by William Rossi

For Beth, Bob, Joe, and Will

Editors’ note—“Making Walden and Its Sandbank” originally appeared in The Concord Saunterer Vol. 30 n.s. (2022), 10–58. Significant differences between the essay as originally published and the essay as it appears here include the addition of relevant manuscript images and other illustrations; revisions to the transcriptions originally printed in an appendix and here placed facing their corresponding manuscript surfaces; and regularization of the terms used to describe Thoreau’s different types of revision, intended to distinguish the manuscript’s major versions more clearly from other, intermediate stages of composition.

Detail from Volume 6, p. 4 of HM 924, the manuscript of Walden.

With the exception of “Ktaadn and the Maine Woods” (1848), Thoreau’s current reputation as an environmental writer rests entirely on the work of his last decade, the second half of his career. Scholarly recognition, beginning in the mid 1980s, of the extent of his engagement and expertise in contemporary natural history as well as the distinctive character of the Journal (then being reedited from manuscript with restored Walden period notebooks coming to light), laid the groundwork for the ecocritical focus of so many studies that followed over the next two decades in the wake of Lawrence Buell’s Environmental Imagination and Laura Dassow Walls’s Seeing New Worlds (both 1995). While modified by shifts in literary theory and an ever-growing sense of urgency regarding environmental justice, race, and climate change, the impetus of this good work continues with a new generation of scholars whose studies have encompassed topics such as materialism and ineffability in the later Journal and “Ktaadn”; environmental time and experimental knowing in the unfinished Kalendar project; and the entanglements of embodiment, ecology, and race in the late natural history and abolitionist writings.¹ Surprisingly, though, we still lack sufficient understanding of the pivotal moment of Thoreau’s career, roughly from 1848 to 1852, out of which the mature thinker and stylist so carefully explored by these scholars emerged.

The period is usually framed by the commercial failure of A Week on the Concord and Merrimack Rivers (1849) and early 1852, when Thoreau seems to have renewed work on the Walden manuscript he is believed to have shelved. This transition, culminating in the re-creation of the book and the “reinvent[ion] of the writer,” is often associated with the dramatic transformation of the Journal that occurred in the interim (Walls, Henry David Thoreau 273).² Although perhaps not as “astonishingly productive” as he had been at the Pond and for the two years following his return to “civilized life again,” Thoreau was far from inactive during this period (Richardson, Henry Thoreau 154; Walden 3). Yet the motivation behind his mid-career shift—what made the pivotal period “pivotal”—remains something of a mystery. As the late Bradley Dean, a scholar and editor-without-peer of Thoreau’s manuscripts, used to say: “Something happened” (Brad favored 1851 as “the” moment), “but what?”

Clearly, as the post-1850 Journal testifies abundantly, whatever precipitated the shift involved a deep reorientation to place and a fresh determination to live what he had earlier characterized as a “natural life.”³ Largely because the transition also involved a heightened commitment to the study and practice of professional natural history, it has come to be described as Thoreau’s “turn” or, more strongly, “conversion to science,” a move influenced, Walls has argued, by his reading of Alexander von Humboldt in the early 1850s (Walls, Henry David Thoreau 273–312; Thorson, Walden’s Shore 273).⁴ In the now standard narrative, this crucial development leads to a renunciation of his earlier idealism, a rejection of Emersonian correspondence, and a resistance to viewing nature symbolically, thereby facilitating the emergence of a recognizably modern Thoreau out of the cocoon of American Victorianism.⁵ But if “conversion” misleadingly suggests an abrupt change of direction that overlooks Thoreau’s decade-long poetic investment in scientific theory, there is no denying the pronounced empiricism and greater environmental density of his writing that we find registered in the post-1850s Journal as well as in additions to the later Walden chapters drafted at this time.⁶

As it happens, an extraordinary event recorded in the spring 1848 Journal—a substantial draft of the famous vision of a thawing sandbank in the Deep Cut—sheds bright light on this pivotal period. Although underscored in the “Historical Introduction” to Princeton Journal 2, and thus known to Thoreau scholars ever since that volume appeared in the mid-1980s, the wide-ranging significance of this draft has never been explored.⁷ But when considered in connection with his use of the Journal at this time, particularly in the composition of Walden, this remarkable episode provides considerable new insight into the “turn” that Thoreau was even then in the process of making. Not that the draft by itself clears up the mystery. But the story of its development proves useful in marking a defined path into, if not entirely through, the enigma. At the same time, it serves to call into question several notions we have come to take for granted about Thoreau’s development, both as a naturalist and a writer, in relation to Walden: the “central idea” and narrative climax of which—that “there is nothing inorganic”—this celebrated moment epitomizes (Richardson, Henry Thoreau 256; Walden 308).⁸

For one thing, the very existence not merely of the first flowering but, as I will argue, really the core of the passage in its final form, a year before the publication of A Week and some four years before the passage is supposed to have entered the manuscript, complicates the standard narrative of Thoreau’s and his book’s development. Considering how the passage’s climactic function presupposes the shape and broad thematics of the narrative as a whole, its early development provides insight into the status of Walden at a time when, having finished a complete draft at the Pond, Thoreau was in the midst of revising his manuscript with an eye to publication. Secondly, drawing on the archive of Walden materials at the Huntington Library promises to shed new light on how Thoreau worked to construct both the passage and the book, particularly on the role played by the Journal, a role hitherto neglected in studies of the book’s composition. While, generally speaking, the development of Thoreau’s use of his Journal from the Walden period to the 1850s is accurately described as a shift from a more or less disposable draft book to a carefully preserved, open-ended record of his local excursions, this useful generalization fails to capture the complexity of the Journal’s relation to the Walden project during these years. Finally, reading the development of the famous passage in the context of Thoreau’s engagement of science well before his so-called “turn to science” reveals a philosophical and quite public dimension of that engagement missing from the more common portrayal of Thoreau as a proto-professional researcher following his private scientific instincts.

Writing Life

From the early months of his residence at Walden Pond until his death, virtually all of Thoreau’s environmental writing exhibits a preoccupation with translating life and vitality, his own and nature’s, into writing. We see this ideal of literary vitality expressed, for instance, in his effort to inscribe seasonality and seasonal transformation in A Week, Walden, “Autumnal Tints,” and the Journal; in his notion that pages of the Journal are most successful when “allied to life” or as bearing the “fugacious ethereal qualities” of wild fruits; or when he envisions the possibility of literature animated by words “so true and fresh and natural that they would appear to expand like the buds at the approach of spring” (January 27^th, 1852; Journal 4:296; “Wild Apples” 267; “Walking” 208). Small wonder that this ideal would also come to inform his notion of science.⁹ To be sure, Thoreau seems to have been fully aware of the delicate art required to keep words alive, of the tension, inherent in the act of experiencing a life in or of nature, on the one hand, and making lively writing, on the other. In the early 1850s, as his natural history explorations were beginning to intensify, he counseled himself not to “tread on the heels of your experience” but rather “[b]e impressed without making a minute of it. Poetry puts an interval between the impression & the expression—waits till the seed germinates naturally” (July 25, 1851; Journal 3:331).¹⁰ But the same aesthetic self-awareness about writing life can be detected earlier, too, as in the couplet, “My life has been the poem I would have writ / But I could not both live and utter it,” where a seemingly straightforward lament belies an awareness that the very life problem the speaker protests is at the same time being “uttered,” that, paradoxically, the couplet manages to perform precisely what it claims is impossible (August 28, 1841; Journal 1:324; A Week, 343).

The post-1850s Journal, of course, represents the most extended, self-reflexive and radical inscription of this ideal, as several have argued. Within Thoreau studies at least, the Journal is now widely recognized as a deliberately wrought, if unconventional, literary work in its own right, “both a work of art and a critical work,” fashioned by “an unprecedented art of seeing, being affected [by the world] and knowing” (Davis 39).¹¹ But prior to the mid-1980s, it was commonly regarded as little more than a diary and portable storage unit, a collection of insights and observations to be culled at some later date for literary works then in progress. This view seems to have coalesced out of a number of facts, rumors, and fantasies: the example of Emerson, who described his notebooks as a “Savings Bank”; Emerson’s influential eulogy, which stressed Thoreau’s peculiar character over any literary talent he might have possessed; the cultural mythology of “Thoreau,” generated by the writer himself; and, related to this, his posthumous reception as an idiosyncratic sometime-mystic, scribbling naturalist. To the extent that literary ornithologist Bradford Torrey, a co-editor of the 1906 Houghton Mifflin edition of the Journal, recognized Thoreau as “a writing man” like himself, for instance, the Journal provided a “place of practice” and “a daily record of things thought, seen, and felt” (1906 Journal I:xx).

Yet, much as the recent study of the Journal has expanded our appreciation of Thoreau’s formal experiment as a means of writing the life of Concord environs and his experiential relation to it, the diary-cum-storehouse model still conditions our view of its relation to Walden’s development. Although partly because that relation has never been questioned and partly because we now focus on the Journal as a separate (albeit un-focusable) work, this particular form of blindness is also a legacy of Sharon Cameron’s insightful, provocative reading of the Journal “against” Walden. In order to define the Journal as both “autonomous” and “primary” rather than subservient to Thoreau’s published writings, Cameron constructs a mutually exclusive relation between the two works based largely on the premise that the Journal is “private.” In that space, she argues, liberated from having to consider any audience but himself, Thoreau could explore his relations to nature “unmediated by the social world,” including the literary conventions that constrained its representation in Walden (25–26, 46).¹² As I will argue, however, rather than the competitive relationship Cameron portrays, Thoreau developed a practice that fostered an intimate collaboration between the Walden and the Journal, one that also complicates the notion of privacy usually associated with the latter. More on this collaboration below. Suffice it now to say that Thoreau’s use of the Journal not only to record but, bit by bit, to construct the life he lived at Walden differed significantly from the role it played in making A Week.

Thoreau put his finger on the difference in summer 1845, not long before completing the first draft of A Week:¹³

Critics have been divided over whether to take this reflection as a prescient observation of the “practical” difficulty readers would experience with A Week, as a prediction of that book’s failure to attain “popular influence,” or as an astute diagnosis of a compositional limitation Thoreau would work to overcome as he revised it (Journal 2:453–54; Fink 216–17). Whichever it is matters less now than noting the difference between a practice of collecting polished thought-gems to be assembled later in some kind of “series,” as he had done before moving to Walden, and the way, once settled there, he began to use the Journal as a draft book uniquely suited to his second project. For, whether by accident or intention, quite early in his residence at the Pond Thoreau seems to have discovered in the Journal a means of mediating life into the “poem” he would write. This is evident in the way many passages that would find their way into Walden “are clearly ‘worked up’ and given a literary treatment.” And this same characteristic, what the Journal 2 editor, Robert Sattelmeyer, describes as the “virtually immediate transformation of personal experience into literary vignette,” continues to feature in the post-Walden Journal throughout the period of Thoreau’s work on the book (Journal 2:455).

Revisiting the Making of Walden¹⁴

What then was the status of the Walden manuscript at this point? According to J. Lyndon Shanley, who identified and sequenced seven partial drafts written between 1846 and 1854, publishing a text of the first version in the late 1950s, Thoreau was indeed close to completing a publishable text of Walden in 1849. But he “never finished preparing [it] for publication at this time because A Week did not sell.” Because Shanley found only “a few scattered [Walden-related] items in the journals of 1850–51” and then an abundance of new draft material beginning in late January 1852, he concluded that while Thoreau may have “thought of Walden from time to time” after A Week appeared, “it is certain that he did not write any significant amount of new material in the manuscript until January or February of 1852,” when he became “engrossed in it once more” (Making 30–31).¹⁵ Accordingly, Walden is believed to have been composed in “two major bursts”—1846–49 (versions A–C) and 1852–54 (D–G)—with “the second half of the book . . . much more extensively developed in the later versions,” its dense environmental description and darkly introspective central chapter, “Higher Laws,” serving to balance as well as to mute the initial social critique in “Economy” (Buell, Environmental Imagination 126, Sattelmeyer, “Remaking” 60).¹⁶

As I have already suggested, the weak link in the conclusions Shanley and his heirs have drawn involves the role of the Journal in Walden’s composition.¹⁷ As chief inspector of Thoreau’s “masterpiece,” Shanley had a stake in countering the myth that the book was essentially a compilation of Journals kept at the Pond. His diligent archival work and analysis of Thoreau’s efforts to improve the style and structure of his book through a succession of stages demonstrated clearly, he argued, that instead of “the immediate, random, and intermittent notes of a journal,” Walden was the work of an artist, one who exercised great “care and skill in revising sentences and paragraphs,” and who, over the years, achieved “an artistically satisfying” and unified structure. Far more than a simple record, the book represented “a reflected-on and consciously shaped re-creation of his experience.” (Making 23, 34, 74).¹⁸

Important and authoritative as this study has been, it was considerably circumscribed in two respects, one material, the other theoretical. In the first place, Shanley’s ability to appreciate the role played by the Walden period Journal was limited by the truncated 1906 Houghton Mifflin edition available to him. Not only had the publisher omitted several passages from the Walden and later Journal notebooks that seemed to “duplicate” already published material, but the editors were also unaware of the existence of three notebooks Thoreau used from April 1846 to September 1850, a four-year period now filled by over two-hundred printed pages in the Princeton Edition.¹⁹ In addition to the deficiency of the record available to him, Shanley’s quasi-New Critical conception of the Journal as a diary, an artless transcript of the author’s experience, limited his ability to accurately gauge the status of the book Thoreau hoped to publish following A Week.

In reconstructing the first (A) version, for instance, Shanley noted a number of segments as “missing” from that version because they show up only in later drafts or, in some cases, not at all. While a few of these, such as the core sandbank passage and the Poet-Hermit colloquy that follows “Higher Laws” can now be found in one of the fragmentary, restored notebooks unknown to him, several do in fact exist in the 1906 Journal, often, as mentioned, quite close to their final published form.²⁰ The simple explanation for why Shanley failed to note the obvious connection between such passages in the 1906 Journal and the Walden manuscript is that, focused primarily on cases where Thoreau “broke up” a Journal passage, “developed it, and finally ordered it,” he read them as mere raw material: spontaneous autobiographical observations not yet shaped by reflection and conscious literary design (Making 23). Yet, strikingly, a set of passages—the encounter with the Irishman John Field and his wife followed by the narrator’s famous woodchuck sighting and his confession of harboring two “instincts”—appears not only in essentially the same form as in the published text but in the same sequential order, thereby implicitly serving the same transitional and thematic function, moving the “plot” from “Baker Farm” to “Higher Laws” (August 23, 1845; 1906 Journal I:383–84; 1906 Journal I:384–85).²¹

Again, according to Shanley, Stephen Adams, and Donald Ross Jr., Thoreau began “to write out the first version late in 1846 or early in 1847,” by which time his draft contained material “that eventually wound up” in every chapter except “Conclusion,” and, “by and large, in [what would be] the final order” (Making 24; Adams and Ross 51).²² However, when we go on to correlate the revisions Thoreau began to make “in great detail” during summer 1848 to make the B/C draft with material that appears in Journal notebooks he filled after leaving the Pond but that for one reason or another was not included in the 1906 edition, it becomes clear that within just a few years Thoreau had made substantial progress in filling out that “final order.” In addition to the John Field-woodchuck-instincts sequence, drafted at the Pond, these include nearly verbatim iterations of the “John Farmer” meditation that concludes “Higher Laws,” the retrospective summary that would eventually appear in “Conclusion” (“I learned this by my experiments in the woods, of more value perhaps than all the rest . . .”), and material used in several of the later chapters, including “The Ponds,” “Brute Neighbors,” “Housewarming,” “Former Inhabitants,” and “Winter Animals”—all before December 1850 (Walden, 221–22, 323–24).²³

According to Ronald Earl Clapper, like the sandbank passage, with the exception of two passages added to version D, none of this material appears in the manuscript before E.²⁴ In this respect, then, the spring 1848 sand foliage draft and its revisions, which do not appear in the Huntington manuscript until versions F and G (early 1854), typify the problematic discrepancy between the evidence of Thoreau’s work on Walden in the Journal, and the representation of that work in Shanley’s arrangement of partial drafts and Clapper’s genetic text. On the one hand, like the majority of passages in Walden that were drafted in the Journal, the sand foliage draft, which dramatically expands Thoreau’s brief description of the phenomenon in version A, exhibits the same “worked up” character described by Sattelmeyer (Journal 2:455). On the other hand, the authority of Shanley’s taxonomy and the genetic text created from it led Richardson, for instance, to interpret “the dramatic expansion of the description of the clay bank” as evidence that only in late 1853, after its long “incubation,” was Thoreau’s book “finally reaching maturity” (Richardson, Henry Thoreau 311, 312).²⁵

We can get help resolving this dilemma by consulting the indexes Thoreau prepared for his Journal notebooks, which are printed as appendices in the Princeton Edition. Indexing in some form seems always to have been an integral part of keeping a journal for him. But because, in the period between 1850 and 1854, the indexes were compiled in real time—as he filled each volume rather than after he had completed it—they yield considerable insight into how and when he used the Journal in composing draft for Walden.²⁶ So meticulously did he track Journal contents during the second phase of the book’s composition, for example, that the index for a single large notebook takes up twenty-six pages (Journal 8:317–42). By contrast, for the Walden period and the immediately post-Walden years, when he used the Journal primarily as a draft book, the indexes are comparatively sparse. This makes sense considering how fervidly Thoreau worked at this time, literally consuming notebook pages in the heat of book and essay composition.²⁷ For why bother to index what has already been transferred, whether physically removed or recopied, to a draft-in-progress? Yet this is what we also find in the early 1850s regarding material obviously written with Walden in mind. When working on “The Ponds” and the loon chase in “Brute Neighbors” in fall 1852, for example, of over thirty passages he drafted in the Journal between late August and early December, only one appears in Thoreau’s index.²⁸ Thus, even as the Journal had become its own project dedicated to inscribing Thoreau’s regular local excursions, he continued to use it as a draft book, albeit differently. Now, in order to preserve his Journal volumes intact, he either copied draft onto separate pieces of paper or entered it directly onto the Walden manuscript; both practices are evident in the Huntington papers.²⁹ Consequently, the very absence of such passages from the indexes—their conspicuous omission from his real-time practice when every other item on the pages where they occur is included—provides a kind of image in relief of Thoreau’s work on his book.

Equally significant is a pattern that can be detected in Walden-related draft that is indexed, which tends to be material that winds up in later chapters. Like his use of the Journal to “utter” his Walden experience while living it, this is a practice that began at the Pond. There, in the first index he made, Thoreau grouped entries from three notebooks he had finished filling in February and late March 1846 under such headings as “beans,” “John Field,” “savage life,” and “Spring.” Then, by numbering each item to indicate its order, he effectively created an outline with the index of the second half of the book (Journal 2:388). As Shanley notes, “neither the original order of the list nor the numbered order [of items in it] corresponds exactly to the order of the material” in version A (Making 100). Yet, besides confirming Thoreau’s conception of a book thus early, the outline nonetheless delineates an ordered sequence that (with the exception of various bird and animal life sketches he would later distribute among “Brute Neighbors,” “Winter Animals,” and “Housewarming”) remained essentially unchanged through publication eight years later. Not only is the seasonal structure clearly present, but so is the placement of particular episodes and segments, such as the encounter with John Field, moral qualms about fishing (“savage life”), and the pond survey.³⁰ And Thoreau continued this practice, using his indexes to keep track of material to be incorporated later, for the duration of his work on Walden. It can be seen, for instance, in what he labeled “old inhabitants” in the first index, made at the Pond, and “Sand foliage,” the heading he used in those of the early 1850s (Journal 2:388; Journal 4:506, 508).³¹ Yet no sign of the passages to which either of these headings refers appears in the Huntington manuscript until the E and F versions, respectively.

All in all, taking into account the mediating role Thoreau’s Journal played in the composition of Walden complements the taxonomy of discrete stages established by Shanley and Clapper, depicting a more fluid, dynamic, and above all continuous process. Certainly the burst of drafting that appears in the January and February 1852 Journal, which Shanley used to date version D, is striking and just as certainly indicates a renewed push on Thoreau’s part to complete his book. But when we observe portions of well over sixty Walden paragraphs drafted in the Journal during the supposed two-year hiatus, together with the pattern of indexing I have described, Shanley’s confident assertion, long accepted by scholars, that Thoreau “did not write any significant amount of new material in the manuscript” between the commercial failure of A Week and early 1852 simply does not hold up (Making 31). To be sure, it makes for a dramatic story to imagine that one day in November 1851 Thoreau pulled down “the old manuscript he had put away two years before. . . . shook off the dust, and set to work.” But the evidence of his persistent labor over the long haul—subject, of course, to creative ebb and flow and worldly interruptions—reduces the likelihood that there was ever such a single day, or any dust (Walls, Henry David Thoreau 312).³²

Making Walden, it appears, was less like incubating an egg, in Richardson’s metaphor, than constructing a house, except that of course Thoreau had to make, find, and borrow all the materials himself. Once founded, framed, raised, roofed and roughed-in, some few rooms, such as “Reading” and “The Bean-Field” were finished quickly; others, like “Sounds,” “Solitude,” “Baker Farm,” and “Economy” finished mostly (though Thoreau never ceased making small additions to the latter); while the initial designs of others were altered later in the process the better to fit the builder (“Where I Lived, and What I Lived For,” “The Village,” “Higher Laws”). Along the way, starting in spring 1848, Thoreau began to install a fireplace whose mantle would be the earth itself.

Into the Weeds

This brings us back to the sand foliage draft, which we can now examine in greater detail. Wary readers will want to lace up their boots at this point, however, because the weeds grow thick and close in this here part. Indeed, if we include an unpublished leaf with which those from Thoreau’s 1848 Journal are grouped in Huntington MS 924, we are faced with at least four levels of text and revision to pick through (see below for a transcription). Fortunately, because the Princeton Edition has been edited to reproduce the “current” text, that is, as Thoreau wrote the Journal from day to day rather than as he later revised it, the original level of composition is clearly identified. But since our interest lies not only in the Journal as an autonomous work but also as a node in the growth of another project, we must first parse and try to date the levels of revision that went into building Walden’s sandbank.

The five-page document is fascinatingly complex. At first glance its pages appear a welter of script written in two different hands and two shades of brown ink, one grayish, the other dark reddish. In the left-hand margin of the first four pages, next to segments of text that have been separated by horizontal lines, Thoreau has written numbers, non-sequentially, in both pencil and ink. Where, finally, his description of the thawing process continues from the bottom of the fourth page onto the fifth, the sentences have been squeezed between the lines of an entirely unrelated, preexisting text, lacking any other space in which to write.³³ In all, and together with the unpublished leaf to be described shortly, these pages comprise all-but-identical iterations of the first three paragraphs of Walden’s famous passage as well as the lead sentence of the fourth.³⁴

As in the published text, this draft begins with what Robert Thorson identifies as a precise rheological description of the thaw and flowing process, including the sand-clay mixture and materials, its colors, fluid mechanics, and diurnal expansion and contraction in accordance with fluctuations in the intensity of heat and light.³⁵ As in the published text, the scope of Thoreau’s description then widens to encompass analogous forms that the mixture appears to assume (“the forms of vines . . . the feet & claws of animals . . . of the human brain or lungs or bowels & excrement of all kinds”), before arriving at its inductivist conclusion: “Thus it seemed as if this one hill side contained an epitome of all the operations in nature.”

For the most part, although a few new observations have been added, Thoreau’s revisions of the 1848 Journal draft primarily refine and contextualize that original draft. He comments, for instance, on the rarity of the sand foliage phenomenon “before railroads were built,” notes how the emerging forms resemble “excrement [in the now obsolete sense of any type of outgrowth] of all kinds,” and imagines that the process is “destined perchance to be the subject of admiration to future geologists” (Journal 2:577–78). But this is not to imply that the revisions are merely literary. The concrete specificity of the new observations—one in particular, to be discussed below—indicates that Thoreau must have reexamined the phenomenon on-site, whether in the same season or another is impossible to determine from the document alone. The signal difference between the two stages of composition, however, is that in the first Thoreau’s observations and reflections stand alone, as yet unconnected and unordered, as if the important thing was to get them down fresh. The marginal numbering in pencil of these standalone segments clearly occurred sometime afterward.³⁶

Dating the Core

So, if not in version F, when did Thoreau make the sequence of revisions that established the sandbank core? We can begin to answer this question by noting that the appearance of sandbank draft in the Journal evidently coincides with his decision to try to publish Walden.³⁷ When we include the existence of the second-half outline in the Walden notebook index, described above, and consider the abundance of Walden-related draft in the Journal through mid-1850 as well as evidence from the indexes that he deliberately stored material for later insertion, it becomes clear that by the winter of 1849/1850 his book was more complete than we have realized. If we factor in the presence of the core sandbank passage, then the entire narrative arc of Walden, with its climactic epiphany, would have been in place. Having negotiated a deal with Emerson’s publisher, James Munroe, in February 1849 to publish both books and being obliged to pay the printing expenses of A Week out of sales, Thoreau would have been on the hook to finish the second book sooner rather than later. It’s hard to imagine that the experience of seeing a forthcoming Walden advertised in the page proofs of A Week he was correcting in March and April of that year would not have been mingled with some anxiety as well as excitement.³⁸ And, indeed, he was already building interest in the second book, having chosen to deliver all nine of the 1848–49 season’s lectures from the Walden manuscript, despite having plenty of other material available.³⁹ This push might explain the apparent rapidity with which the 1848 draft was revised.

We can approach the question from another angle by viewing Thoreau’s work on the core in the context of the passage’s construction as a whole. Like many other parts of Walden, the four-paragraph sandbank passage grew by accretion rather than having been successively rewritten in toto or holistically revised over time. Typically, with much of the book, once the structure of a chapter, section, or paragraph had been established, Thoreau expanded it by inserting new material, whether social critique, anecdote, natural history description, quotations, or transcendental reflection. This practice is what gives so many of Walden’s paragraphs the dense, sometimes bizarrely digressive, fragmented, and polysemous character that continues to delight many readers and exasperate many others.

Viewed from the standpoint of its overall composition, the development of the sand foliage passage epitomizes this process. As we have seen, in the core the passage begins to assume what will be its final form in diction, syntax, and structure. Thereafter Thoreau added a few new descriptive details as he continued to observe the phenomenon annually, when possible, inspecting anew and more closely its physical features and micro-processes.⁴⁰ But apart from the basic “foliage” analogy and the sense of standing in the presence of the “Artist who made the world and me,” most of its lyrical generalizations—the elaborate analogies of “leaf” and “lobe” forms, rapid shifts of perspective from the granular to planetary, and the final rhapsodic peroration to vital force and plasticity—were developed after the core. Like the initial 1848 draft, much of this later material was composed in the Journal, and all of it after December 1851.⁴¹

Given his customarily accretive process of composition, then, we can conclude that the revisions that completed the core must date from no later than December 1851, when the earliest of those later additions was drafted in the Journal. Since, like all other Journal draft related to the sandbank phenomenon (with one exception), those revisions were based on observations made on-site and in season, as their concrete specificity attests, Thoreau must therefore have returned to revise the original spring 1848 account when the foliage flowed again either the following year, or, at the latest, two years after, in late winter/early spring 1850.⁴²

“Glancing Like Lightning”

So much for the what, when, and how of the sandbank core in relation to the Journal and the status of Walden. How does this assessment affect our understanding of Thoreau’s pivot and his “science?” Besides marking the turn considerably earlier, the story of the sandbank draft confirms a metaphorical pivot in the true sense—a turn that carries an object in a continuous action, as in basketball—rather than a stark change of direction.⁴³ On the one hand, that is, Thoreau’s literary representation of the event as a revelation of formative forces shaping both his own creative life and that of the planet aligns with his earlier expressed desire simultaneously to live and to write the “poem” he would utter. On the other, his use of the Journal as both a draft book and an instrument of empirical inquiry anticipates the integral role it will play in the empiricist immersion that has so far defined his engagement with science.

Yet the core sandbank drafting also makes clear that, from the start, Thoreau’s interest in science went far beyond natural history, at least as that discipline came to be narrowly defined by Thomas Henry Huxley and other promoters later in the century.⁴⁴ In A Week he argues that, instead of trusting to blind fact collection, what is needed in science, is to make more “systematic approaches” to the significant or “central fact,” thereby combining “constant and accurate observation with enough of theory to direct and discipline it” (363–64). Appropriately enough, the particular “theory” of “development” (or evolution) that concerned him at the time, one that patently informs his account of the sand foliage, as other critics have noted, was also known as the “theory of creation.”⁴⁵ Although the question of the origin of life, or abiogenesis, is now a study distinct from that of evolution per se, even Darwin could not entirely separate the origin of species and speciation from that of life itself, as his antagonists saw clearly. In the evolutionary debate that preceded the publication of On the Origin of Species, triggered by the appearance of Vestiges of the Natural History of Creation shortly before Thoreau moved to Walden, these two issues were inextricably connected. Consequently the very existence of evidence that might support a theory of natural origins, whether in the deep past or the present, was a primary bone of contention.⁴⁶ Indeed, responding to his scientific critics, Robert Chambers, the anonymous author of Vestiges, speculated that one reason that “no such phenomena as the starting of new life or passage from one form to another has ever been witnessed” by English naturalists was that they did not look for it. It was not a significant “fact” to them, as Thoreau might say; lacking the “idea,” they were not “prepared” to see it, as he put it later (“Autumnal Tints” 256–59). Likewise with regard to transmutation, given the current “strong prepossession” in favor of species immutability, Chambers argued, perhaps even if such a phenomenon were observed it would be “explained away on some other supposition, or, if presented, would be disbelieved and neglected” altogether by naturalists ([Chambers], Vestiges lii).⁴⁷

The full story of Thoreau’s engagement in the Vestiges controversy lies far beyond the scope of this essay.⁴⁸ But, in keeping with our hyper-focus on the sandbank core, we can briefly take the measure of his engagement by examining a little-noticed detail integral to the passage’s comprehensively organic vision. It’s a strange detail, involving the formation of blood vessels which the narrator claims to “see” in the flowing sandy mixture, an observation easily written off as transcendental fantasy or the product of analogical imagination run amok. Yet, in addition to beautifully exemplifying Thoreau’s “extraordinary gift for microcosm,” the analogy directly underpins the ontologically radical assertion to which it leads, that “man” is “but a mass of thawing clay” (Buell, Literary Transcendentalism 188). Once copied into the manuscript and revised only slightly thereafter, this extraordinary description would remain essentially unchanged through the completion of Walden: