Textual Commentary


Mitchell's Attitude Toward Reveries of a Bachelor

Thirty-three years after he first published Reveries of a Bachelor (1850), Donald Grant Mitchell attempted to reconstruct its history in a new 1883 preface. Despite the stunning commercial and critical success of Reveries of a Bachelor, Donald Grant Mitchell seemed a bit embarrassed by--yet protective of--the book. As Mitchell tells the story in the 1883 preface, he began writing the book to prevent readers from discovering that he was the author of The Lorgnette, a series of satirical sketches of New York City that he was publishing under the pen name John Timon. By labeling Reveries "almost an accident," Mitchell showed his own need to detach himself from the work, to treat his most famous work as something that happened almost spontaneously, something that reflected youth culture more than it did his own intention ("A New Preface," 1883, xvi). By examining Mitchell's defensive story of the book's origins, we can gain insight into both how Mitchell approached authorship and how an editor should handle Mitchell's work. He makes his detachment from Reveries more explicit later in the 1883 preface, when he labels the work one of his "youthful whimseys" (xxi). Mitchell's rather dismissive attitude toward his text raises a crucial question: If Mitchell disclaimed responsibility toward authorship, then how should one approach the task of establishing intention? As the Historical Commentary and Historical Collation make clear, the text underwent a number of changes over between 1850 and 1907, mostly accidentals (punctuation, capitalization, and so forth) rather than substantives (word choice). What is one to make of the changes that occur as the volume is reprinted, even revised, in the 1870s, 80s, and 90s, several decades after its initial appearance? Can we assume that Mitchell authorized the emendations to new editions for which he wrote updated prefaces, or do these changes reflect choices made by Scribner's editors or printers?

In his prefaces to Reveries Mitchell does describe his attitude toward revision, but these statements are themselves vague and evasive. Moreover, his approach to revision shifts throughout the text's publication history, posing the problem of which intentions we should take into account. When a new edition of the text was issued in 1863, Mitchell wrote in the preface that he had determined not to make any changes, since they would destroy the text's youthful spirit:

My publisher has written me that the old type of this book of the Reveries are so far worn and battered, that they will bear no further usage; and, in view of a new edition, he asks for such revision of the text as I may deem judicious....I began the revision. I scored out word after word; presently I came to the scoring out of paragraphs; and before I had done, I was making my scores by the page.

It would never do. It might be the better, but it would not be the same. I cannot lop away those twelve swift, changeful years that are gone....I have determined not to touch the book. A race has grown up which may welcome its youngness....For me those young years are gone. (x-xi).
Such a declaration would seem to support the view that the 1850 first edition best reflects Mitchell's intentions. Twenty years later, in preface to the 1883 edition of Reveries, Mitchell echoed his sentiment that the book was an artifact of another time and mood, but with this version he acknowledged that he had made some emendations:
...when my publisher sends me the old sheets for revision, I am in the same quandry which beset me twenty years ago. I may make, and have made, a few verbal emendations--a little coy toning down of over-exuberance; and I have put here and there a short patch of homely words into the place of some garish bit of color: yet when I come to deal with the sentiment of the book, and to question its good balance, or lack of balance, I am even farther removed from the capacity for sound and fair judgment than twenty years ago. More than then--and by great odds, more--the book wears for me the illusions and the fleeting prismatic hues which bubbles always wear, and which youth is always used to blow, and to follow with eager eye, till the iridescence be gone, and the bubbles too! (4-5)

Once again, Mitchell detached himself from his work, as if it were a dream that he could not re-envision, rather than a text to be revised. In speaking of his "coy toning down," Mitchell was himself coy--does he mean that he corrected spelling errors? removed commas? rewrote entire sentences and paragrraphs? Without direct information from Mitchell or his publisher, we must answer these questions by analyzing the data provided in the Historical Collation.

Analysis of the Historical Collation

Versions collated

Given how many times Reveries was reprinted, one could spend several years collating the multiple versions of "Smoke, Flame and Ashes." For the purposes of this edition, however, I chose to collate those versions that seemed most significant to the text's history and that were available at the University of Virginia's library. Thus, I collated the sketch's first published appearance, which was in the September, 1849 issue of Southern Literary Messenger; its October, 1850 publication in Harpers New Monthly Magazine; its first appearance in book form in the first Scribners edition of Reveries of a Bachelor; the 1852 illustrated edition of Reveries; the 1869 Scribners Reveries, which reflects the changes the text underwent in the revised edition of 1863; the 1877 "new" edition of Reveries; the 1884 edition of Reveries, which reflects the emendations made in the 1883 revised edition; and an 1890 edition of Reveries. I also examined the text's 1907 appearance as part of the Edgewood Edition of The Collected Works of Donald Grant Mitchell, and found no variation from the 1890.

In collating these versions of the text, I have found over 500 variants scattered among 33 pages (according to the numbering of the 1850 first edition), most of them accidentals. By analyzing the Historical Collation, I have been able to give tentative characterizations of each version (I will offer fuller analysis of each version later in this essay):

Choice of copy text

Because no manuscript for the sketch has been located, an editor must carefully consider what to choose as copy text. For "Smoke, Flame and Ashes," three early versions of the text present themselves as the possible copy-text: the sketch's September, 1849 publication in Southern Literary Messenger, its October, 1850 appearance in Harpers New Monthly Magazine, and its initial publication as part of a book in Scribners Reveries of a Bachelor, which was published in December of 1850. Although W.W. Greg, whose "The Rationale of Copy Text" is a foundation for textual scholarship (Studies in Bibliography, Vol. 3, 1950-1951) would probably argue that one should use the Southern Literary Messenger sketch as the copy-text, since this version of the text would be closest to Mitchell's manuscript. Applying Greg's principles, such prominent editors as Fredson Bowers and G. Thomas Tanselle have chosen the first appearance in print as the copy-text, even when dealing with a work that shifts as it moves from periodical to book form.

But in the case of "Smoke, Flame and Ashes," the reliance on the first publication for copy text poses significant problems, since the sketch evolved as it made its way from magazine to book form. When we are presented with a work that originally appeared in magazine form but then was re-published as part of a book, can we automatically assume that the version closest to the manuscript is the first magazine appearance? How do we know that Mitchell did not re-copy and revise his work in preparing it for book publication? I believe that Mitchell did in fact submit a new manuscript of his work to Scribners-- not only for the practical reason that it would be difficult for him to write changes in the cramped margins of Southern Literary Messenger, but because it seems unlike him to submit a rag-tag stack of papers--part hand-written manuscript, part pages taken from a magazine--to Scribners for publication, given his desire to control the publication process as much as possible. Moreover, the substantive changes that the text underwent as it moved from Southern Literary Messenger to Harpers to Scribners indicate that he revised it in preparation for each new publication, further suggesting that he produced a new manuscript for the publication of "Smoke, Flame and Ashes" in the 1850 Reveries. I believe that this new manuscript would have been closest to his intentions.

For this reason, among others, I have decided to adopt the 1850 Scribners first edition of Reveries of a Bachelor not only as the copy text, but as the "best text." While the 1850 Reveries contains corrections and revisions that Mitchell made following the original publication of "Smoke, Flame and Ashes," it also preserves Mitchell's aim in presenting the bachelor narrator's "floating Reveries which have, from time to time, drifted across my brain" (First Preface ). To represent the narrator's dreaming, Mitchell often uses dashes and punctuation marks; indeed, his biographer Waldo Dunn commented that dashes characterized Mitchell's style. Consider, for instance, Mitchell's description of the bachelor-narrator listening in on his tenants' evening activities: "My tenant, meantime, in the other room, I can hear now and then,--though there is a thick stone chimney, and broad entry between,--multiplying contrivances with his wife, to put two babies to sleep" (17.5-8). Although the dashes after "then" and "between" are omitted from the Southern Literary Messenger Harpers, and 1877 versions of the text, they are present in the 1850, 1869, and 1884 versions. Another feature of Mitchell's style is his use of capitalization to make key words stand out. While Southern Literary Messenger describes how "Something...had suggested to me the thought of----marriage," both Harpers and the 1850 Scribners capitalize the "M" in marriage, giving additional weight to the word and emphasizing the bachelor's fascination with and terror of marriage (18.8).

Not only does the 1850 Scribners first edition best reflect Mitchell's style, but it includes corrections that were made to clarify or crystallize important moments in the text. Some of these changes were previously made in Harpers, but the 1850 Scribners includes further corrections that seem characteristic of Mitchell (supporting the theory that Mitchell wrote a new manuscript for the Scribners first edition). Both Harpers and the 1850 Scribners intensify a comment found in the Southern Literary Messenger, when the bachelor narrator pledges to "pursue the thought wherever it leads, though it lead me to the ---". Later versions of the text place a "d" before the dash, making it more explicit that the oath involves the devil (18.13). Likewise, the 1850 often adds punctuation, bringing the text more into accord with Mitchell's stop-and-start style and making his meaning clearer. Whereas the Southern Literary Messenger includes the sentence "Not beauty fading;--that now that your heart is wrapped in her being would be nothing," both Harpers and the 1850 Scribners wrap the appositive phrase "that, now that....her being, would" in commas (44.6). Most significantly, Mitchell revises substantives in both the Harpers and first Scribners edition. Consider the following example: "Your friend--poor fellow!--dies:--never mind, that gentle clasp of her fingers, as she steals behind you, telling you not to weep--it is worth ten friends!" (32.22). In the Southern Literary Messenger, Mitchell uses "all friends" rather than "ten," but apparently he thought better of placing the wife above all else (Jesus, the boss, etc.). Similarly, in describing the joy that children bring, Mitchell wrote in the Southern Literary Messenger that "flower, tree, sunlight, are all dead things," but he revised the phrase in Harpers and all succeeding versions, replacing "sunlight" with "gun" (33.17). Not only do these changes suggest that he was trying to make his style more sentimental and contemplative, but they show that he was involved in revising his text as it was re-published.

Since the Harpers October of 1849 version of the text includes many corrections also present in the Scribners first edition, one might ask why the earlier version of the corrected text is not adopted as the copy text. But the Harpers version seems to abide by house style in its attempts to standardize the text. For instance, Harpers corrects "ricketty," spelling it as "rickety" (17.25). Mitchell seems to habitually place the comma inside the parentheses, as in "country,)", but Harpers shifts the comma to the outside of the parentheses (17.11). Whereas Mitchell loads his text with dashes so that he can replicate the changing thoughts and dreams of the bachelor narrator, Harpers omits many of these dashes; compare, for instance, Harpers "I'll not flinch; I'll" to the Southern Literary Messenger and 1850 Scribners "I'll not flinch;--I'll."

Moreover, between September of 1849, when it was published in Southern Literary Messenger, and December of 1850, when the first Scribners edition was released, the text evolved, as Mitchell polished the sketch. For instance, Mitchell seems to have experimented with different ways to open the first part of the sketch, "Smoke--Signifying Doubt" (19.1-4):

Southern Literary Messenger, September 1849
--Hum,--a wife! A wife?--hum!

Why? And pray, my dear sir, why not--why? Why not doubt? Why not hesitate? Why not tremble?

Harpers New Monthly Magazine, October 1850
Ay, a wife--thought I.

A wife!--and why?

And pray, my dear sir, and my gentle lady, why not--why? Why not doubt--nay, tremble?

Reveries of a Bachelor, Scribners, 1850
A WIFE?--thought I; --yes, a wife!

And why?

And pray, my dear sir, why not--why? Why not doubt; why not hesitate; why not tremble?

The most important change was to the title itself. When it was published in magazine form, the sketch was known as "A Bachelor's Reverie." But when it was included in Reveries of a Bachelor, a book that collected four reveries, the sketch was re-named "Smoke, Flame and Ashes." This renaming brings up an important issue for the textual editor, since it suggests that the nature of the text itself changes as it moves from book to magazine form. In my electronic edition, I have chosen to focus on the initial appearance of the text in book form, since this version was best known by American readers and reviewers and since it reflects the evolution of the text.

Justification for the facsimile edition

Even if we take the 1850 as the copy text, viewing it as the authority for the handling of accidental, we would still have to consider how to approach substantives. Given Mitchell's vague statements in his prefaces, it is difficult to determine to what extent he was involved in making changes to the text; we have only his coy remark in 1883 that he toned down excesses of language. Even if we could establish Mitchell as the authority for particular emendations, we would have to ask whether the 1883 Reveries of a Bachelor were the same as the 1850. If Mitchell regarded Reveries of a Bachelor as the product and the reflection of youth, then wouldn't it in fact violate his intention to incorporate an older man's change, to age the text by bringing it into the 1880's? Therefore, I have decided against emending the text, and instead I am presenting "Smoke, Flame, and Ashes" as it appeared in 1850, when it was first published in book form. I base this decision not only on Mitchell's attitude toward the text, but on careful examination of the data collected in the Historical Collation. In making this decision, I have paid special attention to two post-1850 versions of the text: the 1877 "new" edition, in which a number of variants are present, and the 1883 "new and revised" edition, in which Mitchell claimed to have made revisions.

With the 1877 version of the text, it seems that an editor attempted to standardize punctuation and grammar, even when such changes detracted from Mitchell's contemplative style. For instance, we can see that an effort was made to hyphenate compound words, to place titles in quotation marks or italics, to regularize spelling according to American standards, and to remove dashes and commas. Thus, the 1877 adds a hyphen so that "twisted headed" becomes "twisted-headed" (22.26), puts Literary World in quotation marks (27.9), replaces "holydays" with "holidays" and "lett" with "let" (23.12; 19.13), leaves out the comma in "which, by strange metonomy" (22.18), and omits the dash in "And then,--to come to breakfast" (25.3). In many cases, the 1877 changes reflect those made when the sketch appeared in Harpers twenty-seven years before, suggesting that in both cases editors were attempting to make the text conform to standard grammatical practices and to suppress some of Mitchell's idiosyncracies. We can hypothesize that Mitchell did not approve of some of these changes, which aimed more for directness than for the meditative, meandering style that Mitchell cultivated. Between 1877 and 1884, some of the emendations made to the 1850 edition were restored. For instance, while both the 1850 and 1884 versions read "Then, she [the imagined wife] will be forever talking of her fortune; and pleasantly reminding you on occasion of a favorite purchase,--how lucky that she had the means," the 1877 version removes the dash before "how," shifting the emphasis from how the husband perceives his wife to how she talks to the husband (23.25). Likewise, both the 1850 and the 1884 describe "a man who loves his wife, as a wife should only be loved!", but the 1877 omits the comma (37.20).< Given my analysis of the 1877 Scribners, as well as Mitchell's declaration in the 1863 preface that he chose not to revise the text, I find it difficult to trust this version as a source of substantive emendations.

But with the 1884 Reveries, Mitchell included a new preface in which he stated that he had made "a few verbal emendations" to the text ("Third Preface"). But, as I have already argued, he viewed the text as an artifact of another era, so that these changes seem to violate the text's integrity. Moreover, some of the untrustworthy emendations from the 1877 version remain. Reviewers admired the book's sentimentality, calling it a "sweet and tearful product of the heart" (Arthur's Home Gazette). Mitchell himself claims in his 1850 preface the book contains "whimseys, and reflections" ("First Preface"). But in the 1884 version, some of these tears have dried up. Consider the following sentence from the 1850 version:

Shall a man who has been free to chase his fancies over the wide-world, without lett or hindrance, shut himself up to marriage-ship, within four walls called Home, that are to claim him, his time, his trouble, and his tears, thenceforward forever more, without doubts thick, and thick-coming, as Smoke? (19.12-20.4)
In the 1884 version of the text, the phrase "his trouble, and his tears" is replaced by "his thought," as Mitchell tones down the bachelor's emotional response to the prospect of marriage. Similarly, the 1850 describes the comfort that a wife can offer to her husband following his mother's death in this way: "But you are not homeless; you are not alone: she is there;--her tears softening yours, her smile lighting yours, her grief killing yours; and you live again, to assuage that kind sorrow of yours" (33.6-9). The 1884 version omits the phrase "her tears softening yours," softening the sorrowful tone. Of course, both of these examples also reflect the attempt to streamline the text by cutting out words and phrases. A similar change occurs when the phrase "reason with philosophy" is replaced in the 1884 version by "philosophize" (43.9). In my view, these changes in substantives diminish both the power of Mitchell's language and his intention in writing the text. Thus, I have decided not to adopt the substantive emendations in this edition.

In editing this text, I have been tempted to make some corrections, especially those that are reflected in later versions of the text. For instance, the 1850 includes "Will you have pleasant evenings at your home now" (48.10). Every other version includes a question mark at the end of the sentence, but in order to preserve the text as it first appeared in book form, I have decided to maintain the period at the end of the sentence. Likewise, both the 1849 Southern Literary Messenger and the 1850 Scribners include the word "recal," which both Harpers and the 1877 Scribners correct by adding another "l" at the end (18.4). Such idiosyncracies seem important to the flavor of the text, to Mitchell's attempt to use a refined, meditative style that both looks back to British essayists and forward to the new American language. In any case, such errors or oddities will not interfere with an intelligent reader's comprehension of the text. Rather than aiming for "correctness," I strive to represent the text as it first appeared in book form.

My interest in the first edition of Reveries is in part authorial, in part sociological. I believe that this version best represents Mitchell's initial intentions in writing "Smoke, Flame and Ashes," since it presents the corrections and changes the text underwent as it made its way into book form without mangling the text by including changes that Mitchell might have made from a more distant perspective. But the facsimile also is useful because it makes available the version most important to Melville, Hawthorne, Dickinson and other authors of the American Renaissance. Moreover, the facsimile would best serve scholars interested in reconstructing readers' responses to Reveries of a Bachelor, given that the Beinecke Library holds a number of letters written by fans to Mitchell during the 1850s, and that most of the reviews of the book came out in the early 1850s.

Even though I am presenting a digital facsimile of the first edition, I also want to make it possible for readers to re-construct the changes that the text went through. By doing so, they can get a sense of changing literary tastes, of different house styles, and of the unstable nature of a literary text. Therefore, I am experimenting with making variants available through the tagging of the text itself (so far this experiment has only been carried out for the first part of the text, "Over a Wood Fire"). Of course, readers can access the story of the text's changes in a more traditional way by consulting the Historical Collation.