Heterogeneity, Legislative History, and the Costs of Litigation: A Brief Comment on Bruhl’s “Hierarchy and Heterogeneity”

Anuj Desai*

Should lower federal courts rely on legislative history as a source of interpretive authority in statutory cases? And, should the answer to that question depend on a different weighing of factors than answering the same question as to the United States Supreme Court? These are two of the normative questions that Aaron-Andrew Bruhl raises in his recent Cornell Law Review article “Hierarchy and Heterogeneity: How to Read a Statute in a Lower Court.”1 In addressing these questions, Bruhl argues that “[s]tatutory interpretation is a court-specific activity that should differ according to the institutional circumstances of the interpreting court.”2 He points to three kinds of differences among courts: “(1) the court’s place in the hierarchical structure of appellate review, (2) the court’s technical capacity and resources, and (3) the court’s democratic pedigree.”3 Building on these three differences, Bruhl tentatively sketches out an argument that lower federal courts “should be extremely wary of delving into legislative history.”4 In particular, one factor Bruhl focuses on is the resource disparities that “make the use of legislative history more problematic the lower one goes in the legal hierarchy.”5

In this brief Comment, I want to make two points. First, to weigh the merits of Bruhl’s ideas, we need to know the answer to the descriptive question that parallels the normative one: Are the lower federal courts in fact using legislative history? Second, if one of the principal arguments that “hierarchy” matters is that courts at different levels have different decisional capacities (such as resource differences), Bruhl’s argument depends, at some level, on empirical assumptions about the actual costs of both researching legislative history and litigation generally.

Let’s turn to the first point. In literal terms, the answer to the question of whether lower federal courts use legislative history is of course “yes.” So, perhaps more precisely, the question is whether the lower federal courts rely on legislative history infrequently enough to change the weighing of factors in Bruhl’s analysis. For if the lower courts are already deciding most cases without resort to legislative history, many of the potential benefits of a conscious policy of having them do so are significantly reduced, while one possible rule-of-law cost of treating the lower courts differently from the Supreme Court—public confidence that all judges in our system are engaged in the same task— remains the same. Bruhl alludes to scholarly work that is beginning to answer the descriptive question of how the interpretive process is actually operating at the lower-court level, but he argues that the “substantial value [of the positive work] cannot be fully realized without a normative framework that will let us evaluate whether lower courts are doing well and, perhaps, tell them how to do better.”6 My point is just the opposite: the force of any normative argument that the lower federal courts should rely on legislative history less than the Supreme Court depends in part on whether they are already doing this.7 For if the lower courts are already relying less on legislative history, the benefits of advocating that they do so are largely limited to a justification of existing practice. On the other hand, telling lower federal courts to interpret statutes differently from the Supreme Court would contradict the common American conception of what courts do,8 something that might (again, another unanswered empirical question) undermine confidence in the courts.9

My second point also has the merits of Bruhl’s normative argument turn on empirics. Bruhl notes that courts at different levels of the judicial hierarchy enjoy different levels of resources. Bruhl thus appears to assume that properly researching legislative history requires resources and that a court with more resources will do a better job than one with fewer resources.10 As he mentions briefly, this includes the resources that the parties and their lawyers bring to a case.11 In contrast to the run-of-the-mill case in the lower federal courts, most Supreme Court cases involve a bevy of sophisticated lawyers and amicus briefs, along with, the overwhelming majority of the time,12 the Solicitor General of the United States, the premier appellate litigation shop in the country.13

If we view Bruhl’s proposal through this lens, we can see that the proposal might have a distributive impact.14 At the Supreme Court level, even if the parties are not equally matched in terms of resources, it is probably the rare case when both sides (including the amici) cannot comprehensively research and analyze the relevant legislative history. In contrast, lower down the hierarchy, a disparity in overall resources might well lead to a disparity in the resources necessary to research legislative history properly. If so, this disparity can skew the law systematically in favor of “the haves.”15

To understand why this might be, consider a schematic scenario in which one party (or one type of party) systematically has the resources to research legislative history and the other does not.16 For simplicity’s sake, let’s call the former “the richer party.” Assume further that the plain meaning of the statute is arguably ambiguous between two meanings, M1 and M2, but that without any legislative history the court would resolve the ambiguity in favor of M1. Assume further that in both cases the legislative history directly supports M2 and that, if confronted with the legislative history, the court would resolve the statutory ambiguity in favor of M2. More importantly (and perhaps controversially), let us assume that M2 is in fact “the correct answer” (whatever that might mean). So, my scenario assumes that with legislative history, the court decides the case correctly, and without it, the court gets it wrong.

We might then compare two stylized examples, one in which the richer party favors M1 (the hypothetically “wrong” result) and the other in which the richer party favors M2 (the hypothetically “correct” result). If the richer party favors M2, then that party will likely cite the legislative history that it finds. At that point, the lower-court judge is at least on notice that the legislative history might actually matter, thereby possibly leading the judge to inquire further, even if the poorer party fails to do so. Of course, given my starting assumptions, it won’t matter, since the judge will simply find that the legislative history does in fact favor M2 and rule in favor of the richer party. If the richer party favors M1, however, the only party with the information that the legislative history supports M2 has no incentive to disclose that fact, and the court might thus never even consider the legislative history, thereby interpreting the statute with meaning M1 and (again) deciding the case in the richer party’s favor.17 Since I’m assuming for the limited purposes of this brief Comment that legislative history leads to better results (i.e., M2 is in fact a better interpretation than M1),18 a rule that barred legislative history in the lower courts altogether would of course exacerbate the problem broadly by leading the judge to get it “wrong” (i.e., deciding in favor of M1) in both cases. On the other hand, from the standpoint of distributive equity, one might argue that getting it “wrong” in both cases results in a better system overall: allowing lower courts to rely on legislative history leads to the richer party winning every time (i.e., the court would decide in favor of M1 if the richer party favors M1 and in favor of M2 if the richer party favors M2).

Even if one cares about this inequity, though, to determine the likelihood of it occurring requires some more concrete information about the actual costs of both researching legislative history and litigation in general.19 If researching legislative history is both cheap enough in absolute terms and, at the same time, a small enough proportion of the overall costs of litigation in a particular court, then this might well be a nonissue. On the other hand, if the cost of researching and analyzing legislative history is either high enough that certain types of parties simply do not have the resources to do it at all or a large enough proportion of the overall costs of litigation given the stakes in a given case, it might matter.20


* Professor of Law, University of Wisconsin Law School. The author also currently serves as a part-time Commissioner at the Foreign Claims Settlement Commission, United States Department of Justice, Washington, D.C. The opinions expressed here are those of the author and do not reflect those of the United States Department of Justice. Thanks to Andy Coan and Aaron-Andrew Bruhl for helpful comments.


WLR Online: Volume 2013, No. 3


  1. Aaron-Andrew P. Bruhl, Hierarchy and Heterogeneity: How to Read a Statute in a Lower Court, 97 Cornell L. Rev. 433 (2012).

  2. Id. at 433.

  3. Id.

  4. Id. at 495.

  5. Id. at 476.

  6. Id. at 499.

  7. It may be, as Adrian Vermeule argues, that certain empirical questions necessary to answer the normative question can never be answered, Adrian Vermeule, Judging Under Uncertainty: An Institutional Theory of Legal Interpretation 3 (2006), but this one strikes me as well within the realm of social scientific inquiry.

  8. Marbury v. Madison, 5 U.S. 137, 177 (1803) (“It is emphatically the province and duty of the judicial department to say what the law is.”).

  9. This argument naturally parallels the argument that all courts—not just the United States Supreme Court—should have the power of constitutional review. However, some of the empirical considerations that Bruhl rightly raises in the context of the use of legislative history would not arise when debating proposals to have lower courts approach constitutional questions differently from the Supreme Court. Of course, my reference to “the American conception” of the nature of courts is a (perhaps not so) subtle allusion to the fact that in much of the civil-law world, a special “constitutional court” is the sole judicial body with the power to invalidate statutes. Alec Stone Sweet, Governing with Judges: Constitutional Politics in Europe 32–34 (2000).

  10. Embedded in this assumption is also an implicit assumption that this disparity matters. Even if we assume—rightly, I think—that the lower federal courts have fewer resources than the Supreme Court, this does not necessarily make the use of legislative history more problematic at the lower-court level. It depends on the absolute level of the resources, not just the relative level. So if all courts have sufficient resources and competence to do a proper analysis of legislative history, then their relative differences are irrelevant. I suspect this is not the case, but the actual numbers can matter. More importantly, they can change over time, as the costs of research and lawyering change over time.

  11. Bruhl, supra note 1, at 470–72.

  12. See Margaret Meriwether Cordray & Richard Cordray, The Solicitor General’s Changing Role in Supreme Court Litigation, 51 B.C. L. Rev. 1323, 1353–54 (2010).

  13. In his testimony before the Senate Judiciary Committee to become solicitor general, Paul Clement referred to the Office of the Solicitor General as “the finest law firm in the Nation.” Confirmation Hearing on the Nomination of Paul D. Clement to Be Solicitor General of the United States, 109th Cong. 5 (2005) (statement of Paul D. Clement).

  14. As will become clear, my primary focus here is the cost savings to lawyers (and thus, in turn, to their clients) that would come from a rule barring legislative history. For the purposes of this brief Comment, then, I bracket the broader costs to the judicial system, because I assume (again, solely for the purpose of this brief Comment) a party-driven approach to litigation.

    Besides the increase in litigation costs, numerous arguments have been made against the use of legislative history. I will mention just five: (1) it is unconstitutional, see, e.g., John F. Manning, Textualism as a Nondelegation Doctrine, 97 Colum. L. Rev. 673 (1997); (2) it undermines the goal of fair notice to those regulated by the law; (3) collective intent is incoherent, see Max Radin, Statutory Interpretation, 43 Harv. L. Rev. 863, 870–71 (1930); (4) legislative “intent” is irrelevant, see Antonin Scalia, A Matter of Interpretation: Federal Courts and the Law 31 (1997); and (5) even assuming “intent” is the goal, legislative history is unlikely to lead to better results, see Vermeule, supra note 7, at 107–08 (2006). For purposes of this short Comment, I am ignoring (1) through (4), assuming (5) is wrong (i.e., assuming that the use of legislative history will marginally improve decisions), and am concerned almost exclusively with the costs of litigation as a problem. This increase in costs is a variant of a point made famous by Justice Robert Jackson. His concern was broadly with ready access to the underlying materials of legislative history, a problem he described as particularly acute for the small town lawyer or what he called the “average law office.” See generally Robert H. Jackson, The Meaning of Statutes: What Congress Says or What the Court Says, 34 A.B.A. J. 535, 538 (1948) (remarks before the American Law Institute, May 20, 1948), reprinted with revisions as Problems of Statutory Interpretation, 8 F.R.D. 121 (1949); see also United States v. Public Utilities Comm’n. of Calif., 345 U.S. 295, 319 (1953) (Jackson, J., concurring); Schwegmann Bros. v. Calvert Distillers Corp., 341 U.S. 384, 395 (1951) (Jackson, J., concurring). For an updated look at Justice Jackson’s “lament” in light of modern technological developments, see Richard A. Danner, Justice Jackson’s Lament: Historical and Comparative Perspectives on the Availability of Legislative History, 13 Duke J. Comp. & Int’l L. 151, 190–94 (2003). To be clear, I am not claiming any of my assumptions are true, simply maintaining them for the limited purpose of engaging with Bruhl’s argument.

  15. Marc Galanter, Why the“Haves” Come out Ahead: Speculations on the Limits of Legal Change, 9 L. & Soc’y Rev. 95, 149 (1974).

  16. We might think of the second party as being represented by the quintessential “average law office” in Justice Jackson’s “lament.” See supra note 14.

  17. Of course, as I noted in the text, my example is highly stylized. It may well be that different parts of the legislative history point in different directions on a question of statutory interpretation. In such circumstances, as Bruhl points out, a mismatch in the lower courts raises the specter that the richer party will selectively use legislative history to its advantage, even if a more careful analysis of the legislative history would point the other way. See Bruhl, supra note 1, at 475–76.

  18. See supra note 14.

  19. See, e.g., David M. Trubek et al., The Costs of Ordinary Litigation, 31 UCLA L. Rev. 72, 91–92 (1983).

  20. By referring to the idea of “costs,” I mean it in the broadest sense to include not just literal monetary costs (e.g., billable attorney time), but also the expertise necessary to research and properly analyze legislative history. In some circumstances, the “cost” of such expertise might of course be monetized by hiring co-counsel with the requisite expertise.