I.

Much as they hate to admit it, translators are not writers. At their best, they are great readers, the ultimate appreciators. No one knows the curves and angles of a writer’s prose like the translator, who handles every comma and clause with almost indecent familiarity, while attempting an incredibly delicate and painstaking maneuver: the transfer of a million tiny balls from one hand to another without dropping a single one. Some novels take longer to translate than to write. The translator secretly suspects that this is always the case, in the same way that forging a signature takes so much longer to perfect than the signature itself. Raymond Queneau’s We Always Treat Women Too Well, written in a Frenchified faux-Gaelic, took translator Barbara Wright years to translate; Stendhal wrote The Charterhouse of Parma in fifty-two days and Richard Howard worked for twenty-eight weeks to translate it. Even the least complicated translation requires such an effort of sympathetic comprehension and sustained concentration that translators must hope their finished product is not only true to the original, but true in a distinctive way; that their own sensibility is obliquely evident. In the case of the retranslation of classics, the stakes are even higher. For once, the translator’s work can be compared to the work of other translators, and the translation is likely to become the principal object of critical scrutiny. Here, at least, it seems plain that translation plays an essential role: that it has the power to transform an iconic work of literature.

*

“In a village of La Mancha the name of which I have no desire to recall”; “In a village in La Mancha, the name of which I cannot quite recall”; “In a certain village in La Mancha, which I do not wish to name”; “Somewhere in La Mancha, in a place whose name I do not care to remember.” Which is it? The first line of Don Quixote may not be as reflexively quotable as “It was the best of times, it was the worst of times,” or “It is a truth universally acknowledged, that a single man in possession of a good fortune, must be in want of a wife,” but it would seem to have the ring of familiarity. And yet there is no definitive version, no inevitable progression. Is the familiarity in the precise words, or in the chain of associations? In this case, the answer is probably the latter. It doesn’t matter whether it’s “a village in La Mancha” (the popular choice) or “somewhere in La Mancha” (closer to the Spanish), or whether the narrator “cannot quite recall,” “[does] not care to remember,” or “[has] no desire to recall” the village or the place. The words, so long as they move smoothly, are just vehicles for meaning, the slight shifts in sense and emphasis nearly imperceptible.

The translations of the first sentence of Don Quixote cited above are by Samuel Putnam (Modern Library, 1949), John Rutherford (Penguin, 2000), J. M. Cohen (Penguin, 1950), and Edith Grossman (Ecco, 2003), respectively. Grossman’s new translation, published in the fall of last year, came just three years after Rutherford’s. In the last half century, Don Quixote has been translated six times; Walter Starkie’s 1957 Signet translation and Burton Raffel’s 1999 Norton translation complete the list. And these are hardly the only versions available to readers. In the past ten years alone, a bewildering array of editions of Miguel de Cervantes’s seemingly infinitely repackageable novel have appeared. Charles Jarvis’s 1742 translation was republished by Oxford University Press, with an introduction by Milan Kundera. P. A. Motteux’s 1705 translation (considered by most critics to be the work of a team of hacks) was reprinted by Everyman’s Library. Tobias Smollett’s 1755 translation was reissued by Farrar, Straus & Giroux with an introduction by Carlos Fuentes (Fuentes also wrote an introduction for a Barnes & Noble reprint of J. M. Cohen’s translation). Smollett’s translation was reissued again by London’s Folio Society with illustrations by Quentin Blake; Chapel Hill produced a facsimile edition of Thomas Shelton’s original 1612/1620 translation. And these are just the unabridged, unadulterated editions. It is also possible to find at least one cartoon version (Acclaim Classics Illustrated) and any number of children’s editions.

A new translator of Don Quixote knows what he or she is up against, but the general assumption by critics and publishers is that the greater the work of literature, the more translations it can support. Still, three new renditions in five years (Raffel [1999], Rutherford [2000], and Grossman [2003])? It is not enough to claim to be recapturing the classic for a new generation of readers. There must be something else—a fresh vision, deeper scholarship, a sharper eye, a more dexterous hand. Don Quixote belongs to a select body of works that have been translated so many times that the desire to produce the truest possible version of the original inevitably wrestles with the desire to stand out in a crowded field, to be faithful more skillfully than anyone else (faithfulness, of course, being an elusive quality in translation, with translators regularly obliged to strive for the spirit rather than the letter of a phrase or expression). Most of the others, perhaps unsurprisingly, are works in verse: The Odyssey, The Inferno, Goethe’s Faust. Don Quixote presents fewer opportunities for large-scale inventive reworking, but there is always scope for variation. Every translator comes up with hundreds, if not thousands, of brilliant small solutions that no previous translator managed to find; every translator also makes a few glaring errors of judgment or interpretation. But what do all these tiny differences add up to?

*

The history of the translation of Don Quixote mirrors the history of translation itself. In the seventeenth and eighteenth centuries, translators were more concerned with salable storytelling than literary fidelity; it was perfectly acceptable to pass off a playful paraphrase as an honest representation of the original. The early translations of Don Quixote were usually hasty attempts to get the book into the hands of an eager readership—scholars may argue whether Cervantes’s great work was the first novel, but it was undoubtedly one of the first international bestsellers. Thomas Shelton, who translated the two volumes of Don Quixote almost immediately after their Spanish publications in 1605 and 1615, boasted that he had finished the first volume in forty days; Tobias Smollett, whose translation appeared in 1755, is sometimes accused (like Motteux and others) of basing his text on the work of previous translators or hiring ghost-translators to do the real work. It wasn’t until the late nineteenth century that Quixote translators approached their labors in a scholarly and methodical fashion; John Ormsby is the great exemplar of that groundbreaking though often rather ornate approach. In the mid-twentieth century, scholarship was still respected, but lengthy footnotes tracing Cervantes’s references to obscure novels of chivalry were replaced by briefer notes clarifying aspects of language and life in seventeenth-century Spain, and translators allowed themselves to break up Cervantes’s sometimes tortuously long sentences (following the lead of Martín de Riquer, whose 1955 edition is still the standard Spanish text). This is the quintessential hybrid method, and perhaps still the best, with Samuel Putnam as one of its most successful exponents.

One of the reasons Don Quixote has been translated so often is that it is a relatively easy work to tackle. Forty days may still seem a stretch, but Edith Grossman managed to complete both volumes in two years. When translating Don Quixote, there is no need to devise a way to handle iambic pentameter or terza rima, and contemporary translators inevitably marvel at the clarity and pithiness of Cervantes’s language. It is notable that Spanish in the early seventeenth century was much more similar to contemporary Spanish than seventeenth-century English is to English today. (Interestingly, this means that Shelton’s 1612/1620 translation of Don Quixote is more antique-sounding than Cervantes’s 1605/1615 Spanish.) But the underlying reason Don Quixote lends itself so easily to translation is that its enduring influence has less to do with its language than with its iconic hero, its symbolic importance as the great proto-novel, and the readiness with which it lends itself to the purposes of its readers (as Harold Bloom puts it in his introduction to Grossman’s translation, it “will sustain any theory you bring to it, as well or as badly as any other”). One of Cervantes’s most devoted readers, the great turn-of-the-twentieth-century Spanish writer and philosopher Miguel de Unamuno (author of the grandly titled Our Lord Don Quixote), expressed it this way:

Eminently fit for translation, [Don Quixote] loses nothing of its force and poetry when put into any other language whatever… Even more, I go so far as to believe it gains in translation, and that if it has been understood better outside of Spain, it is in good part because a preoccupation with the language has not veiled its beauty in foreign lands.

Unamuno was obsessed with applying the tragic themes of Don Quixote—and particularly the delusions and suffering of Don Quixote himself—to Spanish history at a time when Spain’s fortunes were at a low ebb, and he believed his fellow countrymen were too caught up in convincing themselves of Cervantes’s genius as a stylist to see the true significance of the novel. Foreign readers, reading in translation, would be less apt to view the book as a sacred text, and would feel freer to apply its truths to their circumstances. In Unamuno’s view, Cervantes’s prose is transparent and easily translatable, less worthy of note than the novel’s themes and conceits. Unamuno may exaggerate a bit—his interpretation served his own political ends—but his essential observations are sound. The translation of Don Quixote is relatively simple, language-wise. What begins to become more and more important—especially as the novel accumulates layers of scholarship—is that the translator function not only as interpreter but as guide, that he demonstrate a deep understanding of the novel and the history of how it’s been read and interpreted over the years.

II.

Most casual readers are drawn to Don Quixote because it is reputed to be the classic adventure story—not only a great work of world literature, but the first buddy quest. Still, it is hard to find anyone who has read the novel all the way through for fun (though most are able to give a plausible synopsis, so indelibly are Don Quixote and Sancho Panza stamped on our collective consciousness). On the surface, the book is a succession of exciting encounters—Don Quixote against the wicked Yanguesans, Don Quixote and the galley slaves, Don Quixote on the magic flying horse, Don Quixote (most famously) tilting against windmills—but even by the standards of the times, these aren’t involving dramas in the ordinary sense, with twists and surprise endings. This is made evident, ironically, when they are set against the interpolated tales—long, unrelated stories tacked onto the novel’s first volume. Most critics (and even seventeenth-century readers) judged these tales harshly, as attempts to pad the novel. They are indeed standard-issue, cribbed from other sources, but hackneyed and gratuitous as they may be, they do possess that narrative narcotic, the pull of plot. Don Quixote’s adventures, by contrast, are at once outrageous and monotonous—teeth are lost and ribs bruised at a cartoonish pace, and defeat is blamed on “evil enchanters.” The monotony is refreshing at first—of course, this is what real adventures are like, at once trumped-up and prosaic—but the rote fantastic explanations are tedious, and as the adventures become more and more elaborate (in the second volume, a duke and duchess snare Don Quixote and Sancho in a series of extravagant practical jokes), the fascinations of the ordinary are lost and the whole adventure comes to seem a long-drawn-out contrivance.

In a controversial lecture series on Don Quixote given at Harvard in 1951-52 (published as Lectures on Don Quixote [1983]), Vladimir Nabokov took great pains to deflate Cervantes’s reputation. He makes his point of view plain from the beginning: “Don Quixote has been called the greatest novel ever written. This, of course, is nonsense.” He believes that Cervantes’s humor has not aged well (“The Don is certainly not funny”), and that the novel’s form is primitive. Any comparison of Cervantes to the other great writer of his age seems ludicrous to him: “The only matter in which Cervantes and Shakespeare are equals is the matter of influence, of spiritual irrigation.” And yet, he gradually reveals a grudging respect for Cervantes’s central creation, the unfunny Don, and winds up his lecture series with a graceful (if slightly ungracious) tribute: “We are confronted by an interesting phenomenon: a literary hero losing gradually contact with the book that bore him; leaving his fatherland, leaving his creator’s desk and roaming space after roaming Spain. In result, Don Quixote is greater today than he was in Cervantes’s womb.”

Some may find Nabokov’s judgment too harsh, but his lack of reverence is bracing, and allows for a sharper, more vital appreciation of the novel. Once the expectation that it must be enjoyed as a conventional adventure story is dispelled, it is easier to take pleasure in its subtler charms. The classic divide in Cervantes scholarship over the centuries has been between critics who read the novel as tragedy and those who read it as comedy. And yet the best way to approach it today may be as something else—something not quite a story at all, which must satisfy in a conventional narrative way, but as a nexus of self-reflexiveness, in which Don Quixote and Sancho’s bantering inaugurates the modern trope of playful self-awareness, and Cervantes stares out at the reader, his frankness and curious, impatient intelligence as riveting as the nonfictional testimonies of Herodotus or Montaigne. Cervantes’s game-playing begins in earnest at the start of the novel’s second volume, when Don Quixote and Sancho sit down with a neighbor, the Bachelor Sansón Carrasco, and discuss the reception of the first volume, supposedly authored by a Moor named Cide Hamete Benengeli, in one of the most striking and disconcertingly modern scenes in literature.

“And by any chance,” said Don Quixote, “does the author promise a second part?”

“Yes, he does,” responded Sansón, “but he says he hasn’t found it and doesn’t know who has it, and so we don’t know if it will be published or not; for this reason, and because some people say: ‘Second parts were never very good,’ and others say: ‘What’s been written about Don Quixote is enough,’ there is some doubt that there will be a second part; but certain people who are more jovial than saturnine say: ‘Let’s have more quixoticies: let Don Quixote go charging and Sancho Panza keep talking, and whatever else happens, that will make us happy.” [Grossman]

This back and forth—which delights today’s reader at least as much as Don Quixote’s charging and Sancho Panza’s talking—gradually takes over the second volume, and even becomes the driving force of the action when Don Quixote and Sancho meet the practical-joke-playing duke and duchess:

“Tell me, my dear squire: this master of yours, isn’t he the one who has a history published about him called The Ingenious Gentle Don Quixote of La Mancha, and isn’t the mistress of his heart a lady named Dulcinea of Toboso?”

“He’s the very one, Señora,” responded Sancho, “and that squire of his who is, or ought to be, in that history, the one named Sancho Panza, is me, unless I was changed for another in the cradle, I mean the printing press.” [Grossman]

After this introduction, the duke and duchess proceed to devise a whole series of adventures for poor Don Quixote and Sancho, but the metatextual contortions reach their peak when Don Quixote meets a man who claims to have known a different Don Quixote, the hero of a false second volume of Don Quixote (this volume actually existed; it was written by an unknown author looking to capitalize on the success of Cervantes’s first installment), and Don Quixote must fight to prove that he is the real Don Quixote. Forget tilting at windmills—this is where the novel’s true excitement lies.

III.

Despite the surface differences imposed by the literary conventions of different eras—the studied artfulness and floweriness of the late nineteenth and early twentieth centuries (Constance Garnett’s Dostoeyevsky, C. K. Moncrieff’s Proust), the easy smoothness at mid-twentieth century (David Magarshack’s Dostoeyevsky, Samuel Putnam’s Cervantes)—the central tenets of translation haven’t changed much. Translations are expected to be as faithful as possible, but also perfectly readable. Translators make ceaseless efforts to balance these two contradictory imperatives, but in the end, it is almost always readability that triumphs: unfamiliar idiomatic phrases are replaced with simpler constructions, cryptic references are spelled out, sentences are restructured. And yet, a growing number of translators are testing the bounds of readability—and, sometimes inadvertently, the meaning of faithfulness. They want their texts to read naturally, but they want to provide a jolt of strangeness, too, to provoke an awareness of distance and foreignness.

This ambitious project has its roots in the postmodern desire to expose the inner workings of a text, and is spelled out explicitly in the writings of translation theorists who take their cues from Walter Benjamin and George Steiner. As Lawrence Venuti, author of one of the most recent full-length studies of translation, puts it, the ideal is “a translation that is both readable and resistant to a reductive domestication.” Richard Pevear and Larissa Volokhonsky’s translations of Dostoeyevsky and Lydia Davis’s translation of Swann’s Way are among the most prominent examples of works undertaken in this spirit. The translators’ attempts to cleave as closely as possible to the original may mean retaining vestiges of foreign sentence structure (The Way by Swann’s, for example, was Davis’s preferred title—and the title used in the United Kingdom—for her translation of Proust), preferring literal translations of idiomatic phrases, or leaving foreign references unexplained. Note the differences between a brief passage from Pevear and Volokhonsky’s 2001 translation of Dostoeyevsky’s The Idiot and David Magarshack’s 1955 version:

About my mother there’s nothing to say, she’s an old woman, reads the Menaion, sits with the old crones, and whatever brother Senka decides, so it goes. [P&V]

*

Mother’s all right. She’s an old woman. Reads the lives of the saints, sits with her old women, and what my brother says goes. [Magarshack]

Both are essentially readable (though readers unfamiliar with the Menaion have to resort to the endnotes in the P&V version), but Magarshack’s version has a cleaned-up feel, with its short, neat sentences and omission of proper nouns (Menaion, Senka). In general, the antidomestication approach is riskier. When it is successful, it gives the work a texture it lacks in a more standard translation, but when it fails it introduces a clumsiness that wasn’t present in the original text and draws gratuitous attention to the translation. What it often sacrifices is consistency—the cohesion of voice that comes naturally in the original, but which must be surreptitiously constructed in the translation. It is this invisible work that allows readers to be lost in a book, to surrender themselves, to forget that they are reading a translation. The contemporary translator doesn’t quite want his audience to forget; he wants the reader to be carried along smoothly but not mindlessly, to provoke an awareness of currents beneath the surface of the English text.

This inclination is pervasive, and influences the work of the three most recent translators of Don Quixote, though they state their aims more simply. Edith Grossman, whose translation fits this contemporary mold best, explains that as the translator of mostly twentieth-century novels (by Gabriel García Márquez and Mario Vargas Llosa, among others), she approached Don Quixote with trepidation, afraid that she would have to adopt an “anachronistic, somehow-seventeenth-century voice” in order to translate it, but was ultimately reassured that Cervantes’s Spanish was modern, and that he could be translated “the way I translated everyone else.” What this means, in effect, is that she produces a relatively literal translation, which is what antidomestication usually boils down to in practical terms. It is this literalness—and particularly her inspired marriage of direct translations of epithets and proverbs (“the mother who bore you,” “the dogson”) with punchy contemporary syntax (“‘Read it aloud, your grace,’ said Sancho. ‘I really like things that have to do with love.’”), especially in the dialogues between Don Quixote and Sancho Panza—that gives her text its distinctive tone.

Lively as it is, Grossman’s new translation looks positively staid beside Burton Raffel’s 1999 translation, which rivals Motteux’s cavalier eighteenth-century translation in the liberties it takes. His position is that Cervantes’s “uniquely magnificent” language is impossible to reproduce in English, so he will play fast and easy with everything but syntax, which he believes is the hallmark of Cervantes’s prose. The result is the most eccentric modern translation of Don Quixote, crammed with jarring juxtapositions and awkward constructions (“as soon as I grabbed for my pigsticker they began to really let me have it so hot and heavy that my eyes couldn’t see”), but also remarkably clear where other translations are murky; Raffel is never satisfied with smooth-sounding compromise solutions. John Rutherford takes a wobbly tack somewhere in-between Grossman and Raffel, though he begins with the grandest declaration of purpose: “I… had to reject the notion, implicit in the Puritan tradition, that the translator must never impose anything of his own on his translation.” This, in his opinion, leads to “a surreptitious imposition of [the translator’s] own dullness.” Instead, he will try to “live in [Don Quixote’s] world… and only then search for the English words with which to describe what I found in my imagination.”

As many critics have pointed out, there are two levels of rhetoric in Don Quixote. One is the high-flown speech Don Quixote adopts, with which Cervantes mocks the period’s popular novels of chivalry. The other is the colloquial, conversational Spanish spoken by Sancho, and occasionally by Don Quixote. The former is relatively easy to translate, since it is rote courtly palaver in both languages, and the frills cover up any slips. Ironically, the simplicity of the conversational Spanish is much more difficult to capture, because its effect depends on small shadings of tone, and humor often obscure to twenty-first-century readers. Yet Sancho’s exchanges with Don Quixote are the bedrock of the novel, meaning that the success or failure of any translation is heavily dependent on the rhythm and ease the translator is able to give these exchanges. After one of Don Quixote and Sancho’s ill-fated early adventures in which Don Quixote attacks a flock of sheep, believing it to be an enemy army, knight and squire tally their losses.

“Did you say that the saddlebags are missing, Sancho?” said Don Quixote.

“Yes, they’re missing,” responded Sancho.

“Then we have nothing to eat today,” replied Don Quixote.

“That would be true,” responded Sancho, “if these fields didn’t have the wild plants your grace says you know about, the ones that unfortunate knights errant such as your grace use to make up for shortages like this one.”

“Despite that,” Don Quixote responded, “now I would rather have a ration of bread or a large loaf and a couple of sardine heads…” [Grossman]

*

“You’re missing your saddlebags, Sancho?” said Don Quijote.

“They’re missing, all right,” answered Sancho.

“Then today we’ll have nothing to eat,” responded Don Quijote.

“But that would only happen,” replied Sancho, “if we didn’t have, out in these fields, all those herbs your grace said you knew, which Destiny provides for the unfortunate knights errant like you.”

“Just the same,” replied Don Quijote, “I’d take more pleasure, right now, in a chunk of good white bread, or even a loaf of coarser stuff and a couple of sardine heads…” [Raffel]

*

“You have lost your saddle-bags, Sancho?” said Don Quixote.

“Yes I have,” replied Sancho.

“So we have nothing to eat today,” replied Don Quixote.

“We wouldn’t have,” replied Sancho, “if it wasn’t for those herbs you say you know all about growing in the fields, the ones that unfortunate knight errants like you go and pick to make up for lack of food in fixes like this.”

“For all that,” replied Don Quixote, “I would sooner have a two-pound loaf of white bread or indeed an eight-pound loaf of bran bread and a couple of dozen salted pilchards…” [Rutherford]

It is impossible to pick out a single passage that faithfully mirrors the strengths and weaknesses of each translation, but these three excerpts may be used as field samples. Reading them straight through quickly, one might say there is little difference between them (except perhaps for Rutherford’s startling “two-pound loaf… or indeed an eight-pound loaf… and a couple of dozen salted pilchards”). On closer examination, however, there are a few suggestive variations. Take Raffel’s “They’re missing, all right,” which has a jauntier and more irreverent tone than “Yes, they’re missing” or “Yes I have.” Throughout, Raffel’s tendency is to punch up the novel’s prose, often with contemporary turns of phrase. This is a tricky gambit, and it sits particularly awkwardly with his convoluted, comma-packed syntax (“so let your grace be told, also, since it’s not my place to give orders, that there’s no way I’m going to draw my sword”; “Your grace wasn’t so bad off, getting to hold in your hands that incomparable beauty you talked about”). When Raffel stumbles, he falls off a cliff. When Rutherford stumbles, he loses himself in a fog of insipid, meandering clauses, of which “go and pick to make up for lack of food in fixes like this” is a minor example; his Anglicisms are similarly limp (“‘Oh, dear, dear me!’ cried Sancho” (compare to Grossman’s “‘Woe is me!’”); “Sancho was frozen with fear at the sight of them, and even Don Quixote had the wind up”). And none of the three translators has much luck with Sancho’s jab at Don Quixote. Grossman’s literal “unfortunate knight errants such as your grace” and “shortages like this one” are clunky; Raffel brazenly invents a clause, “which Destiny provides.” In the final sentence of the passage, the translators must go further and venture an interpretation of an ambiguous phrase in Spanish. Here, Raffel shines. “A chunk of good white bread, or even a loaf of coarser stuff” makes more sense than Grossman’s “ration of bread or a large loaf” or Rutherford’s overly precise “a two-pound loaf of white bread or indeed an eight-pound loaf of bran bread.”

As a result of this analysis, the reader might conclude that Raffel’s translation is the best, or at least the most entertaining. There are certainly arguments to be made for it, but on the large scale it is so uneven as to be comical. It is not the individual errors of judgment that sink him, but his inability to corral his inventiveness. Perfect smoothness may no longer be the standard for translation, but there must be consistency within a broad register, prose of a calculated texture in which the inevitable infelicities don’t stand out like knots. This is why Grossman’s translation works, despite its clunky literalness. Grossman makes the bumps seem intentional, vital rather than awkward, and revealing instead of distracting. This requires more work than it might seem at first; she mostly reserves her literalness for nouns and Sancho’s endless proverbs, and streamlines Cervantes’s seventeenth-century syntax (compare this to Raffel’s decision to modernize vocabulary and preserve syntax).

And yet Grossman’s translation will probably supplement, rather than supplant, the best of the previous generation of translations (notably Samuel Putnam’s smooth, assured 1949 Modern Library edition). Her tendency to opt for an updated faithfulness leads her into too many murky spots, where she hasn’t properly untangled the Spanish. The first sentence of the prologue, much less frequently quoted than “Somewhere in La Mancha,” is a good example. The sentence isn’t particularly elegant in the original, but it reads clearly enough:

Desocupado lector: sin juramento me podrás creer que quisiera que este libro, como hijo del entendimiento, fuera el más hermoso, el más gallardo y más discreto que pudiera imaginarse.

Grossman’s rendition is not only clumsy, but also hard to understand.

Idle reader: Without my swearing to it, you can believe that I would like this book, the child of my understanding, to be the most beautiful, the most brilliant, and the most discreet that anyone could imagine.

Compare it to Putnam’s, which may not be perfect, but is entirely comprehensible.

Idling reader, you may believe me when I tell you that I should have liked this book, which is the child of my brain, to be the fairest, the sprightliest, and the cleverest that could be imagined…

The frequency of these lapses and the defensive tone of Grossman’s introduction (“It is important for you to know that ‘final’ versions are determined more by a publisher’s due date than by any sense on my part that the work is actually finished”) undermine her successes and erode the reader’s trust in the translation, though an excellent general introduction by Harold Bloom goes some way toward giving the volume a more considered heft.

Success or failure aside, in a sense the efforts of most Quixote translators are beside the point. Or rather, their translations are simply further appreciations of the novel, like so many critical essays, novels, and volumes of nonfiction. In the end, the mass of works inspired by Don Quixote—by Fielding, Dostoeyevsky, Flaubert, Melville, Borges, Kundera, Coover, and on and on—has come to nearly outweigh the original, until it is possible to argue, like Nabokov, that Cervantes’s Knight of the Sad Countenance has slipped his tether and cantered off into the sunset of literary legend. It is hard to pretend that a new translation will do more than polish or recast the story—interestingly, most reviewers of Grossman’s translation spent more time presenting their own appreciations of the novel than judging Grossman’s work—and readers would probably be better served if translators focused their efforts on the many novels languishing untranslated. But Don Quixote remains a proving ground, for readers and writers as well as for translators, and the desire to test oneself against it and make it one’s own is an impulse that continues to be richly rewarded.