[Many readers of Jeffers, both academic and general, take it for granted that Random House “suppressed” or censored The Double Axe. James Shebl originally argued this position in 1976 In This Wild Water: The Suppressed Poems of Robinson Jeffers, and William Everson and Bill Hotchkiss shape their 1977 edition of The Double Axe and Other Poems on this assumption. However, the manuscript material for The Double Axe poems and the documents that reflect Jeffers’ struggle to shape this collection, both before initially submitting it to Random House and after Random House returned it to him for further consideration, suggest that Jeffers himself was conflicted about these poems and this collection. If we simply treat the alterations to the collection as forced on Jeffers by Random House, we mistake the matter and we obscure crucial aspects of these poems and this collection.
This piece, which appeared originally in TEXT 7 (1995) pp.435-457, draws on some of the evidence that problematizes the claim that Random House censored The Double Axe to explore what was at the time a current editorial question. The piece does not, for that reason, fully engage or develop the implications of Jeffers’ changing positions on the poems and their presentation. It does, though, indicate that we cannot adequately understand these poems or the collection if we insist that the changes in phrasing and content in the collection as finally published are simply the result of Random House coercing Jeffers. This piece, then, is a kind of preliminary clarification, from which a more adequate critical account of The Double Axe in its multiple forms and Jeffers’ World War II poetry could be developed (the documents needed to develop such a critical account can be found in Volume Five of The Collected Poetry of Robinson Jeffers).
This piece, as noted in the Acknowledgements, developed from a somewhat differently focused conference paper delivered at a Society for Textual Scholarship conference. That piece can be accessed using this link: “Censorship and Revision: Politics, Publication, and Robinson Jeffers Changing Sense of The Double Axe.” Two other pieces, one from thirty years ago and one from a year ago, also bear on The Double Axe”:
“To Date or Not to Date: Jeffers’ ‘Pearl Harbor'”
“‘Mornings in Hell’: Jeffers’ Struggle with Politics and History in The Double Axe”]
With modern works it seems we must often decide which is more valid: a work’s final manuscript or what it became through publication. Cases such as Dreiser’s Sister Carrie show that these decisions can have real impact on both the shape and details of what readers encounter, and theorists have advanced various rationales for preferring in general either the more private world of the manuscript or the more public world of the published book. Recently Peter Shillingsburg has suggested that these various rationales can be divided into two groups, one deriving from what he terms “author-centric” assumptions, the other from “socio-centric” assumptions. Author-centric theories, as the term implies, posit writing as a largely autonomous act carried out by an agent capable of singleness of intention or (effectively the same thing) a structured set of them. In these views the work’s integrity derives from the author’s act of writing, not the act or fact of publication, since publication becomes primarily the transmission of something already essentially complete. Even an author’s use of editorial queries, proof corrections, etc. to complete a work would do little to relocate the work’s authenticity, since the process would still be governed by the author’s own intentions, not the editor’s or the publisher’s. Conversely, in socio-centric views, writing is not a complete, selfauthenticating act but a series of gestures toward, and interactions with, the structures of the social world (including the realities of market). As such publishing becomes primary since it transforms the raw material, the incomplete art, of writing into the finished product of a book. From these perspectives a work’s integrity is inseparable from its status as social object and the social process that produced it. In effect, socio-centric positions tend to see the writer not as an authoring subject, a solitary or privileged agent, but as an authoring object (or rather the subjective dimension is understood as a product of other forces), and this author-object and others collaborate in the publishing process.
Shillingsburg’s essay reveals affinities between seemingly competing practices and dichotomies between seemingly related ones. In the process he suggests that our theories, though internally consistent and powerful, may derive less from secured principles and self-evident axioms than from a desire, on the one hand, to affirm the writer as a fully self-knowing, transparent, and controlling subject capable of singleness of vision and, on the other, a desire to dissolve this subjective power (and intentionality with it) into the grammars and dynamics of social, political, and economic realms. This might seem a step toward abandoning the rock of editorial rigor (or the desire for it) for the deconstructive swamp, but Shillingsburg wants us, I think, to consider another option: that the dichotomy he traces may partly derive from the way both camps tend similarly to oversimplify (though in opposite ways) the processes of writing and publication. At the systematizing level of theory what tends to be lost is the possibility that the author might have multiple and conflicting intentions of different orders, some deriving from his or her subjectivity, some from the contextual factors that partly shape that subjectivity, some from writing as a self-reflexive activity, some from writing as a social gesture. What tends to be lost, that is, is the way writing and texts may be simultaneously and dialectically author-centric and socio-centric phenomena. If so, Shillingsburg’s analysis points to the possibility that these two theoretical tendencies should be viewed as complementary rather than competing perspectives, each necessary and relevant (though in different ratios as we face different examples and address different purposes).
Robinson Jeffers’ controversial collection The Double Axe (1948) offers a specific example of an interaction between writing and publishing that might lead author-centric and socio-centric editors to different conclusions but where the textual issues and their critical implications are likely to be understood fully only if the work is treated both author-centrically and socio-centrically. In October 1947 Jeffers sent Random House a typescript he assumed would be the setting copy for his eleventh major trade collection. By the time the book appeared half a year later, he had dropped ten poems, altered others, and restructured the collection. The changes were Jeffers’ own, but he made them at least partly in response to complaints from his editor, Saxe Commins, who was offended by the “frequent damning references to President Roosevelt” and Jeffers’ claim that U.S. leaders had tricked the country into what they knew was an unnecessary war. The Double Axe, thus, exists in two “finished” forms, one more purely authorial and the other more social but each in some way authorized by Jeffers. Deciding which is the “real” version turns in part on whether the publishing process subverted Jeffers’ intentions or helped him realize them and in part on whether he felt himself coerced or viewed his editor’s comments as advice. Shillingsburg’s discussion, though, suggests. as well that editors working from author-centric assumptions would be more likely to see the publisher’s role as an intrusion since introducing any intentions not clearly and solely the author’s own would, almost by definition, taint the text’s integrity (especially when the alterations can be traced to a publisher worried about turning politically unpopular poetry—soon after a major war—into a marketable product) and that those working from sociocentric assumptions would be more likely to see the publisher’s actions as legitimate, since the publisher’s agenda would be an important feature of the social interplay that produced the actual work, the published book. With The Double Axe, though, justifying the typescript because it is “authorial” or the book because it is “social” obscures not only Jeffers’ actual sense of the relationship between private experience and public expression, and thus his sense of writing and publishing in general and of writing and publishing this book in particular, but it also obscures the actual agenda of Random House. That is, with The Double Axe the author-centric scholar’s desire to explicate the creative subject can only be realized by attending to the socio-centric dimension, and the socio-centric scholar’s desire to recover the text as a social and ideological object can only be realized by attending to the author-centric dimension.
In 1947 Jeffers’ days on the cover of Time were fifteen years past, but he was still an important part of Random House’s claim to having become a major literacy firm. Jeffers first established his reputation in the mid and later 1920s with a series of collections published by Boni & Liveright, and as early as 1929 Random House co founder, Bennett Cerf, was looking to publish Jeffers. He tried to arrange a limited edition of what became the title poem of Dear Judas (1929) when Liveright wanted to postpone that collection against Jeffers’ wishes. When Liveright unexpectedly met Jeffers’ demands, Jeffers had to cancel his arrangement with Cerf, but this did lead to Random House issuing a limited edition of the sequence Descent to the Dead in 1931 and situated Cerf to outbid other more established firms and sign Jeffers outright when Liveright went bankrupt in 1933. The day (October 7, 1947) that The Double Axe typescript reached Random House, Cerf wrote the poet’s wife, Una, to say that the arrival “elicited cheers from the entire editorial department.” Cerf was likely exaggerating and almost certainly wrote without reading the work, but his timing and tone indicate that Jeffers’ standing at Random House meant he would have assumed his work would be published as a matter of course (with only the usual copy editing) and that Random House would have assumed the same thing. If so, it was a delicate matter for the firm to ask for major changes. (In this respect the case of The Double Axe differs significantly from such widely explored cases as Sister Carrie and Crane’s The Red Badge of Courage, where relatively inexperienced. not yet established writers were pressured to reshape their books.)
Commins was likely the first at Random House to read The Double Axe. His October 13 memo to Cerf details Jeffers’ treatment of the war and laments the poet’s decision to “personif[y] his bitterness in Roosevelt and the whole human race” instead of (as with earlier work) “disguis[ing it] in horses and hawks and incestuous relationships.” He found Jeffers’ “hatred” of Roosevelt “insane” and inexplicable and concluded that it had turned him into not only “an out-and-out champion of isolationism” but also “a wildly prejudiced slanderer.” Commins also sketches, though at times more by implication than explicit statement, the “questions of policy” the manuscript raised and stressed that these would need to “be considered with the utmost care.” Most simply he feared the book could give “the impression” that Random House was “tolerating angry and irresponsible statements about America and more particularly Roosevelt” and that this could damage the firm’s reputation. But he also worried that refusing it would “deny” Jeffers “the right” he’d earned “to deliver his last dicta on Man,” especially as they re1ated to “his central argument that mankind is not important in the universe.” Commins’ rejection of Jeffers’ politics might make this last point seem only some sort of concern that the literary community would conclude Random House placed profit or convenience ahead of merit or an attempt to hedge his attack in light of Cerf’s admiration for Jeffers. But his later comments (as we’ll see) do indicate that Commins believed publishers had a professional obligation to those who had proved themselves to be major writers. Moreover, his own earlier admiration for Jeffers and his sense of himself as Jeffers’ editor, not simply Random House’s employee, meant he probably had some concern for what he perceived to be Jeffers’ interests. Commins’ and Jeffers’ correspondence both before and after The Double Axe seems to show that each had a genuine regard for the other, and they had worked together with apparent good will on all of Jeffers’ previous Random House collections. They would also choose to work together again. It was Commins who encouraged Jeffers to assemble his next and final collection, Hungerfield and Other Poems (1954), even though Jeffers had by then largely slipped from public view. Commins was seemingly motivated by both his concern for the grieving poet (Una Jeffers had died the year before he proposed the collection) and his regard for the poetry. This is not to minimize Commins’ distaste for Jeffers’ interpretation of World War II, but his worry that the new book—especially the indictments of Roosevelt—would damage Jeffers’ reputation by highlighting his “tight and narrow thinking” and casting him in the image of a “Cassandra-like prophet” may have been genuine. At the least, public reaction to the book shows he was right on this point.
For Commins The Double Axe seemed a no-win situation-for both Random House and Jeffers—but so did not publishing it. It is, thus, no surprise that he closes his memo by grasping for another option—getting Jeffers to alter the book:
I don’t see how we can do anything but protest to Jeffers about the Roosevelt and isolationism passages that are manifestly obnoxious. If we can’t make him see reason, we’ll have to take a strong position on principle. If he does take out the objectionable passages, we will then have a book obscurantist enough to please the dwindling Jeffers following. This book has made me dwindle!
But as Commins here seems to anticipate, at most Jeffers could (as he would to some extent do) mute the references to Roosevelt; he could not remove so fundamental a thread as the isolationist critique without reducing the book to incoherence, and in this paragraph Commins may already have suspected that, unless Jeffers withdrew the book, he and Cerf would have to choose between publishing it substantially as it stood (politically anyway) and refusing to publish it.
When Commins wrote Jeffers two days later he, not surprisingly, moderated his tone, casting himself as a concerned and friendly admirer hoping to be of service by raising issues the poet might have overlooked in the heat of composition. In spite of this tactical shift and the presumably false claim that he wrote “entirely on his own responsibility” without Cerf’s knowledge, the letter repeats the memo’s points and reflects the same confusion about what to do. Commins warns that the “damning references” to Roosevelt and “bitter charges” about the war raised questions of taste and fairness and that the material might cause Jeffers “to be linked with reactionary elements.” But even though he quotes phrases and cites passages he neither stipulates what should be done nor specifies the consequences for letting the text stand: “I am writing,” Commins says, “with the hope that for the sake of your book and the effect it will have that you can temper these references before we think of beginning composition.” This hints that Random House might refuse the book but stops short of saying so, and Commins further muddies the signal in the next sentence by stressing Jeffers’ “rights” and implying that the firm was ready to publish the book as it stood: “Please understand that this is in no way, and I can’t make this too emphatic, an attempt to intrude upon your rights as a free artist.” In part Commins may have thought that an appeal to fairness and professional self-interest would more likely persuade Jeffers than a threat, but the memo to Cerf shows that he may also have skipped the ultimatums because he and Cerf didn’t yet know what they’d do if Jeffers wanted the book published as it stood. And in any case, the letter is a milder “protest” than Commins’ memo to Cerf promises. Commins only hints that Random House might “have to take a strong position on principle” and offers no sense of what that “position” might be.
Commins’ tentativeness and Jeffers’ own relationship with Random House make it unlikely that Jeffers saw the critique as an attempt to dictate changes or a threat to reject the book. Moreover, the timing—and success—of his other major project of this period, a translation of Euripides’ Medea, gave him even less reason to think he needed to placate Random House. For years Dame Judith Anderson had wanted to play the role of Clytemnestra in a major production of The Tower Beyond Tragedy (1925), Jeffers’ adaptation of The Orestia. When a producer asked her to play Medea, she agreed, but only if Jeffers would do the script, and in early 1945 she convinced him to put aside his other work to draft a version of Medea. Although the Random House edition of Jeffers’ Medea received mixed reviews when published January 1946 (ironically Commins wrote one of the positive ones), the production was a sensation. Catalyzed by Anderson’s performance, it revived popular and critical interest in Jeffers’ poetry and gave him sudden prominence as a dramatist. Jeffers had even made a rare trip east for the opening and, thus, shared in the excitement of the first performance and the reviews that followed. Significantly, the production opened October 20, 1947—one week after Commins drafted his memo to Cerf, five days after his letter to Jeffers, and either shortly before or after he returned The Double Axe typescript to Jeffers. Jeffers thus encountered Commins’ plea to moderate his book for the sake of his reputation all but simultaneously with Medea’s dramatic success.
Medea’s success may have led Jeffers to conclude that Commins’ fears were unfounded or made them even more telling since he now had more to lose; either way, it strengthened his position and increased the pressure on Random House to publish the book—with or without changes. Certainly Una Jeffers’ December 8, 1947 letter to Commins indicates she and Jeffers weren’t worrying about Random House refusing the book or tampering with it—in spite of the objections. With the deadline for the spring catalogue approaching, Commins had written to ask whether he could plan on having the typescript (preferably revised) in time for spring publication. Commins to Jeffers. December 4. 1947. Una Jeffers’ reply partly explains Jeffers’ delay in answering Commins’ October queries:
Actually Robin has never unwrapped the ms. of “The Double Axe” since you gave it back to us at Random House. He had a very sharp attack of flu after we got home and, since he has more or less recovered, has been struggling with an article [Poetry, Gongorism and a Thousand Years] he promised to write the N.Y. Times while we were there. He says it would be certainly six weeks from now before he could hand it over to you.
and then adds,
Another thing, Saxe, do you think your firm wishes to publish this book—and, if they do, will push it properly? Robin’s view of politics and a sick world differs so much from yours….It would be best to tell us now if you’d prefer to skip this particular book & would in no way alter our friendly feelings toward Random House!
These comments, which suggest Jeffers may have felt some sympathy for Commins’ plight, indicate that Jeffers assumed he was the one controlling the situation, and Commins’ eventual reaction to Jeffers’ reworking of the collection suggests this was by and large the case.
When Commins received the revised typescript, which became the setting copy for the collection, in early February 1948, he faced essentially the same issues he’d faced in October 1947. Although Jeffers had substantially restructured the collection, he had kept much that Commins had identified as politically problematic, while dropping or altering work that hadn’t been at issue; overall he had left the book’s political tone and judgments intact. Acknowledging the typescript, Commins tried to be positive. His letter praises the “immense improvements,” but his disappointment is clear. He admits he “still disagree[s], and vehemently” with the “interpretations of recent world and political events and the causes underlying them,” but even so he pushes for only two minor changes: he asks Jeffers to “consent” to dropping an allusion to Roosevelt’s paralysis and adds, “I do wish I could persuade you to take out the word ‘little’ to describe Truman.” (Jeffers apparently thought the requests trivial and petty. In his reply he explains that “‘little'” wasn’t a “gratuitous insult” since he’d meant it “in a historical sense” and not as a comment on Truman’s height, but he then adds, “However—to show you what a good fellow I am—Write ‘Harry,’ if it really matters to you.”) Commins’ requests—and his tone indicates he knew this—were essentially beside the point. The two slight changes wouldn’t alter how the book would be perceived politically, and Random House had apparently already decided to publish it, with or without the concessions. The decision was probably partly pragmatic; the revised typescript indicated Jeffers couldn’t or wouldn’t change the book’s political character, and the publicity from Medea had increased the risk in refusing it. The firm’s sense of professional obligation was likely another factor. In his letter, Commins characterizes Jeffers’ position (and by implication his own) as “matters of opinion” and “consequently open to debate,” and from this concludes that the firm has “a moral obligation to present them in your terms and on your responsibility.”
But in addition the letter shows that Commins had hit on a strategy for publishing the book without implicating Random House in its politics:
…lest there be a misapprehension about the differences of views between us [i.e. Commins, Cerf, and Random House on the one hand, Jeffers on the other], it occurred to me to write a publisher’s note…as a statement of our position….Tell me candidly how you feel about it. At best it is an honest statement of my viewpoint and at worst it will serve to underline certain passages which otherwise might even go unnoticed. Since both of us are responsible for our convictions and we must stand by them, why not have them out in the open?
The “Note,” eventually used on both the dust jacket and immediately following Jeffers’ own “Preface,” reads in part:
Random House feels compelled to go on record with its disagreement over some of the political views pronounced by the poet in this volume. Acutely aware of the writer’s freedom to express his convictions boldly and forthrightly and of the publisher’s function to obtain for him the widest possible hearing, whether there is agreement in principle and detail or not, it is of the utmost importance that difference of views should be wide open on both sides.
The “Note” echoes Commins’ original memo to Cerf, which suggests that he, and likely Cerf as well, took Jeffers’ “rights” as an author seriously enough (that they had, that is, an “author-centric” view of writing even as they factored in the social and commercial realities of publishing) that they would likely have published The Double Axe with or without changes, with or without Medea, so long as they’d come up with something like the “Note” to distinguish Jeffers’ politics from theirs.
I am not trying to argue that Commins’ reactions had no impact on The Double Axe. Commins forced Jeffers to reconsider a collection he’d assumed he had already finished. What I am arguing is that Commins and Random House lacked the power to compel Jeffers to change the collection or, at the least, that Jeffers assumed they lacked it. If this is so, we must look primarily to Jeffers to understand what happened and why. We must, that is, consider the project of The Double Axe from an author-centric perspective if we are to understand the socio-centric dynamic of publication.
For Jeffers the poems of The Double Axe were deeply troubling ones. They were poems of despair at the war’s violence, and they were also, in the extremity of the reactions they document, poems which threatened, as he realized, the validity of the aesthetic that had generated all his major work. The aesthetic (itself partly a product of his earlier struggle with the trauma of the First World War) posited nature’s inscrutable power and flux as a kind of ultimate order that could, when viewed with proper reverence, comprehend the human world’s irrational violence within its beauty. Especially in the later 1920s Jeffers had worked to demonstrate that we experience “nature” (including its human and social dimensions) as pain if we focus on our own immediate stake in it but that we can experience it (at least temporarily) as a transcendent redemptive beauty if we look beyond ourselves to the whole. By the mid-1930s Jeffers had concluded that a second European war was inevitable and that the U.S. would be drawn in. Faced with this impending violence, he found it increasingly difficult to affirm nature’s beauty and redemptive power. Increasingly, as the work of Be Angry at the Sun (1941) demonstrates, he alternated between almost confessional poems where the speaker laments his failure to transcend the pain and randomness of the present and poems of didactic assertion. Both responses contradicted his poetics and earlier claims. U.S. entry into the war intensified the anguish, and the war poems of The Double Axe reflect his growing despair and anger (especially at what he saw as the self-serving rhetoric of those in power), which reached its peak in the 1944 narrative “The Love and the Hate.” Originally titled “My Corpse Is Huge,” it tells the story of a young soldier who revives his decaying body, returns from a Pacific atoll, and revenges himself on his father for selling him into the war; the soldier also indicts those who had led “the decent and loyal people of America” into the war “To feed the vanity of a paralytic and make trick fortunes / For swindlers and collaborators.”
“The Love and the Hate” and the shorter poems drafted before it were enough for an ample collection, and notes for an undrafted “Preface” on the May 1944 manuscript of “Invasion” show that Jeffers considered assembling a collection during the war (more than three years before he actually assembled The Double Axe) that would have featured this narrative. He may have waited because he sensed that the war poems could only be seen during the war itself as acts of protest (which in part they were) and not as his attempt to face the crisis the war represented for him. The notes, though (which show him apologizing for “The Love and the Hate,” even as he was still writing it) suggest other factors as well. In them Jeffers worries that his focus on the war’s “wasted effort” and “ridiculous dreams” has become “obsessive” and casts about for a way to justify his response. He settles on the excuse that the poem was “forced” on him and regrets that his inability to “choose” his “subjects” has forced him into what he sees as a “feminine role.” These notes and poems, that is, show Jeffers caught between his desire to confess his failure to transcend pain and his need to minimize the threat of this failure to his aesthetic as he perceived it and wanted his readers to perceive it. As a result these notes and poems show him looking for a way to present the war poems that would admit his crisis yet cast it as a gesture beyond himself that would somehow affirm the world’s difficult beauty that he’d argued for in his earlier work. At least in 1944 the only way he could seemingly think to do this amounted to the paradoxical gesture of denouncing his own subjectivity in order to turn it into a kind of object lesson—a kind of a failure that somehow proved its own success, and his delay in assembling the volume suggests he found this approach, finally, unconvincing and recognized that “The Love and the Hate” (by itself or in the context of the other war poems) would simply stand as a confession of struggle and failure.
After the war Jeffers somewhat recovered his faith in nature’s redemptive capacity. In “The Inhumanist” (drafted mostly in 1946 and the last piece written for The Double Axe) he uses the persona of an elderly, solitary ranch caretaker for a series of meditations that assert a tempered version of his earlier claims. This second narrative gave him a significantly different set of options for his collection. He could, for one, have organized a collection around “The Inhumanist,” omitting “The Love and the Hate” altogether. This would have focused attention on the new narrative’s philosophical claims, cast the earlier short poems as extensions of it (not poems of doubt), and largely erased the dilemma implicit in the undrafted 1944 “Preface” in the process. Instead, he kept “The Love and the Hate,” combining it and “The Inhumanist” into a single two-part “poem”: The Double Axe. In this way he projected the earlier narrative, now the first half of the longer piece, as the cathartic suffering necessary to renew faith and vision, not a poem of horrified obsession, and he asked readers to see the assertions of serenity in “The Inhumanist” as deriving from, and authenticated by, this earlier suffering and not merely as theoretical assertions. Combining the narratives gave Jeffers a way to mediate his earlier and conflicting impulses to confess and yet also distance himself from what he was confessing. It was a way to record his recovery without erasing what he was recovering from and to cast his struggle with the war as proof of his vision’s power over failure, not its power in failure.
But Jeffers’ decision to join the two narratives indicates that The Double Axe (book and poem) was also a compromise between, not a resolution of, the competing intentions that shaped the writing of individual poems and his impulses for presenting them. A sheet with workings for one of a series of tables of contents preceding the typescript shows him, for example, categorizing poems as “post-war” and “war-free” as he apparently sought the right balance between these groups and the war poems. Another, the version immediately prior to the typescript, shows him arranging the short poems into a sequence (arranged more or less chronologically) titled “Mornings in Hell”; this would have explicitly identified the short poems as a record of his struggle for transcendence rather than statements of it or indictments delivered from its security. In fact, the various conceptions of the book, traced from the 1944 notes, through the provisional tables of contents, to the typescript and final Random House version, show that each successive recasting was not only a different compromise between the impulse to present the poems as records of experience and to present them as doctrine but that the impulse to treat them as doctrine became progressively stronger. Had Jeffers assembled the 1944 version, it would have presented the poems as essentially a reflection of his struggle to cope with the war. A book organized as “The Double Axe” and “Mornings in Hell” would still have dramatized that struggle but shown him imposing an interpretive schema, gaining distance from his experience, and thereby some power over it. In the typescript version, adding a “Preface” and dropping the section title “Mornings in Hell” accentuates the political and poetic ideological claims while yet allowing the short poems (still somewhat chronologically ordered) to function as an implicit drama of failure and doubt. And finally in the Random House version, revising the “Preface,” dropping the short poems that most explicitly confess his doubt (“The Blood Guilt,” for instance) and some that simply document his anger, and reordering the short poems to obscure their chronological unfolding pushes the drama of self into the background and brings the book’s argumentative claims more clearly to the fore.
The character of the shift from the 1947 typescript to the Random House version can be glimpsed in the contrast between their “Prefaces.” In the Random House “Preface,” Jeffers states that the book “is to present a certain philosophical attitude, which might be called Inhumanism,” but in the typescript “Preface” he claims that the book, in particular “The Inhumanist,”
…presents…a new attitude, a new manner of thought and feeling, which came to me at the end of the war of 1914, and has since been tested in the confusions of peace and a second world-war, and the hateful approach of a third; and I believe it has truth value.
In both the poems are cast as illustrations of an “attitude,” and this downplays their personal dimension. But in the typescript “Preface” this “attitude” is a “manner of thought and feeling” linked to both circumstance and emotion. In the Random House “Preface,” it is a “philosophy,” and Jeffers underscores this by using the rest of the passage to introduce its tenets. It is, then, no surprise that he dropped, for the Random House version, short poems that focused on his earlier despair or that stressed too simply his disgust with current events. Such reactions, though tolerable in a collection illustrating how his “new manner of thought and feeling” had emerged from his “confusion” and the “hateful” contemporary realities, would have undercut his claim that his “philosophical” “attitude” could allow one to face the worst of the political world with equanimity. Neither is it surprising that he retained the political parallel to his “Inhumanist” position, the isolationist critique that so upset Commins, nor that he muted the attacks on Roosevelt, which (insofar as they could be construed as simply personal spite or rage) similarly compromised his claim to equanimity and so detracted from his desire at this later point to use Roosevelt to exemplify larger historical processes. In effect, in 1944 Jeffers was inclined to present the work as a poetry of experience, after Medea and Commins to present it as a poetry of idea, and in between, including the typescript, to explore different ratios of the two (but with each successive version placing increasing emphasis on the realm of idea at the expense of the realm of experience).
Whatever the aesthetic merits of the actual and potential versions of The Double Axe, Jeffers’ evolving sense of how to present this material shows that he was aware from the start of the gap between what it meant for him to write the poems and for him to publish them. He was, that is, implicitly engaged with publishing as a public and social act—even in the almost purely confessional volume he might have assembled in 1944. One cannot, that is, claim that it was Commins who distorted or compromised the integrity of Jeffers’ poems as private, or authorial, expression by leading or forcing him to consider how the public might react to the work. Jeffers had been doing this from the moment he began trying to assemble the volume, as the example of “Tragedy Has Obligations” suggests. In this poem, dated June 1943, Jeffers anticipates Germany’s defeat and exhorts Hitler to play the role of tragic victim, to rise to the “obligations” he has incurred, rather than become simply “pitiful.” The poem, clearly finished since it exists in a clean draft apparently copied from earlier workings, is one of Jeffers’ strongest from this period, yet it appears in none of the draft tables of contents, which suggests that he never considered publishing it. He did not, that is, need Commins to tell him that a poem allowing Hitler to appear as history’s tragic (even potentially noble) dupe would be intolerable to an American audience—even in 1944 when neither poet nor audience had yet learned the full extent of Hitler’s policies—and would destroy any chance of the other poems being effective. For Jeffers, then, the act of publishing was an attempt to mediate and balance his sense of writing as a private gesture and as a public gesture, and it is this dynamic that explains his willingness to recast the typescript. As he grew more distant from the urgency of his reaction to the war, more distant from the doubts and anger of his “Mornings in Hell,” he became increasingly inclined to shift the balance between private and public and to select and order the poems in ways that would obscure the terms of their composition in order to foreground the lessons he thought he had subsequently learned. As such his own desire to accentuate his increasingly ideological sense of his poetic position led him to moderate his outrage. Here, again, one must admit Commins’ role in triggering the final reworking of the collection but still conclude that Jeffers was neither motivated nor directed by Commins’ scruples—even if he did agree to let Truman be “Harry” rather than “little.” I should also note that “Poetry, Gongorism and a Thousand Years,” the essay Jeffers wrote for the New York Times after Medea opened, after receiving Commins’ critique of the typescript, and just before recasting it assumes, if the process I’ve outlined is generally correct, quite particular importance as an implicit rationale for the final form of The Double Axe and can be read not simply as Jeffers’ general explanation of his practice to that point (as it has typically been read) but as a statement of the revised aesthetic that emerged as he recovered his creative equilibrium after the war.
If, as I am arguing, Jeffers recast The Double Axe typescript freely and for his own purposes, one could then construct two different arguments for the basic textual validity of the Random House version of The Double Axe: From an author-centric perspective one could claim that this published form most completely realizes his final intentions, and from a socio-centric perspective one need only note that the final version was the one that actually entered circulation and that the typescript has little more claim to being a social product than the other forms the various preliminary tables of contents show the collection might have assumed. And yet neither of these arguments is fully satisfying, since one could also claim that the typescript version more fully reveals the presence of Jeffers in his own work and as well, if paradoxically so, more fully reveals him enmeshed in, as the product of, social structures and events. In other words, we can imagine at least some author-centric and socio-centric practitioners wanting to argue for the typescript as the work’s preferred form, though perhaps choosing not to do so.
In part this impulse to argue for a form of the collection that our textual theory suggests should be secondary stems from a basic peculiarity of the typical poetry book, a peculiarity that can be clarified by turning briefly to Hershel Parker’s argument in Flawed Texts and Verbal Icons. Parker notes that writers returning to revise earlier work (on their own or at an editor’s request) often deviate from their original sense of a project and produce clearly authorial revisions that are, nonetheless, aesthetically problematic, even unfortunate, and he argues that we should distinguish between the times when writers function creatively and those when they function editorially. Quite reasonably, that is, he asks that we posit a relationship between the impulse that generates the text, its imaginative materials (settings, characters, imagery), and its structure and that we should prefer these relationships to remain congruent as the text evolves through both its composition and subsequent reworkings. For Parker writers risk compromising this congruence when they revise at too great a remove from their original and originating impulses, especially when forced to rethink a book for reasons distinct from those original impulses.
Parker’s analysis offers a way to argue that an earlier form of a work may in some cases be textually preferable to a later form, even when the author has been fully involved in each, and this rationale might plausibly be applied to The Double Axe. Jeffers did, after all, shape the Random House version at a considerable chronological and psychological distance from the impulses that generated the material in it and in doing so altered, even compromised, those impulses. But Parker’s model derives primarily from his analysis of novelists, and the difference between a novel and a collection of poems has implications for understanding The Double Axe. Although composing a novel is a lengthy process, a draft is, or is supposed to be, a single process of composition. A collection of poems, though, represents multiple compositions; each poem is essentially an individual realization of its own equation of impulse, materials, and structure, and because these acts are also at root independent (the poet typically writes pieces without a clear sense of the book they will form), a poetry collection can only—unlike a novel—be created retrospectively and at a distance from the intentions that governed the original compositions. As such, it is both a product of multiple compositions and of different orders of composition—of writing and of selecting/arranging—that will differ in nature if only because writing poems is more apt to be governed by the poet’s engagement with his or her immediate experience, while constructing a volume will be governed more by the poet’s sense of audience, role, and ideological or commercial reasons for making the poems public. Novels, of course, also mix sets of intentions that derive from writing as a self-reflexive act and an act directed toward the realities of production, but with novels these sets tend to interact more clearly and fully from the beginning, while for poetry one set is more apt to direct composition, the other presentation. In a collection these two sets may blend to some extent, to some extent conflict, but will not be identical. In the case of a poet like Ginsberg the gap between the experience that generates a poem and the ideology of its presentation may be relatively narrow. For poets like Eliot, Jeffers, and Pound who transform the autobiographical into texts not specifically personal, the gap will necessarily be wider, and the potential dissonance between the generating experiences and ideology of presentation potentially more troubling, fundamental, and significant. (Whitman might be worth considering in this regard as a poet who constructed elaborately public representations of a self that often differs from the private energies directing the individual pieces, yet who asks the reader to see the poems as transparent self-revelation.)
In any case, a poet’s decisions about what poems to present and how to present them make the book an interpretation of its individual pieces, and this transforms the collection from a transparent encounter with what the poems might have meant to the poet during composition to an encounter, at least partly, with the poet’s sense of what the poems might mean to a particular audience. As such, reading novels and reading collections of poems are qualitatively different. As Parker shows, we prefer a novel’s episodes, scenes, and stylistic gestures all to contribute to a coherent set of intentions; we judge the elements by their contribution to the whole and assume that the ordinate intentions are those governing the whole novel, not its pieces. With a collection, though, each poem has its own textual dynamic and integrity, but so does the collection, and the compositional and organizational acts are usually separate in time and may well function at cross purposes. We cannot, though, simply prefer the original and individual poems to the structures the writer later creates for them, since the poet’s act of selecting and arranging work for publication functions itself as a kind of composition. A collection, that is, resembles a movie, in which individual scenes have been adjusted, rearranged, recontextualized after the actual moment of filming. As such, it is both an interpretation of existing work and in part a new work, and like a movie the collection becomes a context shaping our perception of individual units by obscuring certain original intentions, highlighting others, imposing still others.
The implication of Parker’s model, when applied to poetry, is that each act of constructing a collection is potentially illegitimate, since each act of collecting is inevitably and essentially retrospective and governed by aims differing in kind from those that generated the material, and if this is the case, it also suggests that we cannot, then, conclude that one act of collecting is more valid than another simply on the basis of its being closer in time to (or further from) the original compositions, since we are dealing with a qualitative rather than a quantitative issue. Thus, for The Double Axe at least, to prefer the selection and arrangement that least distorts the intentions originally governing the poems would lead us not simply back to the typescript but to arranging the short poems as “Mornings in Hell,” then to the 1944 conception, and finally to the poems in some hypothetical uncollected state. Jeffers did not, that is, suddenly realize or pervert the author-centric dimension of individual poems when he encountered the socio-centric reality of publishing in the form of Commins’ political scruples and sense of market; rather, each act of collecting was a different way to realize and “pervert” originally independent pieces. Moreover, one could argue in this case that Jeffers had already compromised the agency of primary importance from an author-centric perspective, already compromised his subjective integrity, in choosing the war as his focus (or being chosen by it); the poems, that is, were enmeshed from their inception in the social, and their didactic edge had already made them a public, even adversarial, gesture. Each version of the book, then, and the individual poems inscribe conflicting public and private impulses, original and secondary intentions; each is a compromise.
But none of this yet answers the question whether the typescript or Random House edition (or for that matter one of the earlier plans Jeffers considered) should be the preferred form of The Double Axe. If the evidence did indicate that Commins coerced Jeffers into reworking his collection by threatening to suppress it, most editors, whatever their textual persuasion, would opt for the typescript. Jeffers did, after all, view it as complete and final when he sent it off; he prepared it without outside meddling; and it is the version that most forcefully reveals the intricate conflict between his desire to transcend history and politics yet his fixation with them. But Jeffers was not coerced, though it would be textually convenient to think he was, and the published version (in spite of, because of, without regard to Commins) most fully resolves Jeffers’ desire—already implicit in the 1944 notes—to translate the poems’ personal dimension into a public, ideological gesture. In other words, each version is sufficiently purposeful, realized, and stable to deserve being read, and each version more richly reveals Jeffers as both authoring subject and object when read in counterpoint with the other and when both are read against the other forms he considered for these poems.
In practice this dilemma can, of course, be resolved easily enough by preparing editions of both versions or an edition of one with an apparatus allowing readers to construct the other. But this solution does not resolve the question of whether one should understand texts as primarily authorial or social constructs or whether one position or the other can be used, finally, to determine the most legitimate version of a text. My sense is that the nature and history of The Double Axe suggests we can only adequately account for either the nature of writing or the nature of texts if author-centric and socio-centric modes are treated as complementary and dialectical, not as dichotomous. I am· suggesting, that is, that we should construe the typescript and Random House versions of The Double Axe as two performances of a single work, each realizing a central cluster of elements and dynamics under different conditions, with a different sense of occasion, and with different results—much as we might see each of Ellington’s recordings of, say, “East St. Louis Toodle-oo” as textually valid and authentic and yet continue to explore (even disagree over) their relative critical merits and implications. What we cannot do, I would suggest, is advocate one form at the expense of the other by virtue of its supposedly being more “author-centric” or “socio-centric” or even by assuming that an author-centric or socio-centric rationale somehow more adequately accounts for a work’s production, form, and transmission. We are both subject and object, creating and created, and if that unfortunately complicates our attempts to create and our attempts to understand what others have created, it also enriches; it is what makes the more than game worthwhile.
David Nordloh and James West were both kind enough to take me to the woodshed for my discussion of The Double Axe at the STS meeting; both reminded me that critical arguments are not necessarily textual ones. Neither should be held accountable for the quite different scope and direction of this piece. In addition I wish to thank Professor Nordloh for his criticisms of the first draft of this essay.
The quotations from the various letters and memos of Bennett Cerf, Saxe Commins, Robinson Jeffers, and Una Jeffers appear here with the permission of Jeffers Literary Properties, Random House, the Butler Library of Columbia University, and the Harry Ransom Humanities Research Center of the University of Texas.
 See “An Inquiry into the Social Status of Texts and Modes of Textual Criticism,” Studies in Bibliography, 42 (1989), 55-79 (especially 72-75). In the first pages of this essay my goal is not to offer a critique of Professor Shillingsburg’s “Inquiry” but rather to set up a simplified, generalized version of one part of his model in order to use it to explore my thoughts (as an ad hoc editor) after attending my first Society for Textual Scholarship meeting. I was puzzled by the tendency of people at times to resort to formulations that seemed overly dichotomous, particularly as regards the nature and existence of “intentions.” Some (champions of the “author) seemed to see “intentions” as absolute (however much they might disagree as to the concept’s meaning and application); others (those who believed in “society”) simply denied that such things existed; and both camps alluded to their theoretical champions. The disagreements are, of course, real and important, but my purpose here is to suggest that our textual scholarship and the criticism we base immediately on it is in practice often much less dichotomous than our theoretical contests—and that it should be. At another level, I would like to think that this essay picks up Professor Shillingsburg’s invitation to try out his perspective in practice.
 The typescript’s table of contents is with the Jeffers papers at the Harry Ransom Humanities Research Center, the University of Texas, Austin (hereafter noted as HRC). These papers also include typescripts for the deleted poems (some with Commins’ queries) numbered to correspond to the typescript’s table of contents, drafts of tables of contents preceding the typescript, annotations that document Jeffers’ various steps in restructuring the typescript into the published form, and workings for several different “Prefaces.” This material can be used to recover the collection’s evolution. A more detailed description of these versions than I can manage here is forthcoming in the Robinson Jeffers Newsletter, and a complete record of the revisions to specific poems will appear in the fourth and final volume of The Collected Poetry of Robinson Jeffers, ed. Tim Hunt (forthcoming, Stanford University Press).
I should also add that Jeffers did not rework his earlier collections in this manner. In several cases he added poems to collections in press, and in one case he substituted a newer poem for an earlier one, but in no case did he substantially alter the contents or organization of a volume after sending copy.
 Saxe Commins to Robinson Jeffers, October 1S, 1947, HRC.
 The first study of Jeffers’ adjustments to The Double Axe was James Shebl’s In This Wild Water: The Suppressed Poems of Robinson Jeffers (Pasadena: Ward Ritchie Press, 1976). Shebl puts the responsibility for the changes on Commins and Random House. He stops short of saying the book was censored, but finds it difficult to explain why else Jeffers “allowed” the poems to be “excised” (p. xi). William Everson and William Hotchkiss follow Shebl’s lead in their edition of The Double Axe (New York: Liveright, 1977), printing the ten poems Jeffers dropped from the typescript as an addendum labeled “Suppressed Poems.” Conversely, Robert Ian Scott, in the commentary to his collection of previously unpublished Jeffers poems, What Odd Expedients (Hamden, CT: Archon Books, 1981) “find[s] no evidence that Random House suppressed any poem by Jeffers, or that leaving these poems out of The Double Axe confused anyone but Shebl.” In Scott’s view, Jeffers dropped the poems because they “complicated” the book and were “apt to confuse readers” (12).
 Letters documenting Jeffers early dealings with Cerf and Random House are in the Random House Collection, The Butler Library, Columbia University (hereafter noted as RHC). Descent to the Dead was included in Give Your Heart to the Hawks (1933). This collection was set in type by Liveright but actually issued by Random House (from the Liveright plates) after Liveright went bankrupt
 This is not to suggest that Random House had any compunctions about altering Jeffers’ punctuation, but the adjustments, whether appropriate or not, were in no way attempts to modify the tone or content of the work.
Jeffers’ earlier titles, especially the Modem Library reissue of Roan Stallion, Tamar and Other Poems (1935) and the 600-page The Selected Poetry of Robinson Jeffers (1938), were a prominent part of the Random House backlist—further evidence of Jeffers’ position.
 Donald Pizer explores how Crane’s and Dreiser’s lack of status and experience complicates our assessment of these works in “Self-Censorship and Textual Editing,” in Textual Criticism and Literary Interpretation, ed. Jerome J. McGann (Chicago: University of Chicago Press, 1985), 144-61.
 Commins to Jeffers, October 15, 1947, HRC.
 In a note (November 18, 1947) to Anderson, Una Jeffers notes that Jeffers “interrupted…a thing he was writing just when it was going full tilt” to begin “the Medea.” The “thing” Jeffers set aside may have been either the narrative ‘”The Love and the Hate” or a subsequent narrative unit later recast and incorporated into the narrative “The Inhumanist.” The Judith Anderson Jeffers Collection, The Tor House Foundation, Carmel, California.
 Commins to Jeffers. December 4. 1947. HRC.
 The revised typescript does not survive, but it was in large part simply the original typescript rearranged with some sheets newly typed, some emended. The revised typescript had a one paragraph “Note” instead of the “Preface” eventually used. Jeffers substituted the published “Preface” after Commins sent him the draft of the “Publisher’s Note” that Random House had decided to add to the collection. In the letter accompanying this “Preface,” see below, Jeffers also agreed to modify two phrases. Commins on this authority apparently altered the revised typescript.
 February 12, 1948, HRC.
 February 19, 1948, RHC.
 The Double Axe and Other Poems (New York: Random House, Inc., 1948), p. ix.
Although Jeffers worried that the “Note” would “make every reader think of politics rather than poetry” and found this “deplorable,” he instructed Commins in his letter of February 19, 1948 (RHC) to “put it in by all means, if it is a matter of conscience.”
 I explore the connection between Jeffers’ sense of nature and his aesthetic in ‘The Problematic Nature of Jeffers’ Tamar,” in Centennial Essays for Robinson Jeffers, ed. Robert Zaller (Newark, Delaware: University of Delaware Press, 1991), pp. 85-106 and in “Nature, Narrative, and Knowing: Robinson Jeffers and the Mode of Roan Stallion, which accompanied The Yolla Bolly Press (Covelo, California: 1990) edition of Roan Stallion.
 In the “Note” that introduces Be Angry at the Sun Jeffers “lament[s] the obsession with contemporary history that pins many of these pieces to the calendar, like butterflies to cardboard.” This was the first time (The Double Axe would be the second) that he felt compelled to explain, to apologize for, what the reader was about to encounter. The Modern Library Roan Stallion. and the 1938 Selected Poetry do have introductions, but these were written at the request of Random· House. I presented a paper (“The Thickening Empire: Jeffers’ Struggle with History”) on this “Note” and its implications for the collection at the May 1990 American Literature Association Convention.
 Commins found this passage offensive and Jeffers eventually modified it to read “To feed the power-hunger of politicians and make trick fortunes/ For swindlers and collaborators.” See The Double Axe, p. 27, and The Collected Poetry of Robinson Jeffers, Volume Three, p. 234.
 The notes are at the bottom of the first of the three manuscript sheets for the poem, titled in this early version, “Eve of Invasion,” HRC. In a December 7, 1945 letter, Jeffers also writes: “Meanwhile I have been writing a narrative poem called ‘Rene Gore’ [early title for “The Love and the Hate”], and many shorter pieces, which together will make a book to be published in due time,—probably next fall. I hope to add something to the book before that.” The letter appears in The Selected Letters of Robinson Jeffers: 1897-1962, ed. Ann N. Ridgeway, (Baltimore: The Johns Hopkins University Press, 1968), p. 303.
 Actually the typescript version included one lyric, “Look All Around You,” written later and inserted at the last minute as page 154 1/2. This was one of the ten poems Jeffers deleted from the Random House version. It is perhaps worth noting as well that the typescript included several poems written in the half year before the U.S. entered the war; these too were dropped.
 The Double Axe, p. vii.
 Everson and Hotchkiss print a version of the typescript “Preface” in their edition of The Double Axe. See pp.171-72.
 Jeffers does admit in the Random House “Preface” that the “first part” of the book (i.e. “The Love and the The Hate”) “bears the scars” of having been written during the war but then asserts that “the poem is not primarily concerned with that grim folly.” The disclaimer is true in a sense, but also somewhat misleading. It is also worth noting that the draft of what became the Random House “Preface” is less forceful than what he finally printed. After typing in the claim that the poems “present … the rejection of human solipsism and recognition of transhuman magnificence,” he adds in pencil above the line (then crosses out): “(But I ought rather to have said, some limping progress toward this attitude.)”
 In spite of his anger at Roosevelt, Jeffers’ sense of his actual historical significance and Hitler’s as well seems to have been shaped by Spengler’s prediction that the West would be entering a period of Caeserism. Significantly (see “Tragedy Has Obligations), he found it easier to view Hitler as a tragic pawn enacting an inevitable historical process than to view Roosevelt this way.
 Everson and Hotchkiss include “Tragedy Has Obligations” in the “Suppressed Poems” section of their edition of The Double Axe (pp. 158-59) even though the poem was not part of the typescript Jeffers originally submitted to Random House. Certainly they are right in assuming that Commins would have found the poem extremely troubling.
 The New York Times (January 18, 1948). sec. 6, p. 6.
 Hershel Parker, Flawed Texts and Verbal Icons: Literary Authority in American Fiction (Evanston: Northwestern University Press, 1984).
 Jeffers’ decision, for instance, to combine “Pearl Harbor” and “West Coast Black-Out,” poems written six months apart, into Parts I and II of “Pearl Harbor” and to use the later unit as Part I significantly alters, I’ve argued elsewhere, the way we are likely to read these units. See, “To Date or Not to Date: Jeffers’ ‘Pearl Harbor,'” Robinson Jeffers Newsletter 16 (1989), 15-18.
 Were I to edit the collection, I would choose the Random House version as the basic text for the reasons I have already cited but would stress both the textual validity and critical importance of the other completed form of the work—the typescript. Actually, to be more accurate, I would adopt the revised typescript that served as the basis of the Random House edition. I would, that is, accept as authorial Jeffers’ decision to delete the ten poems and restructure the collection but reject the few subsequent adjustments in phrasing (“little Truman” for “Harry Truman,” for instance) that appear in the Random House edition and which Jeffers accepted primarily to placate Commins. These adjustments were perhaps relatively indifferent matters to Jeffers, but they stem from Commins’ agenda for the project. not Jeffers’ own.