When did my colds a forward spring remove?

When did the heats which my veins fil*And one more to the plaguie Bill?

The biographer argues that these lines were written in reaction to Donne's having been fired (and as a result thrust into poverty) by Sir Thomas Egerton, whose secretary he had been, for having secretly married Egerton's niece, and that what the lines mean is that Egerton should not have reacted so to the marriage because Donne's love for his wife had no ef signification. The deconstruction of Kafka has led to such apercus as that in "In the Penal Colony" "Kafka portrays the fall of logos into time with the gusto of a Harpo Marx." Allen Thiher, "Kafka's Legacy," 26 Modern Fiction Studies 543 (1980). Then there is Derrida's deconstruction of "Before the Law," in which he finds s.e.xual significance because "door" in French (not in German of course, the language in which Kafka wrote) is porte; the Latin phrase ante portas is a medical expression for premature e.j.a.c.u.l.a.t.i.o.n; and the man from the country never succeeds in entering the door to the law. See Derrida, "Devant la loi," note 16 above, at 143. There is even a deconstructive interpretation of the Odyssey. Frederick Ahl and Hanna M. Roisman, in their book The Odyssey Re-Formed 161166 (1996), argue that while Odysseus's false tales are indeed false, his narration to the Phaeacian court of his adventures after he left Troy is also false; no one knows where he was or what he was doing during those ten years, so he might have made it all up.

fects in the real world and so should not have bothered Egerton.61 The problem with this interpretation is that it diminishes the poem, which is less interesting if taken to be about Donne's private career disappointments than about lovers' renunciation of the world, daringly a.n.a.logized to the renunciation of the world by holy anchorites.62 Richard Baines, a contemporary of Christopher Marlowe, wrote shortly after Marlowe's death that Marlowe had boasted of being an atheist. Although Baines may have been lying, there is corroboration for what he reported.63 Yet it does not follow that Doctor Faustus is blasphemous rather than orthodox. Even if Marlowe was an atheist, he may have wanted to write an orthodox play. He may have thought that such a play would be better, more interesting, more dramatic, or more popular than an atheistic one-indeed, a play believed to be atheistic would have been suppressed. He may simply have wanted to stay out of trouble; atheists were still being burned at the stake in late sixteenth-century England. Knowing that Marlowe probably was an atheist may make us more alert for un-Christian undertones in Doctor Faustus-more sensitive to Faustus's oscillation between skepticism and faith-than we would be if we thought Marlowe had been the Archbishop of Canterbury. But we cannot determine the meaning of the play by reference to his beliefs.

Possible att.i.tudes toward fidelity to an author's conscious intentions thus range from a narrow intentionalism at one end, through formalism in the middle, to deconstruction (and postmodernism more generally) at the other end.64 Generally, but as we shall see not invariably, the interpretation of statutes and const.i.tutions should lie closer to the intentionalist John Stubbs, John Donne: The Reformed Soul: A Biography 171173 (2006).

See Brooks, The Well Wrought Urn, note 30 above, at 1117.

For judicious summaries and a.n.a.lysis of the evidence, see David Riggs, The World of Christopher Marlowe 327329, 336339 (2004); J. B. Steane, "Introduction," in Christopher Marlowe, The Complete Plays 9, 1215 (Steane ed. 1969).

The spectrum becomes a loop in the influential critical writings of William Empson, a brilliant, free-interpreting close reader (too close at times, according to John Crowe Ransom, The New Criticism 121130 [1941]) who often based his readings on highly speculative reconstructions of authorial intent.

*end than literature should. A poet tries to create a work of art, a thing of beauty and pleasure. He either succeeds or fails. If he succeeds we do not care how ba.n.a.l his intentions were, and if he fails we do not care how elevated they were. A legislature is trying to give commands, not only to the persons whose conduct the legislation is intended to regulate but also to the judges who will be applying the legislation in specific cases. A command is a communication, to be decoded in accordance with the sender's intentions. If a message is garbled in transmission you ask the sender to repeat it; that is intentionalism in practice. If you cannot reach the sender, you try to glean from everything you know about him and the circ.u.mstances of the failed message what he would have done had he been on the spot.65 Again the correct a.n.a.lysis is an intentionalist one. One thing that gives intentionalism its purchase in literary criticism, besides the misguided search for authoritative literary interpretations, is the Romantic fascination with the personality of the artist-with the work of art as emanation rather than artifact. That is why there is resistance to the idea that Shakespeare's plays were written by "the Stratford man" rather than by someone like Francis Bacon or the Earl of Oxford who had a more impressive resume. That a statute is an emanation is a condition of its authority.

Notice how in describing the legislative text as command or communication I am doing what the deconstructionists denounce: privileging the spoken over the written word. That is, I am thinking of the written word on the model of speech, a less problematic form of communication. At the same time I am doing what the New Critics denounced: refusing to treat the legal text as a fascinating, multifaceted artifact and instead treating it as an attempt to set up a path of clear communication between author and reader. Some statutes are open-ended but most are not, whereas open-endedness is characteristic of great literature. Remember that works of literature are called great because they transcend boundaries of period and culture. Their property of meaning different things to different peo 65. This is not a novel approach to interpretation. It was Aristotle's theory of statutory interpretation (see Rhetoric, bk. 1, ch. 13), and John Adams articulated it with great clarity shortly before the Const.i.tution was promulgated. See Leonard D. White, The Federalists: A Study in Administrative History 130 (1948) (quoting John Adams, Works, vol. 8, pp. 1112 [1853]).

ple is not problematic because it is not the function of literature to lay down rules of conduct.

A related point is that much great literature-this is a central insight of the New Critics-achieves an equipoise, rather than a resolution, of opposing forces. Recall Marvell's exquisite ambivalence about Cromwell and Yeats's about the Easter Rebellion. To come down on one side or the other-for Cromwell or against him-is not a response that the poem invites. Statutes may be ambivalent too, through failure of foresight or pressure to compromise. But when a statute is drawn into litigation, the court must adopt a reading that will favor one side of the lawsuit or the other. It cannot revel in ambivalence.

Meaning and message are more likely to diverge in literature than in law. One can extract a clear and definite message from either type of writing by paraphrase. The message of Doctor Faustus is that if you sign a pact with the devil you will be sorry in the end. At the level of message most works of literature are clear, but also ba.n.a.l; what makes them unclear is that we are not interested in staying at that level-we seek a deeper meaning. But the message level is the only interesting level of a statutory or const.i.tutional text, unless the text is to be merely a launching pad for judicial flights of fancy. That is why the Peller-Tushnet interpretation of the age-35 provision in the Const.i.tution seems obtuse or provocative rather than ingenious.

Many contemporary critics rebel against being chained to the text. They are "in it for what they can get out of it, not for the satisfaction of getting something right."66 The public would be horrified by such a conception of the judicial role. The preoccupation of literary intentionalists with the idea of authoritative interpretation (for them, the interpretation that is faithful to the author's intent) is displaced from its proper object. The problem of legitimacy need not arise in literary criticism, but is central to law and government. The literary intentionalist is a lawyer manque; the legal New Critic is a literary critic manque. The former demands a type of constraint on interpretation that law rather than literature re 66. Richard Rorty, Consequences of Pragmatism (Essays: 19721980) 152 (1982). And recall the quotation from Hartman in note 19 above.

*quires;67 the latter seeks a freedom of interpretation that literature but not law allows.

For a sophisticated intentionalist like Juhl-who concedes the inexhaustibility of literary interpretation, the possibility of unconscious intention, and the dangers in relying too heavily on historical and biographical materials as clues to an author's intent68-almost the whole significance of intentionalism is to rule out interpretations that could not possibly be referred to the author's conscious or unconscious mind. Mainly these are interpretations that either imply prophetic gifts (for example, interpreting The Trial as an allegory of the police state, the "Fourth Eclogue" as an allegory of the birth of Christ, or "The Second Coming" as an allegory of the Munich Pact) or contradict unequivocal biographical or historical data. Juhl ill.u.s.trates the first point with a pa.s.sage from "The Love Song of J. Alfred Prufrock": "Would it have been worth while . . ./To have squeezed the universe into a ball." Suppose, says Juhl, that we knew Eliot had never read Marvell's "To His Coy Mistress," which contains the lines, "Let us roll all our strength and all / Our sweetness up into one ball." Then, Juhl argues, we could not regard the allusion to Marvell as part of the meaning of Eliot's poem.69 Why not? Why not say that Eliot had made a lucky hit-had, by his choice of words, accidentally enriched the meaning of his poem for readers who remembered Marvell's poem and would contrast the vigor of the lover's solicitations in that poem with Prufrock's hesitations?

Hirsch points out that there is an "advocacy system in [literary] interpretation as in law. The advocates have the task of bringing forward evidence favorable to their side and unfavorable to their opponents . . . But without a judge all those relevant pieces of evidence float uselessly . . . [Therefore] unless advocates sometimes serve as judges, none of this activity will actually contribute to knowledge." Hirsch, note 28 above, at 197. But there is no mechanism for appointing authoritative judges of disputes over literary interpretation; nor would such a mechanism be welcome. And self-appointed judges with no authority are not a close a.n.a.logue to official judges. Only occasionally does the literary community anoint a critic to be an authoritative judge of literary quality, as T. S. Eliot was anointed during his lifetime.

See Juhl, note 28 above, at 135, 151, 225230. See also Hirsch, note 28 above, at 22 ("it is very possible to mean what one is not conscious of meaning") and his discussion, in "Counterfactuals in Interpretation," note 54 above, of the importance of distinguishing dominant from local intentions and the spirit from the letter of a work.

69. Juhl, note 28 above, at 5859.

"The aim of literary interpretation is to reveal those features which make the work a good literary work . . . Coherence and complexity are criteria of an author's intention which override whatever the author himself may have to say about the emergent aesthetic features of a work after he has delivered the text."70 The "author's intention" does no work in this pa.s.sage, but the pa.s.sage does identify the key methodological innovation of the New Criticism, which was to adopt a hypothesis of total coherence-to a.s.sume that no detail of a work of literature is an accident, that everything contributes in some way to its meaning and emotional impact.71 Through studying the details of a work and their interrelations, which the a.s.sumption of total coherence impels one to do, one begins to understand how great literature casts its spell. In "The Wild Swans at Coole" we are made to feel the mystery and remoteness of nature by a surprising yet somehow "right" juxtaposition of the words "cold" and "companionable" in a description of swans on a lake: Unwearied still, lover by lover, They paddle in the cold Companionable streams or climb the air; Their hearts have not grown old; Pa.s.sion or conquest, wander where they will, Attend upon them still.

The same interpretive technique-that of attributing significance to every detail-is, when used on statutes, a familiar source of error. Statutes and const.i.tutions are written in haste by busy people who often are neither able nor conscientious, and we are not privileged to ignore the hasty or hackneyed provisions and reserve our attention for the greatest. Legislative texts are not the products of a single mind but of a committee-the legislature, whose numerous members may have divergent objectives- and so they may contain meaningless repet.i.tions and inconsistencies. To 70. Olsen, note 59 above, at 51.

71. Robert Penn Warren, "A Conversation with Cleanth Brooks," in The Possibilities of Order: Cleanth Brooks and His Work 1, 15 (Lewis P. Simpson ed. 1976), quoting Brooks. The presumption is reb.u.t.table: there are radically imperfect works of literature, as argued in Hershel Parker, Flawed Texts and Verbal Icons: Literary Authority in American Fiction (1984).

*suppose that every word in a statute must be significant-that every statute is a seamless whole-misconceives the nature of the legislative process and can lead to nonsensical interpretations.

A similar mistake would be to suppose that the interpretive task for a court is to make the statute or const.i.tutional provision before it the best rule of law the court can imagine because legislators can be a.s.sumed to strive to make the best rule of law they can, as Yeats strove to make "The Wild Swans at Coole" the best poem he could. That is to be unrealistic about legislative intention, and the practical consequence is to subst.i.tute the judge's will for that of the legislature-which is more serious than the subst.i.tution of a critic's will for Wordsworth's or Yeats's. If a statute's intended meaning remains inscrutable after the court has made its best effort to determine that meaning, the court will have no choice but to impress a meaning upon the legislation, and then Dworkin's approach may be best because in that case the judges really are the legislators. That is a common enough case,72 but it is still exceptional.

There is a feedback loop between interpretation and legislation. If interpretation is too erratic, the making of new legislation is discouraged because legislators cannot predict the effects of the legislation they pa.s.s. There is no similar loop between critics and authors. Legislatures, moreover, usually produce a publicly available "legislative history" to help guide judicial interpretation. Writers of literature sometimes offer interpretations of their own works, but for reasons discussed earlier these interpretations cannot be a.s.sumed to be reliable. Sometimes writers preserve their unpublished drafts, which are comparable to the unenacted bills that often precede an enacted statute. But though invaluable in understanding the process of creative composition, writers' rough drafts rarely dispel interpretive fogs; that is the lesson of Stallworthy's study of the Yeats ma.n.u.scripts. It is harder to extract the writer's intentions than those of the legislature. Often it is attempted by facile psychoa.n.a.lyzing on incomplete data-the fate that has befallen Kafka, among many others.

And the legislature has at least decided to enact the law being interpreted. Many literary works (including The Trial and Billy Budd) were left unfinished at their author's death, making it uncertain whether what 72. See Posner, note 2 above, ch. 3 ("The Judge as Occasional Legislator").

we have was intended to be read. Kafka left instructions to destroy all his unpublished works, though how seriously these instructions were intended to be taken is unclear.

Related to the fact that uniformity is more important in legal than in literary interpretation is the fact that multiple interpretations of the same work may be an equilibrium for the literary marketplace but a disequilibrium in law. A reader's interpretation of a literary text is affected by what he knows; and different readers know different things. There is no way in which a person who has spent a lifetime studying Shakespeare can convey his entire understanding of the Shakespearean context to a nonspecialist; Shakespeare's plays will always mean different things to specialists and nonspecialists. Eliot's fourth quartet, "Little Gidding," will yield additional meaning to the reader who recognizes that "The dove descending breaks the air / With flame of incandescent terror" refers to the bombing of England in World War II, a reference not apparent from the poem itself but a.s.sumed by every student of Eliot's poetry. "The Wild Swans at Coole" will yield a richer meaning to a person who has read a lot of Yeats's poetry than to one who has read little or none of it; the former know that in Yeats's poetry "swan" has not just its ordinary meaning but also an esoteric meaning as a symbol of pride, courage, and power73 and that "cold" has a strongly positive connotation as in "cold/ Companionable streams" or in his epitaph, "Cast a cold eye / On life, on death." The Trial takes on more hues of meaning when read in conjunction with Kafka's legal parables, such as "The Problem of Our Laws," than when read in isolation. But it is incorrect that you cannot understand a Yeats poem or a Kafka novel without having read extensively in, and having reflected profoundly upon, other works by these authors. The specialist is not the only reader who possesses the authentic meaning of a work of literature. Authors do not write for a tiny coterie of professors.

The question of the breadth of the reader's freedom to interpret-or, equivalently, the limits of Cleanth Brooks's dictum that "the poet sometimes writes better than he knows"-is acutely posed by Paradise Lost. We do not need evidence, extrinsic to the text, of Milton's religious be 73. Esoteric, but not unique: compare the lines on the dying swan in Yeats's poem "The Tower" with Socrates' explanation for the "swan song," in Phaedo 84e85b.

*liefs to recognize that Satan is not the hero of Paradise Lost (although he is a more sympathetic figure than G.o.d, in part because he uses persuasive speech-with his fellow devils and with Eve-whereas G.o.d issues commands rather than trying to persuade anyone) and that Adam and Eve were justly punished for committing treason against G.o.d; the poem is strongly didactic.74 But Milton may have written better than he knew because what he knew was that Adam and Eve's disobedience d.a.m.ns them but what he wrote was that it saves them. Before Eve wanders off for a bit to be by herself and falls into Satan's clutches, she and Adam are condemned for eternity to be naked vegetarians whose only occupation is pruning trees and bushes. Life in the Garden of Eden as depicted by Milton is a bore until Eve decides that she'd like to spend some time alone and falls in with Satan and sees an opportunity to get smarter by eating the forbidden fruit and until Adam has to choose between Eve and Eden and chooses her in the most moving pa.s.sage in the poem: ". . . with thee / Certain my resolution is to die; / How can I live without thee, how forgo/Thy sweet converse and love so dearly joined,/To live again in these wild woods forlorn?" (IX.908909). The reference to "sweet converse" is noteworthy: the relation between Adam and Eve, formally patriarchal, is actually companionate.75 Human beings did not evolve to live in the Edenic state; in its monotony and lack of challenge, such a life is as unthinkable and ultimately unbearable for Adam and Eve as Ogygia was for Odysseus. The poem ends with Adam and Eve's escape: "The world was all before them .../They hand in hand with wandering steps and slow,/Through Eden took their solitary way" (XII.646, 648649).76 The human adventure begins.

Readers who accept the scientific rather than the religious worldview might thus read Milton as having all unconsciously (his genes, which told him that human beings are not adapted to the environment depicted in See, for example, Margaret Olofson Thickstun, Milton's Paradise Lost: Moral Education (2007).

See Barbara Kiefer Lewalski, "Paradise Lost and Milton's Politics," 38 Milton Studies 141, 160163 (2000).

"[H]and in hand" is another clue to the companionate nature of the relationship between Adam and Eve. For adults to walk hand in hand, and thus side by side, implies an egalitarian relationship.

the Garden of Eden, warring with his religious beliefs) made Eve the hero of the poem. (Eve would then be the female counterpart of Prometheus, who in Greek mythology saves man by disobeying Zeus and is grievously punished.) That would be a heterodox, to some a blasphemous, interpretation, but it would be defensible despite the absence of the kind of extrinsic evidence of an author's skeptical beliefs that we have for Christopher Marlowe-more defensible than the interpretations that Robin West and Richard Weisberg have impressed on the works of literature that they discuss. Brooks's dictum opens wide the gates to legitimate loose interpretation of literary works. The judge's reasons to feel constrained in his interpretation of legal texts do not bind the reader of a literary text.

But this is to oversimplify the contrast between the two interpreters. The judge has some reasons for loose interpretation that the literary critic does not have. A legislature is not a single mind, and the determination of collective intent is often problematic and sometimes impossible. Moreover, much legislation reflects compromise rather than consensus, and one way to achieve compromise is to use general language, in effect shifting to the courts the task of completing the legislation. Neither intentionalism nor New Criticism provides helpful a.n.a.logies here.

Granted, there is committee authorship in literature too.77 For example, Hamlet as we experience it is a collaboration of Shakespeare, the authors of the sources on which he drew, the early printers who garbled his texts, the actors whose ad-libs may have found their way into the texts,78 modern editors, and the producers, directors, and actors who put on performances of the play. Committee authorship is in fact the norm rather than the exception in drama, opera, and film, and this is worth emphasizing because advocates of strict interpretation of statutes sometimes deny the existence of collective intent-deny that a work with more than one author can have an intent that might be used as a guide to interpretation. There is no such thing as a collective mind-that is true. All that collective intent can signify is agreement. But that is something; the legislators who vote to pa.s.s a bill may agree on what they expect it to accomplish, 77. See note 8 above.

78. Masten, note 8 above, at 371, notes Hamlet's request to the players to be allowed to interpolate some lines, apparently a common Elizabethan practice.

*and their expectation if known by a court asked to interpret the law may be a valuable aid to interpretation.

The principle of stare decisis (decision in accordance with precedent) deflects statutory and const.i.tutional interpretation from framers' intentions; there is no counterpart in literature. When a particular statutory or const.i.tutional interpretation has become entrenched in a long line of decisions, a court will be loath to abandon it, whether because of the reliance that has been engendered, because the court wants to maintain the appearance that law is objective and impersonal, or because the court does not want to encourage legislators to be hasty and careless by leading them to think that judges will make any needed legislative revisions simply by reinterpreting legislation to keep it abreast of the times. The result of a long period of judicial interpretation, such as that undergone by the Const.i.tution of 1787 or the Sherman Ant.i.trust Act of 1890, may be a body of doctrine that bears less relation to the intentions of the framers than a modern interpretation of an old work of literature will bear to its author's intentions.

While the only (though not a bad) reason for an interpretation of a work of literature that makes it mean something different from what it meant to its original audience is to get greater pleasure and insight from the work, a free interpretation of a statute or the Const.i.tution may be necessary to avert serious harms to society. This is just to repeat that more is at stake in judicial than in literary interpretation. The greater stakes may point in either direction-toward more literal interpretation or toward freer interpretation. Radical literary critics would like their literary interpretations to have consequences in the political world,79 but "as things stand now in our society, interpretations of literary works, no matter what their emphasis and independently of the motives of those who produce them, do not connect up strongly with the issues being debated in the larger political arena."80 79. As one of them writes, "Literary theory has come to be identified with the political left . . . In one sense, theory has been the continuation of radical politics by other means." Terry Eagleton, "Discourse and Discos: Theory in the s.p.a.ce between Culture and Capitalism," Times Literary Supplement, July 15, 1994, p. 3. But Eagleton sensibly doubts that literary theory has any political consequences.

80. Stanley Fish, Professional Correctness: Literary Studies and Political Change 51 (1995).

With power to deprive citizens of their property, liberty, or even lives comes responsibility for the wise exercise of power as well as a duty to obey the legal limitations on its exercise.81 The circ.u.mstances of an individual case may rightly lead the judge to decide in a way that cannot easily be squared with the text of the statutory or const.i.tutional provision being interpreted. This is an example of the force of discretionary, equitable, or political considerations in judicial decision-making. Such considerations are elements of the law, as we saw in chapter 3, just as rules handed down by higher authority are.

Ignoring historical context can sometimes do more damage to the understanding of a "timeless" work of literature than of a statute or a const.i.tutional provision. The cultural distance between such a work and its readers is apt to be greater than that between a statute and the judge called upon to interpret it. The legal texts that are authoritative in the American legal system were written by American lawyers no earlier than 1787 and usually later. Works of literature are often much older, often foreign, often written by inhabitants of a radically different cultural milieu from that of the modern American reader. To read literature composed in a different culture without being aware of the difference is reckless. Imagine trying to read Shakespeare without knowing how such words as "brave" and "fat" have changed meaning. "Brave" in Miranda's exclamation "Oh, brave new world / That has such people in't!" (V.1.185) in The Tempest does not mean courageous; it means making a fine appearance. Gertrude's remark during Hamlet's duel with Laertes that Hamlet "is fat and scant of breath" does not mean that Hamlet is overweight but either that he is sweaty or, less probably, since he has been practicing fencing steadily, that he is out of condition. A reader of Shakespeare will also get into trouble if he doesn't know that the English thought Danes were given to excessive drinking; that is the "custom / More honored in the breach than the observance" (I.iv.1516)-that is, more honorably rejected than followed- to which Hamlet refers. Imagine not realizing that a woman's marrying her brother-in-law was deemed incestuous, or trying to read The Merchant of Venice thinking that Jews had the same position in Elizabethan society as they do in ours, or reading in Yeats's poem "Lapis Lazuli"

81. Cf. Robert M. Cover, "Violence and the Word," 95 Yale Law Journal 1601 (1986).

*"They know that Hamlet and Lear are gay" and thinking the poem is about h.o.m.os.e.xuality.

Readings based on ignorance or error may occasionally be better, in the sense of imparting a greater resonance to a work of literature, than readings scrupulously confined by historical knowledge. I gave a few examples earlier; there are others. Several mistranslations in the King James version of the Bible seem superior on literary grounds to the originals, though maybe it is only habit and tradition that make us think so. Mark Twain's careless revising of Pudd'nhead Wilson may be an example of serendipitous error.82 But serendipity is rare. The introduction of a random element is unlikely to improve a work of literature; and except in the rare instance where the reader is as talented a creative writer as the author, he is unlikely to improve the work by using it as a mirror for his own insights.

Still, all things considered, usually a judge should pay more attention to legislators' conscious intentions, insofar as they can be known, than a literary critic should pay to the author's conscious intentions. There is thus no inconsistency in being an intentionalist judge but a formalist literary critic-which ill.u.s.trates my remark at the beginning of this chapter about the field dependence of interpretation.

Nor is there any contradiction between being an intentionalist judge with regard to statutes and const.i.tutional provisions and a formalist judge when dealing with contracts, reluctant to allow extrinsic evidence, including testimony about the contracting parties' conscious intentions, to change the meaning suggested by the contractual text. The literary critics Walter Benn Michaels and Stanley Fish argue that because interpretation is inherently contextual, intrinsic (or what I am calling formalist) legal interpretation is a fantasy.83 Fish notes, for example, that a court will allow Herschel Parker, note 71 above, at 5, disagrees. He asks: how can we take seriously as meditations on race, or genetics, or slavery the chapters that Mark Twain wrote before he decided to make the false "Tom" a black and that he included without change after making him black? See id. at 135145. The answer is that context shapes meaning. Mark Twain may have retained these chapters because, relocated in a work about blacks and whites, they acquired new meaning.

Michaels, note 56 above, at 2629; Fish, There's No Such Thing as Free Speech, and It's a Good Thing, Too 141156 (1994).

evidence that the trade to which the contract pertains (for example, maritime shipping or cotton factoring) attaches a special meaning to words used in the contract, a meaning that would not be apparent to an ordinary reader. So, says Fish, the parties are allowed to contradict the written contract after all.84 A vital distinction is overlooked. Trade usage can be established by disinterested testimony to a reasonable degree of certainty; consulting trade usage is like consulting a dictionary. The concern behind the rule limiting extrinsic evidence in contractual interpretation is that written contracts would mean little if a party could try to persuade a jury that while the contract said X, the parties had actually agreed, without telling anybody or writing anything down, that the deal was Y. That concern is not engaged by objective evidence, such as trade usage, and therefore such evidence is not barred.85 The principles of contractual interpretation depend on the purposes of contract law rather than on any general theory of interpretation.

Chain Novels and Black Ink Scholars who believe that legal texts can be a.n.a.logized to literary texts rarely specify which literary genre provides the best a.n.a.logy to law. Ronald Dworkin is an exception; he has specified the genre-the chain novel.86 One author writes chapter 1. This sets a certain direction because the next author must write chapter 2 in such a way that it seems to grow out of chapter 1, so that the two chapters will seem like work of the same hand; likewise with the subsequent chapters. Each author thus has less freedom than the one before. Dworkin suggests that the judge who must first interpret a const.i.tutional text is like the author of chapter 2, while a judge asked to interpret a const.i.tutional text on which additional meaning has been grafted by previous judicial interpretations is like the author of one of the subsequent chapters. Not so. First of all, the chain novel as Dworkin describes it places no constraint on authors of subsequent chapters.

84. Fish, note 83 above, at 148.

85. See, for example, A.M. International, Inc. v. Graphic Management a.s.sociates, Inc., 44 F.3d 572, 575 (7th Cir. 1995).

86. Dworkin, Law's Empire, note 29 above, at 228250, 313.

*Each author can in the first sentence of his chapter kill off all the existing characters and start anew. Of course this would not be thought cricket, but that just means that the writing of a chain novel is a more complex practice than Dworkin's description of it. It is thus unclear what in the legal interpretive process he is a.n.a.logizing it to.

The a.n.a.logy is also defective because it puts the judges who interpret the Const.i.tution on the same level as the framers of the Const.i.tution: the framers just got the ball rolling. Even if the author of the first chapter of a chain novel could, despite my first point, exclude some possible sequels, all the chapters would be equally authoritative. But decisions interpreting an authoritative legal text, such as the Const.i.tution or a statute, occupy a lower rung than the text.

Even as a description of common law rather than of statutory or const.i.tutional law, the chain-novel a.n.a.logy is misleading. First, chapter 1 in the evolution of common law doctrine is likely to be highly tentative-more like a preface or introduction. Second, the "authors" of the subsequent chapters are not bound to adhere to the directions set by the author of chapter 1. If acc.u.mulating experience shows that chapter 1 took a wrong turn, the judges can discard it. Third, the common law is merely the set of legal concepts created by judicial decisions, and as with any concept the precise articulation is mutable, can be refined, reformulated. The concept is inferred from the decision (more often from a sequence of decisions) but exists apart from it. The literary critic, the biblical exegete, and the judge engaged in statutory and const.i.tutional interpretation all have the difficult task of interpreting a fixed text.

A final distinction between judicial and literary interpretation should be noted. A judiciary is a hierarchy. The judges at the top-the Justices of the U.S. Supreme Court and, in matters of state law, of the state supreme courts-have a great deal of interpretive freedom; the lower-court judges, where most legal interpretation is done, have much less. The situation in literature is different. There have been times when great writers were the foremost literary critics-think of Dryden, Pope, Samuel Johnson, Coleridge, T. S. Eliot. They might be a.n.a.logized to Supreme Court Justices. But nowadays literary criticism is dominated by a professional community, largely academic, of nonwriters-full-time specialists in criticism. They use their specialized knowledge and their influence in hiring, promotion, and publication to determine both what writings shall be cla.s.sified as literature and the outer bounds of permissible interpretation of those works. The parallel is a church's effort to fix the canon and meaning of sacred works.87 But a hierarchical church, like the hierarchical judiciary, has better prospects for circ.u.mscribing interpretive freedom than a nonhierarchical professional community has. It is no surprise that the promiscuous variety of literary interpretations exceeds anything one could readily find in a body of judicial opinions. It is also not a political problem, as interpretive anarchy in law would be.

The skeptical vein in literary criticism, and the interpretive theories that nourish it, show how difficult the interpretation of texts can be and by doing so should make lawyers, judges, and legal scholars more cautious, self-conscious, and tentative about the process of interpreting legal texts. But no specific techniques or discoveries of literary criticism, and no literary a.n.a.logies, such as that of the chain novel, can ease the legal interpreter's task. A good literary critic is a careful, thorough, scrupulous, informed, logical, and practical reader of literary texts, and a good lawyer is a careful, thorough, scrupulous, informed, logical, and practical reader of legal texts. They are both close readers, but of different materials. Their strength as close readers comes from immersion in a particular body of texts rather than from mastery of a theory of interpretation. So while Sanford Levinson may be right that "there are as many plausible readings of the United States Const.i.tution as there are versions of Hamlet, even though each interpreter, like each director, might genuinely believe that he or she has stumbled onto the one best answer to the conundrums of the texts,"88 he is implying a commonality of interpretive problems that does not exist.

Several facts about Hamlet open it to different interpretations: it was written more than 400 years ago (and is thus almost 200 years older than our Bill of Rights) and in another country; the text is corrupt; there is no evidence of the author's intentions beyond what is in the text; and it was written to be performed rather than read. In addition, plays are inherently open-ended because actors' inflection, timing, and body language, and Kermode, note 44 above, ch. 8.

Levinson, "Law as Literature," 60 Texas Law Review 373, 391 (1982).

*details of the set, are usually not prescribed by the text of the play and can alter the meaning of the lines.89 The test of time implies that great literature will usually be open to divergent interpretations; part of the fascination of Hamlet is the number of interpretive puzzles it poses. And we do not care much about what Shakespeare thought he was trying to accomplish-partly because we do not know, partly because we doubt that he fully knew. The Const.i.tution is open to different interpretations because of its multiple authorship; the apparent decision of the framers to leave certain issues open through the use of general language; the social, economic, legal, political, and inst.i.tutional changes that have occurred since the Const.i.tution was drafted; and the lack of agreement on how free a judge should feel in interpreting a const.i.tutional provision and what weight he should give to previous interpretations of it. The puzzles about the Const.i.tution are so different from those about Hamlet that it is unlikely that a Hamlet scholar would have anything useful to say about interpreting the Const.i.tution or a const.i.tutional scholar anything useful to say about interpreting Hamlet.

I thus take exception to Charles Fried's effort to infer the intelligibility of the Const.i.tution from that of Shakespeare's Sonnet 65: Since bra.s.s, nor stone, nor earth, nor boundless sea, But sad mortality o'ersways their power, How with this rage shall beauty hold a plea, Whose action is no stronger than a flower?

O, how shall summer's honey breath hold out Against the wrackful siege of battering days, When rocks impregnable are not so stout, 89. Consider the "reversal of values" achieved by the nineteenth-century English actor Henry Irving, who first made Shylock a thoroughly sympathetic character. "The elopement scene in his production [of The Merchant of Venice] closed in a whirl of music, lights and color, with Jessica and Lorenzo caught up in a crowd of masquers as it swept across the stage. The curtain dropped briefly; when it rose . . . 'the stage was empty, desolate, with no light but a pale moon, and all sounds of life at a great distance-and then over the bridge came the weary figure of the Jew.' He was bearing a lantern, returning (though he did not realize it yet) to a deserted house. Then the curtain fell again, without a word having been spoken." John Gross, Shylock: A Legend and Its Legacy 149 (1992).

Nor gates of steel so strong, but Time decays?

O fearful meditation! Where, alack, Shall Time's best jewel from Time's chest lie hid?

Or what strong hand can hold his swift foot back?

Or who his spoil of beauty can forbid?

O, none, unless this miracle have might, That in black ink my love may still shine bright.

Fried argues that the poem's premise is the intelligibility of writing and that it has been triumphantly vindicated by time, for in spite of being 400 years old, he implies, the poem poses no interpretive problems for the contemporary reader.90 He has overlooked those problems, and not only by sub silentio modernizing the spelling and punctuation.91 In writing "spoil of beauty" he has (again without acknowledgment) used an emended version of Shakespeare's original text, which reads "spoil or beauty."92 And he has overlooked the note of dubiety sounded by "unless" and "may" in the concluding couplet and by the possible pun in "might." Shakespeare is not so confident as Fried supposes that "black ink" can survive time's ravages.93 There may be a graver misinterpretation. It has been suggested that the reference to "black ink" is contemptuous and that Shakespeare, care Fried, "Sonnet LXV and the 'Black Ink' of the Framers' Intention," 100 Harvard Law Review 751 (1987).

On the uncertainties of meaning created by the erratic punctuation of Shakespeare's sonnets as originally published, see Theodore Redpath, "The Punctuation of Shakespeare's Sonnets," in New Essays on Shakespeare's Sonnets 217 (Hilton Landry ed. 1976). To compare the original with the modern spelling and punctuation of Sonnet 65, see Shakespeare's Sonnets 5859 (Stephen Booth ed. 1977), where the original and modern versions are printed side by side. I have quoted the version of the sonnet published in Fried's article.

Booth, in Shakespeare's Sonnets, note 91 above, at 247, argues against the emendation. On the inescapability of literary criticism in textual emendation, see G. Thomas Tanselle, "Recent Editorial Discussion and the Central Questions of Editing," 34 Studies in Bibliography 23 (1981).

Compare Fried, note 90 above, at 756, with Shakespeare's Sonnets, note 91 above, at 247; Murray Krieger, A Window to Criticism: Shakespeare's Sonnets and Modern Poetics 170172 (1964); Philip Martin, Shakespeare's Sonnets: Self, Love and Art 153155 (1972); Kenneth Muir, Shakespeare's Sonnets 66 (1979); Rodney Poisson, "Unequal Friendship: Shakespeare's Sonnets 18126," in New Essays on Shakespeare's Sonnets, note 91 above, at 1, 11.

less of publication, saw the "miracle" elsewhere: "The poet knows that through his poetry, or the poetic consciousness, he establishes, or focuses, a supernal reality, or truth, what we may call a 'poetic dimension,' that cannot otherwise be attained; and of this the written poetry ('black ink'), though it be necessary, is really subsidiary, the carrot to the donkey, but not the journey's purpose."94 Shakespeare "asks how time can be stayed and to allow anything to survive. He resorts to the potentially miraculous powers of poetry to preserve his love."95 I will set aside these quibbles and concede that Fried has shown that a great work of literature can, because of the timelessness of its theme, be universal without being ambiguous. The implications for the interpretation of the Const.i.tution are obscure. Some of Shakespeare's works may not pose acute interpretive difficulties, but others do, as do works of literature considerably more recent. If we pursue the misguided quest for literary a.n.a.logies to problems of legal interpretation, we shall have to ask in every case whether the particular statutory or const.i.tutional provision we're interested in is, in point of interpretive difficulty, more like Sonnet 65 (or what Fried thinks Sonnet 65 is like) or more like other and more ambiguous literary works, such as Billy Budd, written by an American little more than a century ago. On such questions Fried is silent. Literary theory provides no more comfort for the legal Right than for the legal Left. One can no more argue the interpretability of the Const.i.tution from Sonnet 65 than one can argue the inscrutability of the Const.i.tution from Hamlet.

Interpretation as Translation I said at the beginning of this chapter that the debate over interpretation, having reached white-hot intensity in the 1980s, has cooled considerably. There was a new development of note in the 1990s, however: the proposal to model legal interpretation on translation from one language to 94. G. Wilson Knight, The Mutual Flame: On Shakespeare's Sonnets and The Phoenix and the Turtle 86 (2d ed. 1982).

95. Dympna Callaghan, Shakespeare's Sonnets 124 (2007).

another,96 especially translation of literary works, where the problematics of translation are greatest.

James Boyd White argues that since "no sentence can be translated into another language without change," translation can only be "the composition of a particular text by one individual mind in response to another text,"97 and a judicial opinion interpreting a provision of the Const.i.tution should be viewed in the same light. The premise is incorrect. Some sentences can be translated into another language without any loss of meaning: instructions for a.s.sembling a kitchen table, for example. The provision of the Const.i.tution that the President must be at least 35 years old, or the provision that each state is ent.i.tled to two senators, can be "translated" from eighteenth-century linguistic, political, and social understandings into those of the twentieth century without loss of meaning, even though longevity has increased and the method of choosing senators has changed.

This is more than a quibble. It shows that literal translation is not an oxymoron and places on White the burden of showing that legal enactments are more like poems than like instructions. It also suggests that translation requires choices that may be neither right nor wrong.98 Agamemnon's usual t.i.tle in the Iliad is anax andron. A literal translation might be "Supreme Leader of the Warriors," but that is stilted. If we wanted to make the Iliad sound modern we could translate the term as "Supreme Allied Commander" or even "Chief Honcho" (this would be like playing Hamlet in modern dress). If we wanted to preserve the sense of antiquity, of cultural distance, we could leave the term untranslated, as in "Kaiser Wilhelm" (versus "Emperor William"). We might compromise with "Lord Agamemnon," but this sounds a bit British imperial, like "Lord Curzon" or "Earl Mountbatten." The choice among these possi See James Boyd White, Justice as Translation: An Essay in Cultural and Legal Criticism (1990); Lawrence Lessig, "Fidelity in Translation," 71 Texas Law Review 1165 (1993). For criticism of these "translation" theories, see Sanford Levinson, "Conversing about Justice," 100 Yale Law Journal 1855 (1991); Levinson, "Translation: Who Needs It?" 65 Fordham Law Review 1457 (1997).

97. White, note 96 above, at 250, 254.

For striking examples from Kafka, see Osman Durrani, "Editions, Translations, Adaptations," in The Cambridge Companion to Kafka 206,214219 (Julian Preece ed. 2002).

*bilities, none clearly right or clearly wrong but none satisfactory either, depends on the effects the translator is aiming at and the intended use of the translation and hence the intended audience. Although White expects reflection on the difficulties of literary translation to engender humility in the judge faced with the task of "translating" an eighteenth-century doc.u.ment into the culture of today, an alternative inference from the practice of literary translation is the translator's freedom. If a translator can choose between a literal and a free translation, why not a judge between a literal and a free interpretation?

Building, like White, on the lawyerly intuition that const.i.tutional and statutory interpretation must, if it is to be legitimate, be faithful in some sense to the const.i.tutional or statutory text, Lawrence Lessig argues that faithful translation cannot be literal because the cultural significance of words changes. In my example, "Lord" may be an adequate literal translation of anax andron, but it means something so different to modern Americans (Lord Peter Wimsey? Lord Haw-Haw? Lord Acton?) that to affix it to "Agamemnon" alters Homer's meaning for American readers. To preserve meaning in an altered social context, we might have to choose a literal mistranslation.99 This shows that "the translator is empowered to change text."100 The same ought to be true, Lessig argues, for judicial interpretation of the Const.i.tution.

There is something to this view, certainly. As the biblical scholar Bart Ehrman explains, "The only way to make sense of a text is to read it, and the only way to read it is by putting it in other words, and the only way to put it in other words is by having other words to put it into, and the only way you have other words to put it into is that you have a life, and the only way to have a life is being filled with desires, longings, needs, wants, beliefs, perspectives, worldviews, opinions, likes, dislikes-and all the other things that make human beings human. And so to read a text is, See, for example, Eugenio Benitez, "On Literal Translation: Robert Browning and the Agamemnon," 28 Philosophy and Literature 259 (2004); Amel Amin-Zaki, "Religious and Cultural Considerations in Translating Shakespeare into Arabic," in Between Languages and Cultures: Translation and Cross-Cultural Texts 223 (Anuradha Dingwaney and Carol Maier eds. 1995).

Lessig, note 96 above, at 1191. This may not differ from White's claim that the translator must produce a new text.

necessarily, to change a text."101 But this is "to change a text" in a metaphoric sense, or in other words to interpret. The translator, in contrast to the reader of the original, untranslated work, is authorized to change text in the literal sense, but for two reasons neither of which is applicable to legal interpretation. The first is the absence of exactly equivalent words in the language into which a work is being translated. The second, which is related, is the need to make the translation a good "read." Translation thus involves two stages. In the first, the translator interprets the original, that is, decides the meaning. In the second, he expresses his interpretation as best he can in the language of the readers of the translation.102 The first stage, interpretation, has a counterpart in law, of course, but I hope I have persuaded the reader that the interpretation of different kinds of text, specifically the literary and the legal, has no fruitful commonalities. The second stage in translation, expression of the translator's interpretation in a new language, is no part of legal interpretation.

Consider what the best translation of a foreign work of literature or philosophy would be if the only objective were to convey its full meaning clearly-including the shades of meaning conveyed by the form or style of the work-without regard for readability or emotional impact. It would be a literal translation with numerous footnotes explaining the ways in which the translation might mislead. Untranslatable words and phrases (such as anax andron) might be left in the original and their meaning explained by bracketed paraphrases. Anachronisms, false cognates, conventions, changes in the cultural, linguistic, and historical context-all would be patiently explained. The result would be charmless and copious, but the loss of cognitive content would be minimal. The problem of translation would not be "solved," however, and not only because an important part of the meaning of a work of literature is emotional and is killed by literal translation. The challenge of translation is not to achieve fidelity to the original but to strike a compromise between the desire to preserve the original meaning of the translated work and the desire to interest, delight, Bart D. Ehrman, Misquoting Jesus: The Story Behind Who Changed the Bible and Why 217 (2005).

Raymond van den Broeck, "Literary Conventions and Translated Literature," in Convention and Innovation in Literature 57 (Theo D'haen, Rainer Grubel, and Helmut Lethen eds. 1989).

*stir, or even just economize on the time of some target audience. The choice between a fluency that effaces the sense that one is reading a translation and so makes the author present to us and a "translationese" that preserves the sense of the author's foreignness103 will depend upon the nature of the intended audience,104 the character of existing translations, the translator's talents, and the author's preferences if he's alive and controls translation rights.105 The choice has nothing to do with interpreting legally operative doc.u.ments written in the interpreter's language.

An even deeper problem with the "translation" metaphor for statutory and const.i.tutional interpretation, stressed by Sanford Levinson,106 is that of verifiability. The accuracy of a translation can be determined by a person fluent in both languages, and such persons can be identified with considerable confidence. Whom do we trust to know the meaning of a disputed statutory or const.i.tutional provision and thus to verify the accuracy of its "translation" into a modern setting? No one; if judges or professors were trusted to determine what such provisions mean, const.i.tutional law and theory would not be such disputatious activities.

Lawrence Venuti, The Translator's Invisibility: A History of Translation (1995). Venuti argues against the tendency in translations into English to achieve fluency at the expense of accuracy, a tendency he attributes in leftist fashion to English cultural aggressiveness. He advocates "foreignizing" translations. See, for example, id. at 20.

Which is why careful translations of the same work into the same language can differ so greatly-as strikingly shown by the anthology Homer in English (George Steiner ed. 1996).