The texts of Typee, Omoo, and Mardi presented in this volume are those of the Northwestern-Newberry Edition of The Writings of Herman Melville, edited by Harrison Hayford, Hershel Parker, and G. Thomas Tanselle and published by the Northwestern University Press and The Newberry Library in 1968 (Typee and Omoo) and 1970 (Mardi). Those texts were prepared according to the standards established by—and they have received the official approval of—the Center for Editions of American Authors of the Modern Language Association of America. (See its Statement of Editorial Principles and Procedures, revised edition, 1972.)
The textual problems with which the Northwestern-Newberry editors have to deal in Typee are not characteristic of the textual problems that arise in Melville’s other books. In the first place, Typee is the only instance, among Melville’s books, in which the English rather than the American edition was the one set in type directly from his manuscript. John Murray’s London edition, which appeared in February 1846, was set from the manuscript, and the New York edition of Wiley & Putnam, published about a month later, was set from proof sheets of the London edition. Because the London text is the one closest to the manuscript (which is not known to survive), it becomes the basic text adopted by the Northwestern-Newberry editors, who then alter it so as to try to eliminate those features (such as English spellings) that had been introduced by the English publisher. There seems little likelihood that Melville had a chance to go over the proofs of the American edition, and the variant readings in it do not appear to be authorial.
The other difference between the textual situation offered by Typee and that of Melville’s other books is the existence of a revised edition subsequent to the original English and American editions. John Wiley, of the American publishing firm, was uneasy about Melville’s treatment of the missionaries in the South Seas and his references to political and sexual matters. As a result of pressure from the publisher, Melville made a number of revisions and excisions that eliminated or softened his comments in these areas. In this process he cut out some three dozen pages, consisting of all of Chapter 3 (except the first half of the first sentence), all of the Appendix, and substantial passages in eight other chapters and the preface (in this edition: 10.7–24; 11.17–27; 14.14–21; 14.31–17.17; 36.26–41.11; 149.12–151.14; 200.13–202.15; 211.34–212.6; 214.36–215.10; 217.7 note; 217.7–218.2; 219.37–220.15; 222.16–223.8, 10–13, 32–41; 224.24–26; 225.10–11, 19–27; 226.24–38; 227.11–33; 229.40–234.34; 235.4–236.5; 238.1–36; 271.17–272.12), in addition to many smaller excisions. The resulting “Revised Edition,” containing also a newly written postscript, “The Story of Toby,” was published in August. Murray in London added “The Story of Toby” to his edition but did not otherwise alter the text—with the result that, throughout Melville’s lifetime, the version of Typee commonly available in England was unexpurgated, whereas the one available in the United States was the “Revised Edition.” In preparing the revised version, Melville made some changes of other kinds as well, and he professed to be pleased with the overall result, claiming in a letter to Murray on July 15, 1846, that his deletions gave “a unity to the book which it wanted before.” This statement, however, would seem to be a rationalization: there is no doubt that the impulse to bowdlerize was the publisher’s, not Melville’s; and Melville in his next book, Omoo, does not exhibit a changed attitude toward missionaries. The Northwestern-Newberry editors therefore do not accept Melville’s revisions dealing with religious, political, and sexual matters, arguing that in these instances he was acting under external pressure; they do, of course, accept into the text his revisions of other kinds that appear to have been made independently. Thus the text presented here is the full, not the shortened, text of Typee, incorporating only those revisions that one can argue Melville made of his own choice and not simply because he was acquiescing to his publisher’s demands.
Omoo and Mardi have simpler textual histories, conforming to a pattern that occurs several more times in Melville’s career. In each case the American edition was set from manuscript and the English edition from proofs of the American. It was the English edition, however, that was actually published first. In the uncertain climate of international copyright in the 19th century, many English publishers of American works felt on stronger ground if their editions preceded the American editions. As a result, Omoo was published by Murray in London at the end of March 1847 and by Harper & Brothers in New York at the beginning of May; and Mardi appeared in Richard Bentley’s London edition in mid-March 1849 and in the Harper edition in New York about a month later. The fact that the English editions came out first does not of course alter the fact that they are one step further removed from the manuscripts than are the American editions. Therefore the basic texts of these two works adopted by the Northwestern-Newberry editors are those of the American editions.
Whenever there is a difference between the American and English texts and both readings seem to be authorial (as opposed to those situations in which one of the readings is clearly erroneous or is almost certainly the work of someone in the printer’s or the publisher’s office), the question arises as to which reading supersedes the other. Theoretically either could be the later one: Melville could have made alterations on the proofs for England that he had not made on the proofs for America; or he could have continued to read over the proofs for America, and make revisions on them, after sending off one set of proofs to England. In the case of Omoo and Mardi, it seems that the proofs were sent to England promptly, but only in Omoo is there evidence suggesting that Melville may have made additional revisions after that time. There are in fact relatively few differences between the American and English texts and little reason in most instances to depart from the American text. The Northwestern-Newberry editors do correct some erroneous readings in the American text, but in general Omoo and Mardi present an editor with few textual problems.
The standards for American English continue to fluctuate and in some ways were conspicuously different in earlier periods from what they are now. In nineteenth-century writings, for example, a word might be spelled in more than one way, even in the same work, and such variations might be carried into print. Commas were sometimes used expressively to suggest the movements of voice, and capitals were sometimes meant to give significances to a word beyond those it might have in its uncapitalized form. Since modernization would remove such effects, this volume has preserved the spelling, punctuation, capitalization, and wording of the Northwestern-Newberry edition, which strives to be as faithful to Melville’s usage as surviving evidence permits.
The texts in this volume follow exactly the Northwestern-Newberry Typee (4th printing), Omoo (3rd printing), and Mardi (3rd printing), with one exception: in Omoo at 498.25–26 the reading of the original American edition—“considerately”—replaces the erroneous “considerably.” (That typographical error in the Northwestern-Newberry edition is to be corrected in future printings.) The present edition is concerned only with representing the texts of these editions; it does not attempt to reproduce features of the typographic design—such as the display capitalization of chapter openings. In this volume the reader has the results of the most detailed scholarly effort thus far made to establish the texts of Typee, Omoo, and Mardi. Errors corrected for the second printing of this volume: 470.19, revisted; 1325.3, had. Error corrected eighth printing: 796.16, by (LOA).