Autobiography of Mark Twain, Volume 1: The Complete and Authoritative Edition. Edited by Harriet Elinor Smith et al. Berkeley: University of California Press, 2010. 736 pp. $45. ISBN 978-00520-26719-0.

Autobiography of Mark Twain, Volume 2: The Complete and Authoritative Edition. Edited by Benjamin Griffin, Harriet Elinor Smith et al. Berkeley: University of California Press, 2013. 776 pp. $45. ISBN 978-00520-27275-1.

With two of three projected volumes now in print, the University of California Press edition of Mark Twain's autobiography has been a publishing phenomenon with few parallels in the world of scholarly editing. Although the first volume contains little material not already published, it shot to New York Times bestseller status in 2010, remained there half a year, and ended up selling more than a half million copies. That is an astounding figure for any volume of edited documents and is even more impressive when one considers that an online version of its text has been available for free since the book came out.

The impressive sales of the first volume doubtless had much to do with public fascination with Mark Twain, arguably the most revered and widely read of nineteenth-century American writers. Many readers apparently expected the volume to be filled with shocking material that had been long suppressed. That was a mistaken assumption that had grown out of the story that Mark Twain did not want his autobiography published until a century after his death. The story was basically true, but his wish was not honored.

Many things connected with Samuel L. Clemens—better known to the world as Mark Twain—are complicated and full of surprises, and that is certainly true of his autobiography. To begin with, what he called his "autobiography" is not what most people mean by that term—namely an introspective history of one's life related in chronological order. Clemens consciously rejected that approach in favor of a discursive collection of reminiscences of events and acquaintances from his past, arranged in order of the moments when he recorded them. Moreover, he wanted his autobiography to "become a model for all future autobiographies" and be admired "because of its form and method . . . whereby the past and the present are constantly brought face to face, resulting in contrasts which newly fire up the interest all along, like contact of flint with steel." Spontaneity was one of his goals, and he predicted he would never run of out subject matter:

I shall talk about the matter which for the moment interests me, and cast it aside and talk about something else the moment its interest for me is exhausted. It is a system which follows no charted course and is not going to follow any such course. It is a system which is a complete and purposed jumble—a course which begins nowhere, follows no specified route, and can never reach an end while I am alive.

Not surprisingly, the resulting "jumble" has not been universally admired. When a large portion of it was published in book form in 1924, a critic for the Dallas Morning News said it was

surely the most strangely put together of all memoirs. It is a work of shreds and patches, a hodge-podge of miscellaneous papers and dictations written in various parts of the world . . . and in as many different manners as could well be found in a book not an anthology. Still worse, there is no sort of chronology; each new chapter begins with whatever subject happens to come into the muser's mind.

What concerns us here, however, is not the merits of Clemens's method, but rather its relevance to the challenges faced by his editors, of whom there have been many over the years (including the present reviewer).

Although Clemens began consciously experimenting with autobiographical writing when he was in his early forties, he did not get down to serious composition until three decades later, by which time he had settled on dictation to stenographers as his preferred method of composition. During a two-year period beginning in early 1906, he conducted a series of about 250 dictation sessions that produced the 500,000-word manuscript he regarded as his formal autobiography. He had written other autobiographical pieces earlier but did not regard them as parts of that autobiography. All published versions of the autobiography go back to the pages transcribed from the dictations, but several versions have also incorporated parts of the other material.

One of Clemens’s goals was to be completely candid in the descriptions and opinions he expressed. To help make that possible, he wished to delay publication of the autobiography until a century after his death, so he would be, in a sense, "speaking from the grave." He eventually concluded that complete honesty was still impossible but nevertheless remained determined not to have his autobiography published in book form until long after he died.

Clemens was serious about delaying publication of his autobiography, but he was also a pragmatist who recognized the work had a pecuniary value. He needed money, so he assented to having selected passages published in magazine form even before he had completed his dictations. In September 1906 the fortnightly North American Review, edited by George Harvey, began serializing twenty-five installments under the title "Chapters from My Autobiography"; the installments were later syndicated in newspapers. Selected by Harvey with Clemens's help, those articles are the only parts of Clemens's autobiography published under his direct supervision.

Part of the impetus behind Clemens's undertaking to dictate his autobiography in 1906 was the entry into his life of Albert Bigelow Paine, an author and magazine editor who asked to write his biography. Impressed by Paine's recent biography of cartoonist Thomas Nast, Clemens not only assented to Paine's request but also invited him into his home so he would have ready access to his papers. At Paine's suggestion, he hired a stenographer to transcribe his dictations and welcomed Paine to sit in on the sessions, in which Paine helped provide focus. After Clemens died in 1910, Paine became his literary executor. As sole editor of what became known as the Mark Twain Papers, he exercised almost complete control over the voluminous manuscripts, including the dictations. Through the remaining quarter century of his own life, Paine would go on to publish a monumental biography of Clemens, several shorter biographies, and edited collections of Clemens's letters, speeches, notebooks, and previously unpublished and uncollected literary works.

Perhaps the most notable of Paine's later editions was Mark Twain's Autobiography, a 195,000-word, two-volume work published by Harper in 1924. It incorporated about 40 percent of the dictations, including many portions that had already appeared in the North American Review. It also included scraps of Clemens's earlier attempts at autobiography. Paine more or less adhered to Clemens's wish to arrange passages in the order in which they had been dictated, but he also included material Clemens had not considered part of his autobiography proper. Moreover, as was typical of Mark Twain books Paine edited, he made little effort to explain how he selected the material, what was left out, or what editorial principles he applied to it. Among the omitted portions were diatribes that might have offended Clemens's targets or their descendants.

Paine's introduction assured readers "that positive mistakes of date and occurrence have been corrected" but said nothing about the numerous substantive changes he made in the texts. So far as anyone knew, Mark Twain's Autobiography was definitive because Paine allowed no one else to see Clemens's original manuscripts.

After Paine died in 1937, Bernard DeVoto succeeded him as editor of the Mark Twain Papers. One of DeVoto's first priorities was to publish parts of the autobiography Paine had omitted. Selecting only what he regarded as the most interesting material, he published the 110,000-word volume Mark Twain in Eruption: Hitherto Unpublished Pages About Men and Events in 1940. This book included some material that had appeared in the North American Review but nothing from Mark Twain's Autobiography. In contrast to Paine, DeVoto ignored Clemens's preferred arrangement, instead organizing passages under topical headings such as "Theodore Roosevelt," "Andrew Carnegie," "The Plutocracy," and "Various Literary People." As these titles might suggest, the book included many of the vitriolic passages about people Paine had left out.

By DeVoto's estimate, his and Paine's editions used three-quarters of Clemens's dictations. In 1959 a third editor, Charles Neider, published The Autobiography of Mark Twain, a completely reorganized volume that incorporated material from all earlier editions, adding about 30,000 words not previously used. Neider arranged the material in the traditional form of a cradle-to-grave narrative. This was the antithesis of what Clemens had wanted, but it pleased most readers and has remained in print almost continuously for nearly six decades. In 1999 the board of the Modern Library ranked it as one of the greatest nonfiction books of the twentieth century.

Additional autobiography editions that appeared in later years were mostly reprints of the old North American Review articles. The most notable of these was edited by Michael Kiskis, who titled his edition Mark Twain's Own Autobiography (1990 and 2010) to differentiate it from earlier editions that Clemens did not oversee. The 29-volume Oxford Mark Twain edition of 1996 included a volume titled Chapters from My Autobiography, which contains facsimile reproductions of the original Review pages.

After all these autobiography editions, the question now to be asked is what is different about the University of California Press's Autobiography of Mark Twain. The obvious first answer, of course, is that the volumes are being prepared with the same high level of scholarly exactitude that its editors at the University of California's Mark Twain Project apply to all the volumes they edit. None of the earlier editions adhered to standards remotely comparable. Indeed, DeVoto even boasted about having modernized Clemens's punctuation by deleting thousands of commas and dashes.

Given the large portions of the autobiography that were already published, the first two volumes of the California edition do not contain a great deal not previously seen. What makes their text more important is our knowing that we are reading Clemens's precise words, not a timid or opinionated editor's approximation of them. Most of the edition's corrections are technical, but occasional zingers jump off the page. An example is a scathing remark about James W. Paige, the inventor of a typesetting machine in which Clemens invested and lost a large fortune. Paine's Mark Twain's Autobiography quotes Clemens as saying, "Paige and I always meet on effusively affectionate terms, and yet he knows perfectly well that if I had him in a steel trap I would shut out all human succor and watch that trap till he died." The new edition's corrected text reads a little differently: "he knows perfectly well that if I had his nuts in a steel trap I would shut out all human succor and watch that trap till he died."

In sharp contrast to the earlier autobiography editors, those of the Mark Twain Project follow several bedrock principles. The first of these is to publish Clemens's works as close to what he intended as possible. The trick, of course, is knowing exactly what he intended. This is especially difficult to know in the case of the autobiography, as he left few explicit instructions for it. One clear instruction, however, was to publish the autobiography a century after his death, which occurred in 1910. The Project complied with that wish. That obvious issue aside, Clemens's intentions always include such matters as leaving his punctuation and word choices as he wrote them. They also include publishing his works in their entirety and not omitting anything he himself did not mark for deletion. This principle alone sets the California edition far apart from its predecessors.

Another bedrock principle of the Mark Twain Project is achieving complete transparency. Through various methods, its editors make it possible for readers to understand virtually every editing decision made in the texts. This strategy ties into the Project's overriding goal to ensure that future scholars will never need to edit the same texts again. Whenever anything is left out of a text or an alteration is not explained, readers cannot be sure how the printed text differs from the original. Readers of the Project's autobiography need not wonder about such questions, as nothing is ever left out, and Clemens's writing is never altered without clear and full explanations.

Though it may not quite rank as a bedrock principle, another of the Project's aims is to provide all its complex editorial apparatus in ways that do not interfere with reading the texts. That requires leaving pages uncluttered and putting annotations and editorial explanations elsewhere—either in appendixes or on the Project's website.

A unique challenge to the editors of the autobiography has been determining the authority of multiple typescript versions of the dictations. Four typescript copies of almost every session have survived, so it has been necessary to determine the sequence in which each was typed to know which of them reflect Clemens's final intentions. The problem is complicated by handwritten notes made on the pages by stenographers, Clemens himself, and later generations of editors. In her excellent introduction to the first volume, Harriet Elinor Smith discusses how the Project editors learned to decipher the typescripts, supporting the discussion with numerous facsimiles of sample pages. The Project's website at www.marktwainproject.org contains detailed explanations of every sequence of typescripts for the autobiography's first two published volumes.

Despite placing most of the technical apparatus for the autobiography online, the printed volumes themselves are heavily loaded with explanatory material. In volume 1, for example, only 468 of the book's 736 pages contain Clemens's own writing. There are 58 pages of introduction and about 180 pages of explanatory notes. The rest of the pages contain technical notes, bibliography, and a detailed index. The proportions in the second volume are about the same, except that in the absence of an introduction, there are more pages of Clemens's writing. Also, because the first volume contains about 140 pages of Clemens's predictation writings, the proportion of dictated texts in the second volume is much higher.

Marvels of scholarly thoroughness, the endnotes concisely explain almost every subject in the texts that would benefit from additional information. Of particular interest are identifications of persons Clemens named or merely alluded to. Paine once described Clemens’s dictated reminiscences as bearing “only an atmospheric relation to history.” It is thus not surprising that many annotations address contradictions between what Clemens said and what is actually known about his life.

When the third volume of Autobiography of Mark Twain is published next year, all earlier editions will cease to have any value. Some readers, however, will find the new edition's three bulky volumes overly cumbersome. To meet their needs, the University of California Press is also publishing scaled-down paperback "reader's editions," shorn of much of the scholarly editorial apparatus and weighing less than a third of the hardback editions. This move will doubtless ensure that the Mark Twain Project’s unparalleled edition will reach the widest possible audience.