor. 

For example, I spent several days agonizing over the Chadwyck-Healey microfilm publication of Swan Sonnenschein’s correspondence files before going in desperation to the basement of Messrs. Allen and Unwin, where the original documents were at last transcribed. The experience of working with photoduplicated holographs can convince an editor that letter-writers guard their own spirit and compel us, as it were, back to the source material.

Transcription Is Translation

Admitting our practical inability to replicate holographs photographically for editorial purposes, but also desiring to capture the spirit of the holograph in our transcriptions, editors ought first to acknowledge that transcription is, in Leon Edel’s phrase, translation from one medium into another. 

A few editors are not prepared to go that far. C.L. Cline, who altered the text of George Meredith’s letters to increase readability, writes: “I do not think it would have been difficult to replicate holograph in the medium of print with the Meredith letters.” J.T. Boulton denied a fundamental difference between media, though he has made concessions to a few differences, considered trifles, in the letters of D.H. Lawrence. Neither Anne Saddlemyer nor Michael Millgate faced an insurmountable barrier to the printed replication of letters written by Synge or Hardy. C. Richard Sanders does “not acknowledge, psychologically, a fundamental difference between holograph and print. I think that the difference is superficial and could be ignored altogether.” 

But these are minority views. Most of the canvassed editors confirmed my conclusion from a survey of editions: that the regular form and spacing of print, its size on the page and the bound pages themselves, are all intrinsic elements of the printed medium that are foreign, even contrary, to the corresponding elements of the holograph. The difference between them is fundamental because the spirit and, for some editors, the meaning of the message are rooted in idiosyncrasies of characters, spacing, and the physical properties of the holograph as a whole.

Say what your will, transcription is translation. To discover a practical means of translation, editors might be guided by Professor Sanders’ experience of the Carlyle letters. “We retain slips, misspellings, and careless errors for the sake of fidelity to text,” he writes, “but I now believe they are more of a distraction than anything of positive value. About all that we can come up with in studying them is that Jane was more careless and a poorer speller than Thomas. If I were starting again, I would normalize the text.” 

Against Pedantry

The failure of Professor Sanders to detect Carlyle’s personality in erratic wrist movement and orthography might be ascribed to lack of imagination.

By contrast, Edwin H. Miller confesses that he is “a psychological critic, and most interested in the latent level.” Professor Miller believes that the difference between holograph and print is so great that “it is impossible to reproduce a letter exactly under any circumstances.” He transcribed a normalized text of Walt Whitman’s letters not out of insensitivity to the most subtle qualities of holograph, but because these qualities are not translatable. “People looking for hidden secrets will have to go to the originals,” he writes, echoing the sentiments of another psychological critic and editor, Karl Beckson. The word Professor Miller summons to characterize diplomatic transcription is not psychology but pedantry.

Kathleen Coburn has urged that literary correspondence radiates “some sort of warmth that ought not to be damped with the cold water of pedantry.”⁶ The pedantic posture of some editors has also annoyed Dan Laurence: “I would like the text to be as close to transcription in facsimile as I can make it without irritating the reader with extraneous signs of the pedant’s presence, or trying to carry scholarship to the point of noting that there are inkblots in the margin, an uncrossed ‘t’ on line 17, and a dropped letter in a typewritten word!” 

Shaw’s editor thus objects to the paraphernalia of The Sacred Text — the brackets, squiggly lines and textual notes — and in so doing he objects, perhaps unwittingly, to The Sacred Text itself. Editors do not invent symbols and assign valuable space to textual notes merely to make themselves visible. The symbols and notes, which make many printed texts unreadable, are structural support for diplomatic transcription. 

Herman Melville’s editors found it impossible to “reproduce his orthography without typographical obtrusiveness and a resulting irritation to the reader. A text of the letters which Melville wrote rapidly and spontaneously would end up a typographical monstrosity” (p. xxv). 

Leon Edel adds: “To reproduce every hen-scratch in collections of letters running into thousands is not only pedantic, it is futile.” Thus Professors Edel and Sanders, model editors of opposed traditions, seem to occupy common ground. 

Edel continues: “The correspondence needs to be characterized; and if it is sown with ampersands by the bushel the editor should decide whether he wants to have his page look like a series of potato mounds or whether he wants to bestow on the reader the legibility that print is bestowing.”

Gordon Haight, correctly regarded as a very different kind of editor, supplies a comment about George Eliot’s letters which reflects Professor Edel’s viewpoint: “To reproduce in type all the vagaries of manuscript is neither feasible nor desirable; and though I tried to provide an accurate text, it was never an exact transcription.”

Transcription Is Characterization

To characterize a correspondence is an editorial alternative to paraphernalia. From experience of editing Virginia Woolf’s letters, Nigel Nicolson may direct us to an agreeable and practical perception of editorial responsibilities: “The principle, I believe, is to convey the raciness of a letter, as distinct from something intended for print where she was very careful), but without putting unnecessary strain on the reader’s understanding of the passage. One ought to hear the writer’s voice in reading a printed letter, and this depends largely on retaining her original punctuation; but it is legitimate, I think, for the editor to supply very occasionally a punctuation mark to make the passage intelligible, without pedantic sics or square brackets …. A letter should be read, as it is written, as cursively as possible, to convey sense, mood and spirit, and the intrusiveness of an editor with niggling amendments can often prevent this.”

So to characterize a literary correspondence, to translate a holograph into print, the editor must make judgment calls on behalf of the letter-writer, for the benefit of the reader. It may be admitted that editorial paraphernalia is more harmful to the spirit of the correspondence than responsible, even silent normalization of text. 

For comparison, Edwin Miller and Thomas Lewis identified the Harvard University Press publication of Emerson’s Journals as an unfortunate but logical outcome of diplomatic transcription. “It seems an outrage,” writes Professor Miller, “to make it almost impossible to read perhaps the greatest autobiography in American literature.”

It is equally outrageous to maim and cripple the correspondence of great writers through transcription that aspires to replicate every holographic eccentricity. If the aspiration is excusable, the results are not.

My Modest Proposal

As an editor responsible mainly to letter-writers and readers — not merely to critics and scholars — I hereby propose ten ground rules for transcription of literary correspondence. Bound by common interests and experience, editors of the future can do it as one. 

I made an appeal like that forty years ago in a conference of textual scholars, and of course nothing at all happened. At the time there was no digital age dawning, no economic or cultural reason to innovate, so here I am with the same seven guidelines I proposed back then. They still seem right. 

Guidelines eight-ten are new. They’re for editors like me who transcribe for the digital object rather than the printed page. That’s how all literary legacies will be kickstarted once the old fogies see the light, or retire.

1. Edit for the People. Because editors devote years to the concentrated study of a letter-writer, achieving expertise and mastery of the life and work, they earn tacit authority to identify what Nigel Nicolson called an author’s “voice.” The duty of an editor is to make that voice as audible as possible, to as many different people as possible. Editors, listen up: Do not wrap your mummy. There is no mummy.
2. Any Half-wit Can Copy. What is “possible” is defined not by what Edwin Miller calls “mindless and pedantic” replication of the holographic medium. It depends rather on realistic appraisal of both platform and medium. Instead of engaging in futile experiments to exceed the limits of print (whether on paper or film or screen), editors should “translate” holograph for a fundamentally different platform and medium. Editors: Do not stick to your knitting. Unlock value.
3. Put a Finger in the Dyke. It is agreed that holographs include myriad eccentricities, not all of which contribute to a reader’s understanding; not all of which mean anything at all. Editors should prioritize eccentricities which, in the words of Merrell Davis and William Gilman, are “characteristics within the author’s intention that seem important to the reading and understanding of the letters.”⁷ The select eccentricities should be translated for the new medium if possible; the rest should be discarded. Editors: Keep the good stuff. Toss the crap.
4. Enhance Coherence. Eccentricities to ignore are careless or erratic spelling and any punctuation and capitalization due to slips of the pen. For example: the fusion or elision of characters which result from hasty penmanship; a full stop lengthened by a heavy hand to resemble a dash; a large lower case character that resembles upper case. Eliminate distracting eccentricities which don’t add any value. For example, transcribe “&” as “and” and “wd” as “would.” Editors: Make your letters fun to read. Don’t annoy readers.
5. Declutter. In deciding further which eccentricities to represent in print, do not transcribe any that create typographical clutter. For example, crossed out words which don’t contribute to final meaning, multiple underlinings, skewed lines and crammed line breaks. Editors should normalize silently, transcribe intentional errors verbatim, and describe glaring eccentricities in notes. Editors: Do your own housekeeping.
6. Don’t Be Absurd. Editors should acknowledge the silliness of holding an author responsible for a secretary’s, bookseller’s, or journalist’s mistakes. They should silently normalize all typed and printed source texts. Editors: Give your letter-writer the benefit of the doubt.
7. Be Transparent. In prefatory matter, editors should explicitly describe their methods of transcription. Typical differences between holographic and transcribed texts, including omissions and alterations, should be noted so that readers can see how and why the correspondence has been “characterized.” Editors: Open your kimono.
8. Avoid footnotes. In digital editions, there is no reason to pepper text with footnotes that interrupt flow and fragment attention. No recipient of a literary letter ever had to read footnotes, so why should our readers? Instead, use hypertext to make letters explain themselves — by linking words and phrases to online references and related data elsewhere in the correspondence. Editors: Let the medium be the message.
9. Include Facsimiles. The economics of print publishing discourage the inclusion of pictorial content in large editions of letters. A few plates are tolerated, but these are often worthless decorations. In digital editions, there are no such constraints. Therefore every letter should be linked to an image of the holograph it translates so that readers who want to examine handwriting can have at it. Editors: Empower your readers.
10. Program Behavior. No printed edition of letters has ever behaved in any way at all. It just sits there, saying nothing, making nothing, forcing readers to do the exploration and discovery and synthesis on their own. Is that good? It is not. A literary correspondence should be made from active text, run by intelligent systems, so that it senses and responds to human users. Programming behavior into correspondence is a technical hack, for sure; but textual scholarship can ensure that the results are brilliant. Starting with nuanced, self-aware, dynamic transcriptions. Editors: Kickstart your literary legacy in the digital age.

Conclusion

Readers of this post may someday observe my guidelines being practiced in the forthcoming Letters of George Moore: all digital, all powerful, all free. The edition will have taken half a century to edit, but that’s okay. Much of what makes it worthwhile could not have been finished sooner.

In urging universal acceptance of Sensible Text for all readers, I mentioned the outrage of two distinguished editors over the crippling of Ralph Waldo Emerson’s Journals.

Emerson reportedly wrote: Raphael paints wisdom, Handel sings it, Phidias carves it, Shakespeare writes it, Wren builds it, Columbus sails it, Luther preaches it, Washington arms it, Watt mechanizes it.

In the twenty-first century we shall add: Editors transcribe it.

1. 
   The Letters of Anthony Trollope, ed. Bradford Booth (Oxford: Oxford Univ. Press, 1951), p. xi. ↩︎
2. 
   The Correspondence of Henry David Thoreau, eds. Walter Harding and Carl Bode (New York: New York Univ. Press, 1958), p. xix. ↩︎
3. 
   The Letters of W.B. Yeats, ed. Allan Wade (London: Rupert Hart-Davis, 1954), p. 16. ↩︎
4. 
   Byron’s Letters and Journals, I, ed. Leslie Marchand (Cambridge, Mass.: Harvard Univ. Press, 1973), p. 28. ↩︎
5. 
   The Letters of Herman Melville, eds. Merrell Davis and William Gilman (New Haven: Yale Univ. Press, 1960), p. xxv. ↩︎
6. 
   The Letters of Sara Hutchinson from 1800 to 1835, ed. Kathleen Coburn (London: Routledge and Kegan Paul, 1954), p. xii. ↩︎
7. 
   The Letters of Herman Melville, p. xxvi. ↩︎