Universal De Novo review

George Washington Law Review

Volumn 77, Issue 2, 2009, Pages 308-365

  Oldfather, Chad M ^a  

^a Universal de Novo Review ^*

Author keywords

[No Author keywords available]

Indexed keywords

EID: 67649553910     PISSN: 00168076     EISSN: None     Source Type: Journal    
DOI: None     Document Type: Article

References (227)

1. See, e.g., RICHARD D. FREER, INTRODUCTION TO CIVIL PROCEDURE § 14.7 (2006) ("This intrusive review is appropriate because the determination of what the law is, is not something on which the trial court would have especial expertise .. It is a dry question of research, which the appellate court may actually be better equipped than the trial court to perform.");
2. see also ALLAN IDES & CHRISTOPHER N. MAY, CIVIL PROCEDURE 1056-57 (2d ed. 2006) (implying that the standard is based on competence, despite similar training in the law among district and appellate court judges). Other introductory texts attribute the de novo standard to the need to ensure uniformity. See, e.g., THOMAS D. ROWE, Jr. et al., CIVIL PROCEDURE 300 (2004) (stating that de novo review of questions of law "reflects the importance of attempting to assure uniformity in the definition of generally applicable legal standards").
3. As Professor Caminker noted in concluding his analysis of why lower courts are bound to follow precedent, "Familiarity dulls curiosity." Evan H. Caminker, Why Must Inferior Courts Obey Superior Court Precedents?, 46 STAN. L. REV. 817, 873 (1994).
4. See generally Ann Woolhandler & Michael G. Collins, Judicial Federalism and the Administrative States, 87 CAL. L. REV. 613, 658 (1999) ("[D]e novo review actually gives the courts a greater role in lawmaking than does deferential review.").
5. See generally CASS R. SUNSTEIN, ONE CASE AT A TIME: JUDICIAL MINIMALISM ON THE SUPREME COURT 4 (1999) ("Decisional minimalism has two attractive features. First, it is likely to reduce the burdens of judicial decision .. Second, and more fundamentally, minimalism is likely to make judicial errors less frequent and (above all) less damaging.").
6. See infra text accompanying notes 180-84.
7. See generally DANIEL J. MEADOR et al., APPELLATE COURTS 1040-50 (2d ed. 2006) (reprinting and commenting on Donald P. Lay, A Proposal for Discretionary Review in Federal Courts of Appeals, 34 Sw. L.J. 1151, 1155-58 (1981);
8. Bernard G. Barrow, The Discretionary Appeal: A Cost Effective Tool of Appellate Justice, 11 GEO. MASON L. REV. 31 (1988); and DANIEL J. MEADOR, APPELLATE COURTS: STAFF AND PROCESS IN THE CRISIS OF VOLUME 168-71 (1974)).
9. There has been an enormous amount of commentary concerning the use of unpublished opinions over the past decade. For a small sampling, see Martha Dragich Pearson, Citation of Unpublished Opinions as Precedent, 55 HASTINGS L.J. 1235,1275-93 (2004);
10. Michael B.W. Sinclair, Anastasoff Versus Hart-The Constitutionality and Wisdom of Denying Precedential Authority to Circuit Court Decisions, 64 U. PITT. L. REV. 695, 705-10 (2003);
11. Lauren Robel, The Practice of Precedent: Anastasoff, Noncitation Rules, and the Meaning of Precedent in an Interpretive Community, 35 IND. L. REV. 399, 402-04 (2002).
12. Bush v. Gore, 531 U.S. 98 (2000).
13. See id. at 109 ("Our consideration is limited to the present circumstances."). to See generally RICHARD L. PACELLE, Jr., The ROLE OF THE SUPREME COURT IN AMERICAN POLITICS 9-31 (2002) (describing the issues raised by judicial policymaking in light of Bush v. Gore).
14. For an analysis of the nature of this "duty to decide," see Chad M. Oldfather, Defining Judicial Inactivism: Models of Adjudication and the Duty to Decide, 94 GEO. L.J. 121, 161-80 (2005).
15. The Chevron doctrine, arising out of Chevron U.S.A. Inc. v. Natural Resources Defense Council, Inc., 467 U.S. 837, 843-44, 864-66 (1984), requires courts in certain circumstances to defer to agency interpretations of their governing statutes and is rooted in considerations of institutional competence and separation of powers. See, e.g., Note, The Two Faces of Chevron, 120 HARV. L. REV. 1562, 1562 (2007).
16. See infra Part II.
17. See infra Part H.B.
18. See infra notes 79-81 and accompanying text.
19. See infra notes 169-71 and accompanying text.
20. See infra notes 172-76 and accompanying text.
21. See generally Chad M. Oldfather, Appellate Courts, Historical Facts, and the Civil-Criminal Distinction, 57 VAND. L. REV. 437, 444-66 (2004) (describing and critiquing justifications for deference to trial-level factfinding).
22. See id. at 497-98 (describing a study revealing that appellate courts frequently disregard factual determinations of the jury in sufficiency-of-the-evidence reviews).
23. The primary example involves questions of "constitutional fact." See generally Henry P. Monaghan, Constitutional Fact Review, 85 COLUM. L. REV. 229 (1985);
24. see also generally Adam Hoffman, Note, Corralling Constitutional Fact: De Novo Fact Review in the Federal Appellate Courts, 50 DUKE L.J. 1427 (2001) (defining and critiquing the version of the constitutional fact doctrine applied by the Supreme Court).
25. See Charles Alan Wright, The Doubtful Omniscience of Appellate Courts, 41 MINN. L. REV. 751, 778-79 (1957) (discussing power maximization and concluding that it is a result of decisions purportedly in the public interest, not hunger for power).
26. See 1 STEVEN ALAN CHILDRESS & MARTHA S. DAVIS, FEDERAL STANDARDS OF REVIEW § 2.14 (3d ed. 1999);
27. see also MARY BETH BEAZLEY, A PRACTICAL GUIDE TO APPELLATE ADVOCACY 15 (2d ed. 2006) ("De novo review is sometimes referred to as plenary review because it allows the court to give a full, or plenary, review to the findings below.").
28. See 1 CHILDRESS & DAVIS, supra note 22, § 2.14 ("This [appellate] role is more accurately described as one of no particular deference.");
29. BEAZLEY, supra note 22, at 15 ("When courts apply the de novo standard, they look at the legal questions as if no one had yet decided them, giving no deference to legal findings made below. When this standard is applied, the reviewing court is willing to substitute its judgment for that of the trial court or the intermediate court of appeals.").
30. See Salve Regina Coll. v. Russell, 499 U.S. 225, 232 (1991) ("Independent appellate review necessarily entails a careful consideration of the district court's legal analysis, and an efficient and sensitive appellate court at least will naturally consider this analysis in undertaking its review."); cf. 1 CHILDRESS & DAVIS, supra note 22, § 2.14 ("Even de novo review, then, seems to call for specific and aggressive allegation of error by the appellant, who must also keep in mind that free review is not strict scrutiny and will lack the skepticism built into that constitutional review test.").
31. See 1 CHILDRESS & DAVIS, supra note 22, § 2.14 ("[T]he appellate court is acting within a body of established law binding on it under principles of judicial hierarchy, stare decisis, and sometimes law of the case.").
32. See id. ("When the appeals court has full review of a legal issue, it has no license to venture freely into other issues of fact or the case as a whole.").
33. See id. ("[W]hat is meant [by de novo review] is merely appellate power, ability, and competency to come to a different conclusion on the record as determined below."); see also, e.g., Watzek v. Walker, 485 P.2d 3, 6 (Ariz. Ct. App. 1971) ("This Court on appeal will normally not disturb a trial court judgment if there is any reasonable evidence supporting it. Yet, in reviewing questions of law, we are not bound by the findings of the trial court but are free to draw our own legal conclusions from the evidence presented." (citations omitted)). In this sense it is distinct from trial de novo as an appeal mechanism. See MARTIN SHAPIRO, COURTS: A COMPARATIVE AND POLITICAL ANALYSIS 37 (1981) ("[W]e find many other legal systems that use or have used trial de novo at [sic] the standard mode of appeal. Where appeal is by trial de novo, the appellate court simply hears the whole case all over again. Trial de novo is found in most nonliterate societies, no doubt in large part because of the difficulty of preserving trial court findings of fact to serve as a basis for appellate decision." (citation omitted)).
34. See HARRY T. EDWARDS & LINDA A. ELLIOTT, FEDERAL COURTS STANDARDS OF REVIEW 24 (2007) ("[D]e novo review requires nothing more of an appellate court than that it decide legal issues as would the first level decisionmaker .. It adds nothing of its own to the appellate process, but rather simply mandates that an appellate court apply the substantive standards governing resolution of the legal question at issue.").
35. The content of the abuse of discretion standard-to the extent it is even appropriate to speak of a single standard in this context-is considerably more elusive than that of de novo review. See generally 1 CHILDRESS & DAVIS, supra note 22, § 4.21 (discussing how the abuse of discretion standard "appears to differ between contexts"). Still, the standard entails at least some deference to the trial judge, which implies a range of affirmable rulings, rather than a single one. See id. ("It is clear, of course, that the abuse of discretion phrase is meant to insulate the judge's choice from appellate second-guessing. It also should be clarified that in most cases the appellate court may not treat the phrase as equal to an error of law if that is meant to imply free review; surely some deference is due under an expression of abuse of discretion, and if indeed independent review is appropriate in certain situations, courts might then say that an abuse test is not applicable there.").
36. See HENRY M. HART, Jr. & ALBERT M. SACKS, THE LEGAL PROCESS 349 (William N. Eskridge, Jr. & Philip P. Frickey eds., 1994) ("Some critics have concluded that 'law' cannot be distinguished analytically from 'fact.'"); MEADOR ET AL., supra note 6, at 223 ("The distinction [between questions of law and questions of fact] has bedeviled the courts for decades. There are, of course, some clear-cut situations. But in many instances there is no bright line, and decisions are difficult to reconcile.").
37. 1 CHILDRESS & DAVIS, supra note 22, § 2.13 ("To be sure, at its outer limit the line is both clear and uncontroversial: when the appeals court is examining a 'purely legal' conclusion, such as whether the First Amendment can broadly protect gestures or applies to the states, it is not hard to recognize the court acting in a law-making role .. The other end also may be neat: what Alice did yesterday is regarded as a pure fact and is usually delegated to the trial court."); see also EDWARDS & ELLIOTT, supra note 28, at 7 (articulating further the spectrum spanning from legal precepts to historical facts).
38. See 1 CHILDRESS & DAVIS, supra note 22, § 2.13 ("In spite of some clarity at wicks' ends, the area in between these 'pure' examples gets increasingly sticky, and the courts cannot resolve the tricky differences by bare citation to the general rule of appellate lawmaking or by pointing to the ubiquitous but quixotic law-fact distinction. An error of law can involve anything from simple reliance on an overturned case, to a complex mix of the evidence at hand with the effort to state some substance and guidance that will apply beyond the facts of the particular case. The latter mission may be considered lawmaking, but it is not undoubtedly so, and the appellate court in its action may be either just doing its job or usurping the trial judge's job.");
39. EDWARDS & ELLIOTT, supra note 28, at 8 ("As the labels suggest, mixed findings or mixed questions generally defy ready categorization as either law or fact. Consequently, the fact/law paradigm falters as a method for determining whether appellate review should be de novo or deferential.").
40. See BEAZLEY, supra note 22, at 15 ("Courts apply the de novo standard not only to questions of law, but also to mixed questions of law and fact. A mixed question of law and fact is often characterized as a question about whether certain agreed-upon facts meet a legal standard."). Standards of review relating to mixed questions differ from one jurisdiction to the next, and not for any apparent philosophical or jurisprudential reasoning. At least one commentator has argued that "[t]here seems to be no rhyme or reason" to the jurisdictional variations. See Randall H. Warner, All Mixed Up About Mixed Questions, 7 J. APP. PRAC. & PROCESS 101, 107 (2005). Others offer explanations grounded in practical and policy considerations. See, e.g., Or-nelas v. United States, 517 U.S. 690, 700 (1996) (Scalia, J., dissenting) (noting that when a mixed question is presented on appeal, the standard of review "depend[s] upon essentially practical considerations"); Kent S. Scheidegger, Habeas Corpus, Relitigation, and the Legislative Power, 98 COLUM. L. REV. 888, 905 (1998) (examining three instances of mixed questions and concluding that the court considered "the choice of the standard of review [to be] a policy question"). Some courts and commentators have attempted to articulate guidelines for allocating review authority, such as by considering whether the question at issue is "primarily" or " essentially" a question of fact or law and by considering who (judge or jury) is better situated to make the determination. Warner, supra, at 107, 109-11. Still other courts have adopted a process of bifurcated review, pursuant to which the appellate court defers to the trial court's resolution of the factual aspects of the issue while allowing for independent review of its legal aspects. 1 CHILDRESS & DAVIS, supra note 22, § 2.18 (explaining the "split inquiry" and citing two examples: Price v. Wain-wright, 759 F.2d 1549, 1551 (11th Cir. 1985), and United States v. McConney, 728 F.2d 1195, 1200-01 (9th Cir. 1984) (en banc)).
41. See generally 1 JACOB A. STEIN, STEIN ON PERSONAL INJURY DAMAGES § 4:47 (3d ed. 1997) (citing BMW of N. Am., Inc. v. Gore, 517 U.S. 559 (19%), as providing "guideposts" for de novo appellate review of punitive damage awards and, as examples of the application of the standards, Cont'l Trend Res., Inc. v. Oxy USA, Inc., 101 F.3d 634 (10th Cir. 1996), Lee v. Ed-wards, 101 F.3d 805 (2d Cir. 1996), Utah Foam Prods. Co. v. Upjohn Co., 930 F. Supp. 513 (D. Utah 1996), and Rush v. Scott Specialty Gases, Inc., 930 F. Supp. 194 (E.D. Pa. 1996)).
42. See generally 1 CHILDRESS & DAVIS, supra note 22, § 2.19 (describing the history and uncertain application of the constitutional fact category); see also generally Monaghan, supra note 20, at 238 (describing presuppositions behind constitutional fact review and arguing "that constitutional fact review at the appellate level is a matter for judicial (and legislative) discretion"). Generally speaking, constitutional facts are those that implicate constitutional rights and are to be reviewed de novo in order to ensure the policy objective, among others, of uniform and effective application of the Constitution. See 1 CHILDRESS & DAVIS, supra note 22, § 2.19 ("[Factfindings on decisive constitutional questions should be, for party protection and constitutional integrity, left for independent appellate review."). For example, in Bose Corp. v. Consumers Union of United States, Inc., 466 U.S. 485, 502 (1984), the Court stated that "the constitutional values protected by the [actual malice] rule make it imperative that judges . . . make sure that it is correctly applied."
43. Professor Meador has described this function as requiring the court "[t]o correct error in the trial proceedings and to insure justice under law to the litigants." MEADOR ET AL., supra note 6, at 2. Professor Meador noted, however, the diminishing dominance of the error-correction function in the modern court: "This is the historic basis for the intervention of an upper court. Though still central to the mission of an appellate system, it is no longer viewed as the sole or even (in the eyes of some) as the primary purpose." Id. Elsewhere, Meador, along with Professors Carrington and Rosenberg, have observed that "[t]he traditional appeal calls for an examination of the rulings below to assure that they are correct, or at least within the range of error the law for sufficient reasons allows the primary decisionmaker." PAUL D. CARRINGTON, DANIEL J. MEADOR & MAURICE ROSENBERG, JUSTICE ON APPEAL 2 (1976). In addition, they emphasize the role of the appeal mechanism in assuring trial judges that their correct decisions will be endorsed by the larger system and providing similar assurances to the parties and the public. Id.
44. According to Professor Meador, the lawmaking function requires a court "[t]o enunciate and harmonize the decisional law of the jurisdiction." MEADOR ET AL., supra note 6, at 2. This function now occupies a more prominent place both because legal realism undercut the classic view of courts finding, rather than creating, law, and because the complexity of modern society requires more law, and more refined law, than legislatures are capable of providing. See id. Meador identifies additional functions, including the supervision of trial courts and further-ance of the perception of systemic legitimacy. See id. at 2-3; see also ROBERT J. MARTINEAU, MODERN APPELLATE PRACTICE § 1.10 (1983) (arguing that "[d]oing justice" in both "the individual case" and "in the context of the entire legal system" ought to be regarded as a function of appellate courts).
45. See MEADOR ET AL., supra note 6, at 2. But see Wright, supra note 21, at 779 ("The controversial question is whether appellate courts have a second function, that of ensuring that justice is done in a particular case.").
46. See generally Chad M. Oldfather, The Concept of Error in Civil Appeals (Oct. 16,2008) (unpublished manuscript, on file with author), for a discussion of the nature of review for error and a more complete treatment of conceptions of the error correction function.
47. See, e.g., Commonwealth v. Craft, 669 A.2d 394, 398 (Pa. Super. Ct. 1995) (Johnson, J., concurring) ("As an error-correcting court, we are generally limited to determining whether the trial judge has committed either an abuse of discretion or an error of law in the handling and disposition of a case.").
48. See MEADOR ET AL., supra note 6, at 3 ("[A]n appeal spreads responsibility, thereby making difficult or sensitive decisions more liveable for the decision makers and the mistakes more tolerable for all concerned."); see also SHAPIRO, supra note 27, at 49 ("For appeal allows the loser to continue to assert his rightness in the abstract without attacking the legitimacy of the legal system or refusing to obey the trial court.").
49. See ROSCOE POUND, APPELLATE PROCEDURE IN CIVIL CASES 3 (1941) ("That hasty, unfair or erroneous action may be reversed by a court of review holds back the impulsive, impels caution, constrains fairness and moves tribunals to keep to the best of their ability in the straight path.").
50. See MEADOR ET AL., supra note 6, at 2 ("A recognition of [law making] as a purpose of appellate review is a product of modern legal realism. Under the earlier view of courts as simply finding and declaring the law, this would have been perceived as merely a by-product or an incidental feature of appellate adjudication."); see also Paul D. Carrington, The Function of the Civil Appeal: A Late-Century View, 38 S.C. L. REV. 411, 416 (1987) (observing that at the time of the creation of the present structure of the federal judiciary "[t]he perceived role of the appellate court was to correct the errors of the trial court in applying the law to the facts" and "[n]o one thought appellate courts necessary or useful in making law or policy").
51. MEADOR ET AL., supra note 6, at 2 ("Some now think of this law enunciating function as a more important justification for appeals than the error correcting purpose."); see also Wright, supra note 21, at 779 ("Everyone agrees, so far as I know, that one function of an appellate court is to discover and declare-or to make-the law.").
52. There are exceptions. Some litigants, such as public interest groups and repeat players, are often more concerned with convincing the court to articulate a favorable legal standard than with getting a particular result in the case before the court.
53. This wju not always be the case. If the appellant's argument is that the trial judge applied a manifestly incorrect legal standard, such as by rendering a decision pursuant to a case that has been overruled or a statute that has been repealed, then the claim would implicate only the error correction function.
54. See Richard S. Arnold, Unpublished Opinions: A Comment, 1 J. APP. PRAC. & PROCESS 219,222-23 (1999) (arguing that virtually every appellate judicial decision requires "a conclusion of law with precedential significance").
55. This is not to suggest that trial court departures from the appellate court's preferred rule will always be deemed erroneous in the sense that they will result in a reversal of the trial court's decision. "Harmless error" will not result in reversal, but nonetheless remains error. See Chapman v. California, 386 U.S. 18, 22 (1967) ("We conclude that there may be some constitutional errors which in the setting of a particular case are so unimportant and insignificant that they may, consistent with the Federal Constitution, be deemed harmless, not requiring the automatic reversal of the conviction."); Fed. R. CIV. P. 61 ("The court at every stage of the proceeding must disregard any error or defect in the proceeding which does not affect the substantial rights of the parties.").
56. In the Seventh Circuit, one formulation of the abuse of discretion standard famously requires that to be overturned the decision under review must "strike [the reviewing court] as wrong with the force of a five-week-old, unrefrigerated dead fish." Parts & Elec. Motors, Inc. v. Sterling Elec, Inc., 866 F.2d 228, 233 (7th Cir. 1988). Of course, appellate review for abuse of discretion "must vary with the nature of the lower court decision." Lawson Prods., Inc. v. Avnet, Inc., 782 F.2d 1429, 1437 (7th Cir. 1986). "[T]he variety of matters committed to the discretion of district judges means that the standard is necessarily variable." EDWARDS & ELLIOT, supra note 28, at 67.
57. As William Lucy has pointed out, the list of things we expect from adjudication is long-including "impartiality, consistency, predictability, fairness, justice, rationality and legitimacy"-and exceedingly difficult to accommodate. William Lucy, Adjudication, in THE OXFORD HANDBOOK OF JURISPRUDENCE AND PHILOSOPHY OF LAW 206, 206 (Jules Coleman & Scott Shapiro eds., 2002). Any judicial practice will rest on some set of understandings with respect to each of these expectations, which in turn "brings into play many of the central questions of ancient and contemporary legal and political philosophy." Id. at 207.
58. See KARL N. LLEWELLYN, THE COMMON LAW TRADITION: DECIDING APPEALS 24 (1960) ("The deciding is done under an ideology which in older days amounted to a faith that there is and can be only one single right answer. This underlies such ideas as 'finding the law' and 'the true' rule, and 'the' just decision.").
59. I have in mind here something close to full determinacy, which, as Larry Solum defines it, holds "with respect to a given case if and only if the set of legally acceptable outcomes contains one and only one member." Lawrence B. Solum, Indeterminacy, in A COMPANION TO PHILOSOPHY OF LAW AND LEGAL THEORY 488, 490 (Dennis Patterson ed., 1999). This includes both metaphysic and epistemic components-the former relating to whether there is law, the latter relating to whether it is knowable. See id. at 498 (discussing Ken Kress, A Preface to Epistemological Indeterminacy, 85 Nw. U. L. REV. 134 (1990)). As Dan Coenen has observed, the doctrine of de novo review appears based on an understanding that law is determinate in both senses. See Dan T. Coenen, To Defer or Not to Defer: A Study of Federal Circuit Court Deference to District Court Rulings on State Law, 73 MINN. L. REV. 899, 911-12 (1989) ("This accepted approach to the appellate process apparently rests on a shared understanding that law exists, that the 'duty' of judges is 'to say what the law is,' and that doctrinal coherence and fairness for similarly situated litigants demands consistent application of legal rules. The rule of de novo review, whatever its source, runs deep in our history; independent appellate inquiry into questions of law has marked our republic's legal system from its earliest days." (citations omitted)).
60. See supra note 38 and accompanying text.
61. As Paul Carrington has observed: A chief concern [in 1891 when the Evarts Act was adopted] seems to have been a mistrust of the professionalism of the judiciary and of the capacity of individual judges to apply correctly law that was presumed clear and, thus, amenable to application .. Such a legal system required an effective appellate system for one reason: The perceived role of the appellate court was to correct the errors of the trial court in applying the law to the facts. Carrington, supra note 43, at 416; see also id. at 423-24 (discussing the formalist assumptions predominant in 1925).
62. See, e.g., Cass R. Sunstein, Must Formalism Be Defended Empirically?, 66 U. CHI. L. REV. 636, 660 (1999) (describing case law arguments in "the debate over formalism," in particular "the suggestion, central to legal realism, that the decision how to read ambiguities in law involves no brooding omnipresence in the sky but is an emphatically human judgment about policy or principle"). Of course, as Michael Steven Green points out, this story contains elements of caricature. See Michael Steven Green, Legal Realism as Theory of Law, 46 WM. & MARY L. REV. 1915, 1984-85 (2005).
63. See, e.g., Green, supra note 55, at 1917 (noting that the assertion that we are all Realists has been repeated so often "that it has become a cliche' to call it a 'cliche""); see also Solum, supra note 52, at 501 ("[T]he indeterminacy debate has made it clear that almost no one defends a strong formalist claim that the law determines every aspect of the outcome of every case .. If the claim that the law is radically indeterminate turns out to be silly, it is also the case that the strong formalist claim that the outcomes of cases are completely determined by the law is just as implausible.").
64. Indeed, the fact that law declaration is now regarded as one of the twin functions of appellate courts stands as evidence of broad acceptance of the proposition that law is "un-derdeterminate." Per Solum: "The law is underdeterminate with respect to a given case if and only if the set of legally acceptable outcomes is a non-identical subset of the set of all possible results." Solum, supra note 52, at 490. This stands in contrast to indeterminacy, which holds "with respect to a given case if the set of legally acceptable outcomes is identical with the set of all possible results." Id.
65. See infra Part II.
66. See, e.g., Gilman v. Philadelphia, 70 U.S. (3 Wall.) 713, 724 (1865) ("It is almost as important that the law should be settled permanently, as that it should be settled correctly.").
67. Commentators have offered various formulations of what might make a case "hard." Solum suggests that a hard case is one in which "the outcome is underdetermined by the law in a manner such that the judge must choose among legally acceptable outcomes in a way that changes who will be perceived as the 'winner' and who the 'loser.'" Solum, supra note 52, at 490. Frederick Schauer enumerates various types of cases that might fall into the category, including those where applicable rules provide no answer, where the rules provide an answer that conflicts with the rules' underlying purposes, where applicable rules provide contradictory answers, and where the rules provide a clear answer that is consistent with their purpose yet still difficult to accept. Frederick Schauer, Easy Cases, 58 S. CAL. L. REV. 399, 415 (1985). I take no position regarding whether the notion of error correction is inappropriate in the context of every such case. I do contend that the notion of error correction probably does not make sense with respect to, at the very least, some substantial subset of such cases.
68. Judge Richard Posner suggests that this dynamic will be present across an even larger segment of cases: There is almost always a zone of reasonableness within which a decision either way can be defended persuasively, or at least plausibly, using the resources of judicial rhetoric. But the zone can be narrow or wide-narrow when formalist analysis provides a satisfactory solution, wide when it does not. Within the zone, a decision cannot be labeled "right" or "wrong"; truth just is not in the picture. Richard A. Posner, The Role of the Judge in the Twenty-First Century, 86 B.U. L. REV. 1049,1053 (2006).
69. One could articulate a rule-consequentialist justification for error correction akin to that developed below in Part II.B, in which fulfillment of the institutional error correction function would not depend on appellate courts having the ability or opportunity to correct error in all instances. The possibility of such a justification does not of course detract from the point that review in such cases does not itself involve error correction.
70. See Dan Simon, A Psychological Model of Judicial Decision Making, 30 RUTGERS LJ. 1, 11 (1998) ("Opinions are overstated, rigid, seemingly inevitable. The rhetorical style is that of closure. The judge is depicted as having little choice in the matter: the decisions are strongly constrained by the legal materials." (citations omitted)).
71. See LLEWELLYN, supra note 51, at 26.
72. Perhaps the most prominent example of relatively recent vintage is the "joint opinion" in Planned Parenthood of Southeastern Pennsylvania v. Casey, 505 U.S. 833, 845-16 (1992), in which Justices O'Connor, Souter, and Kennedy voted to uphold "the essential holding of Roe v. Wade" "[a]fter considering the fundamental constitutional questions resolved by Roe, principles of institutional integrity, and the rule of stare decisis . . . ."
73. Thanks to Larry Solum for suggesting this term. The notion that I am after is akin to what Ken Kress has called "epistemological indeterminacy." See Kress, supra note 52, at 138-39. As Kress draws the distinction: Metaphysical indeterminacy speaks to whether there is law; epistemic indeterminacy, to whether the law can be known. We might say that the question of abortion in a particular jurisdiction is metaphysically (or ontologically) determinate (at some particular time) if there is a right answer to the question whether a woman has a right to an abortion in that jurisdiction. It might nevertheless be epistemically indeterminate whether women have that right, because the right answer is not demonstrable, or because there is no method for determining the right answer, or because there is great controversy among lawyers or other persons about what that right answer is.
74. Thus, the situation I posit is distinct from one where the legislature could not have foreseen that the precise question presented would arise, such as where the question concerns the applicability of a statute passed in the first half of the nineteenth century to a situation involving automobiles. In that situation, again assuming that legislative intent is the established guidepost for the analysis, it would arguably be more appropriate to regard the question as one without a correct answer.
75. For a discussion of the problems associated with such a determination, see Jonathan R. Siegel, Textualism and Contextualism in Administrative Law, 78 B.U. L. Rev. 1023, 1025-32 (1998).
76. See id. at 1026-27.
77. See infra Part II.A.3.
78. For a discussion touching on the articulability of legal rules, see Chad M. Oldfather, Writing, Cognition, and the Nature of the Judicial Function, 96 GEO. L.J. 1283, 1318-39 (2008).
79. This is not always a substantial demand in the sense that it requires a lot in the way of detail. Vague, inarticulable concepts can be law. The notion of reasonableness, for example, is a legal standard that depends for its content on a host of intuitive judgments. "Articulability" does not always equate to "precise articulability" or "comprehensive articulability"-the court can have a loose or vague standard in mind (such as reasonableness), but if it is unable to convey adequately that standard in words, then the same problems arise.
80. One could, for example, imagine an originalist defense of the practice akin to the argument Judge Richard Arnold made against the practice of issuing nonprecedential opinions. See Arnold, supra note 47, at 225-26. Arnold argued that the original understanding of the phrase "judicial power" as used in Article III of the Constitution compels the conclusion that virtually every decision rendered by an appellate court includes some conclusion of law that has, and ought to be accorded, precedential significance. Id. at 226. For a brief time, Arnold's arguments had the force of law in the Eighth Circuit. See Anastasoff v. United States, 223 F.3d 898, 899 (8th Cir.), vacated as moot, 235 F.3d 1054, 1056 (8th Cir. 2000) (en banc). A parallel argument that the "judicial power" in the case of appellate courts necessarily includes de novo review of every legal question presented seems plausible, though not without problems. The judicial systems in the states as they existed at the framing of the Constitution were very different from the federal system created in the Judiciary Act of 1789, which in turn became the model for state judiciaries. See WILFRED J. Rrrz, REWRITING THE HISTORY OF THE JUDICIARY ACT OF 1789, at 5-6, 35-38 (Wythe Holt & L. H. LaRue eds., 1990) (discussing the differences among the systems). In similar fashion, one might argue that any right to an appeal includes only a right to error correction and not a right to law declaration.
81. See FRANK M. COFFIN, THE WAYS OF A JUDGE 58 (1980) (arguing that an appellate court's "only
82. See Caminker, supra note 2, at 846-47; Wilfred Feinberg, Unique Customs and Practices of the Second Circuit, 14 HOFSTRA L. REV. 297, 300 (1986); Wilfred Feinberg, The Office of Chief Judge of a Federal Court of Appeals, 53 FORDHAM L. REV. 369, 385 (1984).
83. See Lewis A. Kornhauser & Lawrence G. Sager, Unpacking the Court, 96 YALE L.J. 82, 97-98 (1986). This holds if we assume that the probability of each judge on a panel reaching the "correct" outcome is greater than chance. Id. at 97. Relying on the Condorcet Jury Theorem, Kornhauser and Sager ask us to imagine a jar filled with marbles of only two colors (black and white), and to consider each judge's decision to be a draw of a single marble. Id. at 97-98. If the marbles are distributed in proportion to the probability of an individual judge reaching the correct outcome (with white marbles corresponding to the correct outcome and black marbles corresponding to an incorrect outcome), then the more draws (i.e., votes), the more likely the majority of draws will be white (i.e., reflect the correct outcome). Id. For a thorough analysis of the appropriateness and limitations of applying the theorem in this context, see Adrian Vermeule, Many-Minds Arguments in Legal Theory 2-6 (Harvard Law Sch. Pub. Law Research, Paper No. 08-02, 2008), available at http://ssrn.com/abstract= 1087017.
84. See William M. Richman & William L. Reynolds, Elitism, Expediency, and the New Certiorari: Requiem for the Learned Hand Tradition, 81 CORNELL L. REV. 273, 278 (1996) (describing the emphasis in "the Learned Hand model" on individual contemplation and face-to-face interaction).
85. Some contend that such interaction never did occur. See, e.g., RICHARD A. POSNER, HOW JUDGES THINK 2 (2008) ("The difficulty outsiders have in understanding judicial behavior is due partly to the fact that judges deliberate in secret, though it would be more accurate to say that the fact that they do not deliberate (by which I mean deliberate collectively) very much is the real secret." (citation omitted));
86. id. at 34 ("Remember that judges do not engage in much collective deliberation over a case (in fact less than most juries do)."); Peter M. Tiersma, The Textualization of Precedent, 82 NOTRE DAME L. REV. 1187, 1230-31 (arguing that historical prevalence of unanimous opinions actually evidences less debate among justices); Adam J. Hirsch, Cognitive Jurisprudence, 76 S. CAL. L. REV. 599, 630-32 (2003) (noting multimember courts' lack of "collegiality and deliberation" and arguing that more interaction might lead to qualitatively better lawmaking).
87. See MEADOR ET AL., supra note 6, at 385-419.
88. See Richman & Reynolds, supra note 77, at 296-97.
89. Id. at 295-96.
90. See RUGGERO J. ALDISERT, OPINION WRITING § 3.4 (1990) (commenting on practices that allow "a one-person opinion to emerge from a multi-judge court").
91. See Ellen E. Sward, Appellate Review of Judicial Fact-Finding, 40 U. KAN. L. REV. 1, 13-14 (1991) (describing the argument, but concluding that the appellate court's competence advantage, if any, "is slight").
92. Viewed in this light, the analysis also suggests the possibility that the "law" specialization of appellate courts could potentially be a detriment, in much the same way that other specialized bodies are criticized for reaching suboptimal decisions in some respects.
93. See Caminker, supra note 2, at 845-46.
94. Id. at 846 ("District courts deal primarily with case management and factfinding and have thus developed expertise in these areas. By contrast, both tiers of appellate courts deal primarily with legal issues and have thus developed a corresponding expertise. The assignment of different primary tasks and its effect on procedures and experience thus ensure that appellate court legal rulings are likely to be 'better' than district court rulings." (citations omitted)).
95. See EDWARDS & ELLIOTT, supra note 28, at 76.
96. Caminker, supra note 2, at 846 n.117.
97. This is not to suggest that being closer to the facts will generally position a court to make superior legal detenninations. In fact, as explored in greater detail in the next subpart, the opposite is probably true.
98. See, e.g., Robert C. Owen & Melissa Mather, Thawing Out the "Cold Record": Some Thoughts on How Videotaped Records May Affect Traditional Standards of Deference on Direct and Collateral Review, 2 J. APP. PRAC. & PROCESS 411, 412 (2000) (observing the perceived relationship between the "cold" record and "reasoned decisionmaking");
99. see also LESTER BERNHARDT ORFIELD, CRIMINAL APPEALS IN AMERICA 85 (1939) ("The cold printed record inevitably must give an incomplete and sometimes distorted picture of the case.").
100. Cf. Oldfather, supra note 71, at 1312-14 (exploring the "Unconscious Thought Theory" in the context of a broader survey of psychological research, suggesting that attempts to provide written justifications for decisions may negatively affect the quality of the underlying decision).
101. This is one of the central critiques offered by advocates of the "public law" model of adjudication. For a brief overview, see Chad M. Oldfather, Remedying Judicial lnactivism: Opinions as Informational Regulation, 58 FLA. L. REV. 743, 752-54 (2006).
102. This idea can be traced back at least to Aristotle. David J. Seipp, Our Law, Their Law, History, and the Citation of Foreign Law, 86 B.U. L. REV. 1417, 1437-38 (2006).
103. See Caminker, supra note 2, at 852.
104. See id. at 850-51.
105. Id. at 851.
106. Id.
107. See id. at 851-52.
108. See id. at 852-53.
109. loo See Sward, supra note 83, at 14. Here we see the operationalization of the assumptions that law is both ascertainable and articulable.
110. Cf. Steven Shavell, The Appeals Process as a Means of Error Correction, 24 J. LEGAL STUD. 379, 425 (1995) (questioning the value of harmonization of law); Amanda Frost, (Overvaluing Uniformity, 94 VA. L. REV. (forthcoming 2008) (questioning the inherent and practical value of uniformity in the interpretation of law).
111. See STEPHEN C. YEAZELL, CIVIL Procedure 628 (6th ed. 2004) ("Discovery and the pretrial process now dominate the procedural landscape. Most rulings entered at this stage do not immediately produce final judgments, and many will never produce final judgments because the case settles. Under such circumstances the final judgment rule does not just defer but eliminates appellate review.").
112. Id.
113. See id.
114. New State Ice Co. v. Liebmann, 285 U.S. 262,311 (1932) (Brandeis, J., dissenting) ("It is one of the happy incidents of the federal system that a single courageous State may, if its citizens choose, serve as a laboratory; and try novel social and economic experiments without risk to the rest of the country.").
115. See generally RICHARD A. POSNER, THE FEDERAL COURTS: CHALLENGE and REFORM 53-123 (1996) (describing and attempting to explain, based in part on the increased amount of law, the tremendous growth in caseloads in the federal courts).
116. The view that such decisions are problematic is surprisingly long-lived. See John P. Bor-ger & Chad M. Oldfather, Anastasoff v. United States and the Debate over Unpublished Opinions, 36 TORT & INS. L.J. 899, 900-04 (2001).
117. See POSNER, supra note 106, at 166 ("Despite the vast number of published opinions, most federal circuit judges will confess that a surprising fraction of federal appeals, at least in civil cases, are difficult to decide not because there are too many precedents but because there are too few on point.").
118. Cf. THE FeDERALIST No. 22, at 112 (Alexander Hamilton) (George W. Carey & James McClellan eds., 1990) ("To avoid the confusion which would unavoidably result from the contradictory decisions of a number of independent judicatories, all nations have found it necessary to establish one court paramount to the rest . . . [t]o settle and declare in the last resort, a uniform rule of civil justice.").
119. 11o See, e.g., Charles E. Wyzanski, Jr., A Trial Judge's Freedom and Responsibility, 65 HARV. L. REV. 1281,1297-98 (1952) (noting that, in trial courts, "the pace is quicker, the troublesome issues have not been sorted from those which go by rote, the briefs of counsel have not reached their ultimate perfection").
120. But see Caminker, supra note 2, at 826-28 (imagining a system in which lower courts are not bound by the legal determinations of higher courts).
121. In reality, it seems likely that, as Karl Llewellyn pointed out, lawyers and judges vacillate between these two views of the force of precedent as necessary to suit their purpose in a given case. See KARL NICKERSON LLEWELLYN, THE BRAMBLE BUSH 66-69 (1960); see also Walter V. Schaefer, Precedent and Policy, 34 U. CHI. L. REV. 3, 12-16 (1966) (outlining various factors that will affect the extent to which a decision is given precedential effect by subsequent courts).
122. See Brad Hooker, Rule Consequentialism, in STANFORD ENCYCLOPEDIA OF PHILOSOPHY para. 1 (Edward N. Zalta ed., 2008), http://plato.stanford.edu/archives/fall2008/entries/conse quentialism-rule/ (describing rule-consequentialism as a theory by which morality of a rule is judged "solely in terms of the goodness of [its] consequences" rather than in terms of the rule itself in the abstract).
123. See id. para. 4 (describing an act-consequentialist decision procedure as requiring that, "[o]n each occasion, an agent should decide what to do by calculating which act would produce the most good").
124. Under a rule-consequentialist decision procedure, "fa]t least normally, agents should decide what to do by applying rules whose acceptance will produce the best consequences. .." Id. Adrian Vermeule defines the distinction in a similar way. See ADRIAN VERMEULE, JUDGING UNDER UNCERTAINTY: AN INSTITUTIONAL THEORY OF LEGAL INTERPRETATION 5 (2006).
125. Of course, the effects outlined in this Section will vary somewhat depending upon just how much present courts are constrained by past decisions. Some have suggested that stare decisis does not, or at least should not, provide much of a constraint at all. See, e.g., Michael Stokes Paulsen, Does the Supreme Court's Current Doctrine of Stare Decisis Require Adherence to the Supreme Court's Current Doctrine of Stare Decisis?, 86 N.C. L. REV. 1165,1166 n.3 (2008) (enumerating works in which Paulsen has questioned the doctrine of stare decisis in constitutional law). If we were to do away with precedent as any sort of formal constraint on judicial decisionmaking (with it still providing some amount of logical constraint), then we are left with error-correction-based rationales for de novo review, which are in turn rooted in the assumptions about institutional competence that I have called into question above, and which counsel against universal de novo review. On the other hand, if we have a strong doctrine of stare decisis, pursuant to which precedents clearly bind and are difficult to dislodge, then the resulting high costs of an erroneous initial decision likewise counsel against universal de novo review. Viewed from this angle, then, probably the strongest case for universal de novo review occurs where the binding nature of precedent is somewhere in the middle, such that bad precedents are relatively easy to distinguish or overrule.
126. See Jeffrey O. Cooper & Douglas A. Berman, Passive Virtues and Casual Vices in the Federal Courts of Appeals, 66 BROOK. L. REV. 685, 721-22 (2000).
127. I am going to assume that it is meaningful to talk about "better" and "worse" legal rules and resolutions of cases and will do so without staking any claim as to the identity of the criteria based on which to make the distinction. That said, the discussion that follows will necessarily touch on some ways in which a rule might be suboptimal.
128. Lon L. Fuller, The Forms and Limits of Adjudication, 92 HARV. L. REV. 353,364 (1978).
129. See id.
130. See id. at 385-86; see also STEPHAN LANDSMAN, THE ADVERSARY SYSTEM 2-3 (1984) (describing arguments for a "neutral and passive decision maker").
131. See Abram Chayes, The Role of the Judge in Public Law Litigation, 89 HARV. L. REV. 1281, 1284 (1976) (noting examples of cases affecting nonparties, including cases involving school desegregation, employment discrimination, prisoners' rights, antitrust, securities fraud, bankruptcy, union governance, consumer fraud, housing discrimination, electoral reapportionment, and environmental management).
132. See id. For an overview of the public law model, see Oldfather, supra note 11, at 145-49.
133. See, e.g., Owen M. Fiss, The Supreme Court, 1978 Term-Foreword: The Forms of Justice, 93 HARV. L. REV. 1, 26 (1979).
134. See id.
135. See Oldfather, supra note 11, at 152-53. For a thorough consideration of the signifi-cance of participation, see Lawrence B. Solum, Procedural Justice, 78 S. CAL. L. REV. 181, 273-305 (2004).
136. For tongue-in-cheek support of poor appellate lawyering, with examples, see Alex Kozinski, The Wrong Stuff, 1992 BYU L. REV. 325, 326-33.
137. Cf, e.g., Ryan S. Goldstein et al., Specialized IP Trial Courts Around the World, INTELL. PROP. & TECH. L.J., Oct. 2006, at 1, 1 ("The Federal Circuit, which has subject matter jurisdiction over patent, trademark, US International Trade Commission, and other cases, was formed to have the type of specialized expertise necessary to adjudicate intellectual property cases and provide guidance to lower courts through its opinions."); see also id. at 3-4 (describing arguments for and against specialized courts).
138. Cf., e.g., Mirit Eyal-Cohen, Preventive Tax Policy: Chief Justice Roger J. Traynor's Tax Philosophy, 59 HASTINGS L.J. 877, 879 (2008) ("When Congress established the [predecessor to the federal Tax Court].. it initially considered it a technical, tax expert tribunal.. ." (citation omitted)). One response to the argument that the judges on a given panel might lack the subject matter expertise to resolve a complex, specialized issue could be that such a possibility is contrary to the very nature of law. That is, that ideals such as "rule of law and not of men" presuppose a body of law that is accessible to and understandable by not only those with legal training but also to educated laypersons. On this view, there neither can nor should be questions of law that are amenable to resolution only by judges possessing an appropriate base of specialized knowledge.
139. See Mitu Gulati & C.M.A. McCauliff, On Not Making Law, 61 LAW & CONTEMP. PROBS. 157, 162, 173-74 (1998).
140. See infra Part II.A.3.
141. See supra note 11 and accompanying text.
142. Peter J. Smith, New Legal Fictions, 95 GEO. L.J. 1435, 1468 (2007).
143. Adrian Vermeule characterizes these sorts of questions as "'trans-scientific': although they are empirical in principle, they are unresolvable at acceptable cost within any reasonable time frame." VERMEULE, supra note 115, at 3. Vermeule reasons: It may be that the costs of acquiring the data needed to answer the empirical questions are prohibitive or at least greater than the benefits to be gained from choosing the best interpretive doctrine. It may also be, more simply, that the needed data cannot be obtained (or at least not obtained in full) at any cost, because uncertainty is objective and irreducible.
144. Even where apposite social science evidence exists, courts have a mixed track record of incorporating it into their decisionmaking. For an overview of the debate over courts' use of empirical research in formulating legal rules, see Smith, supra note 133, at 1445-49.
145. See Posner, supra note 61, at 1053 ("For the judge, the duty to decide the case and to do so, moreover, with reasonable dispatch is primary."); see also Oldfather, supra note 11, at 124.
146. See VERMEULE, supra note 115, at 154 ("Stalemate is not tolerable for judges who must actually choose interpretive rules on which decisions will be based today and tomorrow and the next day rather than a generation on. From the standpoint of the institutionalist judge, then, the hard question is what to do in the short run, given the absence of necessary information and the limited capacity of boundedly rational judges to absorb whatever information is present.").
147. For an overview of justiciability doctrines and justifications thereof, see generally Jonathan R. Siegel, A Theory of Justiciability, 86 TEX. L. REV. 73 (2007).
148. See supra note 4 and accompanying text.
149. Smith, supra note 133, at 1449.
150. As Vermeule points out, even relevant information about a problem can be useless, and even misleading, until the decisionmaker has gathered other pieces of information as well. VERMEULE, supra note 115, at 162 ("The benefit of gathering data may just as easily prove to be nil-not even helpful-until the amount of data reaches some discontinuous threshold, perhaps a very high threshold. Just as knowing only the first digit of a phone number is essentially useless, so too, if we currently know 5 percent of what we would have to know to choose between the alternative rules of statutory precedent, then we might increase our knowledge tenfold and still only know half of what we need to."). Such a dynamic will not always exist, of course, and it will often be the case that action based on partial information is quite sensible. See id. at 162-63.
151. Id. at 38 ("The concern here is that vivid costs in particular cases may trigger cognitive failings in both theorists and judges, causing them to overreact to the specifics of particular cases while ignoring the overall systemic effects of the interpretive rules they defend or adopt." (citation omitted)). Karl Llewellyn noted this phenomenon as well. See LLEWELLYN, supra note 51, at 43-44.
152. As broken down by Professor Schauer, a court offering a reason for a decision takes the issue at hand to a higher level of generality and makes an implicit commitment to be bound by that reason in future situations. See Frederick Schauer, Giving Reasons, 47 STAN. L. REV. 633, 640 (1995). Because a reason, being more general than the specific question under consideration, will have implications extending beyond the specific situation under consideration, the process of giving reasons can commit the decisionmaker to decide unanticipated future cases in what may turn out to be a suboptimal fashion. See id. at 649, 651. Thus, the process of giving reasons limits a reason-giver's ability to take full account of the context in which a future decision will arise, which in turn suggests that reason-giving is appropriate in contexts where abstraction and generalization are desirable, but not where particularization and contextualization are important. See id. at 653-54.
153. See Cass R. Sunstein, Problems with Rules, 83 CAL. L. REV. 953, 956-58 (1995).
154. Frederick Schauer, Do Cases Make Bad Law?, 73 U. CHI. L. REV. 883, 884 (2006).
155. This stands in contrast to the rulemakers who are not constrained by the need to resolve a particular dispute. See id. at 891.
156. See Jeffrey J. Rachlinski, Bottom-Up Versus Top-Down Lawmaking, 73 U. CHI. L. Rev. 933, 942 (2006) (noting the effects of the "availability heuristic" on judicial lawmaking). Rachlinski notes a series of other pathologies inherent in case-based rulemaking, including distortions in the rulemaker's "sense of scale and proportion," id. at 945, relative difficulties in statistical reasoning, see id. at 946, and increases in subjectivity, see id. Similarly, Adam Hirsch has explored the implications of "selective search" on the formulation of rules. See Hirsch, supra note 78, at 605-10. Because of the impossibility of considering all potential analyses of a problem, courts and other rulemakers focus only on what appear to be the most promising subset. The particulars of the fact pattern before the court will affect the identity of that subset, thereby potentially affecting the content of the chosen rule. See id. at 605-06.
157. Schauer, supra note 145, at 895. Schauer writes: [B]ecause judges (and other rulemakers) in their lawmaking and rulemaking capacity are necessarily engaged in a process of mapping a large array of future events that will be governed by the rules they make, the risk is that judges who have a particular case before them to decide will systematically overestimate the extent to which those future events will resemble the one they are now most immediately confronting. Id. at 896.
158. The facts of the immediate case might be viewed as "framing" or "anchoring" the judge's thought processes. See id. at 896-98.
159. Professor Rachlinski echoes these ideas, noting that "the pervasive message of the psychological literature on judgment and choice is that context influences decisionmaking-often more so than a decision's underlying logical structure." Rachlinski, supra note 147, at 940. The case-by-case approach, inherent in judicial lawmaking, leaves the decisionmaker more susceptible to being swayed by "the quirks and oddities of the individual parties" and other sympathy-generating factors than is likely to be the case with legislative lawmaking. Id. at 941; see also VERMEULE, supra note 115, at 38 (discussing "the distorting force of particulars").
160. Schauer notes that a court in such a situation could reach the wrong result in the particular case in order to generate the best rule, or choose to "make less law" than it otherwise might. See Schauer, supra note 145, at 900-01 ("More commonly, however, the power of the particular is a power with distorting emanations, with courts often announcing the decision rule that will most directly produce the correct result in the particular case even though that rule will produce erroneous outcomes in future cases."). Schauer provides a series of examples, from both the Supreme Court and lower courts, in which the power of the particular seems to have distorted the lawmaking that occurred. See id. at 901-04. He posits that if the problem afflicts the Supreme Court, which has a tremendous ability to screen its cases, then it is likely to be even more of a problem in the lower courts. See id. at 903.
161. See Rachlinski, supra note 147, at 951-55 (outlining some of the advantages of adjudication relative to legislation as a means of lawmaking, including that it is structured to enable courts to learn over time, that the judicial hierarchy allows for the correction of bad results, the existence of parallel authority, and the fact that courts approach problems from different frames).
162. See Schauer, supra note 145, at 906 & n.99.
163. Id. at 909 (arguing that precedent creates "the omnipresent possibility that any mistake will be systematically more powerful than any later attempts to correct it" both because such an effect inheres in the doctrine itself and because of the effects of path dependence). As Adam Hirsch argues, in operation the constraining effects of precedent may be to a large degree a byproduct of bounded rationality. Hirsch, supra note 78, at 612 ("In connection with case law, the abstention heuristic translates into blind adherence to precedent-or, in the vernacular of jurisprudence, formalism-with the same result that rules become stranded in the past.").
164. See Schauer, supra note 145, at 909-11 (exploring the nature of "selection effect" problems and their implications for self-correction).
165. See id. at 906-07 ("Although a dynamic case-based rulemaking system possesses the capacity for change, it is not clear that those changes take place at the right time or that those changes are necessarily or even systematically for the better. Initially, a significant issue is the extent to which rules may be changed with excess frequency just because the cases that prompt change are thought, for the very reason we are considering here, to be more representative than they in fact are.").
166. See George L. Priest & Benjamin Klein, The Selection of Disputes for Litigation, 13 J. LEGAL STUD. 1, 17 (1984) (arguing that cases that are not "close" tend to settle, leaving more difficult cases for adjudication).
167. See Schauer, supra note 145, at 907 ("A rule that gets it right 99 percent of the time may well be a very good rule, but a process that focuses only on the remaining 1 percent may be a process influenced to believe that some of these very good rules are in need of modification.").
168. See id. at 907-08.
169. See Hooker, supra note 113, para. 4 ("[F]ull rule-consequentialism claims that an act is morally wrong if and only if it is forbidden by rules justified by their consequences.").
170. The set of reasons that follows in this Section parallels that identified in Hooker's essay. Id.
171. These limitations will, to a great degree, be the same as those identified in the preceding Section as reasons for questioning whether the exercise of de novo review is appropriate with respect to every legal question raised on appeal. See supra Part I.
172. See supra text accompanying notes 79-81.
173. Id.
174. See Oldfather, supra note 92, at 764-79 (outlining the various ways in which the appellate process has changed in response to increased caseloads).
175. Other aspects of the changing institutional landscape may have differing sorts of implications. For example, a half century ago, Professor Wright argued against the preceding half century's trend toward appellate courts recharacterizing issues as presenting questions of law. See Wright, supra note 21, at 751 ("The principal means by which appellate courts have obtained such complete control of litigation has been the transmutation of specific circumstances into questions of law. Subtle rules about presumptions and burden of proof, elaborate concepts of causation and consideration and the rest, have been devised in such a way that unless the appellate judge handling the case is a dullard, some doctrine is always at hand to achieve the ends of justice, as they appear to the appellate court."). Wright criticized this development on the ground that it would encourage more appeals and undermine public confidence in the legal system while providing no reason to believe that appellate courts would reach "better" or "more just" decisions than trial judges. See id. at 779, 782.Wright was troubled primarily by appellate resolution of issues that fall more toward the center of the fact-law spectrum, see id. at 778, and not the sorts of purely legal questions with which this Article is primarily concerned, see supra notes 30-33 and accompanying text. As a result, the analysis set forth in this Article does not squarely apply to them. Still, it is easy to imagine defending the proposition, based in part on an extension of Wright's arguments, that appellate courts ought to have the ability to decline to engage in de novo review of such issues. See Wright, supra note 21, at 780-81. That would be consistent with the more recent trend, identified by Judge Posner and others, toward appellate courts emphasizing standards of review so as to afford greater deference to lower courts and agencies. See POSNER, supra note 106, at 176.
176. See POSNER, supra note 106, at 174.
177. See Richman & Reynolds, supra note 77, at 275 ("The overall quality of the work of the circuit courts has deteriorated markedly."); Cooper & Berman, supra note 117, at 704-05 ("[G]iven the pressures of burgeoning dockets, the courts could not remain what they once were [T]he particular mechanisms that the courts have adopted threaten the integrity of the bench.").
178. See, e.g., Carolyn Dineen King, Challenges to Judicial Independence and the Rule of Law: A Perspective from the Circuit Courts, 90 MARQ. L. REV. 765, 785 (2007) (discussing the ways in which a motivated panel "can produce a result that is not true to the rule of law, either because it is not faithful to the record in the case or because it does not fairly apply the existing law, without that fact being apparent to anyone other than the litigants").
179. Such opinions are of course meant for issuance only in "easy" cases that do not present the sorts of issues that a court would be inclined to dodge. See Cooper & Berman, supra note 117, at 707-09 (describing arguments for and against use of unpublished opinions).
180. See Gulati & McCauliff, supra note 130, at 162.
181. See Arnold, supra note 47, at 223 (suggesting that courts face the temptation to use unpublished opinions to sweep difficulties under the rug); POSNER, supra note 106, at 165 (same); Patricia M. Wald, The Rhetoric of Results and the Results of Rhetoric: Judicial Writings, 62 U. CHI. L. REV. 1371, 1374 (1995) (same).
182. See Gulati & McCauliff, supra note 130, at 162, 184.
183. For an overview of the literature relating to judicial candor, see Oldfather, supra note 11, at 155-60 nn.144-79.
184. See generally Oldfather, supra note 11 (exploring in detail this presumption).
185. See Gulati & McCauliff, supra note 130, at 192-96 (arguing that the widespread use of judgment orders has numerous "secondary effects," including increases in strategic behavior and decreases in accountability and legitimacy, that potentially outweigh any benefits arising out of efficiency or related considerations).
186. See Oldfather, supra note 92, at 746 (noting that most reform proposals aimed at ad-dressing problems created by the crisis of volume are designed to do so by restoring lost time to judges).
187. See sources cited supra note 7.
188. See supra note 6 and accompanying text.
189. Cooper & Berman, supra note 117, at 730-32.
190. Id. at 723.
191. See id. at 730, 733-35, 738-41.
192. See Schauer, supra note 145, at 914-15.
193. See id. at 915-16.
194. See Gulati & McCauliff, supra note 130, at 175 (suggesting that an appellate court's failure to decide an issue in a hard case does not deny litigants their right to error correction).
195. See id. at 176 n.89 ("A limited amount of time and a complex issue from an unfamiliar area can provide a recipe for disaster." (citing Frank H. Easterbrook, What's So Special About Judges, 61 U. COLO. L. REV. 773, 780 (1990))).
196. We miglyt also think of these situations as resulting in error creation in a more localized sense. Trial courts are generally understood to have a greater familiarity with the facts and equities of a case. At least insofar as one is focused on the error correction function of appeals, it seems appropriate to conclude that district courts will often possess an expertise advantage that, while it is not likely to extend so far as to position a district court to be better able to articulate a general rule of law, might well suffice to enable the district court to be better positioned to reach a result that is more consistent with the governing legal principles, or more "just" on its facts, than an appellate court.
197. See supra note 12 and accompanying text.
198. See Jonathan Remy Nash, Resuscitating Deference to Lower Federal Court Judges' Interpretations of State Law, 77 S. CAL. L. REV. 975, 976 (2004); Coenen, supra note 52, at 899.
199. Thanks to my colleague Michael McChrystal for drawing my attention to this parallel.
200. 1 CHILDRESS & DAVIS, supra note 22, § 4.23.
201. Id.
202. See supra note 12.
203. The justification for this "rule of deference" was that district court judges sitting in the state the law of which was at issue would have greater expertise in its determination. See Coenen, supra note 52, at 904-05. Professor Coenen questioned the legitimacy of this practice, basing his arguments in large part on the justifications for universal de novo review, including notions of appellate court specialization and collaboration. See id. at 920-31. The Supreme Court followed Coenen's recommendations and renounced deference in 1991 in Salve Regina College v. Russell, 499 U.S. 225, 231-32 (1991). However, Professor Nash suggests that the Court has engaged in this sort of deference since Salve Regina and further contends that such deference is appropriate. See Nash, supra note 189, at 991-95. He also explores the parallels between deference on questions of state law and Chevron deference. See id. at 1021. Elsewhere, David Frisch has argued that cases in which appellate courts are bound to apply the law of a foreign jurisdiction due to a contractual choice-of-law clause ought to engage in similar deference to a trial court's reasonable interpretations of that law. See David Frisch, Contractual Choice of Law and the Prudential Foundations of Appellate Review, 56 VAND. L. REV. 57, 95, 111-12 (2003).
204. For a thoughtful article addressing the complex relationship between deference and the impulse toward acontextual legal standards, see generally Paul Horwitz, Three Faces of Deference, 83 NOTRE DAME L. REV. 1061 (2008). As Horwitz observes, "[i]n deferring to other actors, courts open up a space for shared legal and constitutional interpretation by other actors who may be closer to the facts on the ground." Id. at 1066.
205. See supra Part II.A.3.
206. As I have argued elsewhere, judicial opinions can be viewed as a species of informational regulation. The literature in that area suggests that simply making a point doctrinally relevant will raise its salience, which will in turn lead courts to take it into account in their decisionmaking. Cf. Oldfather, supra note 92, at 792-93 (arguing that opinion formats, such as one with a paragraph for the claim to jurisdiction, make judges less likely to make mistakes).
207. See Fiss, supra note 124, at 25.
208. See supra notes 172-73 and accompanying text.
209. I have argued elsewhere that, even though the notion of error correction is an appropriate characterization of one of the institutional functions of intermediate appellate courts, within the context of individual cases such courts ought to regard themselves as engaged in "derivative dispute resolution." Oldfather, supra note 39, at 6. Although the dispute will almost invariably be about whether the trial court erred, approaching the task with an understanding that one's task is to resolve a dispute will, I contend, result in a decision that is more responsive to the parties' contentions and thus more consistent with the overall aims of our adjudicative system, than that likely to be reached by a court that understands its task as mere error correction. See generally id.
210. See Oldfather, supra note 11, at 152-53.
211. See supra notes 169-70 and accompanying text.
212. See Evan H. Caminker, Precedent and Prediction: The Forward-Looking Aspects of Inferior Court Decisionmaking, 73 TEX. L. REV. 1, 54-61 (1994) (discussing the benefits of percolation).
213. See id. at 56 ("In a mature legal system, there quite frequently exists a relatively small number of readily identifiable, plausible interpretations of precedent and sensible doctrinal constructs. In such cases, the independent judgment of inferior courts will not likely bring to the Supreme Court's attention arguments and approaches that would not otherwise present themselves either upon the Justices' (or their clerks') reflection, through briefing by litigants or amici curiae, or through scholarly commentary.").
214. Cf. Harlon Leigh Dalton, Taking the Right to Appeal (More or Less) Seriously, 95 YALE L.J. 62, 72-73 (1985) (addressing this concern in the context of a discretionary review system).
215. See Guido Calabresi, A Common Law for the Age of Statutes 172-73 (1982) (describing the desirability of blanket prohibitions with respect to behavior that we might sometimes be willing to tolerate, but that it appears dangerous to appear to sanction to any extent).
216. See Gulati & McCauliff, supra note 130, at 191-92 (considering such an effect in the context of the use of judgment orders to resolve hard cases). Of course, such disparities already exist because of, among other things, the circuits' different norms with respect to the publication of opinions. See id.
217. See id. at 188-89. Gulati and McCauliff speculate that such an effect might not be altogether bad, as those judges who are less inclined to defer might also be those who are more likely to generate good law. See id. at 189.
218. See id. at 189-91.
219. Any such disparity needs to be considered relative to the baseline created by present practices. At present, Gulati and McCauliff suggest, there are disparities in the frequency with which certain subject matters come before the courts that are a product of litigant behavior. See id. Because society's "haves" will have greater resources and better lawyers, the issues that come before the courts will disproportionately fall within subject areas of law affecting that segment of the population. See id.
220. For a dramatic example, see United States v. Mandel, 602 F.2d 653, 653 (4th Cir. 1979) (en banc), affg per curiam 591 F.2d 1347 (4th Cir. 1979) (affirming convictions with one claim of error denied due to an even division of votes in the en banc court).
221. See supra note 11.
222. See supra note 59.
223. For one such procedural mechanism, see Oldfather, supra note 92, at 794-801 (proposing the incorporation of "framing arguments" into judicial opinions).
224. See supra text accompanying notes 180-84.
225. See Oldfather, supra note 92, at 797-98 (outlining the virtues of transparency as a discipline on judicial behavior).
226. This category is potentially dangerous and would need to be somewhat tightly constructed for the simple reason that there are likely to be some areas of law, such as perhaps tax, in which most or even all the judges on a court lack both expertise and interest. In those situations the risk would be that the court would never make any law in such subjects. There would accordingly need to be something like a rule of necessity that applied to such situations, pursuant to which a panel could pass on an issue only where there are other judges on the court who possess the relevant expertise. But even that might not be satisfactory, since the whole idea runs counter to the notion that generalist judges are a desirable feature of our legal system.
227. There are two variables here-not only do we need to be looking for the sorts of cases where we have reason to believe that appellate courts cannot do as good a job as they normally would, we also need to be mindful of the possibility that those are not also the sorts of cases where we might expect district courts to do worse as well.