An “Unusual Jurisdictional Argument”: The Codifier’s Canon in Ayestas v. Davis

Not every one of a judge’s actions is judicial. Judges also engage in administrative tasks, such as hiring personnel. In Ayestas v. Davis, the Court has been asked to decide a jurisdictional question that rests on whether a particular authority was conferred on judges in their administrative or judicial capacities. This Essay offers a resolution to the jurisdictional issue by looking at the underlying statutory provision. By examining the provision’s codification in the United States Code, interpreted in light of the codifier’s canon, it can be seen that the relevant authority is judicial in nature.

Introduction

It is the nature of the law that sometimes life or death determinations can hinge on technicalities. Such is the case in Ayestas v. Davis, which was argued before the Court earlier this Term. The question in that case is whether the Fifth Circuit erred in the standard it set for withholding resources to investigate and develop an ineffective-assistance-of-counsel claim in the capital habeas corpus context. Arguing that the Court lacked jurisdiction to decide this issue, the Texas Solicitor General, counsel for the respondent, claimed that the authority conferred by the relevant provision, 18 U.S.C. §3599(f), was conferred to the judge in her administrative capacity, and not her judicial capacity.¹1.See Ayestas v. Stephens, 817 F.3d 888 (5th Cir. 2016), cert. granted in part sub nom. Ayestas v. Davis, 137 S. Ct. 1433 (2017). Expressing skepticism toward this argument, Justice Breyer suggested that the placement of the provision among other provisions establishing judicial duties implied that § 3599(f) is also conferred upon the judge in her judicial capacity.²2.Transcript of Oral Argument at 42-43, Ayestas v. Davis, No. 16-6795 (Oct. 30, 2017).

Is such an argument from placement in the Code legitimate? The Court seems to think so. For example, in Yates v. United States, the Court narrowed the meaning of the word “tangible object” in 18 U.S.C. § 1519, in part by citing the fact that the provisions surrounding it in the Code only dealt with documents and records.³3.135 S. Ct. 1074, 1083-84 (2015). But the decision in Yates was not uncontroversial. Soon after the decision was handed down, Tobias Dorsey, a former attorney for the House Office of Legislative Counsel, noted that a little-known provision appearing in Title 18—along with a number of other titles—seems to say that courts may not invoke a statutory provision’s placement or caption in the Code as indicative of legislative intent.⁴4.Tobias A. Dorsey, Some Reflections on Yates and the Statutes We Threw Away, 18 Green Bag 2d 377 (2015). The provision Dorsey pointed to is a legislated canon, a congressional instruction to the courts regarding how they should interpret statutes. That Justice Breyer seems prepared to flout its direction raises deep questions about the relationship between the legislative and judicial branches, including whether Congress can restrict to which tools a judge can reach when seeking to resolve ambiguities in the law.

Given this potential conflict between the Court and Congress, it would be ideal if such invocations of placement in the Code could be reconciled to Congress’s enacted interpretive instruction. In a recent Comment in the Yale Law Journal, I argued that, in fact, they can be.⁵5.Daniel B. Listwa, Comment, Uncovering the Codifier’s Canon: How Codification Informs Interpretation, 127 Yale L.J. 464 (2017). I contend that this interpretive rule—which I call the codifier’s canon—is consistent with certain invocations of a provision’s placement. Specifically, the codifier’s canon allows judges to draw interpretive inferences from codifications decisions made by Congress, but blocks off from consideration any editorial changes introduced by the non-legislative codifiers—that is, the Office of the Law Revision Counsel.

Yates, for example, exhibits such a reconcilable invocation—relying upon organizational features in the Code that were deliberated upon by Congress.

In this Essay, I explain how Justice Breyer’s suggested consideration of placement is also legitimate given a proper interpretation of the codifier’s canon. This Essay proceeds in two Parts. Part I briefly explains the codifier’s canon and the potential role of codification in statutory interpretation. Part II then provides an overview of the relevant issues in Ayestas v. Davis, including the context of Justice Breyer’s remarks. It then examines the statutory history, with particular attention to codification, for evidence of legislative intent. As the analysis of Justice Breyer’s comments reveals, by understanding the function of the codifier’s canon—and the role of codification more generally—one can more accurately and legitimately engage in statutory interpretation.

I. The Codifier’s Canon: Learning from Codification

The Court has often struggled with the impact of codification on substantive questions of law. Maine v. Thiboutot⁶6.448 U.S. 1 (1980). is one remarkable example. There the dissent and majority sparred over the relevance of a statutory phrase that was modified during an early attempt at codifying the federal laws. As the dissent pointed out, the majority relied on a change introduced by the codifiers to vastly expand the set of rights for which Section 1983 provides a cause of action.⁷7.Id. at 16-17. The change was introduced despite significant effort “to expunge all substantive alterations” resulting from the codification process.⁸8.Id. at 19. The fact that such efforts were ultimately unable to prevent the Court from finding the meaning of the statute changes helps to illustrate why Congress saw the need to legislate particular instructions for how the Court should treat elements added by the codifiers. As elaborated below, the codifier’s canon is such an interpretive direction, intended to help courts avoid being misled by editorial decisions that cannot rightfully thought to be expressions of legislative intent.⁹9.Elsewhere I have drawn a distinction between the generic codifier’s canon and the legislated codifier’s canon. See Listwa, supra note 5, at 468 n.21. The generic codifier’s canon is simply the general principle that interpreters should distinguish between editorial decisions in the Code that reflect Congress’s intent and those that instead were introduced by the codifiers. The legislated codifier’s canon is the specific statutory text directing that interpreters ought not, in certain instances, draw inferences from a provision’s caption or placement in the Code. In this Essay, I only discuss that legislated codifier’s canon and thus refer to it merely as the “codifier’s canon.”

The codifier’s canon has been enacted into at least fourteen titles of the United States Code, its language nearly identical in each instance.¹⁰10.Act of Sept. 6, 1966, Pub. L. No. 89-554, § 7(e), 80 Stat. 378, 631 (codified at 5 U.S.C. Front Matter at 10 (2012)); Act of Aug. 10, 1956, Pub. L. No. 84-1028, § 49(e), 70A Stat. 1, 640 (codified at 10 U.S.C. Front Matter at 12 (2012)); Act of Aug. 31, 1954, Pub. L. No. 83-740, § 5, 68 Stat. 1012, 1025 (codified at 13 U.S.C. Front Matter at 1 (2012)); Act of Aug. 4, 1949, Pub. L. No. 81-207, § 3, 63 Stat. 495, 557 (codified at 14 U.S.C. Front Matter at 2-3 (2012)); Act of June 25, 1948, Pub. L. No. 80-772, § 19, 62 Stat. 683, 862 (codified at 18 U.S.C. Front Matter at 5 (2012)); Act of June 25, 1948, Pub. L. No. 80-773, § 33, 62 Stat. 869, 991 (28 U.S.C. Front Matter at 5 (2012)); Act of Sept. 13, 1982, Pub. L. No. 97-258, § 4(e), 96 Stat. 877, 1067 (codified at 31 U.S.C. Front Matter at 6 (2012)); Act of Nov. 3, 1998, Pub. L. No. 105-354, §4(e), 112 Stat. 3238, 3245 (codified at 36 U.S.C. Front Matter at 11 (2012)); Act of Aug. 12, 1970, Pub. L. No. 91-375, § 11(b), 84 Stat. 719, 785 (codified at 39 U.S.C. Front Matter at 7 (2012)); Act of Aug. 21, 2002, Pub. L. No. 107-217, § 5(f), 116 Stat. 1062, 1304 (codified at 40 U.S.C. Front Matter at 7 (2012)); Act of Oct. 22, 1968, Pub L. No. 90-620, § 2(e), 82 Stat. 1238, 1306 (codified at 44 U.S.C. Front Matter at 3 (2012)); Act of Aug. 26, 1983, Pub. L. No. 98-89, § 2(e), 97 Stat. 500, 598 (codified at 46 U.S.C. Front Matter at 10 (2012), applying to subtitle II); Act of Nov. 23, 1988, Pub. L. No. 100-710, § 105(d), 102 Stat. 4735, 4751 (codified at 46 U.S.C. Front Matter at 10 (2012), applying to subtitle III); Act of Oct. 17, 1978, Pub. L. No. 95-473, § 3(e), 92 Stat. 1337, 1466 (codified at 49 U.S.C. Front Matter at 14 (2012)). The version in the title at issue in Ayestas is typical: “No inference of a legislative construction is to be drawn by reason of the chapter in Title 18 . . . in which any particular section is placed, nor by reason of the catchlines [captions] used in such title.”¹¹11.Act of June 25, 1948, Pub. L. No. 80-772, § 19, 62 Stat. 683, 862 (codified at 18 U.S.C. Front Matter at 5 (2012)). On its face, the rule construction appears to prohibit entirely the invocation of the captions or placement of a section in the Code, throwing into doubt the argument alluded to by Justice Breyer. While this is the interpretation assumed by some scholars, a careful reading of the provision’s text and its history reveals a much narrower meaning.

The codifier’s canon only appears in positive law titles of the Code.¹²12.Positive Law Codification, Off. L. Revision Couns., http://uscode.house.gov/codification/legislation.shtml (http://perma.cc/8RAZ-XX6Q). These are titles that have been significantly edited and restructured by the codifier’s office and then enacted into law. The process of positive law codification repeals and replaces the law that the positive law title is intended to codify.¹³13.Id. This is in contrast to non-positive law titles, which are mere editorial compilations of enacted law, but not enacted into law themselves.¹⁴14.Id. Ideally, positive law codification should be done in such a manner so as not to change the meaning of the underlying law in a substantive manner. But that is a herculean task. When deciding where to place a provision in the Code, the codifiers must determine the intentions of Congress when the provisions were enacted—an event that may have occurred decades before. There is a risk that they will make a mistake, resulting in an organizational decision that implies some incorrect meaning. The case of Maine v. Thiboutot,¹⁵15.448 U.S. 1 (1980). discussed above, is a one illustration of these sort of concerns.¹⁶16.See supra notes 6-8 and accompanying text.

The Members of Congress, when enacting the first titles into positive law, were highly cognizant of this risk.¹⁷17.See Listwa, supra note 5, at 476-78 (discussing congressional concerns that codification would introduce inadvertent substantive changes in the law). For this reason, they added the codifier’s canon to the title in order to direct interpreters not to rely on organizational decisions made by the codifiers during the positive law codification process. But note that this same concern does not apply to amendments to the positive law title. Once a positive law title is enacted, it becomes a statute. Any subsequent change must be in the form of a congressional amendment. Thus, every subsequent amendment to a positive law title includes instructions within the statute itself as to how the amendments should be codified. Enacted into law like any other part of the statutory text, these organizational decisions—including where in the Code the statute should be placed—can legitimately be relied upon as evidence of congressional intent.

This gives rise to a simple rule of thumb: “it is legitimate to cite the placement of a provision in a positive law title so long as the provision was enacted after the title itself was passed into positive law.”¹⁸18.Id. at 489. This rule is reflected in the text of the codifier’s canon. For example, the version of the canon that appears in Title 18 directs that “[n]o inference of a legislative construction is to be drawn by reason of the chapter in Title 18 . . . as set out in section 1 of this Act, in which any particular section is placed, nor by reason of the catchlines used in such title.”¹⁹19.Act of June 25, 1948, Pub. L. No. 80-772, ch. 645, § 5037, 62 Stat. 683, 862 (codified at 18 U.S.C. Front Matter at 5 (2012)). This rule was enacted as part of a statute that included the full text of Title 18, as prepared by the codifier’s office. The text of Title 18 appeared in section 1 of the enacting statute.

Thus, by pointing to the title “as set out in section 1 of this Act,” rather than using the language “this Title,” the legislated canon references only the version of the title included in the Act itself. Reading the text literally and narrowly, the prohibition on referencing placement and captions thus applies only to invoking these features of the title as first enacted into positive law.²⁰20.Listwa, supra note 5, at 484.

In other words, the codifier’s canon only prohibits invoking placement or caption decisions made in the context of the positive law codification process. No reference is made to future statutes that direct how amendments to Title 18 should be organized. In the next Part, I apply the codifier’s canon to the law relevant to Ayestas. As I explain, the proper interpretation of the codifier’s canon supports the legitimacy of Justice Breyer’s argument and suggests that Texas’s jurisdictional argument should not prevail.

II. The Judge as Administrator? An “Unusual” Jurisdictional Argument

In this Part, I turn to apply the codifier’s canon to the facts and law of Ayestas. The central question is whether the judge’s authority to grant funds to habeas petitioners for experts and investigators is conferred to her in her judicial or administrative capacity. While seemingly a mere semantic distinction, the consequences are severe for petitioners, as a judge’s decisions to deny funding would be effectively unreviewable if such decisions are determined to be administrative in nature. In the first Section, I outline the relevant law and explain why the question is ultimately one of congressional intent. In the second section, I invoke codification to gain insight into Congress’s understanding the relevant provision. Through the lens of the codifier’s canon, it is clear that Justice Breyer was right to be skeptical of Texas’s jurisdictional argument.

A. A Jurisdictional Boundary: The Administrative and Judicial Capacities of the Court

In 18 U.S.C. § 3599(f), Congress provided that federal courts in post-conviction capital cases involving indigent defendants should fund “investigative, expert, or other services [that] are reasonably necessary for the representation of the defendant, whether in connection with issues relating to guilt or the sentence.” The Fifth Circuit has interpreted the requirement of being “reasonably necessary” as meaning that a petitioner is only entitled to such funding if he or she can demonstrate a “substantial need” for the services, requiring a “substantiated argument, not speculation, about what the prior counsel did or omitted doing.”²¹21.Ayestas v. Stephens, 817 F.3d 888, at 896 (5th Cir. 2016), cert. granted in part sub nom. Ayestas v. Davis, 137 S. Ct. 1433 (2017). By setting this high bar, the Fifth Circuit broke with the Sixth Circuit, setting a standard that makes it much more difficult for indigent death-row inmates to challenge the effectiveness of their trial lawyers through federal habeas petitions.²²22.See Matthews v. White, 87 F.3d 756, 760 n.2 (6th Cir. 2015) (rejecting the “substantial necessity” standard); Wright v. Angelone, 151 F.3d 151 (4th Cir. 1998) (same).

In Ayestas v. Davis, the Court granted certiorari to address whether the Fifth Circuit erred in its demanding interpretation of what “reasonably necessary” means in the context of § 3599(f); but when the case came up for oral argument another issue occupied much of the courtroom discussion. In particular, at least some of the Justices seemed interested in what Justice Breyer referred to as an “unusual jurisdictional argument” raised by the Texas Solicitor General, representing the respondent, Lorie Davis, the Director of the Texas Department of Criminal Justice.²³23.Transcript of Oral Argument at 41, Ayestas v. Davis, No. 16-6795 (Oct. 30, 2017). In the respondent’s brief, Texas argued that the denial of funding under § 3599(f) could not be reviewed by the Court because it was a nonjudicial “administrative” order.²⁴24.Brief for the Respondent at 18-28, Ayestas v. Davis, No. 16-6795 (Aug. 1, 2017).

The jurisdictional argument was not argued in the court below, and no other circuit court has addressed this precise question. However, at least seven circuits have addressed a similar argument, ruling in such a manner as to give credence to Texas’s claim. When jurisdictional issues are raised, these circuits have held that a district court’s determination with regard to Criminal Justice Act (CJA) fee compensation, under 18 U.S.C. § 3006A, is not appealable.²⁵25.See United States v. French, 556 F.3d 1091, 1093 (10th Cir. 2009); United States v. Stone, 53 F.3d 141, 143 (6th Cir. 1995); Shearin v. United States, 992 F.2d 1195, 1196 (Fed. Cir. 1993); Landano v. Rafferty, 859 F.2d 301, 302 (3rd Cir. 1988); United States v. Rodriguez, 833 F.2d 1536,1537-38 (11th Cir. 1987); United States v. Walton (In re Baker), 693 F.2d 925, 927 (9th Cir. 1982); United States v. Smith, 633 F.2d 739, 742 (7th Cir. 1980). But see United States v. Turner, 584 F.2d 1389 (8th Cir. 1978) (allowing appeal where jurisdictional argument not raised); United States v. Ketchem, 420 F.2d 901 (4th Cir. 1969) (reversing denial of expenses for court appointed defense counsel where jurisdictional argument not raised). That conclusion is based on a judgment that such a decision is a mere “administrative act,” not a “judicial decision.”²⁶26.French, 556 F.3d at 1093. In distinguishing the decision as administrative, the courts pointed to two key indicia: (1) the decision is made by the district court through a “non-adversarial” process;²⁷27.Id. and (2) the CJA provided for a particular appeals process, namely a form of “minimal review by the chief judge of the circuit.”²⁸28.Id. In this way, the setting of compensation under the CJA does seem to mirror many other, more clearly administrative decisions made by district judges and subject to the approval of the chief circuit judge, such as the assignment of temporary bankruptcy referees²⁹29.Act of Sept. 19, 1950, Pub. L. No. 81-790, 64 Stat. 866, 866 (codified at 11 U.S.C. § 71(c) (2012)). and the hiring of law clerks.³⁰30.Act of July 5, 1935, Pub. L. No. 74-449, 49 Stat. 1140, 1140. That judges’ hiring decisions are not made within their judicial capacity is confirmed by the fact that absolute immunity does not apply in those contexts. See Forrester v. White, 484 U.S. 219, 220 (1988).

The circuit courts’ decisions were based on an interpretation of 28 U.S.C. § 1291 as granting jurisdiction only for judicial—as opposed to administrative—decisions.³¹31.See, e.g., Matter of Baker, 693 F.2d 925, 927 (9th Cir. 1982). The Supreme Court has ruled similarly with regard to its own jurisdiction on Article III grounds, namely that “[w]hen judges perform administrative functions, their decisions are not subject to [the Supreme Court’s] review.”³²32.Hohn v. United States, 524 U.S. 236, 245 (1998) (citing United States v. Ferreira, 54 U.S. (13 How.) 40, 51-52 (1851)). There is no reason to believe this jurisdictional bar could not be removed, but Congress has yet to do so. Congress has provided no general cause of action to appeal administrative decisions made by individuals within the judicial branch—there is no equivalent to § 702 of the Administrative Procedure Act, which only applies to executive agencies. See 5 U.S.C. § 7019(b)(1)(B) (2012) (defining “agency” to exclude “the courts of the United States”); Id. § 702 (providing a general cause of action for review of “agency action”). One caveat, however, derives from Marbury v. Madison, 5 U.S. (1 Cranch) 137, 174-176 (1803). In Marbury, Chief Justice Marshall made clear that the Court cannot exercise “appellate Jurisdiction” under Article III directly from an executive officer. Thus, any attempt to directly review an agency action would need to fall within the Court’s original jurisdiction to be valid. However, the original jurisdiction of the Supreme Court cannot be expanded. Id. at 174-176. Thus, the Court can only review an executive agency action by way of review of an appeal from a court. The question of what constitutes a “court” for the purposes of defining the Supreme Court’s appellate jurisdiction is an open question that may be resolved this Term. See Brief for Professor Aditya Bamzai as Amicus Curiae in Support of Neither Party, Dalmazzi v. United States, No. 16-961 (U.S. Nov. 14, 2017). However, it is likely the case that a judge acting in her administrative capacity is not a “court” within the relevant meaning. For this reason, Congress could only provide for Supreme Court review of judicial administrative actions by way of a cause of action that first places that claim in some other court, such as a federal district or circuit court. The Supreme Court had weighed similar indicia as the lower courts in judging whether a particular decision is made in a judicial or administrative capacity. Specifically, the Court has looked to whether a decision is: (1) made through an ex parte decision process;³³33.Ferreria, 54 U.S. (13 How.) at 46. and (2) subject to an appeal process other than the traditional judicial hierarchy, such as review by the Secretary of Treasury³⁴34.Id. at 45. or the Secretary of War.³⁵35.Hayburn’s Case, 2 U.S. (2 Dall.) 409, 409-10 & n.* (1792).

Both of these indicia have constitutional valences. The Court has indicated that a decision cannot validly be an exercise of “judicial power” if not acting on issues “presented in an adversary context.”³⁶36.Franks v. Bowman Transp. Co., 424 U.S. 747, 754-55 (1976) (discussing mootness as deriving from the Article III requirement that there be an adversarial controversy). Similarly, with regard to non-traditional appeals processes, the Court has suggested that there would be “constitutional questions” raised by an “arrangement” in which “an entity not wielding judicial power might review the decision of an Article III court.”³⁷37.Hohn, 524 U.S. at 245-46. But while these issues have driven the Court to invoke constitutional avoidance in determining whether a particular duty is administrative or judicial,³⁸38.See id. the Court is fairly clear that the question is ultimately a statutory one: did Congress intend to confer the authority to the judge in her judicial or her administrative capacity?³⁹39.See United States v. Ferreira, 54 U.S. (13 How.) 40, 47 (1851). Ultimately, what the Court is looking for is evidence of congressional intent.

In her brief, the respondent argued that both of the aforementioned indicia augur in favor of finding a funding decision under § 3599(f) to be administrative in nature. But the argument is not particularly strong. First, Texas pointed to the fact that the § 3599(f) motions can be submitted ex parte to the district judge.⁴⁰40.Brief for the Respondent at 22, Ayestas v. Davis, No. 16-6795 (Aug. 1, 2017). But that is not dispositive, since, for example, the determination of whether one is entitled to a mental-health expert is made through a similarly ex parte proceeding, and yet such determinations are appealable judicial decisions.⁴¹41.See Ake v. Oklahoma, 470 U.S. 83, 95 (1968). Second, Texas pointed to the fact that the statute establishes a process by which the amount of funding awarded under § 3599(f) is reviewed by the chief judge of the circuit if it is in excess of $7,500.⁴²42.Brief for the Respondent at 24, Ayestas v. Davis, No. 16-6795 (Aug. 1, 2017) (discussing 18 U.S.C. § 3599(g)(2) (2012)). This argument, while similarly inconclusive, at least has precedent in its favor.

As with regard to CJA compensation decisions, this review structure has been interpreted by the Tenth Circuit as implying that appellate courts do not have jurisdiction to review decisions regarding the amount of funding awarded in response to a § 3599(f) request.⁴³43.See Rojem v. Workman, 655 F.3d 1199, 1202 (10th Cir. 2011). Of course, a lower court’s interpretation may not persuade the Court. But even if it did, the decision would not determine the jurisdictional question in Ayestas. While deciding the amount of funding might be administrative, it may very well be a judicial determination as to whether § 3599(f) services are “reasonably necessary” within the meaning of the statute. In fact, such a distinction was made with regard to CJA fees by the Tenth Circuit, which allows appeals of decisions not to compensate counsel at all.⁴⁴44.See Hooper v. Jones, 536 F. App’x 796, 798-99 (10th Cir. 2013). Thus, there is not an unambiguous case in favor of construing a decision on a § 3599(f) motion to be administrative in nature.

At oral arguments, Justice Breyer expressed skepticism towards Texas’s argument, pointing toward evidence in § 3599(f) that the Congress intended to confer authority to the judge in her judicial capacity. First, he pointed to the plain meaning of the text—which says, “the court may authorize”—as suggesting that a judicial capacity was understood. Then, after a brief exchange with the respondent, he said that the provision in question is “in with other statutes that talk about [the judge’s] judicial duties.”⁴⁵45.Transcript of Oral Argument at 43, Ayestas v. Davis, No. 16-6795 (Oct. 30, 2017). In context, it appears that Justice Breyer was referencing the codification of the statutory provision enacting the provision of funds for investigators and experts—that is, § 3599—alongside other statues that, in the Justice’s mind, unambiguously describe judicial duties.

To which other provisions in the Code Justice Breyer was referring is not altogether clear. For example, he might have been thinking of the other provisions appearing in the same chapter of the Code as § 3599, which all relate to the death penalty. Alternatively, he may have been referring specifically to the other subsections of § 3599, which concern the appointing of counsel to defendants charged with crime punishable by death—a decision that has been held to be reviewable by the Court.⁴⁶46.See, e.g., Martel v. Clair, 565 U.S. 648, 652 (2012) (reviewing what standard courts should apply when determining whether appointed counsel should be replaced under § 3599(e)).

Regardless of which particular part of the Code Justice Breyer had in mind, it is reasonably clear that he was considering an argument based on the placement of § 3599(f) in the United States Code. As such, the argument must be reconciled with the codifier’s canon to be legitimate. The next Section explores the statutory history of § 3599(f), which confirms the legitimacy of Justice Breyer’s argument from codification and strongly suggests that Texas’s jurisdictional argument should not prevail.

B. The Codifier’s Canon Applied: An Argument from Codification

As reflected in Justice Breyer’s comments during oral arguments, a provision’s codification is frequently looked at as evidence of congressional intent. A key lesson for applying the codifier’s canon, however, is that an argument from codification must look to statutory history. Here, statutory history not only confirms the legitimacy of the argument from placement, but also provide further evidence to refute Texas’s argument that decisions to deny funding are made in an administrative capacity.

The codifier’s canon allows the statutory provision at issue in Ayestas—18 U.S.C. § 3599(f)—to be relied upon by interpreters. As outlined in Part I, the relevant question is whether the provision was added after the positive law codification of the title. Title 18 was enacted as a positive law title in 1948.⁴⁷47.Act of June 25, 1948, Pub. L. No. 80-772, § 20, 62 Stat. 683, 862 (codified at 18 U.S.C. (2012)). The USA Patriot Improvement and Reauthorization Act, which first added § 3599(f) to Title 18, was enacted in 2006.⁴⁸48.USA Patriot Improvement and Reauthorization Act, Pub. L. No. 109-177, § 222, 120 Stat. 192, 232 (2006) (codified at 18 U.S.C. § 3599(f)). Thus, the codifier’s canon permits arguments based on the provision’s placement in the Code. Justice Breyer’s line of argument can go forward. But the fact that the provision’s placement can be legitimately invoked without running afoul of the codifier’s canon does not imply that it is necessarily persuasive.

The strongest argument from placement—and the one that Justice Breyer likely had in mind—is that § 3599(f) appears in a section of Title 18 authorizing judges to appoint counsel in habeas cases, a role that has already been implicitly held to be judicial. As alluded to above, in addition to authorizing funds for investigators and experts, § 3599 directs district judges to appoint counsel to indigent defendants for habeas proceedings and provides that the appointed counsel may be “replaced . . . upon motion of the defendant.”⁴⁹49.18 U.S.C. § 3599(e). While no standard is provided in the statute for evaluating those motions, the Court, in Martel v. Clair,⁵⁰50.565 U.S. 648, 652 (2012); see also Christeson v. Roper, 135 S. Ct. 891 (2017) (reviewing a district court’s decision made pursuant to 18 U.S.C. § 3599(e)). held that it should be decided “in the interests of justice,” adopting the same standard utilized by such motions in non-capital cases under 18 U.S.C. § 3006A. By taking the appeal in Clair and judging the district court for abuse of discretion, the Court implied that decisions regarding the appointing of counsel are undertaken in the judge’s judicial capacity. Such a determination is highly reasonable, since a judgement about the propriety of replacing counsel, for example, requires making judgements based on many of the same facts and laws that appear before the judge as part of the core case.

The subsection of interest in Ayestas, § 3599(f), was added to Title 18 by the same statute and as part of the same section of the Code as § 3599(e), the provision at issue in Martel and, more recently, Christeson.⁵¹51.135 S. Ct. 891 (2017). Nothing on the face of the statute seems to suggest that Congress intended for a judge to exercise her authority under § 3955(f) in a capacity different from the capacity in which she is authorized under § 3955(e). In fact, the only plausible distinguishing factor is the somewhat unusual review process for fees and expenses under § 3955(f). In contrast, nothing is specified with regard to appeals for motions under § 3955(e). But it seems farfetched to rely upon that as evidence the Congress intended determinations of whether to award fees and expenses to be administrative in nature.

This is particularly true given § 3955’s statutory history. Although first added to Title 18 in 2006, the statutory scheme had been in force since 1988, when it was enacted as part of the Anti-Drug Abuse Act of 1988.⁵²52.Pub. L. No. 100-690, § 7001, 102 Stat. 4181, 4393-94 (1988). The statute, which directed the section be added to 21 U.S.C. § 848, enacted a section essentially identical to the modern 18 U.S.C. § 3599, but with one major difference—no special review procedure is described. Thus, there is nothing in the original version of the provision to suggest that Congress intended for § 3599(f) to define an administrative duty. Unless one were to concede to an argument from constitutional avoidance, the statutory argument for finding no jurisdiction to review § 3599(f) claims is weak.

Nor is the weight of the constitutional questions particularly significant. Review of fees and expenses above $7,500 by the chief judge is not review by “an entity not wielding judicial power.” A chief judge can wield judicial power even when acting alone, much as the Justices do, for example, in the context of their circuit assignments.⁵³53.See Sandra Day O’Connor, Foreword: The Changing Role of the Circuit Justice, 17 U. Tol. L. Rev. 521, 523-24 (1985). Congress is free to “ordain and establish” the lower courts in any manner that it wishes, making a single judge appellate panel prima facie unproblematic.⁵⁴54.U.S. Const. art. III, § 1. Further, there is nothing inherently nonjudicial about a limited ex parte proceeding. A temporary restraining order, for example, is issued in a judge’s judicial capacity⁵⁵55.See, e.g., Ellakkany v. Common Pleas Court of Montgomery Cty., 658 F. App’x 25, 28 (3d Cir. 2016) (holding that absolute immunity applies to a decision not to issue a temporary restraining order, since such decisions fall within a judge’s judicial capacity). and yet is often done in an ex parte manner.⁵⁶56.Fed. R. Civ. P. 65(b)(1). While the Court has held that parties to a case must be adversarial to satisfy Article III,⁵⁷57.See, e.g., Franks v. Bowman Transp. Co., 424 U.S. 747, 754-55 (1976). this requirement is met even if one of the parties does not participate in some particular part of the proceeding.⁵⁸58.Courts regularly utilize “judicial power,” id., in contexts in which only one of the parties to a controversy participate, either because one of the parties are excluded, like in an ex parte proceeding, or because one of the parties fails to appear. Since the State, in this case through the director of Texas’s criminal justice system, regularly contests funding decisions, there are clearly adversarial parties for the purposes of Article III. Since neither of the potential routes toward constitutional avoidance are plausible, there is little justification for viewing a determination of whether a defendant is entitled to fees and expenses to cover investigators and experts to be anything other than a judicial duty. Thus, congressional intent, as expressed through the codification history, ought to be adhered to and Texas’s jurisdictional argument should not succeed.

Conclusion

The codifier’s canon, enacted into law and appearing in Title 18, directs judges to draw interpretive inferences only from placement and captioning decisions that originate with Congress and not the office of the codifiers. Since the decision to place § 3599(f) in a section of the Code authorizing other judicial duties was made by Congress, Justice Breyer was right to turn to that provision’s organizational context as evidence of legislative intent. In fact, considering the provision’s placement, in light of the statutory history, Texas’s jurisdictional argument is not persuasive.

Daniel B. Listwa is Yale Law School 2019. I would like to thank Professor Jerry Mashaw for his insights on an early draft; Charlie Seidell for alerting me to this oral argument; and the editors of the Yale Journal on Regulation, especially Rachel Cheong and John Brinkerhoff, for their help developing this piece.