Textual Editing and the Cooper Edition
Presented at the 5ᵗʰ Cooper Seminar, James Fenimore Cooper: His Country and His Art at the State University of New York College at Oneonta, July, 1984.
Originally published in James Fenimore Cooper: His Country and His Art, Papers from the 1984 Conference at State University of New York College — Oneonta and Cooperstown. George A. Test, editor. (pp. 10-23).
Copyright © 1985 by State University of New York College at Oneonta.
[May be reproduced for instructional use by individuals or institutions; commercial use prohibited.]
BEFORE examining some of the kinds of discoveries about Cooper’s patterns of revision which the Cooper Edition has disclosed, I begin by discussing generally the problem faced by textual editors, and the kinds of solutions to these problems used in current editorial practice.
I. Why Texts Need to be Edited
In a perfect world, textual editing would be unnecessary. The processes of writing and publishing, in ideal circunstances, might go as follows.
1. Authors would provide their publishers with perfect texts in which no one other than the author — no secretary, copyist, or typesetter — had intervened. In other words, an exact correspondence would exist between the author’s final intention for the work and the text submitted for reproduction. The author would have checked each word and each punctuation mark to insure complete accuracy.
2. The publisher would reproduce the text in such a way that no mechanical intervention (bringing inevitable corruption) would occur between the author’s text and the final public product. For example, photographic or xerographic reproduction, or type set from the author’s digitally prepared input, would eliminate the role of a typesetter who in setting type or executing keystrokes, can change or misread the author’s intention.
3. Before publication, the author would double-check the final product in the minutest detail, inspecting for accuracy both the words themselves and all aspects of punctuation and styling.
4. Finally, in all reproduction subsequent to the first edition, the text would be duplicated photographically or in other ways which avoid resetting of physical type. Thus the typesetters’ or compositors’ inevitable failure to reproduce exactly what is in front of him would be eliminated.
In such a procedure, an author could be assured of communicating meaning to an audience without corruption. But of course, in the real world, such ideal situations are sheerest fantasy. The evidence of literary history suggests that what really occurs is the following:
1. Authors commonly pay little attention to insuring complete accuracy of the texts they send off for publication. Stories are legion of poems dashed off on backs of envelopes, or novels submitted in chaotic condition like the fabled trunks containing Thomas Wolfe’s works. Joyce’s Ulysses was printed in a small shop in southern France where few of the compositors knew English. Their work was further complicated by Joyce’s habit of grafting new verbal felicities onto his text at every occasion when he read proof, requiring the printer to wrestle with his difficult hand to capture his continually developing intentions. (A scholarly edition of Ulysses produced by a team of West German editors made the front page of the New York Times recently.)
In earlier centuries, things were worse. As far as we know, Shakespeare was not directly involved in the printing of any of his plays. His texts at best come from copies of the plays as they were produced by the acting company in which he was partner, no doubt with his original language changed because of the actors altering or forgetting their lines. Among his contemporaries, few had better fortune. For example, the texts of major poems like many of John Donne’s circulated in manuscripts, and thus suffered from the errors of copying out by hand.
In ancient times, matters were worse still. Modern scholars believe that Homer’s Odyssey and Iliad circulated for several centuries in the oral tradition before being written down. No doubt every bard who performed them made changes in response to memory lapse, inspiration, or audience interaction. Probably a similar lineage occurred for the Old Testament. Happily, the dialogues of Plato seem internally consistent enough to appear fairly authentic. But Aristotle suffered a chilling fate indeed — his “books” appear to be redactions of lecture notes taken down by his students.
2. The advent in the West of printing with movable type imposed a significant degree of stability upon textual history, since no longer was every version of a work by definition copied out by hand and subject to numerous unconscious changes. From Gutenberg in the mid-15ᵗʰ century until about a hundred years ago, books were made by skilled workmen called compositors picking each letter, punctuation mark, and blank space out of a sometimes chaotic box of type. Reading from the author’s manuscript or some copy thereof, the compositor would attempt as accurately as possible to replicate in type what he read and kept in his memory as he reached, piece by piece, into his font of type. Leaving aside questions of compositors refusing to set passages with which they might disagree for various reasons, abundant sources for errors remain.
In summary, any one who’s played the old parlor game of passing a sentence quietly around a circle of players, and observing the sometimes comic or grotesque evolutions of meanings in the process of mortal communication, will appreciate how quickly an author’s intention can be distorted when others, even with the best intentions, try to preserve it.
3. The third step in our idealized picture is for the author to check carefully the product of the publisher before releasing the text to the public. Here the record of authors is even more abysmal than in the first stage of submitting clean copy. Few major professional authors exerted much care in checking the compositor’s workmanship. Cooper read proof when time and distance permitted, but he usually was engaged in a new project when the proofs of an earlier work were available. (Remember he produced thirty-two novels and enough travel literature, social commentary, and political and historical writing to fill his complete works out to about fifty volumes — in a space of a little more than thirty years.) Charles Dickens offers another interesting example of how an author might orphan a work after completing it. Since nine of his fifteen books were published in monthly pamphlets exactly thirty- two pages long, when out of London he faced the difficulty of estimating how long his manuscript should be in order to accommodate the prescribed monthly installment. Often he drew upon the services of his literary adviser John Forster to aid him by having Forster cut overlong passages to fit a serial part to exactly the bottom of the thirty-second page. What to do with the many scraps of Dickens’ wit which Forster thus excised is an interesting problem facing editors of the current Dickens edition.
4. Until the introduction of stereotyping into America in the late 1820s, new editions generally required a resetting of standing type. Such new settings might provide an author with the opportunity to revise the earlier text, but the resetting assuredly opened the text to new errors arising from the problems with compositors mentioned above. In some cases with Cooper, scholars who have compared the early and late texts of his novels have found literally thousands of changes in words and punctuation between them. The challenge then is to decide which of these changes represent Cooper’s revisions and which are the errors introduced by inattentive compositors. To grasp how editors deal with the problems of determining which variants in a text they believe to be the author’s work, we turn to the next section which reviews briefly the history of textual editing.
II. Textual Editing: A Brief History
The first “editors” of texts were doubtless those poets and story-tellers whose feats of memory have been attested to by Milman Parry and Albert Lord. In passing dawn from generation to generation long narrative stories like the Iliad, they both preserved and modified an ancient heritage. In the late classical period scholars emerged like Aristarchus of Samothrace who sought to create as authentic a text as possible for Homer. The search for textual authenticity became a dominant concern in the Christian Church as the early forms of the Bible were assembled and compared; modern textual criticisn emerges from this tradition. By the nineteenth century an elaborate technique for comparing genealogies of manuscript texts had been devised. Since all evidence suggested that texts were corrupted by transmission through time, the aim of these genealogies or stemmas was to discover the earliest text or Ur-text which preserved the original intentions in pristine authenticity.
Further problems arose with the possibility for authorial revision introduced by the capability of resetting printed texts. The fashion in editorial circles wavered between two opposite viewpoints for dealing with a text where many printed editions (some of which might be revised by the author) existed subsequent to the first edition:
A. Choose the earliest edition since, as the text closest to the author’s pen, it will suffer the smallest possible number of corruptions inflicted by new typesettings. Unless persuasive evidence existed that a later edition was extensively revised by the author, editions subsecquent to the first played no role in textual editing. The other viewpoint was the mirror image of the first.
B. Choose the latest edition possible. Even though admittedly such an edition might contain compositorial errors, it would logically also have the greatest possible number of authorial revisions.
Both these views agreed on a basic point: whichever edition was chosen, early or late, should be the exclusive authority for the editor. That is to say, no attempt could be made to sort out among (in Cooper’s case) the thousands of variants in wording which existed between early and late editions in terms of which changes were likely to be the author’s and which from some unauthorized source.
Beginning in 1950, a new argument arose in the work of Walter Greg, Fredson Bowers and others. Greg believed that evidence of changes should be gathered from the widest possible range of sources. All of the following were to be considered if extant: the author’s manuscript in all its versions, copies of the author’s manuscript if they affected the transmission of the text, all proofs of the first printing or other printings with which the author was involved, the first edition, and all other editions either where the author made revisions or where the later text became a source for a still later one the author revised. Greg then urged that editors distinguish between two types of variants, namely: accidentals which affect punctuation, styling, or spelling, but not meaning, and substantives which affect meaning — from spelling changes that do so through changes of whole pages. Greg applied a bit of sound psychology about how involved the author was likely to be in originating these changes or variants. He argued that authors like everyone else pay more attention to the meaning of words (the substantives) than to those aspects of their text which are stylistic but do not directly affect meaning (the accidentals). Thus in every stage of revising, the author is more likely to intervene to make a major change of meaning than to alter punctuation or spelling. To create a text which canes closest to what the author really wanted, the editor should take the closest to the author’s hand as the standard for accidentals — spelling, punctuation, paragraphing and the like. If the manuscript exists, use it; if not, then the next closest text to the author’s original hand, most likely the first edition. Here, Greg argued, we have the most trustworthy source for the author’s personal preferences for accidentals. The earliest source preserves those idosyncrasies of punctuation and spelling that typesetters are most likely to alter to suit themselves, and that authors are least likely to bother to revise back to their standards if engaged in recasting the work.
Then, Greg argued, into the matrix of the original accidentals should go all those changes in substantives (in meaning), which the editors can identify with some confidence as being the author’s work. If we are lucky enough to have the manuscript of the author’s own revisions (as we do for Cooper’s The Spy), then we are able to say with certainty which changes in the revised edition set from that manuscript are authorial, and which not. If we lack manuscript authority, a sense of an author’s typical practices of revision — as The Spy manuscript offers abundantly — can guide the editor. Other external sources of evidence are also crucial — errata sheets, correspondence with publishers or friends about problems with printers, proof sheets, publishers’ accounts or their known practices of imposing their own style on authors.
In summary, Greg advocated that editors use their scholarship and taste to produce an eclectic text drawing upon as many sources of possible authorial revision as can be identified. Rather than adhering to just one version of a text, early or late, the editor should preserve the surest evidence of the accidentals as found in the earliest source, while adopting any readings subsequent to the first edition which the editor believes to be certain or likely authorial revisions. At the same time, readings in the texts subsequent to the first edition not believed to be authorial are purged from the eclectic text.
III. Three Examples of Modern Editions of Cooper’s Fiction
I have been involved in editing three well-known Cooper novels, The Spy (1821), The Pioneers (1823) and The Deerslayer (1841). Happily, all three have offered significantly different editorial challenges, and thus provide a range of examples of how modern editorial practice proceeds.
The Pioneers (SUNY Press, 1980) was Cooper’s and America’s first best seller in fiction. Readers anticipating another adventure story like the popular Spy, its predecessor, snapped up the first printing of 3500 copies on the morning of initial sale. In Cooper’s lifetime, The Pioneers was reprinted at least nineteen times in America, thirteen times in England, twice in Paris and once in Germany. Cooper made good use of certain of these reprints to repair the omissions of the first edition. An errata note in that first edition alerts the reader to an unusual number of errors resulting from a yellow fever plague in the summer the book was printed: Cooper and many people associated with his publisher stayed away from New York and its contagion. Proofreading procedures were disrupted, and such errors as printing among others, “to fall” for “to fell” or “moreen” for “marine”, were perpetrated. The evidence of Cooper’s letters, publishing history, and a close examination of the kinds of variants which occurred in later editions permitted the editors of The Pioneers to identify five texts where Cooper demonstrably made revisions. These were:
1. Brief excerpts in a magazine which carried several chapters before the actual book publication.
2. A second New York edition issued just weeks after the first, which previous editors and bibliographers had failed to identify as largely a new and significantly revised edition.
3. The first London edition which contained dozens of corrections not appeasing in any other later edition.
4. An extensively revised edition prepared for the 1832 Bentley Standards Novels edition; though the manuscript for these revisions is not known to exist, the manuscript for The Spy does exist and provides a full record of the kinds of revisions Cooper made to The Pioneers.
5. A lightly revised edition Cooper made for an incomplete new edition of his works by Putnam near the end of the novelist’ s life.
The benefit of applying the Greg-Bowers eclectic editorial policy to this edition of The Pioneers is as follows. A wide variety of revisions which improve verbal felicity, especially in the Bentley edition, are preserved, as are some revisions made for the magazine and first British publication which did not enter the mainstream of subsequent American texts. At the same time, preserving the accidentals of the first edition meant here preserving much of the dialect of the various speakers. Clearly Cooper tried hard to establish a distinctive dialect sound for the diversity of national and racial exemplars assembled in this “Descriptive Tale.” Subsequent editions had done much to mute the distinctive dialect of Monsieur Le Quoi the French shopkeeper, the retired German Major Hartmann, the Yankees Hiram Doolittle and Remarkable Pettibone, the diminutive Cockney Ben Pump, the various blacks, and above all, the Leather-stocking, here in his first appearance. Without imposing modern standards on Cooper’s characteristic punctuation, which aims at catching the rhythms of a speaking voice, the Cooper Edition of The Pioneers also smooths out numerous verbal kinks, with full explanations in the Textual Notes.
The editorial situation in The Deerslayer (SUNY Press, scheduled for 1985) is opposite to Pioneers in all respects. The last of Cooper’s Leatherstocking Tales to be written, The Deerslayer (like much of Cooper’s fiction in the 1840s) failed to excite the popular acclaim of the early work. Though the first edition was reimpressed six times in the author’s lifetime, only one newly set edition was called for, again in the Putnam series and again lightly revised. But for The Deerslayer the manuscript exists, almost in complete form, at the Pierpont Morgan Library in New York City. Thus the editors of The Deerslayer had the opportunity to preserve Cooper’s original intentions in accidentals, even before the first edition ccmpositors modified them. In hundreds of places the editors rejected errors of the first edition typesetters, and at numerous places published for the first time the words Cooper really wrote.
Some examples of these restorations follow. Half way through the story the young Deerslayer is described as “indulg[ingl in a fit of his hearty, benevolent laughter” (285.16-17). A nice enough trait, but not Cooper’s intent, for the manuscript reads “hearty, but silent laughter.” This is perhaps a small point, but a significant difference in character — silent laughter implies a reserve, a reflectiveness the very opposite of the heartiness of the printed text. When Hetty sings to the men on board the boat at 91.15-18 of the first edition, Hurry Harry and her father cease their labor until “the last of the sweet sounds had actually died among the remarkable shores, which, at that witching hour would waft, even the lowest modulations of the human voice, more than a mile.” Remarkable shores indeed, to waft her gentle song more than a mile — but Cooper in fact wrote echoes, not shores. Occasionally, errors in punctuation totally pervert Cooper’s prose. For example, at 87.28-29 when Hurry Harry defends his scalping expedition against the Huron camp, he argues “The savages scalp your fr’inds, the Delawares, or Mohicans, whichever they may be, among the rest; and why shouldn’t we scalp.” Here Hurry justifies his proposed venture by arguing that other Indians scalp the Delawares; why shouldn’t Deerslayer defend his allies by helping slay their enemies. But in the manuscript Hurry’s argument is both more accurate and more powerful: “The savages scalp — your fr’inds the Delawares, or Mohicans, whichever they may be, among the rest — and why shouldn’t we scalp?” Hurry in reality throws in Deerslayer’s face that his savage friends admit scalping to the practices of warfare; why is Deerslayer so fastidious to refuse?
The most important compositorial error disclosed in our edition occurs at the very end, when Deerslayer reluctantly rebuffs Judith’s plea that they marry. Judith attributes Deerslayer’s rejection of her entirely to his moral objection to her sexual inproprieties with the British officers — a view only partly correct, as she fails to understand his reluctance to marry anyone and settle down anywhere. In the printed text at 545.20-22, in a paragraph beginning “Truth was the Deerslayer’s polar star,” Judith parts from him with one final beseeching glance, and then she “read his answer in his countenance; and with a heart nearly broken by the consciousness of undeserving, she signed to him an adieu, and buried herself in the woods.” The crucial word here is “undeserving”: Judith appears to feel heartbroken because she believes she does not deserve her fate — to be abandoned by the only man she has respected enough to love, and thus to have to turn to an uncertain life of prostitution. Several critics have joined in her plaint, and regarded Cooper as callous in visiting upon her so ominous a fate for so slight a flaw. But Cooper did not write the sentence as printed (see Figure 1).
Figure 1.
He wrote “undue erring”: a reading which alters the final paragraph of her appearance with Deerslayer, and in turn our whole sense of her leave-taking. “Undue” — that is, excessive — “erring”: she knows she has been wrong to give herself to Captain Warley, perhaps others; her indiscretion has helped her to lose her only love. And the erring was excessive; it led to her downfall. “With heart nearly broken by the consciousness of undue erring,” she accepts her fate. In Cooper’s version — never before printed — she resigns Leatherstocking not as the trampled-down victim of melodrama, but more as the tragic heroine who recognizes that her fatal flaw has contributed to her grief and downfall.
The last examples come from work in progress on Cooper’s second novel and first great success, The Spy (1821), a tale of heroism and devotion in the Revolutionary War. The Berg Collection of the New York Public Library owns the manuscript of Cooper’s 1830 revisions for the Bentley Standard Novels series edition of the text (see Figure 2).
Figure 2.
These corrections are made on a specially prepared copy of a previous edition with blank sheets interpolated into the text on which Cooper wrote his revisions. This interleaved manuscript of Cooper revisions not only establishes with certainty which variants — substantive and accidental — Cooper made for the Bentley edition, it offers a fascinating record of the kind of attentive care Cooper lavished on improving an early text he considered unusually corrupt. Ten years into his career Cooper had grown as an artist, and was in a position in 1830 to remedy many defects in style which had crept into his fledgling work.
Indeed, so thorough were Cooper’s revisions for Spy that certain assumptions of the Greg theory of editing are challenged. Cooper does in fact pay unusual attention to the accidentals as well as substantives. At numerous points he inserts punctuation marks, especially commas, alters spelling, or makes the kind of small-scale changes — “which” for “that” — which editors might well consider compositorial rather than authorial changes. On a larger scale, he frequently prunes passages which identify speakers in dialogues and which characterize their emotions at length, leaving the dialogue itself to establish mood. Excessive sentiment, as in the death of Isabella Singleton, is dropped. At many places he emends dialect expressions for characters like the Irish washerwoman Betty Flanagan — often repairing the damage of earlier compositors who had ignored his intentions for dialect. Perhaps to prepare the new edition to meet its British audience, he replaces many instances of vulgarity or crudeness of expression among the Americans with normal forms; and he also mutes some of the more anti-British expressions. Colonel Wellmere, whose attempted bigamy drives the heroine’s sister insane, is no longer simply referred to as the “Englishman,” as if that term were synonymous with “villain,” This is not to say that Cooper submits to what he conceives as the nationalistic expectations of the British: footnotes are often added to explain or justify American customs.
Figure 3.
One of these, for example (see Figure 3), explicitly compares the death of the spies Nathan Hale and John Andre. Cooper tells his British readers that the American Hale died for his country amidst the mockery of the British, while the Americans wept for grief even as they executed the British officer, whose only motive was personal ambition.
Cooper does not forget, though, that he’s revising a novel and not writing a social tract. Late in volume two he cancels a long passage of verbose description about the roads of Westchester country, scene of the Neutral Ground of the novel. Responding to the mockery of American institutions, especially modes of transportation, which filled the pages of many British travellers’ accounts of the new country, he at first composed a facetiously sarcastic attack on such misrepresentations. As his comments grow longer and more irritated, he broke off and cancelled the revision. No doubt he recognized that the petulant tone his description was taking on was out of place in a novel; in any case, the Notions of the Americans (1828) had already provided a full account of the country’s glory, for the purposes of disabusing Europeans of their willfull ignorance about America. In place of both passages went a brief, objective narrative transition.
Of great interest too is his revision of the characterization of the pedlar-spy hero, Harvey Birch. Though he still is made to use occasional dialect forms appropriate to his class, often his grammatical lapses are remedied. His motivation — a desire to serve his country known only to Washington himself — is clarified by eliminating references to emotional unbalance. And while his recognition of the personal price he must pay to continue as Washington’s chief secret agent remains clear, the occasional handwringing pathos of his loneliness is muted. For example, in the first edition, he bursts upon the wedding of Wellmere and Sarah Wharton as follows:
It was the pedlar. His sunken and cowering eye no longer avoided the looks of others, but glared wildly around him; and his whole frame was agitated by an exertion that had shaken even his iron nerves. But all these emotions passed away like shadows from a fleeting cloud, and assuming a look of deep humility, and habitual respect, he turned to the bridegroom and bowing low, said —
This unlikely medley of moods is replaced with:
It was the pedlar. His look was bitter and ironical, while a finger, raised toward the divine, seemed to forbid the ceremony to go any further.
The revision shows Harvey in control of the scene, not awash in his own conflicting feelings as in the first edition. Later on the same page, a reference to his appearing “wild and agitated” is simply dropped (see Figure 4).
Figure 4.
The value to our knowledge of Cooper which these revisions have goes beyond their assistance in establishing a sound text of The Spy. Beginning with Mark Twain’s famous attack on Cooper in 1895, critics have often ridiculed his use of the language. Our work on The Deerslayer, which Twain singles out for special abuse, has shown that in every case but one in Twain’s notorious list of Cooper’s alleged misused words, Cooper was in fact using the word or phrase accurately and in a sense allowed in the early nineteenth century. (The one unlocated phrase, “funeral obsequies,” does not occur in The Deerslayer despite Twain’s fun with it.) Now the work on The Spy shows in rich detail that Cooper was a thoughtful as well as careful reviser. He worked painstakingly to reduce unneeded verbiage, enhance the drama, and clarify the motivations of his characters, as well as introducing thousands of small repairs of errors. Cooper did care about the style of his prose; in The Spy he worked hard to improve his first efforts, in the early American editions and most extensively in the Bentley edition discussed above. His style is not the style of Twain, Hemingway, or the late twentieth century; rather it is a style formed on late eighteenth-century models, before the conventions of grammar and punctuation we all learned were fully formed. Realizing that Cooper was concerned enough with his writing to revise The Spy as often and as heavily as he did should help us to understand him better as an artist.
Notes
1. The pages from the autograph manuscript of The Deerslayer are reproduced with the permission of the Pierpont Morgan Library. The pages from The Spy are reproduced with the permission of the Henry W. and Albert A. Berg Collection, The New York Public Library, Astor, Lenox and Tilden Foundations.
