13 June 2016 by Jessica Palmer
“Substantial, Verbatim, (Un)attributed, Misleading?” A Review Article on the Doubleday Dante (2007)
Madison U. Sowell
Brigham Young University
Dante Alighieri. Paradiso: A Verse Translation. Trans. Robert and Jean Hollander. Notes by Robert Hollander. New York: Doubleday, 2007. 916 pp.
This review of Robert and Jean Hollander’s widely publicized “Verse Translation” of Dante’s Paradiso evaluates first the originality and quality of the translation and then that of the scholarly apparatus. The question posed in my title echoes the title of an article written by Professor Wilfried Decoo, cited hereafter.
Using the John D. Sinclair translation, first published in 1939, I just completed my 25th semester of teaching Dante’s Paradiso. Having made thorough use of this bilingual version for decades, I am intimately familiar with its English prose, the opening tercet of which reads thus: “The glory of Him who moves all things penetrates the universe and shines in one part more and in another less.” I then opened the reputedly new “verse translation” by Robert and Jean Hollander and perused its opening tercet: “The glory of Him who moves all things / pervades the universe and shines / in one part more and in another less” (Parad. 1.1-3, emphasis added here and later to underscore differences, usually minor). The only change in wording the Hollanders made was to substitute the arguably more poetic pervades for the more textually accurate penetrates. (N.B. The Italian word being translated is penetra, which subtly introduces theParadiso’s pervasive erotic language. This eroticism, which becomes progressively more apparent, is reminiscent of Ovid’s amorous poetry and the Old Testament’s Canticum canticorum, for which St. Bernard, Dante’s third guide at the end of the Paradiso, composed an allegorical commentary. In a similar vein Dante employs the complementary term vagina a few terzine later in Parad. 1.21.)
For an example of another tercet that proves remarkably unoriginal, given that this translation is being touted as a “new version” of Dante (see dust-jacket blurbby Bernard Knox from the New York Review of Books), I cite Paradiso 1.82-84, which in Sinclair appears so: “The newness of the sound and the great light kindled in me such keenness of desire to know their cause as I had never felt before.” Whereas the Hollanders modify these three verses ever so slightly: “The newness of the sound and the bright light / lit in me such keen desire to know their cause / as I had never with such sharpness felt before.” The translation of “grande lume” as “bright [rather than ‘great’] light” is more pleasing to the ear accustomed to rhyme; however, the “light / lit” juxtaposition is poetically jarring, at least to this ear. Consequently, to promote the Hollanders’ translation as “closest to the Italian” (Tim Parks, The New Yorker, cited on the dust jacket), when grande etymologically refers to greatness, calls into question the expertise in Italian of either the translators or, what is more likely, the dust-jacket reviewer. The fact of the matter is that, from this reviewer’s perspective, the Hollanders’ modifications of Sinclair’s prose repeatedly diverge further from the Italian rather than mimic it more closely.
In each of the above-cited tercets, the Hollanders’ transcription or copying of Sinclair admittedly divides the prose so that it appears as three lines of unrhymed versification. The initial “verse,” however, scans 9 syllables; the second, 8; and the third, 10. In the second example, the three “verses,” if we may so call them, number respectively 10, 11, and 12 syllables. The first canto’s shortest “verse” (Parad. 1.81), on the other hand, contains only seven syllables: “ever fed a lake so vast.” How such translations provide “maximum fidelity to Dante’s text with the feeling necessary to give the English reader a sense of the work’s poetic greatness in Italian” (Doubleday’s dust-jacket and catalogue publicity) remains a mystery. Can anyone at Random House actually read the fourteenth-century Italian original? If so, did they compare it to this modern translation?
For the record, Dante’s terza rima, as is universally known, is hendecasyllabic, with an interlocking rhyme scheme (ABA, BCB, CDC, DED, etc.): 11 syllables to the verse, 33 syllables, or the number of Christ’s years on earth, in each and every tercet; tonic accents fall on the 6th and 10th syllables; the 4th, 7th, and 10th syllables; or the 4th, 8th, and 10th syllables. Given the Hollanders’ lack of rhyme and any obvious metrical pattern and given the dearth of the translators’ explanation of what “verse translation” means, the sentient reader is left to puzzle over what the husband-and-wife team interpret “verse” to signify. Given the absence of any stated principles of prosody, perhaps they mean “free verse translation”? If so, why not state that openly? The answer probably lies in the fact that free verse – unrhymed and without any consistent metrical pattern – is about as far from Dante’s rhymed endecasillabi as one might imagine.
And so it goes with this translation: Sinclair’s “I was in the heaven that most receives His light” becomes, in the Hollanders’ version, “I was in that heavenwhich receives / more of His light” (1.4-5); “fault and shame of human wills” becomes “fault and shame of human wishes” (1.30); “with better words so that Cyrrha may respond” drops the that to read “with better words so Cyrrha may respond” (1.36); “the wax of the world more after its own fashion” becomes “the wax of the world more to its own fashion” (1.42); “on the eternal wheels” becomes “upon the eternal wheels” (1.64); “all natures have their bent according to their different lots” becomes “all natures have their bent / according to their different destinies” (1.109-110), whereas the Italian word being translated issorti, from the Latin sortes, referring to the ancient practice of casting lots.
The final line of Sinclair’s translation of Paradiso’s initial canto describes the conclusion of Beatrice to Dante-Pilgrim, upon whose face she had been gazing: “Then she turned her face again to the sky.” Hollanders’ version starts off identically, but completes it with a slight twist: “Then she turned her face up to the heavens” (1.142). The Italian original actually underscores a different point: she stops looking at Dante to look towards their destination. In other words, “Quinci rivolse inver lo cielo il viso” literally translates thus: “Hence [meaning, ‘from here,’ i.e., from facing Dante] she turned her face towards the sky” (inver <inverso, towards).
The exactness of so many phrases in both translations is troubling, of course, because some might think that such a practice smacks of plagiarism. (See Wilfried Decoo, “Substantial, Verbatim, Unattributed, Misleading: Applying Criteria to Assess Textual Plagiarism,” in Student Plagiarism in an Online World: Problems and Solutions, ed. Tim S. Roberts [New York: Information Science Reference, 2008], pp. 228-43.) They might cite several phrases in a cluster of tercets that remain identical to Sinclair’s, except for occasional minor changes in spelling, punctuation, or capitalization: “drawing near to its desire” (1.7); “that memory cannot follow” (1.9); “much of the holy kingdom” (1.10); “treasure in my mind / shall now” (1.11-12); “O good Apollo, for this [Sinclair: the] labor / make me” (1.13-14); “my breast and breathe” (1.19); “of his limbs” (1.21); “that I may show” (1.23); “shadow / of the blessed kingdom” (1.23-24); “leaves / of which my [the] theme and you [thou] will make my worthy” (1.26-27). Literal translations, however, often parallel one another to some degree, and I personally prefer not to jump to such a harsh conclusion.
The trouble, however, with the Doubleday publication is that this pattern of verbatim citation continues no matter where one searches: “I was changed within” remains the same, word for word (1.67), as do such phrases as “so much of the sky” (1.79), and “here the higher creatures see” (1.106), except in the last case the Hollanders capitalize Here, as it begins a sentence. The sentence Sinclair renders as “She . . . bent her eyes on me with the look a mother casts on her delirious child” changes only slightly to this: “she . . . / bent her eyes on me with just that look / a mother casts on her delirious child” (1.100-102). The thoughtful Dante scholar is reminded of Bonagiunta da Lucca’s comment inPurgatorio 24.58-59: “Io veggio ben come le vostre penne / di retro al dittator sen vanno strette” (Singleton’s translation: “Clearly I see how your pens follow close after him who dictates”). Sinclair would seem to “dictate” while the Hollanders’ pens follow close after.
In seeking an explanation for such close “following after,” this reviewer turned to the “Note on the Translation,” found on pages vii-ix. There we find no mention whatsoever of Sinclair, but instead we read the following cryptic sentence: “Since our goals in translating the third cantica of Dante’s poem are not in substance different from those that animated our translation of the first and second, the reader is asked to consult the similar notices that precede [those] translations” (vii). That recommendation sent us to the “Note on the Translation” that prefaced the Hollanders’ version of the Inferno (New York: Doubleday, 2000). There we learn that their “original intention was to reproduce the John D. Sinclair translation (1939) of Inferno, cleaning up its just barely post-Victorian ‘thee’s and ‘thou’s and other such, to a twenty-first-century ear, outdated usages”; however, they purport to have eschewed that approach, in part because of “a sense of ways in which a prose translation eventually fails to be ‘sayable’” (vii). Rather they claim to have been led “to attempt a new verse translation of the firstcantica, despite our original debt to Sinclair” (ibid.). They go on to state that “the reason our translation seems to reflect [Charles] Singleton’s, to the extent that it does, is that ours, on occasion, and Singleton’s, almost always, are both deeply indebted to Sinclair” (viii).
Based on the evidence adduced above, which could be multiplied a thousand-fold, I would argue that their “original debt to Sinclair” has remained, at least in the case of the Paradiso, an “on-going debt to Sinclair” and that their disingenuous phrase “on occasion” does not adequately clarify the substantial extent to which they copy Sinclair verbatim. Furthermore, their “sayable” verse often appears to be little more than slightly modified prose set up to appear as verse. (For an example of a poetic rendering of Paradiso, see Mark Musa’s translation of the third cantica and my review of it in Speculum 60.3 [July 1985]: 669-71, or see the Laurence Binyon translation, jointly lauded by Ezra Pound and Robert Fitzgerald.)
Given the Hollanders’ purported desire to lay all their cards on the table and create “an honest book” (their phrase, citing Montaigne), they ostensibly fall short even in the Paradiso’s “Note on the Translation.” For example, in speaking of Dante’s penchant for neologisms, they cite Paradiso 13.57, in which Dante coins the verb intrearsi. This verb, they proudly claim, “we have translated with an English neologism, ‘the Love that is intrined with them’” (their italics, viii). While it is true that their dittator Sinclair renders this verse somewhat clumsily as “the Love which with them makes the Three,” it was Singleton (not the Hollanders) who invented the English neologism intrined. In fact, practically identical to that of the Hollanders is Singleton’s translation of this verse, published in 1975: “the Love which is intrined with them.” While as a grammatical purist I can appreciate the Hollanders’ substitution of that for Singleton’s which, I feel strongly that, for the sake of academic honesty, they should not cite intrined as if it were their own “neologism” (literally, new word), as Singleton invented or employed it more than three decades before their translation appeared on the scene. To my way of thinking, that makes intrined no longer “new” in 2007. Likewise, in their note toParadiso 12.125, they state, “We have translated ‘la scrittura’ in the narrowest sense (‘the Rule’).” In reality, their translation of “vegnon . . . a la scrittura” (“come to the Rule”) parrots word for word Sinclair’s “come to the Rule.” In both cases, the more honest approach would be to state “we have adopted [or employed]” Singleton’s [or Sinclair’s] translation of the term in question.
In sharp and vivid contrast, Robert Hollander’s documentation of the sources for his scholarly notes is painstakingly precise. (Such precision, in fact, serves to highlight the dearth of clarity regarding how extensively Sinclair is copied.) Dante scholarship is now, and has been for decades (if not centuries), a highly complex affair, and Hollander’s notes prove this fact repeatedly. Over 70 commentaries are readily available in the Dartmouth Dante Project database (http://dante.dartmouth.edu), ranging from that of Jacopo Alighieri (1322) to Nicola Fosco (2003-2006). The notes to the Doubleday Paradiso, as is evident by myriad citations, draw heavily on that Dartmouth database and, as Hollander himself acknowledges, “are (at times) shorter versions of materials found in the Princeton Dante Project” (xiv). Nevertheless the resulting, generally daunting, avalanche of information will not fail to impress most students: in sheer number of pages the Hollander Paradiso(916 pages) almost rivals that of Singleton (390 pages of text and translation + 610 pages of commentary = 1000 pages). As always, Hollander’s forte lies in his ability to summarize and synthesize various critical viewpoints on any given verse or topic. His mastery of Dantean bibliography is likely without peer, at least on this side of the Atlantic. His “List of Works Cited in the Notes” encompasses just over 50 pages (see pp. 865-915).
Nevertheless almost any strength, when pushed to an extreme, becomes a weakness. And the well-read Dante scholar approaching such garrulous glosses cannot help but question wherein lies the originality of Hollander’s interpretation of the third cantica of the Commedia. It appears to be primarily in the highly selective melding of seven centuries of commentary tradition, with surprising emphasis on G. A. Scartazzini (2nd ed., 1900) as well as on Umberto Bosco and Giovanni Reggio (1979). Certainly Hollander’s approach appears a far cry from the groundbreaking, interpretive commentary of a creative genius such as Tom Phillips. Instead, from start to finish, we find a plethora of statements such as these: “as Scartazzini . . . insists” (p. 13), “Some commentators . . . put forward the notion” (p. 14), “several commentators tried another solution” (p. 21), “As Bosco/Reggio . . . point out” (p. 22), “This citation . . . has not escaped many commentators” (p. 25), “as Bosco/Reggio . . . explain” (p. 148), “As Bosco/Reggio remark” (p. 217), “Bosco/Reggio . . . report that” (p. 220), “Bosco/Reggio . . . are of the opinion” (p. 224), “Scartazzini . . . suggests” (p. 226), “Bosco/Reggio . . . point out” (p. 264), “Bosco/Reggio . . . say” (p. 295), “See Bosco/Reggio” (p. 296), “Bosco/Reggio point out” (p. 346), “See Bosco/Reggio . . . for a summary of the debate” (p. 349), “It is a commonplace in the commentaries to say” (p. 368), “See Bosco/Reggio . . . for the following explanation” (p. 317), “Bosco/Reggio point to” (p. 374), “Scartazzini . . . began the tradition of seeing both meanings” (p. 424), “Commentators agree” (p. 426), “Beginning with Scartazzini” (p. 430), “Scartazzini . . . disentangles the tortuous skein of debate over this line” (p. 446), “Beginning with Scartazzini” (p. 450), “Scartazzini may have been the first to suggest” (p. 470), “Bosco/Reggio . . . point out” (p. 471), “As Bosco/Reggio . . . point out” (p. 492), “Commentators, beginning with Scartazzini . . . suggest” (p. 499), “Bosco/Reggio . . . paraphrase the passage as follows” (p. 521), “As usual, Scartazzini . . . has a lengthy discussion of the variant readings” (pp. 521-22), “Bosco/Reggio . . . propose a reading that is mirrored in our translation” (p. 551), “See Scartazzini . . . for a full discussion of the dispute” (p. 597), “Bosco/Reggio . . . interpret” (p. 602), “Scartazzini, after an exhaustive survey . . . , concludes” (p. 650), “Bosco/Reggio . . . cite” (p. 666), “See Bosco/Reggio . . . who claim” (p. 692), “Bosco/Reggio . . . point out that the commentators are confused” (p. 694), “as Bosco/Reggio point out” (p 717), “Bosco/Reggio . . . allow” (p. 722), “Bosco/Reggio . . . point out” (p. 725), “see Scartazzini” (p. 742), “Scartazzini . . . says”(p. 745), “Bosco/Reggio . . . point out that it is difficult to be certain just what Dante means” (p. 755), “See Bosco/Reggio” (p. 829), “Bosco/Reggio . . rightly express surprise” (p. 831), “See Bosco/Reggio” (ibid.), “We follow Singleton’s interpretation” (p. 832), “see the similar position of Bosco/Reggio” (ibid.), “A good place to begin one’s study . . . is with the extensive gloss of Scartazzini” (p. 837), “Scartazzini . . . points out” (p. 844).
What is surprising, if not distressing, is that, despite the extensive citations of scores of commentators, Hollander often misses the forest (Dante’s basic allusion) for the trees (the myriad filters or accretions of commentators). I cite a pertinent example in Hollander’s verbose and redundant discussion (21 lines long) of aiuola(usually translated as “threshing floor” but as “little patch of earth” by the Hollanders) in Paradiso 22.151. This term is, according to Hollander and (as is his practice) to another commentator (John A. Scott, whom he dutifully cites), “without a doubt . . . without the biblical resonance of Matthew 3:12 and Luke 3:17” (p. 551, here and later). Having rejected the very real possibility (via citation of another authority) of a biblical allusion, Hollander postulates thatperhaps “the word reflects its presence in a phrase found in Boethius’s De consolatione, II.7.” In support he marshals no fewer than nine commentators stretching across the centuries (Scott, Kay, Pasquini, Pietro di Dante, Francesco da Buti, Landino, Daniello, Tommaseo, and Longfellow), ascribing to Scott the notion that Longfellow was “responsible . . . for the [presumably wrong-headed] invention of ‘threshing floor’ in his translation.” Hollander concludes by repeating, though somewhat tentatively, that “Dante probably had in mind Boethius’s description,” now cited, curiously enough, as “Cons. Phil. II.7” (emphasis mine). This last statement would seem to be all the reader truly needs to know of Hollander’s opinion (as filtered through the various commentators). The problem is that it totally ignores Singleton’s succinctly stated gloss on this verse, which traces the word’s origin directly to an obvious Virgilian source: “For the Latinarea (the word from which aiuola derives) in the sense of ‘threshing-floor,’ see Virgil, Georg. I, 178-81.”
At other times the problem with Hollander’s commentary relates not so much to that of “forest versus trees [plural]” but of his selecting a single tree to climb and then sit in with eyes closed to any others, no matter how tall. Again, one example may suffice: his citation of the hotly debated Epistle to Cangrande, “written by no less an expert than Dante himself” and that “cannot easily be denied its Dantean paternity” (p. 12). Certainly Hollander, whose long career as an allegorizing Dantist has been built on the belief that Dante composed that epistle, has the right to persist in his claim that Dante authored it, despite clausular and historical evidence to the contrary. He also has the right to cite only one side in the debate over the epistle’s authenticity – his own book, published in 1993, on the subject. But is it an example of “an honest book” to cite only one side of a debate? No mention of Peter Dronke’s critical analysis of the epistle appears, and Henry Ansgar Kelly’s devastating book-length critique of the subject fails to merit even a footnote.
Furthermore, despite the semblance of scholarly decorum that Hollander’s thousands of bibliographical references would suggest, the commentator’s colloquial speech, for which he is legendary, repeatedly shines through in a host of expressions. (For further commentary on Hollander’s use of inappropriately chatty language in academic venues, see my review of his Studies in Dante inForum Italicum 16.1-2 [Spring-Fall 1982]: 159-62.) I am thinking of his repetition of a supercilious commentator’s derogatory reference to Beatrice as “St. Thomas in drag” (p. xix) or his own claim that “Beatrice will later raise the ante” (p. 124), as if she were involved in a high stakes’ poker game. At times such language reminds us of the Princeton professor’s fascination with cinema and theatre. This is seen in references to Cunizza and Rahab as “two women in starring or major supporting roles” (p. 217); to Henry II of Lusignan, who “keeps the (metaphoric) company of the dirty dozen” (p. 479), or to Chaucer’s Troilus, which “is Paradiso XXXIII done as Some Like It Hot, one of Billy Wilder’s greatest films” (p. 830).
To conclude this analysis of the latest English version of Dante’s Paradiso, I quote the late Tibor Wlassics’ trenchant assessment of the Singleton translation: “The Bollingen [Singleton] Dante should be redefined as a revision of John Sinclair’s translation. Singleton’s text reads as a more or less lightly edited version of the Oxford UP  Dante. Apart from the you-ing of thou’s, etc., the ‘changes’ are curiously inconsistent and rarely improve on the Sinclair” (Lectura Dantis 4 [Spring 1989]: 111). Were the insightful Wlassics still alive, he likely would think the most honest description of the Doubleday version would read similarly. As for its scholarship, Wlassics would be surprised at the summary rejection of his own thoughtful conclusion that “the Cangrande letter . . . can cede its long-abused function as Dante’s own how-to book – to take on the role, of general usefulness, as one medieval manifesto expressing tenets probably close to (a minor) Dante’s way of thinking” (Lectura Dantis 6 [Spring 1990]: 159-60). No such luck, Tibor. No such luck.