[begin page 389]

Mark twain's manuscript of The Prince and the Pauper has survived, but the secretarial transcription that evidently served as printer's copy and the proof sheets that the author saw and revised have both been lost. The textual history of the first American edition, published in Boston by James R. Osgood and Company on 12 December 1881, therefore rests chiefly on the evidence obtained from collations of the author's manuscript, the publisher's prospectus, and the two states of that edition. Additional collations have shown that the first Canadian edition and the three states of the first English edition derive in complicated ways from the first American edition; all later editions are also derivative and embody no authorial revision. This evidence, in conjunction with other documents and letters, makes possible a more detailed account than has previously been available of how Mark Twain saw his book into print.

On 1 February 1881 Mark Twain wrote Osgood from Hartford, “Suppose you . . . run down here & sign, & lug off the MS—which I finished once more to-day.”1 Osgood presumably went to Hartford; in any event he signed the contract for the book on February 9. Whether he actually returned to Boston with the manuscript is uncertain: the contract specified only that the manuscript was to be in his hands “as soon as practicable after the date of this agreement, and not later than April 1st next ensuing.”2 But in less than a month he did indeed have it, for on March 3 Mark Twain wrote to A. V. S. Anthony, who was in charge of hiring illustrators for the book, “Very well then, I do say 'go [begin page 390] ahead' to the artist who is ready to make a couple of drawings on approbation.”3

Mark Twain's manuscript probably became the house copy for the Osgood company, and may have been used by the illustrators as well, but it did not serve as printer's copy. In 1885, Mark Twain remembered that “Osgood had the Prince & Pauper copied, & sent the copy to the printer.”4 The manuscript has only a few minor notes and changes in another hand on the first pages, most of them added by someone in Osgood's office. Osgood himself added “Boston
James R. Osgood & Company” to the title page and moved “All rights reserved” from there to the copyright page.5

The printer's copy, now lost, was apparently a handwritten secretarial copy of the manuscript. Although it was Mark Twain's usual practice to revise printer's copy, he evidently had no opportunity to do so for The Prince and the Pauper before Osgood sent it to the printer: he confined his postmanuscript revisions to proof.

The typesetting of the first American edition was done by the Franklin Press: Rand, Avery, and Company, Boston, who also electroplated the illustrations and, later, the book itself.6 Although the author told one correspondent on 22 April 1881 that his book was “in press,”7 he doubtless meant only that the slow work of producing the illustrations had begun. The records of the Osgood company show that the first two “relief plates” were completed on April 21 and that these continued to be manufactured in batches until October 7.8 The type setting of the book itself may have begun in April, but it seems somewhat more likely that it was delayed until the company had prepared, and Mark Twain had approved, a sizable number of illustrations. On July 2 he asked the company to send him “proofs of the pictures that are thus far completed for my book,” and reminded them [begin page 391] that “Osgood promised, but Osgood forgot.”9 On July 31 Mark Twain told another correspondent, “Proofs of a hundred & fifty of the engravings for my new book came yesterday, & I like them far better than any that have ever been made for me before.”10 The next day, August I, he praised the pictures extravagantly to Benjamin H. Ticknor, Osgood's partner, and tried to arrange for special printings to be neatly bound in boards. At this time he had clearly not seen any proof sheets of the text, for he told Ticknor, “Put titles under the pictures yourself—I'll alter them in proof if any alteration shall seem necessary.” And on August 14 he again praised the illustrations to Ticknor but said (apparently in response to an inquiry from the publisher), “I hain't got no proofs, yet—but there may be some in the post office now.” The next day he had in fact received the first installment of proofs—an incomplete set which omitted several chapters and which contained, moreover, a problem with the illustration in chapter 1. Ticknor must have explained that this illustration had been made too large and would need to be redone—a step that probably required most of this short chapter to be reset. On August 15 Mark Twain wrote him, “Yes, that is the correct idea—do the cut over again; process it down to the required reduction.” He added that he would “have to wait till you send Chap 1 again, & then begin fair & read consecutively—can't begin in the middle of the book.”11

The illustration was eventually redone, but Mark Twain evidently could not resist the temptation to begin reading, even “in the middle of the book,” and he must have read chapters 1, 3, 5, and probably 7 as well, sometime between August 15 and 23. His reaction to the quality of the typesetting was mixed, and he set it forth in a detailed letter to Ticknor while Osgood was in Europe:

Shortly after writing Ticknor in this fashion Mark Twain wrote to Osgood, on August 23, and he was more than annoyed:

This letter makes it clear that Mark Twain was not able to read [begin page 393] consecutive chapters, at least in the initial stages of proof: he looked at chapters 1, 3, 5, and 7, and that “latter mess of God-forever-God-damned lunacy” finally enraged him. At some point before August 23 he evidently received the even-numbered chapters, or at least chapter 6—which he returned “unread,” presuming that the “godamded idiotic punctuating & capitalizing” was as bad there as in the odd-numbered chapters. Despite his tirades to Ticknor and to Osgood, chapters 1, 3, 5, 7, 8, 9, 10, 30, and 31 appeared in the book with their punctuation and word forms drastically altered from the manuscript readings. Chapters 2,4, and 6, on the other hand, are fairly close to the manuscript. It therefore seems likely that alternate chapters were set by two compositors, only one of whom had been carefully or persuasively instructed in the author's preference for his own styling. Mark Twain's protest about the early chapters was ineffective: if the compositors did reset them or follow his corrections on the proofs, they did not succeed in restoring the practice of the manuscript. Nevertheless, he did eventually prevail, and of the thirty-four chapters in the first American edition, twenty-five conform fairly closely to the manuscript in the styling of accidentals.

Mark Twain met with Osgood on August 25 in Boston, and the problems with the printers were probably ironed out at that time. On September 12 he took a trip to Fredonia, New York, to visit his “mother & the rest of that family.” He reported to Mary Mason Fairbanks on September 18 that when he had returned two days previously he had found “a stack” of proofs, “waiting to be read.” He had already read “⅔ of it in proof,” he explained, and he now supposed that his “labors on that work were about ended.”14 Presumably he finished reading and returned the final third of the proofs promptly.

Although there is of course no direct evidence of the revisions Mark Twain made on the proofs during the summer, comparison of the manuscript with the first American edition reveals a number of “literary lapses and infelicities” that were corrected at this time, some of them undoubtedly by the author. For instance, the first American edition substitutes “Hugo” for the manuscript “Hugh” at 241.3 and 241.22. Mark Twain had mixed up the names of his villains intermit-
[begin page 394] tently throughout the work and had already corrected the mistake several times in his manuscript—for example, at 240.5 and 242.5.

In fact Mark Twain generally seems to have continued on the proofs the process of revision that he had begun on his manuscript. He polished his dialogue and narration, often choosing the modern form of a word instead of an archaic one. In addition to the substitution in chapter 3 of “It is” for the manuscript “'Tis” (at 62.21, 64.13, and 64.37), which he mentioned in his letter to Ticknor, the first American edition substituted “since” for the manuscript “sith” at 105.30. Mark Twain had altered “sith” to “since” five times in his manuscript (at 93.11, 140.17, 143.1, 148.16, and 167.17) and was obviously making the same change on proof here.

Also consistent with his manuscript revision are several changes in italic and roman word forms. Mark Twain took great pains when marking for emphasis and often returned to his work to tinker with italics, especially in dialogue. There are eight such changes that probably occurred on proof (see the emendations at 69.14, 78.14 twice, 139.5, 139.15, 164.8, 229.19, and 235.24).

By October 7 the publisher's prospectus was printed and ready for distribution to the canvassing agents.15 It was made up of selected pages that would later appear in the first American edition, as well as four pages of descriptive material inserted by the publisher (including a price list), several ruled pages for subscribers' names, and samples of four bindings. Machine collation of the prospectus against the first American edition reveals that both were printed from plates rather than from standing type.

As Mark Twain indicated in his letter to Mrs. Fairbanks, he had considered his work on the book virtually finished when he returned the proofs to Osgood in late September. Nevertheless, a number of revisions were made in the plates sometime between the printing of the prospectus and the printing of the first American edition.16 Many [begin page 395] of these changes were suggested by William Dean Howells and Edward H. House.

Between September 11 and October 12, Mark Twain sent a set of proof sheets to Howells, who had been commissioned to write a review of The Prince and the Pauper, which appeared in the New York Tribune of October 25.17 Howells had already read the book in manuscript the year before, when Mark Twain gave it to him seeking his reaction and comments.18 On October 12, Howells wrote to him about the proofs:

And again on October 13 he wrote:

Mark Twain's response indicated that he did not think that Howells was meddling. On October 15 he wrote:

Despite Mark Twain's apparent willingness for Howells to make changes without consultation, Howells was actually sending him the proof sheets with queries or suggested alterations, and the author was making the decisions. He had often in the past asked Howells to read his works as an editor and as a friend; Howells had criticized The Adventures of Tom Sawyer before it was published, for example. And [begin page 396] just as he had done with Tom Sawyer, Mark Twain undoubtedly took some of Howells' suggestions, rejected some, and came up with new solutions in other instances.22

Of the problems that Howells specifically mentioned in his letter of October 12, “devils” was altered to “fiends” at 50.15, and “basting” was altered to “beating” at 115.6.23 The word “hick” mysteriously does not appear in the manuscript, the prospectus, the first American edition, or either of the two editions set from American proof sheets (the first English and Canadian editions). Mark Twain may have added it during his first proofreading and then taken it out again before the book was printed.

In the instances already mentioned Howells may have suggested the alternative readings that Mark Twain adopted. But Mark Twain apparently supplied his own new reading for the “thing . . . on p. 154” (p. 150 of this edition) which Howells couldn't cope with changing—the last line of the ballad that Miles Hendon sings (beginning at 148.8). The manuscript reads:

There was a woman in our town,
In our town did dwell—
She loved her husband dearilee,
But another man twice as well,—

In the first American edition the last line of the ballad was altered, in the plates, to read “But another man he loved she,—.”24

In order to identify the other revisions resulting from the “20 such” queries that Howells mentioned in his first letter about the proofs, and the unspecified “passages marked” which he wrote of in his second, we must rely on the physical evidence provided by the altered plates. Alterations in the plates have been discovered by two methods. The first is machine collation of the prospectus against the first American [begin page 397] edition with a Hinman collator. At 90.3, for instance, such collation revealed that “styes” was cut into the plates of the first American edition.

The second method is careful sight inspection of the first American edition pages. The way the type was cut in made many plate changes apparent, because slight differences in type size, alignment, and height often resulted in uneven inking. For instance, an inspection of the first American edition reveals that “magnificent array of” was cut into the plates at 58.11–12.

g bastions and turrets, the huge stone
and its magnificent array of colossal
signs and symbols of English royalty.
be satisfied at last? Here, indeed, was a

First American edition

About thirty plate alterations that were probably suggested by Howells have been identified.25 Although in the absence of the proofs [begin page 398] themselves it is of course impossible to determine their provenance with certainty, most of them reflect the sorts of concerns which Howells expressed in his letters of October 12 and 13. He undoubtedly considered the words “womb” at 96.9 and 267.4, and “misbegotten” at 152.9 and 175.37–38, “rather strong milk for babes,” for example. Other alterations seem aimed at ridding the book of American colloquialisms, and one of the few criticisms that Howells offered in his review of October 25 was that “the effort to preserve the English of Henry VII.'s reign in the dialogue sometimes wavers between theatrical insistence and downright lapse into the American of Arthur's Presidency.”26

Not long after the plate alterations presumably suggested by Howells had been ordered, House also saw a set of proof sheets and made suggestions that resulted in further alterations. House, like Howells, had read the book once before, in manuscript.27 When he saw the proofs he became concerned about an anachronism that he had apparently not noticed the first time: by calling the Hendons baronets Mark Twain was giving them a title that did not exist in the time of Henry VIII. His discovery set off a flurry of activity to find ways to correct or rationalize the Hendons' status without having to tear up the plates of the book once again.

House wrote of his discovery to Mark Twain, who replied on October 21:

[begin page 399]

In a letter to Osgood written the same day, Mark Twain proposed two possible footnotes to follow the words “My father is a baronet” in Miles Hendon's speech at 139.23–24. The longer of the notes, which he canceled before he sent the letter, reads:

Realizing that there might be difficulty in fitting such a long footnote onto the page, he also wrote a shorter version: “*I created all the baronets that occur in this book. My plates were electrotyped & ready for the press before it recurred to my memory that in England there were no baronets in those days.—M.T.”29

Adding a footnote meant that room had to be made for the new matter by deleting lines from the text. In the same letter Mark Twain indicated that he had “succeeded in providing the necessary room” for the shorter footnote, apparently by deleting “A grateful . . . him;” (139.10–11) and making the paragraph at 139.21 run-in. The deletion and the altered paragraphing stood in the first American edition, but he adopted another solution to the baronet problem, and a different footnote was used.

Before Osgood acted on Mark Twain's letter, House wrote proposing the new solution. House's letter included a list of changes to be made where the text referred to the Hendons' rank, and suggestions for at least two alternative footnotes. The author replied to House:

[begin page 400]

“Form B” was almost certainly the footnote that appears in chapter 12 (p. 139):

According to House's later statement, he was responsible for “nearly a dozen” changes in the text, made to adjust it to the Hendons' new status.31 In addition to the footnote, the changes were the substitution of the word “knight” for “baronet” (at 144 caption, 144.18, 145.5, 247.6, 247.9, 289.5, and 330.2), the substitution of “honors” or “show” for “title” (at 285.9 and 331.11), the substitution of “For” for “Sir” (at 266.20), and the insertion of the phrase “one of the smaller lords, by knight service—” following “baronet—” in Miles Hendon's speech at 139.24. Although most of these alterations were cut into the plates, collation of the prospectus against the first American edition reveals that the page with the footnote and the added phrase was entirely reset.

Mark Twain was very pleased with this solution. In his letter to Osgood explaining why he preferred “Form B” of House's footnote, he said:

Mark Twain made no mention of further revision in his correspondence after late October. But alteration of the plates continued even after printing had begun: a second state of the first American edition corrects three errors passed over in the preparation of the first state.33

The first American edition was published on 12 December 1881.34 Four impressions, totaling just over 25,000 copies, were made between mid-November and the end of December. By 1 March 1882 over 21,000 copies had been sold, and soon another impression of 5,000 copies was made.35 Most of the printing for these five impressions was done by Rand, Avery, and Company; the balance was done by John Wilson and Son.36 Sales declined drastically thereafter, for almost 5,000 copies remained unsold by February 1884, when Osgood ceased to be Mark Twain's publisher and all rights, stock, and other material were transferred to Charles L. Webster and Company.37

After becoming Mark Twain's publisher in 1884, the Webster company continued to issue the first American edition of The Prince and [begin page 402] the Pauper until 1891. At first the unbound copies acquired from Osgood were simply cased in Webster bindings with no change of imprint, but later when new sheets were printed the title page bore the Webster company name and the date of issue.38 The change of publishers produced no change in the text. All subsequent American editions derive from the second state of the first American edition and contribute nothing to the present text.

The first English edition, set from American proof sheets, was published earlier than the first American edition to ensure the English copyright. It was printed by Spottiswoode and Company, London, and published by Chatto and Windus on 1 December 1881.39 Mark Twain did not see the printer's copy, nor did he read proof. The text exists in three states, the first set from proofs sent from the United States in late September and October, the second corrected against a copy of the first American edition, and the third further corrected and styled apparently by a Chatto and Windus or press proofreader. Machine collation indicates that all three states of the first English edition were printed from standing type, into which corrections were introduced.40

Because the English edition had to appear before the American edition, it was important to coordinate their production. The English copyright of Mark Twain's preceding book, A Tramp Abroad, also typeset from American proof sheets, had been threatened when the first English edition was published later than the first American edition. A misunderstanding delayed the dispatch of electroplates for the illustrations, and over a hundred pages of text had not reached England when the American Publishing Company brought out the book in the United States in March of 1880 with no advance notice to Chatto and Windus. When the final pages did reach England, Chatto and Windus quickly printed an unillustrated two-volume “Library Edition” and later, when the electroplates came, followed it up with a more expensive illustrated edition. But Andrew Chatto felt that the American Publishing Company had seriously threatened the English copyright by publishing “without giving us sufficient notice” and [begin page 403] complained of the financial burden of having to make two distinct typesettings with no copyright guarantee.41

The following year, remembering all the difficulties with A Tramp Abroad and more concerned than ever about his copyright, Mark Twain wrote to Chatto on 7 October 1881 to assure him that the same thing would not happen with The Prince and the Pauper:

Osgood did send everything to Chatto and Windus in “ample time.” The publishers' records indicate that the first third of the proofs and duplicate electroplates for the illustrations had arrived by September 27. Chatto immediately placed an order with his printers for an impression of 5,000.43 He wrote to Osgood on the same day that they would issue an illustrated edition of the book first, instead of an unillustrated edition as they had done with A Tramp Abroad.44 He wrote to Mark Twain on November 1:

Evidently by the time Chatto wrote, the rest of the proofs had arrived, and the first impression was complete or nearly so, for soon after, on November 3, Chatto and Windus ordered a second impression of 5,000. On the last day of the month, the company ordered a third impression of 5,000, after which no further copies of the first English edition were printed.46

[begin page 404]

The earliest state of the first English edition incorporates the changes that Mark Twain made on the American proofs during the summer, but not all of the changes suggested by Howells in October, in particular those in the first part of the book. Spottiswoode and Company received the first installment of the American proofs and began setting type from them before all of the plate alterations had been made in the United States, but collation indicates that the printers must have received later proofs after they had been corrected or marked to include Howells' changes. In any case Mark Twain had nothing to do with the proofs forwarded to England, which were as a matter of course sent directly from the Osgood company.

However, during the flurry of activity over the American proofs set off by House's discovery of the baronet anachronism, Mark Twain did become concerned about the transmission of the alterations to England. On October 25, after sending House's list of changes to Osgood, he suggested that Osgood cable at least a footnote to Chatto and Windus:

Apparently Osgood did cable the baronet footnote and all of House's related changes to England, for even the earliest state of the first English edition shares these readings with the first American edition.

On November 18 Ticknor informed Mark Twain that he had “mailed Chatto a complete book so that he can look the whole thing over.”48 This copy of the first American edition was probably used by the Chatto and Windus editors to correct the early printing of the first English edition and resulted in the second state. The second state differs from the first in ten substantive readings, nine of which are [begin page 405] taken from the first American edition; for instance, “devils” becomes “fiends” at 50.15, and “make a” is corrected to “make” at 80.9.49 The English editors were not thorough, however, and did not alter in the English edition all of the substantive readings that had been changed in the first American edition; for example, the description of the beggars at 191.22, “diseased ones, with running sores peeping from ineffectual wrappings,” was retained throughout the first English edition, although it had been dropped from the American.

The third state of the first English edition reflects the efforts of a house editor or press proofreader who made further necessary corrections (“art” for “are” at 197.12), fussed with usage (“slowly” for “slow” at 59.1), and corrected for house style as well (“By-and-by” for “By and by” at 54.4). All three states of the first English edition contain a great many house conventions, as well as a number of sophistications and Anglicisms.

Although the first English edition has no primary authority, it is of interest because of its close relationship with various stages of the first American edition, which Mark Twain revised. No subsequent English edition in his lifetime has any bearing on the present text.

Like the first English edition, the first Canadian edition of The Prince and the Pauper was published earlier than the first American.50 It was an unillustrated edition brought out by Dawson Brothers, Montreal, to establish Canadian copyright. Printer's copy was again a set of the American proof sheets. Although printed in Canda, it was set and plated in the United States. The author did not read proof for the edition.

Mark Twain was determined to obtain Canadian copyright in order to prevent a Canadian piracy of The Prince and the Pauper. Piracies of his earlier books had been sold not only in Canada but in the United States, severely undercutting the sales and profits of the American subscription editions for which he received royalties. In late September, Osgood, having read the Canadian copyright act and corresponded with Samuel Dawson, “a thoroughly honorable man and the [begin page 406] most intelligent publisher in Canada,” suggested a plan whereby Mark Twain would put out an authorized edition in Canada, attempting to get Canadian copyright, but would meanwhile assign his imperial copyright to Chatto and Windus, who would be in a better legal position to fight infringers in case the effort failed. In addition, Osgood wrote, Clemens was to go to Canada for several days before and after publication, and they would “arrange to have an edition set up & printed in Canada at the proper time.”51

At first the author agreed: he wrote Osgood on October 2 to “go ahead and set up the types for Canada whenever you please.” But by October 27 he had begun to have second thoughts, having realized that “in setting up and printing in Canada, we run one risk—that the sheets may be bought or stolen, and a pirated edition brought out ahead of us.” He suggested as a solution that a signature here and there be left out of the Montreal printing until a few days before the Canadian publishing date. The next day he advanced another plan to Osgood whereby the first and last signatures would be typeset in Boston and the rest in Canada.

Finally, on November 1, he wrote Osgood, “Derned if I can think of anything to suggest except taking a set of plates to Canada to print from. If that will answer in place of setting up the book there, I should recommend that.—They wouldn't need to be electrotyped, but only stereotyped.” This idea was adopted; the book was set and plated in Boston by Rand, Avery, and Company, and the plates were sent to Dawson Brothers for printing on November 18.53

[begin page 407]

Clemens left for Montreal on November 26 and two days later wrote Osgood from Canada, “Have just returned from visiting Mr. Dawson. He has printed an edition of 275, and they are ready to be put into the paper covers.”54 Although he was greatly concerned about the way the Canadian edition was to be produced, and was in Montreal when it was printed, he had nothing directly to do with its production. Thus the text of the Dawson edition is without authority and is of interest mainly because of its close relationship with the publication of the first American edition.

Collation indicates that printer's copy for the Dawson edition was a late stage of the American proof sheets. Once it was designated as copy for the Canadian edition, it undoubtedly did not leave the house of Rand, Avery, and Company, who were simultaneously working on the first American edition. The composition was quite accurate, and the Dawson text closely resembles that of the first American edition. The Canadian edition lacks the illustrations and the Latimer letter frontispiece and transcription, but otherwise differs from the first American edition in only six substantive readings, two of which (“cornered” at 88.37 and “slums” at 90.3) are manuscript readings that also appear in the earliest state of the English edition, apparently having been changed for the American edition only in a very late stage of proof. The other four variants appear to be due to compositor error—for example, the substitution of “unchartered” for “uncharted” at 100.5.

One odd circumstance of the Canadian edition is that of the three readings that differentiate the first and second states of the first American edition, the Canadian edition shares one reading with the first state and two readings with the second state. Perhaps the compositor first noticed the need for the two corrections in the text as he was setting type for the Canadian edition, and as a consequence they were later, along with the third correction, cut into the plates of the first American edition.55

Although Clemens went to Montreal to establish residency, he was not granted a Canadian copyright. The copyright law required that he [begin page 408] be “domiciled” in Canada, which was interpreted to mean permanent and not merely temporary residency.56 Consequently, as he had feared, two pirated editions of The Prince and the Pauper appeared in Canada. The earliest, the Rose-Belford Publishing Company edition, appeared in early 1882.57 Sometime later in the year the second piracy appeared, published by John Ross Robertson. Both piracies derive from the Dawson edition and are therefore without authority.

The second American edition, set from a copy of the second state of the first American edition, was published by the Webster company in 1892. Clemens was in Europe at the time and had no involvement with the production of the new edition. His only concern with it seems to have been financial.58 Collation against the first American edition reveals only fourteen substantive variants, all of them probably due to compositor error.59

All subsequent American editions published in the author's lifetime derive from the Webster 1892 edition; Mark Twain had nothing to do with their production. The third American edition, called the “Library Edition,” was set from a copy of the second and published by Harper and Brothers in 1896. The fourth American edition, the last published in Mark Twain's lifetime, was set from a copy of the third. Issued in numerous impressions with varied imprints, the fourth American edition was variously called the “Autograph Edition,” the “Royal Edition,” the “Japan Edition,” the “De Luxe Edition,” the “Riverdale Edition,” the “Underwood Edition,” the “Hillcrest Edition,” the “Author's National Edition,” and so on.

Sometime after the “Autograph Edition” was printed, a marked copy of the “Royal Edition” of The Prince and the Pauper was used to correct the plates.60 It does not contain authorized revisions and corrects only those errors introduced into the text of the “Autograph [begin page 409] Edition.” Collation indicates that the proofreader must have drawn his corrections from the 1896 “Library Edition,” because errors that had first occurred in that edition were not corrected. Mark Twain seems to have been consulted only once, about whether to make corrections in the transcription of the Latimer letter frontispiece (see the textual note at 29.6–32).

The second English edition was ordered by Chatto and Windus from Spottiswoode and Company on the same day as they ordered the third impression of the first English edition.61 It probably derives from one of the later states of the first English edition. The third English edition was a Chatto and Windus resetting from a copy of the third state of the first English edition. First printed in 1891, it was initially called the “7/6” and later the “3/6” edition by the publishers. In 1900 Chatto and Windus offered a set of Mark Twain's works for sale by subscription. This set, called the “Author's De Luxe Edition,” was actually the 1899 American Publishing Company edition produced with a dual imprint. The Prince and the Pauper is volume 15 of this set. The last English edition published during Mark Twain's lifetime was printed in 1907 in an impression of 50,000 copies to sell for sixpence each.62

The Text

Modern editorial theory stipulates that a critical text must place before the reader not only the text itself but the evidence and reasoning used by the editor to establish it. As a first step the editor designates a copy-text, the form of the text to be edited—usually the manuscript [begin page 410] or first printing—which, because it is the least corrupt, provides the most satisfactory basis for establishing a text free from unauthorized readings.63 The editor agrees to follow the copy-text in every particular except where he considers emendation justified or required. And he agrees to report and defend all such emendations, so that a reader may if he chooses reconstruct the base from which the editor has departed. The copy-text therefore becomes the source for nearly every substantive and accidental reading in the critical text, and it largely determines the form of the textual apparatus used to report the editor's decisions.

Unauthorized changes made by copyists, editors, and compositors are by this means excluded—usually silently—from the text of this edition, while authorized changes in the printer's copy, proofs, or plates, along with simple corrections supplied by the editor himself, appear in the text as emendations and are so recorded. The copy-text for the present edition is Mark Twain's manuscript for The Prince and the Pauper. This copy-text has been emended in the following ways:

Substantives (Words and Word Order)

(1) Variants in the first American edition considered to be Mark Twain's changes in proof are here adopted. Suggestions and alterations made by William Dean Howells and Edward House and introduced into the first American edition were presumably approved by the author and are likewise adopted.

Authorial changes in proof may be detected by analogous changes demonstrably made by Mark Twain in his manuscript (such as the change of “to't” to “to it,” and “sith” to “since”), by documentary evidence (such as his letter to Benjamin Ticknor about changes of “ 'tis” to “it is”), or by their length and content—criteria which make it unlikely that a compositor or editor had ventured to risk the author's wrath by altering his work (such as the omission of manuscript passages at 191.22 and 237.22–23 and the substitution of “bakeries” for “bookstores” at 132.7).

Letters establish that Howells and House suggested numerous changes that Mark Twain solicited and then adopted, presumably in [begin page 411] proof. Howells' substitution of “beating” for “basting” at 115.6 and House's alterations from “baronet” to “knight” at 144.18 and 145.5 are typical.

(2) Variants in the first American edition that correct simple errors in the manuscript are adopted here. These include corrections of tense and agreement (such as “ordered” for “order” at 328.5 and “houses” for “house” at 49.11), of omitted words and dittography (such as “to and fro” for “to fro” at 107.10 and “after” for “after after” at 324.13), and of misidentification (such as “Hugo” for “Hugh” at 241.3). These corrections would be adopted in any case, but their appearance in the first American edition indicates that the author himself may have supplied them.

(3) Variants in the first American edition that apparently result from errors in transcription or from editorial sophistication are rejected. When it is possible to compare the copy-text with the prospectus, the first American edition, the first English edition, and the first Canadian edition, precise discrimination about even very small variants is possible. For instance, the manuscript reading “splatter” is rejected in favor of the first American edition “spatter” (63.9), because the agreement of the manuscript with the first English edition shows that the manuscript reading was initially typeset correctly and remained unchanged at least through the stage of proof from which the English edition was set, and the change must therefore have been made at a relatively late stage of production, when only an author would think to alter his text.

(4) When Mark Twain transcribed material from a source he often adapted it to fit his text; for instance, he cut inappropriate references to Queen Elizabeth and altered verbs from the past to the present tense in his quotations from Hunt, pp. 143–145. He also made changes that did not materially alter the sense of the passage; for instance, he substituted “from” for “in” in the quotation from Trumbull at 340.7. The copy-text reading is preferred to the original source in every case.

Accidentals (Paragraphing, Punctuation, and Word Forms)

(1) A conservative policy regarding the accidentals of the copy-text has been followed. Old-fashioned spellings (“recal” and “pedlar”) have been retained. Mark Twain's punctuation has been emended [begin page 412] here and there to correct mechanical errors (omitted quotation marks in dialogue, periods instead of question marks), but otherwise has been respected, even when it appears idiosyncratic.

(2) On succeeding stages of his work, Mark Twain often revised italic word forms and exclamation points for emphasis, and such emphasis variants in the first American edition are here adopted as authorial revisions in proof. An exception to this policy is made for nine chapters in the first American edition (1, 3, 5, 7, 8, 9, 10, 30, and 31), because they are so drastically different from the manuscript. The presumption is that in these instances the author's alterations and the compositor's unauthorized ones are inextricably tangled. Moreover, Mark Twain's changes in emphasis there were made in a corrupt text in a vain effort to restore his manuscript punctuation. For example, in chapter 3, where Mark Twain had to turn his “whole attention to restoring” his punctuation, the following emphasis variants occur:

Manuscript: O, he was a prince! a prince! a living prince,
First American edition: Oh! he was a prince—a prince,
a living prince, a real prince, without the shadow of a question, and the prayer of
a real prince—without the shadow of a question; and the prayer of
the pauper-boy's heart was answered at last!
the pauper-boy's heart was answered at last.

There is no profitable way to determine which of these alterations were Mark Twain's. In this and numerous similar instances the editor runs the risk of seriously distorting or misrepresenting the author's intentions, whether he adopts the whole set of variants or tries to extract Mark Twain's revisions from the compositor's. Thus in the nine chapters a conservative policy is followed, and the copy-text is the authority for emphasis.

(3) Mark Twain was not as careful about his spelling, capitalization, and hyphenation, as he was about his punctuation. His work contains outright errors (such as his habitual misspelling “sieze”) and lapses stemming from haste or carelessness (such as the omission of a letter in “straigtway” at 90.16–17). Moreover, he was often pointlessly inconsistent in such matters. He found the chore of hunting down and changing such inconsistent forms distasteful and expected it to be performed by the editors and compositors of his published works. In the manuscript of The Prince and the Pauper, he did make an effort to [begin page 413] do some of this type of correcting; for instance, he went over his manuscript and numerous times added an apostrophe to the word “an” (meaning “if”). He nevertheless left many inconsistencies. Because he expected others to smooth the formal texture of his work, and because inconsistencies can be distracting to a reader, emendations for consistency have been adopted whenever retaining the inconsistency would serve no conceivable purpose and the author's preference is discernible. As a rule, the resolution of inconsistencies is guided by his preponderant usage throughout the work. Out of well over a hundred references to the prince, for example, Mark Twain capitalized only seven times, all in the early pages of his manuscript. The frequency of occurrence within the book is sometimes inconclusive, and in these cases reference has been made to other writings of the same period. And in a few instances, as with the spellings “beggar-boy” and “beggar boy,” Mark Twain seems to have been utterly indifferent, and the editor's choice is essentially arbitrary. In every case, however, the form chosen has the warrant of the author's usage.64

(4) Mark Twain's manuscript contains numerous instances of a device characteristic of his manuscripts of the 1870s and 1880s: when the end of a sentence fell short of the right margin of the page, and there was not enough room on the same line for the following word, he often inscribed a dash to fill out the line. The device was probably a holdover from Clemens' days as a printer—a translation into hand-
[begin page 414] writing of a common newspaper technique for justifying a line following terminal punctuation (and for signaling that no paragraph break was intended). Newspaper compositors would often fill in such spaces with a dash rather than respace the line. The amanuensis who transcribed Mark Twain's manuscript for the printer must have copied at least some of his end-line dashes after terminal punctuation, for a few are preserved in the first American edition. The compositors evidently interpreted such dashes as a justifying device, for almost none of them were typeset. In fact, in five of the six instances where the compositors typeset them in the main body of the text, they probably did so for their own convenience: these five dashes occur at the ends of lines in that edition as well (at 150.21, 164.16, 205.4, 320.30, and 328.32). The sixth instance, at 226.4, is discussed below.

It is not always possible, however, to interpret manuscript end-line dashes after terminal punctuation as a mere justifying device of no further significance. For example, Mark Twain used dashes following terminal punctuation to separate his historical notes from their source citations (pp. 337–342). In addition, he sometimes used them within a manuscript line to represent a pause or a continued thought, or to link a question and response (for instance, at 318.9–10 “Was it round? —and thick?—and had it letters and devices graved upon it?—Yes?” and at 263.8–9 “My father dead!—O, this is heavy news”). The decision about whether to retain Mark Twain's end-line dashes must therefore take into account their literary significance.

Three categories of end-line dashes following terminal punctuation have been identified. The first and by far the largest category comprises all instances in which manuscript end-line dashes clearly have no rhetorical or stylistic function. In these instances, the dashes have been rejected in this edition as superfluous.65

[begin page 415]

The second category comprises instances of dashes in Mark Twain's historical notes. The dashes in this section, some of which occur at the ends of manuscript lines following terminal punctuation, are retained in this edition (as they were in the first American edition) because Mark Twain used them to separate the text of each note from its source citation.

The third category comprises sixteen doubtful cases of end-line dashes following terminal punctuation whose significance is ambiguous. In these cases, the dashes occur in passages of dialogue and internal monologue, where Mark Twain's punctuation tends to be particularly idiosyncratic and rhetorical. The first American edition printed a dash in only one of these instances, at 226.4 (but changed the following word from the manuscript “Here” to “here”). The present edition retains the dashes in six cases in which they seem to serve an identifiable literary purpose,66 but rejects them in the remaining ten as superfluous.67

(5) When Mark Twain interlined revisions in his manuscript, he sometimes inserted new punctuation without deleting the original punctuation. For instance, at 329.9 he wrote “for?”, inserted a caret between “for” and the question mark, and interlined “—who shall solve me this riddle?” above, inadvertently leaving two question marks. Similarly, he sometimes inserted a new word at the beginning of a sentence without changing the capital letter of the word that originally began the sentence to a lowercase letter. For instance, at 326.26 he wrote “He,” added “So” in front of it, and left standing the capital H. In order to avoid excessive listing of these mechanical emendations, such cases of double punctuation and capitalization are reported only in the list of alterations in the manuscript.

(6) In addition to the superfluous end-line dashes and instances of double punctuation and capitalization just discussed, a few mechanical changes are made without notation in the list of emendations.

[begin page 416]

a. Mark Twain's ampersands are expanded to “and.”68

b. Superscript letters are lowered to the line.

c. Mark Twain's chapter headings have been standardized to “CHAPTER” followed by an arabic numeral (in the manuscript Mark Twain designated chapter headings with a variety of abbreviations in upper or lower case); periods and flourishes following headings have been dropped.

d. The headings to Mark Twain's notes (for instance, “Note 1.—Page”) which he varied in minor ways in the manuscript have been standardized to follow the first American edition, and the page numbers of the present edition are silently supplied.

e. The opening words of each chapter appear in small capitals with an ornamental initial letter as an editorial convention.

f. Punctuation following italic words is italicized according to the usual practice, whether or not Mark Twain underlined the mark of punctuation.

Moreover, because it was Mark Twain's intent that they be published as part of his book, the table of contents, chapter titles, list of illustrations, and illustrations and their captions are adopted from the first American edition, although they are styled to accord with this edition.69

V.F.