Drawing extensively on archival research, The Late Cantos of Ezra Pound critically explores the textual history of Pound's late verse, namely Section: Rock-Drill (1955) and Thrones (1959). Examining unpublished letters, draft manuscripts and other prepublication material, this book addresses the composition, revision and dissemination of these difficult texts in order to shed new light on their significance to Pound's wider project, his methods and techniques, and the structures of authority--literary and political-that govern the meaning of his poetry. 



1. Introduction: “The Text is Somewhat Exigeant”: the Lateness of Pound's Late Cantos

2. “I Have Always Loathed Reading”: the Trouble with Philology

3. “To Copy and Amplify”: Totalitarian Scholarship in Section: Rock-Drill

4. “No, That Is Not Textual”: the Immaterial Language of Thrones

5. “A Butcher's Block for Biographers”: On Pound's Quiddity

6. Conclusion: “Ego, Scriptor Cantilenae”: Authority Reconsidered

Appendices and Figures







Michael Kindellan. The Late Cantos of Ezra Pound: Composition, Revision, Publication. London: Bloomsbury, 2017. xx+276p. £85. ISBN 9781474258746. 


review by Massimo Bacigalupo



Michael Kindellan’s study is a brilliant account of a fascinating story: how the writing and publishing of the late cantos came about among countless biographical and textual obstacles and conundrums. Pound emerges in full as the problematic source of the text,  trying to control, direct, compose his life and his major work, and only succeeding in part on all fronts. But the drama of the battle is well worth watching and recounting. Besides marshalling a prodigious amount of chiefly textual facts, Kindellan advances a proposal for a new approach to reading and editing Pound. So his book makes also a difference in our approach to the man and the poem and can therefore be considered a milestone in the long story of Pound studies. By looking so carefully at the text, Kindellan can for once set aside the whole subject of Pound’s “opprobrious politics” (97), of which one hears to no end these days, and takes us in that uncomfortable space which is the pages of The Cantos, where finally we get back to work, watching in amazement as draft becomes typescript becomes print and volume – and endless agonizing discussion and anger and wonder. 

The Late Cantos has four chapters, with apt Poundian phrases as titles: 1. “I have always loathed reading”; 2. “‘To copy and amplify’: Section: Rock-Drill”; 3. “‘No, that is not textual’: Thrones”; 4. “Or true editions?”. Chapter 1 is chiefly about Pound’s misreading and resistance (or animus) towards philology, which he caricatured as the nitpicking of drudges, “obscuring the texts with philology.” The Cantos, largely composed as it is of reports on Pound’s reading, is written by a man who loathes the act of reading as such and instructs a would-be translator “dont bother about the WORDS, translate the meaning” (13). A habitual misreader who inevitably invests quotations and ideograms with  idiosyncratic interpretations, Pound does not allow for any freedom of the sort in his own readers, who must “git the ideaHHHHH” (226), share his convictions, and know what he means regardless of the text and its corruptions. Kindellan, in the impressive theoretical scaffolding of his book, brings up the distinction between “text” and “work” (229), the former the concern of philology, the latter an ideal, even metaphysical, entity, “now in the mind indestructible.” Thus it is a mistake to think of The Cantos as an opera aperta (Umberto Eco’s phrase), i.e., open to infinite interpretations. There is only one meaning, and it is the one Pound intended. And this (the work) “coheres all right / even if my notes [the text] do not cohere.” 

Kindellan in the opening chapter considers closely the strange Chou King cantos (85-87), studded with ideograms, claiming generally that “when Pound offers in-text references to external sources he is not in fact admonishing his readers to consult independently the passages to which he refers; normally, Pound is attempting to relieve us of that obligation” (39). All we have to do is accept his word about what a text proves or means. Now, it is true that if we look at the source we will often discover that Pound’s interpretation is wide of the mark, but one still can defend the analogy of The Cantos with a set of lecture notes for some ten courses (one per installment). As a class, we are supposed to procure the text under discussion. But this teacher, being authoritarian, as doubtless many lecturers are when telling students about “Sailing to Byzantium” or “After Apple-Picking,” expects us to unquestioningly accept his findings. 

Kindellan has studied very carefully the notebooks containing drafts of these cantos and shares valuable discoveries (and unpublished fragments). On pages 29-32 he reproduces a section of a page from Couvreur’s Chou King, Pound’s notes in the relevant spiral notebook, and a page of the proofs of the Milan 1955 Rock-Drill. He does so to prove a rather esoteric point (unconvincing to this reader) about Pound copying in his notes the accents in Couvreur’s transliterations of Chinese and adding these accents irregularly by hand to his typescripts and proofs. But the possibility of seeing the text go through its various longhand, typescript, proof and published incarnations is per se arresting. In his final chapter about projected editions of The Cantos Kindellan envisages, in the footsteps of his mentor Richard Taylor, an online archive in which it would be possible to inspect all of these variant forms, and all materials related to given readings (correspondence, published and unpublished prose, etc. etc.). This is mind-boggling though not impossible since Pound curiously hoarded most of his notes and drafts, and archives over the two hemispheres are rich in materials of all kinds for the inquisitive (“serious,” as Pound would say) researcher. Kindellan is surely one of this elect company, as his painstaking references to archives chiefly in Italy, Germany, England and the U.S. prove. He offers a wealth of material, as well as a method and a theory for other “wandering clerics.” (He is less thorough with his own book. I tried to locate through his index a quotation from Mary de Rachewiltz that had struck me, unsuccessfully, for the quote is on page 193, and the index lists no reference to her after page 142. I hasten to add or repeat that Kindellan’s scholarship is impressive.) He is also witty in a subdued way, quoting for example G. Thomas Tanselle in a note on page 229: “The world of documents is a world of imperfection,” and adding: “This book being no exception!” 

That Pound was witty is proved often in these pages, so in Kindellan he has found an unusually sympathetic reader, who can suspend disbelief long enough to situate himself (and us) on Pound’s wavelength. In any case, for a subject so complicated, intricate and esoteric, Kindellan’s account is extraordinarily error-free from the scholarly and interpretative point of view. (The references to secondary sources, i.e., the whole range of Pound studies, are exceptionally thorough.) Besides, this book is unusually well-written, so that there is hardly a page that does not provide a perceptive comment or quotation.  And since, as Kindellan admits, he does not overtly engage in interpretation, there is little repetition and soul-searching about Pound’s subjects and beliefs. There is just “the event of writing” (99) occurring under our eyes. Conclusions are drawn unblinkingly but not dwelt upon: “These manuscripts ... are also sincere in the modern sense of displaying an absence of ‘dissimulating, feigning or pretence’: ... Pound puts Hitler at heaven’s door, a decision that shows the limitations of sincerity as a sole criterion for ethical care” (93, 95). One could object that only the skilled in fire will discover the Führer in canto 90’s “beer halls” and “furious ... perception.” Pound is tricky and covers his traces on taboo subjects. 

The paranoid character of the late cantos is also addressed by Kindellan: written in a madhouse, secretly disseminated, hidden even from long-standing supporters. The publication of Rock-Drill in Milan by twenty-year-old Vanni Scheiwiller was also intended to spite Laughlin and Eliot. Considering possible publishers, the poet wrote his daughter in November 1954, when he was about to mail the Rock-Drill typescript to Vanni: “certainly NOT Mr Eliot/ Mr Eliot can buy it if he likes/ He has left the Kung/ Anthology out of print for five years ...” (115). However, we find that Pound’s suspicions were partly shared even by James Laughlin, who attributes to Pound’s enemies delay in printing cantos, and communicates that his new printers “have no objections and don’t foresee any pressure from their union, which is not Jew controlled” (110 — letter to Olga Rudge, date not given). Just as Hemingway at the end believed he was watched by the FBI (and he was right), so with Pound, given his polarized circumstances, it is hard to tell truth from paranoia. 

Still, it is surprising to find out that Pound actually feared being caught in error by the professors whom he despised, as if anyone would ever take the trouble of reading Thrones to discover wrong spellings of Greek words! But this Pound believed, while on the other hand insisting on idiosyncratic spellings for poetic effect, as in the famous Leucothea/Leucothoe imbroglio. Kindellan’s discussion of Rock-Drill and Thrones really boils down to illuminating analyses of a few central cruces, like oinos aithìops and xaladines in canto 97, epos vs. eros in 102, “wars after ’70” in 103, Chinese transliterations in 99 (the entire discussion of the Sacred Edict cantos is exemplary), Deianira in 87. Pound is unaccountably scared of scholars: “At any rate it will be a field day for the Spitzes [Leo Spitzer?] and Shitzerald’s [Robert Fitzgerald?], if grampaw don’t correct Mr Storr, and if he corrects him wrong” (to Scheiweiller on Deiainira, 21 April 1955, p. 118). Actually “Mr Storr,” editor of the Loeb Trachiniae, was not in error, only Pound did not know the difference between upper- and lower-case representation of iota subscript, which I’m pretty sure Scheiwiller did. Pound goes on: “at any rat [sic] there is a cedilla under the atah … wich we HAVE got indisputably.” Kindellan comments: “Pound’s publishers never did insert a ‘cedilla’ (iota subscript) under the eta in any of his texts” (118). Here for once he is wrong, unless I misread him: both the Scheiwiller Rock-Drill (31) and the Faber 1964 Cantos (607) do have a rightful iota subscript in Deianira (“Δάνειρα”); only the New Directions text omits it, which is rather strange since it is supposed to be an offprint of the Scheiwiller edition. But if you look closely at page 571 of the 1973 New Directions volume you will see that the Greek in the last two lines is in a different typeface, more like italic, than the Greek at the top of the page. So for some reason the two lines were reset, and here the error (sabotage!?) occurred. (Confirming, if needed, that the current New Directions text is the most unreliable Cantos ever printed.) 

In his final chapter on editions, Kindellan recounts the strange tale of the misguided attempts to correct The Cantos, based on the crazy idea that spellings and even facts had to be checked against reliable sources, and perhaps notes were to be introduced explaining that some statement was not really accurate. We see that Pound did on and off demand a few sparse corrections, but rightly believed that his message was pretty clear despite some inaccuracies (his fear of being faulted by the profs is another matter). But then, the moment Pound ceased to be able to express an opinion on these matters, which was as early as 1959, the “editorial committee” (216-27) set to work. The letters exchanged and quoted here are revealing of the fact that these well-meaning correctors did not understand that an imaginative work is not bound to any external standard but its own processes, and that even lapses of attention and involuntary errors are significant as reflecting the act of writing under given peculiar circumstances. What is more alarming is that the correctors knew so little about Pound’s intentions and sources, so for example in the penultimate page of canto 96 “know the Manuel” (manuel is from Pound’s French source) is “corrected” to “know the Manuale.” A little knowledge is a dangerous thing (here the corrector believed that Pound wanted an Italian word and went on without a scruple to change the text accordingly). But as Mary de Rachewiltz usefully noted in the statement I spoke of above: “it can be safely assumed that the last spontaneous ‘corrections’ made by Pound himself were in Thrones, Canto 103” (i.e. the “wars after ’70” affair, in 1959). So, though one sympathizes with Kindellan’s dream of a hypertext that will contain all the tales of the tribe in their countless incarnations, a more realistic project would be convincing Pound’s publishers to reprint The Cantos as it was before the correctors had their field-day. What they perpetrated is scarcely relevant to the story of the text, though it tells us a lot about (mis)conceptions of it. And about ignorance. 

On the other hand, Kindellan’s knowing and balanced account provides a timely incentive to new scholarship and new readings. What he has accomplished is no less than a change of paradigm. This substantial book has reminded me of the simple and welcome fact that reading in (and about) Pound can be fun.





Michael Kindellan. The Late Cantos of Ezra Pound: Composition, Revision, Publication. London: Bloomsbury, 2017. xx+276p. £85. ISBN 9781474258746. 


review by Alec Marsh 



This is a very valuable book, a real contribution to Pound studies covering an epoch of his work that still remains relatively unexplored, the “Washington Cantos,” Rock-Drill (1956) and Thrones (1959). These last authorized sections of Pound’s epic poem are notoriously difficult even to read, much less understand. Liberally salted with Chinese ideograms, some Egyptian hieroglyphics and even notionally Sumerian writing, reliant on rare and far-out texts, they are also compromised by dozens and dozens of textual anomalies, of which some are outright errors and others deliberate distortions. “Any close scrutiny” of these cantos, “will raise questions about the integrity of the text. Not every canto contains a ‘howler,’ but the theatrical self-consciousness of their status as written artifacts might solicit curiosity about how they were written, revised and published” (234). “'Pound’s opacity' is intentional,” Kindellan claims; above all, he argues, these are “writerly” texts, flaunting their writtenness and daring readers to either read the implied curriculum of sources or to accept the poem as gnomic wisdom writing, its esoteric core hidden beneath exoteric chaff. Scholars usually take the first option; disciples hip to the poems’ encrypted agenda, like Sheri Martinelli or John Kasper, could ignore the philological rabbit-holes because they already knew what was at stake, politically. They knew that “the model of understanding The Cantos tries to realize is one where philological difficulties can be overcome by belief” (18-9). Regardless, these cantos “constantly foreground writerly concerns, making them into topics and subjects of the poetry” (xi). 

Pound’s antipathy to philology is well known; loathing it himself, he was nonetheless the cause of it in others. Aware that Pound would somehow have it otherwise, Kindellan, like the rest of us, must read the poem philologically—as if all the words mattered. As his subtitle suggests, Kindellan hopes that by “looking into their stages of development, from first inception to final published version(s), and extrapolate therefrom new understandings about the processes and procedures that governed Pound’s writing” (xi). But that means trouble: the poet’s early hostility to philology didn’t abate as he grew older, even as the poetry he wrote—especially in these late cantos—would seem to demand a philological approach. But most of us who’ve gone down that road come back baffled, shaking our heads. It’s not just the errors of fact, it’s the perverse twisting of things to their opposite. Why is it, I’ve asked myself more than once; why is it, Pound would change “And we bjJayzus regret your damn bishops/  have lied” perfectly plain in the first typescript of Canto 105, to the nonsensical “And we bijJazus reject your damn bishops” in the published poem (105/ 769); why in the same poem, has the unlikely “Loans from Tibet” replaced the plausible “Loans to Tibet..” already written at the end of canto 87 (87/ 596)? Everyone who has looked closely into these Late Cantos will have their own pet irritations. 

And there seem to be errors Pound cares about and those he doesn’t (218). Why, given the many, many errors, typos, mis and dis-information evident to any careful reader, did Pound obsess about the gaffe over the Franco-Prussian War of 1870 (103/ 755) that prompted him to send distressed telegrams insisting on an errata slip. Yet he almost never roused himself to equally egregious errors noticed at the time by Eva Hesse, Mary de Rachewiltz, Achilles Fang and others. 

Philology seems to raise more questions than it answers. Perhaps, Kindellan proposes—and this is where things get interesting and very unsettling—perhaps Pound’s poem demands a kind of philological anti-philology—the reader’s philological urge can be used to identify the “sometimes coercive structures of intention that inform so much of Pound’s work”—both in a spirit of acceptance and resistance. His researches confirm that the text is unstable; the drafts that Kindellan consults and displays make things more, not less problematic; Pound proves to have a shaky—if not utterly idiosyncratic, if not completely mad—understanding of the Greek and Chinese languages that saturate these late poems. “We have yet to come to terms with the extent to which the most important of Pound’s textual influences were written in languages he could not properly understand,” Kindellan warns (158). Pound’s “maverick sinology” (160), his turn to notional Chinese sounds, rather than Chinese sense evident in Canto 99 produces babble—though Kindellan is too kind to say so outright (see 175). As he points out, many of Pound’s most salient key terms, like Aristotle’s metathemenon, (and xreia, which Kindellan does not address), Frobenius’s paideuma, and even the great ling character, which, (as George Kearns has shown) had not hitherto meant “sensibility,” are simply misappropriated by Pound for his own polemical purposes in the manner of Humpty Dumpty; that is, to show who’s master.

As Pound was more or less indifferent to his mistakes, (with the exception of the matter of wars after 1870), steadily refusing to get interested in correcting or revising his poem, it occurs to Kindellan that at least some of the “mistakes” are not only errors, but the result of decisions made during the composition process. How else to try to grasp the strange reversals of sense between drafts and published poem? “Pound’s method of composition is inextricable from editorial processes, processes that invariably obscure as much as they express” (32). Thus, Kindellan entertains the radical thought that “the poetics of The Cantos might justly be described as the intentional deformation of philological work as such” (16).

This idea of the cantos as anti-philology comports with the startling claim with which Kindellan opens his book; to whit, that Pound “loathed reading” and that “The Cantos as writing could itself constitute an explicit protest against reading” (2; Kindellan’s emphasis). Kindellan is perfectly aware of the irony of this. Many of the late Cantos—for example, the Apollonius Canto 96 and Canto 99, the Sacred Edict canto—are essentially Pound’s reading notes. But Pound’s aversion to reading is constantly expressed in his idea that his poem can save its readers from having to read the books the poet has had to plow through in order to write it. “The rock-drill,” Pound wrote Sheri Martinelli, was necessary to educate his followers till “kindergarten diplomas can be issued.” He was writing Rock-Drill, “to save you some strength, from 30 years war” so that, “the descent into hell is not wasted.”1And yet, because the poem is “written in flagrant contravention of normative standards of attention, diligence and care,” it demands so much more care from its readers that we’re forced back to check his sources (Kindellan 2-3). Thus, the further irony that “Pound’s attempts to circumvent, détourne, [sic] travesty and otherwise eschew the protocols of philology have led to the formation of many more philologists” (3). 

The poem’s “formal incoherence and fragmentation—more often aphasic than articulate” (214) puts us in an odd position as readers. Either the poem does cohere after all, as Pound hoped, or maybe it doesn’t, though Pound may have thought it did. We just have to trust that Pound knows what he’s doing. Kindellan argues that this requires us to imagine a “strong poet-guide and install him in the role of the poem’s dominant authority” (214), placing the reader in the student’s role, “reading with the mind of a grandson” (12; 85/570). Is this tacit assertion of authorial control somehow compensation for the actual, legal loss of authority, when Pound surrendered his legal personhood to “the Committee” in 1946 (209; see also Moody III: 246-7)? Perhaps “the bullish manners of Rock-Drill and Thrones” are “signs of eviscerated legal personhood,” and the “incessant reiterations” a reassertion of self (210). Kindellan doesn’t spend much time psycho-analyzing Pound, but it seems we must. How else to account for the way these Cantos work and fail; their insistence, their errancy? 

After reading Kindellan, one goes back to the poems as the record of a live performance—say, by The Grateful Dead, full of false notes, weird riffs and obsessive repetitions—flaws which, nonetheless, give these cantos their indescribable feeling of spontaneity and rapidity, their fascinating, undeniable power. 

Despite this feeling of spontaneity, Kindellan stresses how relatively unchanged these Cantos are from notebooks to publication—something that will come as a surprise to many. Unlike Oppen, whom Kindellan cites as a counter-example, Pound didn’t fiddle around. Minutia of punctuation and word choice didn’t matter much, it seems. Anyone who looks at the Notebooks can see that Pound cut swatches, or saved chunks of poetry for other uses, but Pound’s open form did not encourage the impatient poet to mess with phrases. His revisions have much more to do with juxtapositions or line breaks than word choice. In other words, Kindellan is adamant that Pound was quite sure about what he wanted to say. In a striking phrase (one of many throughout the book) he writes, “Pound’s texts are unstable not because he was unsure about his meanings; they were unstable because he was sure” (247; Kindellan’s emphasis). To read Pound properly, he concludes, “we have to allow his misconceptions to unfold according to the logic of their errancy” (249); that is, we have to accept them—it’s pointless to contest them. “The model for understanding The Cantos tries to realize is one where philological difficulties can be overcome by belief” (19). In fact, “what Pound’s anti-philological attitude is designed to facilitate is a staunchly author-centric conception of literary production” (22). We believe Pound sincerely meant it all, or we don’t. Intention, it seems, is everything.

But how can we believe or reject Pound’s strange ideas without doing some digging into his use and misuse of sources to understand at least what he intends? When we do this, Kindellan contends, we get less than we bargained for. We constantly have to salvage Pound’s intentions from his expression of them. This becomes extremely difficult in these late cantos for various reasons—he’s a political prisoner, he is under indictment for treason, he’s pretending to be insane and he badly wants to save his country from the Jewish-Communist conspiracy; or, as Yunte Huang has suggested, perhaps because he realized that he suffered from “undeniable logophiliac madness” (50), Pound needs to disguise his intentions. So he uses various codes to get them across to the few who have the will to understand his sub-text. Most don’t. Kindellan doesn’t go so far as to say the poem is a “botch,” but a reader of this work will have to let Pound’s despairing judgment cross his or her mind. That’s the unsettling aspect of this book. 

Kindellan doesn’t cover closely the evolution of every canto, but chooses representative texts and moments. Despite demonstrating the futility of a rigorously philological approach, we learn a lot about the gestation of the poems that is invaluable, regardless of our philological inclinations. I did not know, for example, that William Seagle’s Men of Law (1947) was in the source field of The Cantos, much less that it was the “dominant intertext” for the latter pages of canto 94, after Pound had put down Conybeare’s edition of Philostratus’ Apollonius. This is evident in drafts displayed here. Suddenly, a whole range of references leap into life. The research throughout is impressive, not only Kindellan’s command of the late cantos and the crucial Notebooks that sustain them, but also of Greek and Chinese philology, and the critical literature, especially on philology and textual criticism. 

Given Kindellan’s resort to an “anti-philological” philology, it’s a tad surprising to find him advocating for a variorum edition as he ends his book. Naturally, one is desirable, but except for tracing the “logic of Pound’s errancy” it is unlikely that assiduous scrutiny of source texts will help much in understanding the poem; “in the face of a proliferating and unstable textual record that a variorum would uncover and present for critical inspection, Pound’s meanings would not be affected by variation as such” Kindellan admits (215). There is no question, however, that the poem’s authority would be further vitiated by such close vetting.

If Pound’s poetics are anti-philological, for better or worse, his deeply held politics means that there is much more at stake in the late cantos than twitting the professors and making editors and publishers weep. Pound was presenting his own revisionist history of the world—the true history, condensed so that his ideas could be carried into action. Pound saw his late cantos as a fulcrum, even “a political weapon” as he told a number of correspondents. Using The Cantos as a solid basis, a poem that would not sqush, but give the acolyte four legs to stand on, Pound’s activist readers would move ahead to realign earth with heaven, oikos with polis with kosmos. Sheri Martinelli understood Pound’s purpose well, closing a letter to Pound of 12 December 1958: "it AINT/ WOT’S/ ON/ the paper/ it is/ WOT streams/ FROM/ it/ [/] the/ “message”/ as the Spades put it / [/] it EVOKES/ what in turn will/ re-NEW the same / [/] it is the OPPOSITE/ from/ the WORD."2





Moody, David. Ezra Pound Poet. III: The Tragic Years 1939-1972. Oxford: Oxford UP, 2015.

Pound, Ezra. The Cantos. New York: New Directions, 1998.

Huang, Yunte. "Ezra Pound, Made in China." Paideuma: Modern and Contemporary Poetry and Poetics 42 (2015): 39-56. 





1. Ezra Pound to Sheri Martinelli, 10 June 1954. YCAL 868: 12/4 “Pound, Ezra June 1-10.”

2. Sheri Martinelli to Ezra Pound, 12 December 1958 p.4. YCAL 43: 33/1392