THE SCOPE OF UNITED STATES MAGISTRATE JUDGE AUTHORITY AFTER STERN V. MARSHALL

THE SCOPE OF UNITED STATES MAGISTRATE JUDGE AUTHORITY AFTER STERN V. MARSHALL

This Note examines the impact of Stern v. Marshall—the Supreme Court’s recent decision on the authority of bankruptcy judges—on United States magistrate judges, with a particular focus on two exercises of magistrate judge authority that have been called into question by circuit courts post-Stern. The Note argues that institutional differences between magistrate judges and bankruptcy judges should lead circuit courts to be more permissive of the delegation of tasks by district judges to magistrate judges than the circuits are of expansive bankruptcy judge authority.

Introduction

United States magistrate judges disposed of 1,102,396 matters from October 2013 through September 2014. 1 Admin. Office of the U.S. Courts, Judicial Business 2014, at 1 tbl.S-17, (Sept. 30, 2014) [hereinafter Judicial Business 2014], http://www.uscourts.gov/uscourts/Statistics/
JudicialBusiness/2014/tables/S17Sep14.pdf (on file with the Columbia Law Review).
These matters were spread across 551 full- and part-time magistrate judgeships 2 How the Federal Courts Are Organized: Federal Judges and How They Get Appointed, Fed. Judicial Ctr., http://www.fjc.gov/federal/courts.nsf/page/183 [http://
perma.cc/W3VU-FJJB] (last visited Jan. 29, 2016). As of July 1, 2014, there were 531 full-time magistrate judgeships and thirty-nine part-time judgeships. Peter G. McCabe, A Guide to the Federal Magistrate Judge System, Fed. Bar Ass’n 7 n.4 (Aug. 2014) [here­inafter McCabe, Guide], http://www.fedbar.org/PDFs/A-Guide-to-the-Federal-Magistrate-Judge-System.aspx?FT=.pdf [http://perma.cc/P8BG-A76D].
and included 346,318 preliminary proceedings in federal felony actions, trials of 106,654 Class A misdemeanors, and assorted petty offenses. 3 Judicial Business 2014, supra note 1, at 2. The sheer size of these dockets reflects the success federal magistrate judges have had in fulfilling the promise of the Federal Magistrates Act of 1968 (the Act) 4 Pub. L. No. 90-578, 82 Stat. 1107 (codified as amended at 28 U.S.C. §§ 631–639 (2012)). : “to cull from the ever-growing workload of the U.S. district courts matters that are more desirably performed by a lower tier of judicial officers.” 5 S. Rep. No. 90-371, at 9 (1967).

Neither the Federal Magistrates Act of 1968 nor any of its sub­sequent amending acts mandates the assignment of any duties to magistrate judges. 6 See McCabe, Guide, supra note 2, at 23 (describing “flexibility” of Act). The Act is only permissive: The duties it describes may be delegated to a magistrate judge by a district judge in an individual case or in every case of a particular type, 7 For example, the Southern District of New York (S.D.N.Y.) established a “Magistrates Court,” on which each magistrate judge in the District sits five times a year. See infra Appendix A.1. The judges assigned to sit on Magistrates Court sign all arrest warrants, advise all arrested persons of their rights, and set bail. See infra Appendix A.1. Within the District, this practice is not flexible and district judges have essentially forfeited the authority to sign arrest warrants and set bail for any criminal defendants. Of course, this may be different in other districts: No portion of the magistrate judge system requires the S.D.N.Y. to handle preliminary criminal matters in this way. leaving the decision up to individual district courts and district judges. 8 See McCabe, Guide, supra note 2, at 23 (noting Act “lets each District Court determine what duties are most needed in light of local conditions and changing caseloads”). While there are some powers exercised by magistrate judges in functionally every district, 9 See Fed. R. Crim. P. 41(b) (describing magistrate judge authority to issue search warrants). In several places, the Federal Rules of Criminal Procedure uses the term “magistrate judge” in assigning roles in the criminal procedure context. See infra note 96 (cataloging rules). Rule 1(c) states that “[w]hen these rules authorize a magistrate judge to act, any other federal judge may also act.” Fed. R. Crim. P. 1(c). there is “substantial disparity in usage” of magistrate judges across districts. 10 McCabe, Guide, supra note 2, at 23. The Magistrate Judges Division of the Administrative Office of the United States Courts distributes an “inventory” of magistrate judge duties to each district, describing the positions of the Supreme Court and the various circuits on magistrate judge authority over a wide array of issues. Magistrate Judges Div., Admin. Office of the U.S. Courts, Inventory of United States Magistrate Judge Duties, at ii (4th ed. 2012) [hereinafter Inventory].

The flexibility of the magistrate judge system is seen as one of its principal strengths, 11 See McCabe, Guide, supra note 2, at 23 (“A particular genius of the Federal Magistrates Act is that it does not mandate the assignment of particular duties to Magistrate Judges.”). but this flexibility is not unlimited. This Note con­siders the statutory and constitutional limitations on a magistrate judge’s exercise of authority generally and specifically the exercise of “additional duties” as authorized by the Federal Magistrates Act of 1968 and sub­sequent amendments. 12 See infra note 32 (discussing move of additional-duties clause in 1976 amendments to 1968 Act). While there are some meaningful statutory restrictions, the primary legal challenge to the exercise of particular additional duties is most often that the duty the magistrate judge is attempting to execute is one that is reserved for Article III judges, 13 As the term is used in this Note, Article III judges are those appointed by the President and confirmed by the Senate, including most prominently federal district and circuit court judges, associate justices of the U.S. Supreme Court, and the Chief Justice of the United States. See infra note 16 (discussing requirements of Article III). In general, this Note discusses the contrast between responsibilities undertaken by magistrate judges, who are not selected by a process that involves either Congress or the President, and district judges, who must be selected by the President and confirmed by the Senate. which magistrate judges are not. This Note evaluates the extent to which the Supreme Court’s much-discussed 2011 decision in Stern v. Marshall, 14 131 S. Ct. 2594 (2011). which held it unconstitutional for bankruptcy judges to decide certain state-law counterclaims, should lead to a re-evaluation of the source and scope of magistrate judge authority in relation to Article III. This Note assesses whether such a re-evaluation is already underway in several circuit courts of appeals. This Note makes a salient analytical contri­bution by distinguishing magistrate judges from bankruptcy judges post-Stern and demonstrates why magistrate judge authority should be less troubling under Article III.

Part I of this Note provides background information on the early history of the federal magistrate judge system, its development over the decades following its genesis in 1968, and the role of magistrate judges today. Part I also discusses the additional duty with the most well deve­loped Supreme Court jurisprudence, the authority of magistrate judges to preside over jury selection in a federal felony case. The analytical frame­work developed in the cases on that power is important in assessing the role of magistrate judges more generally. Part II describes Stern v. Marshall and two recent decisions from the Seventh and Eleventh Circuits, which concern particular exercises of the additional-duties power by magistrate judges. This Part reaches some preliminary con­clusions about the considerations the Court deemed important in asses­sing the appropriateness of an exercise of additional-duties authority in the jury selection cases described in Part I and re-evaluates these con­siderations in light of Stern. Part III proposes a framework that should be used to evaluate whether a particular additional duty meets statutory and constitutional muster, with a focus on ensuring the flexibility of the magistrate judge system and the supremacy of Article III district court judges over magistrate judges. This framework is then applied to the duties considered by the Seventh and Eleventh Circuits, rejecting argu­ments to extend the reasoning of Stern to magistrate judges. This Note argues that it would be a mistake to conclude that Stern should lead to a significant re-examination of the permissible duties of magistrate judges.

In elaborating on the work and duties of federal magistrate judges, this Note relies on both traditional sources of legal authority, including Supreme Court precedent and legal scholarship, and five interviews the author conducted with magistrate judges in various circuits. These interviews provided further detail and context for their work, primarily by illustrating the degree of differentiation among the district courts in the use and powers of magistrate judges. In addition to being generally informative, those interviews provided important detail on the relation­ship between district judges and magistrate judges, a significant topic in this Note. A full summary of the interviews is included in the Appendix.

The additional-duties category of the Federal Magistrates Act was intended to allow district courts to “remain free to experiment” in the duties assigned to magistrate judges because this experimentation would increase “time available to judges for the careful and unhurried perfor­mance of their vital and traditional adjudicatory duties.” 15 S. Rep. No. 94-625, at 10–11 (1976). In light of the clear intent of the Federal Magistrates Act and the importance of the system in allowing district judges to prioritize their “traditional adjudica­tory duties,” this Note argues appellate courts should defer to the judg­ment of district courts on whether a particular exercise of magistrate judge authority is properly an “additional duty” in all but the most egregious circumstances.

I. Background

This Part provides background information important to this Note, including the historical development of the magistrate judge system and the doctrinal principles that provide the framework for assessing the role and responsibilities of magistrate judges. Section I.A chronicles the history of adjuncts to federal judges, from the early system of U.S. com­missioners to the magistrate judge system as it stood in 1979. Section I.B describes in detail the Supreme Court’s precedent on the authority of magistrate judges to select juries in felony trials, an issue that has come before the Court three times in the last few decades. Section I.C provides a profile of the magistrate judge system as it exists today, and section I.D sum­marizes the Article III doctrine that underlies any discussion of magis­trate judge responsibilities.

A. U.S. Commissioner System and Early Development of Magistrate System

The federal judiciary has employed adjudicators without Article III protections 16 Article III provides that “[j]udges . . . of the supreme and inferior Courts, shall hold their Offices during good Behaviour, and shall . . . receive for their Services, a Compensation, which shall not be diminished during their Continuance in Office.” U.S. Const. art. III, § 1. Neither magistrate judges nor their antecedent judicial officers have had these specific protections against dismissal and salary reduction. Today, magistrate judges are appointed to renewable eight-year terms by a specific district court, while district judges must be confirmed by the Senate and are given life tenure and irreducible salary. McCabe, Guide, supra note 2, at 7. or responsibilities since the formation of the Republic. 17 See McCabe, Guide, supra note 2, at 8 (“The antecedents of the federal Magistrate Judge program date back to the early days of the republic . . . .”). A 1793 act amending the Judiciary Act of 1789 included bail provisions for criminals brought before a federal judge. 18 Act of March 2, 1793, ch. 22, § 4, 1 Stat. 333. The 1793 act provided that bail may be taken by a variety of judicial officers and “by any person having authority from a circuit court . . . which authority . . . any circuit court . . . may give to one or more discreet persons learned in the law.” 19 Id. This was the first statutory delegation of authority from federal judges to non-Article III officials and these “learned” persons were the precursors to U.S. commissioners. 20 Joseph F. Spaniol, Jr., The Federal Magistrates Act: History and Development, 1974 Ariz. St. L.J. 565, 566. The learned individuals were named commissioners by act of Congress in 1817, and these positions evolved with the federal judiciary when the district court system developed. Id. The commissioner system was first formalized by act of Congress in 1896, and commissioners’ general duties were to issue arrest and search warrants and to administer oaths. 21 See McCabe, Guide, supra note 2, at 8 (describing early system and 1896 commissioner statute). The 1896 act gave commissioners the formal title “United States commissioners” and created a uniform fee system for compensation. Spaniol, supra note 20, at 566. Commissioners slowly developed the authority to handle petty offenses “on property under the exclusive or concurrent jurisdiction of the federal govern­ment,” meaning military bases, designated federal territories, national parks, and roads. 22 McCabe, Guide, supra note 2, at 9. This particular authority developed out of the park commissioner system: Congress created special commissioner positions for several national parks, with the first placed in Yellowstone National Park in 1894. Id. The park commissioners were tasked with handling petty offenses in federal territories and national parks and on federal roads. Id. In 1940, Congress authorized all United States com­missioners, with special designation by their district court, to try petty offenses occurring on any property over which the federal government had exclusive jurisdiction, eliminating the special “park commissioners.” Id.

The commissioner system had significant shortcomings. There was “a great deal of confusion . . . about the procedures and purpose” of preliminary hearings presided over by commissioners, 23 Federal Magistrates Act: Hearing on S. 3475 Before the Subcomm. on Improvements in Judicial Mach. of the S. Comm. of the Judiciary, 89th Cong. 2 (1967) [hereinafter Judicial Machinery Hearing]. and the narrow scope of commissioner trial jurisdiction burdened the district courts with “minor criminal matters.” 24 Id. at 3. As early as 1942, the Administrative Office of the U.S. Courts recommended to the Judicial Conference that all commissioners be allowed to conduct the trial of certain misdemeanors regardless of where they occurred, but these earlier proposals went nowhere. McCabe, Guide, supra note 2, at 9. The commissioners’ general trial jurisdiction was never expanded. See id. (discussing park commissioners and recommendations to extend misdemeanor trial authority to all commissioners). Despite the “great import” of the commis­sioners’ work and the important legal questions commissioners handled, in July 1966, “some 30 percent of the over 700 U.S. com­missioners” were not lawyers and lacked formal legal training. 25 Judicial Machinery Hearing, supra note 23, at 6. For these and other reasons, 26 Commissioners were not salaried, but were rather compensated on a fee schedule that provided set amounts to the commissioners for each action they took, such as drawing a bail bond, issuing arrest and search warrants, and administering oaths. See McCabe, Guide, supra note 2, at 8 (describing compensation). This system led many to conclude that the commissioners were undercompensated and that it was difficult to attract qualified people for the job. See, e.g., Judicial Machinery Hearing, supra note 23, at 6–7 (discussing problem of low compensation). The narrow scope of the role, lack of formalized training for many commissioners, and low compensation were not the only defects in the system. See Peter G. McCabe, The Federal Magistrate Act of 1979, 16 Harv. J. on Legis. 343, 347 (1979) [hereinafter McCabe, Federal Magistrate Act] (listing seven faults in commissioner system cited by reformers). by the 1960s, the commissioner system was in significant need of reform, and the Administrative Office drafted legis­lation to create a completely new commissioner system, which the Senate eventually transformed into the federal magistrates system, 27 See McCabe, Guide, supra note 2, at 9–10 (discussing proposed reforms to commissioner system). designed to replace the commissioners with U.S. Magistrates. 28 Federal Magistrates Act of 1968, Pub. L. No. 90-578, 82 Stat. 1107 (codified as amended at 28 U.S.C. §§ 631–639 (2012)).

The system created by the Federal Magistrates Act of 1968 corrected several of the procedural shortcomings of the commissioner system 29 To be a federal magistrate judge, one is required to be “a member in good standing of the bar of the highest court of a State,” and magistrate judges are compensated at ninety-two percent of the salary of a district court judge. 28 U.S.C. §§ 631(b)(1), 634(a). and featured significantly expanded jurisdiction for magistrates compared to commissioners. 30 See McCabe, Federal Magistrate Act, supra note 26, at 349 (identifying “three basic categories” of magistrate duties). Under the 1968 Act, magistrates had all the powers and duties of commissioners as well as two important additions: (1) authority to try and dispose of “minor criminal offenses” and (2) authority under the additional-duties clause. 31 See id. at 349–50 (listing as additional duties “service as a special master,” “conduct of pretrial and discovery,” and “preliminary review of . . . habeas corpus petitions”). This Note is concerned with the contem­porary scope of the additional-duties clause, today codified at 28 U.S.C. § 636(b)(3). 32 The additional-duties clause, included in the 1968 Act, was relocated to its own section by amendment in 1976, in part to “emphasize[]” that the additional duties are “not restricted in any way by any other specific grant of authority to magistrates.” S. Rep. No. 94-625, at 10 (1976).

While it was successful in remedying several problems of the com­missioner system, the 1968 Act was not without its flaws. In particular, various courts of appeals split on whether district judges could delegate to magistrates the authority to preside over evidentiary hearings in habeas corpus actions. 33 McCabe, Federal Magistrate Act, supra note 26, at 352. The Supreme Court resolved this split in Wingo v. Wedding, holding that the additional-duties provision of the 1968 Act did not allow magistrates to preside over such hearings. 34 418 U.S. 461, 472 (1974) (“[A]lthough § 636(b) provides that ‘additional duties authorized by rule may include, but are not restricted to,’ duties defined in subsection (b)(3), the legislative history of the subsection compels the conclusion that Congress made a deliberate choice to preclude district courts from assigning magistrates the duty to hold evidentiary hearings.” (quoting 28 U.S.C. § 636(b))). Congress res­ponded by expressly granting magistrates this contested power by act in 1976. 35 Act of Oct. 21, 1976, Pub. L. No. 94-577, 90 Stat. 2729 (codified as amended at 28 U.S.C. § 636(b)). Earlier in 1976, the Court ruled that magistrates could conduct preliminary review and oral argument and make a recommendation for a decision on an administrative appeal under the Social Security Act, notwithstanding the objection of a party (in this case, the government). Mathews v. Weber, 423 U.S. 261, 271–72 (1976). The Court held that this function could properly be delegated to a magistrate under the additional-duties language. Id. The ruling of this case was not addressed by the 1976 Act. In addition to overriding Wingo, this 1976 law “affirmed the broad range of duties” magistrates had been performing under the aus­pices of the additional-duties clause of the 1968 Act and moved the clause into its own section. 36 McCabe, Federal Magistrate Act, supra note 26, at 354; see also supra note 32 (noting statutory relocation of additional-duties clause to emphasize its broad scope). One commentator, writing shortly after enactment of the 1976 Act, concluded that the amendments “placed the jurisdiction of magistrates on a much firmer and more uniform basis nationally.” 37 McCabe, Federal Magistrate Act, supra note 26, at 354.

Congress further expanded the jurisdiction of magistrates just three years later, in the Federal Magistrates Act of 1979. 38 Pub. L. No. 96-82, 93 Stat. 643 (codified as amended at 28 U.S.C. § 636). Among several amendments, the 1979 Act added a new subsection to 28 U.S.C. § 636 empowering magistrates to preside over civil trials with the consent of the parties. 39 Id. at 643–44. As written in the 1979 Act, § 636(c) empowers magistrates to, “[u]pon the consent of the parties, . . . conduct any or all proceedings in a jury or nonjury civil matter and order the entirety of judgment in the case.” 40 Id. at 643. This provision occupies a significant portion of the workload of magistrate judges today, with one former magistrate judge reporting in an interview that he tried twenty-five jury trials pursuant to § 636(c) during his four-year tenure. 41 See infra Appendix A.2; see also Admin. Office of the U.S. Courts, Judicial Business 2013, tbl.4.12 (Sept. 30, 2013), http://www.uscourts.gov/file/13204/download (on file with the Columbia Law Review) (presenting data on civil consent cases terminated by magistrate judges: 15,049 total in fiscal year 2012, including 360 jury trials).

The 1979 Act represents the last substantive change Congress has made to the authority of magistrate judges, but in 1990, and after much debate, the title of “United States Magistrate” was changed to “United States Magistrate Judge.” 42 Judicial Improvements Act of 1990, Pub. L. No. 101-650, 104 Stat. 5089 (codified at 28 U.S.C. § 482); see also McCabe, Guide, supra note 2, at 14 (describing motivation for name change, debate, and other proposed titles). Before this change, some districts did not address magistrates as “judge,” contributing to a general lack of appreci­ation for the system and the magistrates. 43 See McCabe, Guide, supra note 2, at 19 (describing “early days” of magistrate system). The addition of “judge” to the title has had tangible impacts: Peter McCabe’s guide to the federal magis­trate system for the Federal Bar Association claims that the lack of the title “judge” contributed to significant uncertainty about magistrates’ role in the early years of the system. 44 Id. McCabe also identifies three factors that have led to the “high quality” of magistrate judges today: “a better salary, a sound judicial retirement system, and addition of the title ‘judge.’” 45 Id. at 20.

With the exception of the retitling, the 1979 Act’s addition of civil consent authority is the most recent example of significant congressional action on magistrate judges. Since then, any change or development in the jurisdiction or authority of magistrate judges has evolved out of a combination of publications from the Judicial Conference of the United States and the Administrative Office of the U.S. Courts, 46 See Magistrate Judgeships, History of the Federal Judiciary, Fed. Judicial Ctr., http://www.fjc.gov/history/home.nsf/page/judges_magistrate.html [http://perma.cc/
6VQM-FGVP] (last visited Jan. 29, 2016) (noting Administrative Office administers magistrate judge system and Judicial Conference determines number of magistrate judgeships and authorizes new positions). The most important publication pertaining to magistrate judges put out by the Administrative Office is the Inventory of United States Magistrate Judge Duties. See generally Inventory, supra note 10. The Inventory is not binding, but according to one magistrate judge, it is the first document magistrate judges refer to when confronted with an uncertain application of authority. See infra Appendix D. The Administrative Office also publishes Suggestions for Utilization of Magistrate Judges, which “offer the courts a set of ‘lessons learned’ on the most effective and efficient way to delegate duties” to magistrate judges. McCabe, Guide, supra note 2, at 24–25.
decisions of the Supreme Court and various courts of appeals, 47 See, e.g., infra notes 53–87 and accompanying text (discussing Supreme Court jurisprudence on magistrate judge authority to preside over felony jury selection). and the practices of individual districts and district judges. 48 The Administrative Office’s Suggestions for Utilization of Magistrate Judges encourages district courts to make court-wide decisions on magistrate judge utilization. See McCabe, Guide, supra note 2, at 25. (describing recommendations from Administrative Office). Conversations with several federal magistrate judges confirm that many decisions about how magistrate judges are used are made at a district level, but individual district judges vary in what tasks and how frequently they delegate to magistrate judges. See infra Appendix B (discussing historical development of magistrate role within districts). In general, there is “substantial disparity in usage of Magistrate Judges among the courts.” McCabe, Guide, supra note 2, at 23. The importance of this last category of development cannot be overstated: Conversations with magis­trate judges confirmed that different districts structure the assign­ment of duties to magistrate judges in significantly different ways based on the preferences of the Article III judges in the district. 49 See infra Appendix B (noting districts develop portfolio of magistrate responsibilities based on historical practice and district needs). Even where district-wide policies dictate the assignment of many specific duties to magistrate judges, district judges within a single district can still exercise discretion in what responsibilities they assign to magistrate judges, par­ticularly under the additional-duties clause, as most exercises of authority under that clause occur at the direction of the district judge. 50 See infra Appendix A (noting one magistrate judge stated additional duties were “whatever came along”).

B. Post-1979 Dispute Over Magistrate Judge Jurisdiction: Selection of Felony Juries

Between 1979 and 1989, there was little further development in the authority of magistrates. 51 In 1980, the Court affirmed the constitutionality of a provision of the Federal Magistrates Act that allowed district courts to refer motions to suppress evidence to magistrates and to decide such motions based on the record before the magistrate. United States v. Raddatz, 447 U.S. 667, 683 (1980) (holding “delegation does not violate Art. III so long as the ultimate decision is made by the district court”). This delegation was held proper despite the objections of the defendant. Id. at 669. While much modern scholar­ship on the scope of magistrate judge authority focuses on the importance of litigant consent, see infra note 82, none of the early cases on the matter, including Raddatz, Wingo, and Weber, considered consent important at all. Raddatz is most important for this Note because of reasoning found in Justice Blackmun’s concurrence. See infra note 80 and accompanying text (describing Court’s later use of Justice Blackmun’s Raddatz concurrence). By this time, the scope of the general authority of magistrate judges to handle both civil and criminal matters was relatively well established, but in 1989 the legitimacy of one particular facet of magistrate authority, the power to conduct jury selection in felony cases, reached the Supreme Court and ultimately resulted in two important decisions. These decisions spawned most of the modern com­mentary on the authority of magistrate judges, and it is for that reason they are discussed in this Note.

1. Gomez v. United States and the Early Jurisprudence on Magistrates and Jury Selection. — When the jurisdiction of magistrates returned squarely to the attention of the Supreme Court in 1989, the issue centered, as it had in Wingo v. Wedding, 52 418 U.S. 461 (1974); see also supra note 34 (quoting from Wingo). on the additional-duties language of the Federal Magistrates Act. In Gomez v. United States, the substantive issue was whether magistrates could conduct jury selection in a felony case without the defendant’s consent under the additional-duties clause. 53 490 U.S. 858, 861–62 (1989). This case was decided just before the retitling of magistrates to magistrate judges, see supra note 42, and for that reason, the judges are referred to as “magistrates” in this section. The Court avoided the constitutional question of whether this exercise of magistrate authority would contravene Article III 54 See Gomez, 490 U.S. at 864 (declining to discuss whether exercise of magistrate judge’s authority contravened Article III on grounds of constitutional avoidance). The constitutional avoidance canon provided the Court with justification to require exceedingly clear legislative intent that Congress meant the Federal Magistrates Act to give magistrates the authority to preside over felony jury selection. See id. (noting aim of Court is to “deduce[]” “alternative interpretation of the additional duties clause” which “poses no constitutional question”). and instead engaged in a lengthy statutory analysis of the Federal Magistrates Act and its 1976 and 1979 amendments. The central holding of Gomez is that “additional duties . . . should bear some relation to the specified duties” laid out in other sections of the Act. 55 Id. The Court then concluded that Congress did not intend to include felony jury selection “among a magistrate’s additional duties” because felony jury selection did not bear a sufficiently close rela­tionship to the specified duties in the Act. 56 Id. at 872.

In Gomez, the Court made reference to the fact that the defendants had objected to the magistrate presiding over jury selection, but the case’s holding is not expressly limited to situations in which the defendant does not consent to the magistrate exercising this authority. 57 The holding of Gomez is based on the Court’s interpretation of the structure of the Federal Magistrates Act itself and its conclusion that jury selection is not an “additional duty” in Congress’s cognizance. Id. at 875–76. The Court found that “the carefully defined grant of authority [to magis­trates] to conduct trials of civil matters and of minor criminal cases should be construed as an implicit withholding of the authority to preside at a felony trial.” 58 Id. at 872. This quote provides an early example of what became a common and important grouping of magistrate authority to enter sentences for misdemeanors and the separate authority to preside over civil trials. 59 See infra text accompanying notes 88–89 (noting “functionalist identification of common issues between [these] two grants of authority”). The Court read its own precedent and the Speedy Trial Act to conclude that “Congress . . . considers jury selection part of a felony trial.” 60 Gomez, 490 U.S. at 873. The Court also noted that Congress did not classify juror voir dire as either a “dispositive” or “nondispositive” pretrial matter when they did so for various enumerated matters in the 1976 amendments. The Court took this to be an implied statement by Congress that it never considered presiding over voir dire to be within the aegis of a magistrate’s authority. Id. at 873–74. Because of the aforementioned “implicit withholding of the authority to preside at a felony trial” 61 Id. at 872. and its conclusion that jury selection was part of the felony trial, the Court held that the Federal Magistrates Act did not authorize a magistrate to preside over jury selec­tion in a felony trial. 62 See id. at 875–76 (“The absence of a specific reference to jury selection in the statute, or indeed, in the legislative history, persuades us that Congress did not intend the additional duties clause to embrace this function.”). The Supreme Court and all the circuit courts of appeals that have faced this issue have concluded that a magistrate judge cannot preside over a felony trial, even with the express consent of both the government and the defendant. 63 See, e.g., id. at 872 (“[T]he carefully defined grant of authority to conduct trials of civil matters and of minor criminal cases should be construed as an implicit withholding of the authority to preside at a felony trial.”); United States v. Dees, 125 F.3d 261, 267 (5th Cir. 1997) (contrasting taking of guilty plea with “exclusive Article III power of a district court to preside over a felony trial”). Considering this in light of the Gomez Court’s holding that jury selection was “part” of the felony trial, one interpretation of the breadth of Gomez is that it foreclosed a magis­trate judge from presiding over felony jury selection with or without the consent of the litigants.

2. Peretz v. United States and Limiting Gomez. — Whether and to what extent the holding in Gomez depended on the defendant’s non­consent to have the magistrate judge preside over felony jury selection created significant uncertainty for courts. 64 Shortly after Gomez, three circuits and one state supreme court faced with the question of the constitutionality of the delegation of jury selection to a magistrate all reached slightly different conclusions on the meaning of the case. See Monique Mulcare, Article III, The Federal Magistrate, and the Power of Consent, 1992/1993 Ann. Surv. Am. L. 297, 307–13 (cataloguing responses to Gomez). After a false start, 65 The Court had earlier granted the government’s petition for certiorari in United States v. France, 886 F.2d 223 (9th Cir. 1989), aff’d, 498 U.S. 335 (1991), in which the Ninth Circuit had reversed a conviction by relying on Gomez, but the Court affirmed that decision 4-4 without opinion, 498 U.S. at 335. The published per curiam order does not indicate which justices were in the majority, but Justice Souter, in the eventual majority in Peretz, the case that followed, did not participate in France, indicating that the split in France matched the eventual 5-4 split in Peretz, less Justice Souter. Id. Certiorari in United States v. Peretz, the case discussed in this section, was granted the same day the decision in France was published. France, 498 U.S. at 335 (issuing judgment January 22, 1991); Peretz v. United States, 498 U.S. 1066, 1066 (1991) (granting certiorari). the Court heard United States v. Peretz with two presented questions: (1) Whether the additional-duties clause permits a magistrate judge to preside over jury selection when the defendant consents; and (2) whether, if the statute does provide this authority, the delegation of jury selection auth­ority from a district judge is consistent with Article III. 66 Peretz v. United States, 501 U.S. 923, 927 (1991) (describing procedural history, France, and questions presented).

Gomez had the support of a unanimous Court, but Peretz was a 5-4 split decision. 67 See id. at 924 (noting opinion written by Justice Stevens, joined by Chief Justice Rehnquist and Justices O’Connor, Kennedy, and Souter, with Justice Marshall filing dissenting opinion joined by Justices White and Blackmun and Justice Scalia filing separate dissenting opinion). In a relatively short opinion for the Court, Justice Stevens clouded the two questions presented, illuminating the extent to which the statutory interpretation issue and the Article III question are inter­related. Specifically, the opinion first concluded that the consent of the defendant “significantly changes the constitutional analysis,” elim­inating the Article III issues present in Gomez. 68 Id. at 932. The Court based this determination in part on earlier holdings that litigants can waive their right to have an Article III judge preside over civil trials and various basic rights of criminal defendants. Id. at 936–37; see also infra notes 110–125 and accompanying text (discussing federal courts doctrine of legislative courts). While litigants may not waive the so-called “structural” protections of Article III, the Court concluded that a magistrate judge presiding over jury selection did not threaten these structural protections because of the high degree of control district judges possess over magistrate judges. Peretz, 501 U.S. at 937. Gomez, as discussed above, was decided on the statutory ground that the additional-duties clause of the Federal Magistrates Act did not embrace magistrates presiding over felony jury selection. 69 See Gomez v. United States, 490 U.S. 858, 875–76 (1989); supra text accompanying note 56 (discussing statutory interpretation issue in Gomez). Because the Peretz Court found that the defen­dant’s consent cured the Article III problem in delegating jury selection authority to a magistrate judge, the Court “attach[ed] far less impor­tance . . . to the fact that Congress did not focus on jury selection” in writing the additional-duties clause. 70 Peretz, 501 U.S. at 932. The exact interaction between the holdings of Peretz and Gomez is somewhat confusing because the Peretz Court decided that it would not give weight to the Gomez Court’s statutory argument since the presence of consent had cured a constitutional pro­blem. Exactly how (or whether) the defendant’s consent “cures” the statutory defect with assigning jury selection to magistrate judges is unclear after Peretz. That said, it is at least clear that as a practical matter, Peretz narrowed the holding of Gomez by allowing a magistrate judge to exercise authority that does not “bear some relation to the [Federal Magistrates Act’s] specified duties” 71 Gomez, 490 U.S. at 864. when the judge has the consent of the parties. 72 The Court does not explicitly overrule any aspect of Gomez and instead the majority argues that its decision is simply an application of the rule from that earlier case. Some commentators have argued that Peretz represented a significant departure from Gomez. See, e.g., Mulcare, supra note 64, at 315 (arguing in Peretz, “the Supreme Court reversed itself”).

Peretz strayed furthest from Gomez on the question of what powers Congress intended to delegate to magistrate judges through the additional-duties clause. Peretz described the express delegation of civil and misdemeanor trials to a magistrate judge in the Federal Magistrates Act as “comparable in responsibility and importance” to presiding over felony jury selection. 73 Peretz, 501 U.S. at 933. The Gomez Court made clear that the possible authority of a magistrate to conduct voir dire was categorically different from the magistrate’s authority to try misdemeanors and all civil cases with the consent of the parties. 74 See supra note 60 and accompanying text (noting Gomez considered jury selection a part of the felony trial). This is the crux of Justice Marshall’s dissent in Peretz. See infra note 75 and accompanying text (discussing Justice Marshall’s dissent).

In no uncertain words, Justice Marshall dissented in Peretz, drawing particular attention to the aforementioned inconsistency with Gomez. While the Peretz Court grouped the authority to preside over voir dire with litigant consent with the clearly authorized power to conduct mis­demeanor and jury trials with litigant consent, Justice Marshall’s dissent drew from Gomez to conclude that Congress intended to cate­gorize felony jury selection with felony trials more generally. 75 See Peretz, 501 U.S. at 943 (Marshall, J., dissenting) (“To hold . . . that a magistrate may . . . conduct jury selection in a felony trial so long as the defendant consents is to treat the magistrate’s authority . . . as perfectly coextensive with his authority in civil and misdemeanor trials—the reading of the Act that Gomez categorically rejected.”). This dis­agreement between the Court and Justice Marshall in dissent identified two crucial reference points for exercises of magistrate consent authority under the additional-duties clause: If a particular exercise of consented-to magistrate authority is more similar to the magistrate judge’s power to preside over misdemeanor and civil trials, then the challenged exercise is within the statutory text. On the other hand, if the consent power is similar to, or a constitutive part of a felony trial, it is both a violation of Article III and outside the scope of the Federal Magistrates Act for the magistrate judge to exercise that power, even with litigant consent. The majority and Justice Marshall agreed that these are the two useful “guide­posts” for determining the appropriateness of a particular exercise of magistrate authority, but disagreed on where felony jury selection falls between them, with the majority analogizing felony jury selection to presiding over a misdemeanor or civil trial and Justice Marshall con­cluding (in line with the Court in Gomez) that felony jury selection is part of the felony trial.

The majority opinion and Justice Marshall’s dissent illustrated two additional concerns central to this Note. First, the relevance of litigant consent to magistrate authority under the additional-duties clause has its origins in Peretz. 76 See Inventory, supra note 10, § 7, at 13 (noting Peretz has led many courts to focus on whether the litigants consent to the exercise of a particular additional duty). Under the Court’s decision, a litigant may waive the right to have an Article III judge select a jury. 77 See Peretz, 501 U.S. at 936 (majority opinion) (listing “basic rights of criminal defendants” that are waivable). Justice Marshall disagreed and argued that jury selection is a “structural guarantee” of Article III and a litigant cannot consent to have a magistrate preside over jury selec­tion because the rule serves “‘institutional interests that the parties cannot be expected to protect.’” 78 Id. at 950 (Marshall, J., dissenting) (quoting CFTC v. Schor, 478 U.S. 833, 850–51 (1986)).

The second basis for disagreement on magistrate authority Peretz embodied is the importance (or unimportance) of who has the ultimate decisionmaking authority. Justice Marshall distinguished juror voir dire from other magistrate judge additional duties because there is no judicial review of jury selection and hence, the magistrate judge makes the “ultimate decision” on the issue. 79 Id. at 951. On this point, the Court was satisfied with an argument drawn from Justice Blackmun’s concurrence in an earlier case: “‘magistrate[s] . . . [are] subject to the Art. III judge’s control,’” and therefore there is no “‘threat to the judicial power or the independence of judicial decisionmaking’” when magistrates preside over felony voir dire. 80 Id. at 938 (majority opinion) (quoting United States v. Raddatz, 447 U.S. 667, 685–86 (1980) (Blackmun, J., concurring)). Justice Blackmun, whose Raddatz concurrence proved persuasive to the majority in Peretz, actually joined Justice Marshall’s dissent in Peretz. See id. at 940 (Marshall, J., dissenting). For more on Raddatz and the particular holding of that case, see supra note 51 (discussing Raddatz). This Note returns to the question of whether magistrates are always “subject” to the control of an Article III judge and the importance of this question in evaluating the constitutionality of a particular exercise of magistrate authority. 81 See infra note 128 and accompanying text (considering proximity of magistrate judges to district judges in context of legislative court doctrine); infra notes 233–239 (discussing unique features of magistrate judge–district judge relationship and implications for Article III review of magistrate judge decisions).

3. Summary and Recent Developments. — Gomez and Peretz are the most prominent cases on magistrate judge authority since the passage of the Federal Magistrates Act, and both cases inspired significant commen­tary. 82 For discussion of the “power of consent” in these cases, see, e.g., Mulcare, supra note 64, at 316 (proposing amendment to Federal Magistrates Act to require defendant provide written consent before magistrate may preside over voir dire); Kimberly Anne Huffman, Note, Peretz v. United States: Magistrates Perform Felony Voir Dire, 70 N.C. L. Rev. 1334, 1357 (1992) (arguing Peretz “dismantled . . . notions of consent, waiver, and Article III structural protections articulated earlier”). For the purposes of this Note, it is useful to understand the importance of three factors weighed by the Court in Gomez and Peretz: (1) whether litigants consent to magistrate authority; (2) whether the magis­trate judge is the “final arbiter” of the matter before her; and (3) whether the matter at issue is more analogous to presiding over civil and misde­meanor trials with litigant consent or to conducting a felony trial.

These three components of Gomez and Peretz were again the focus of a recent Supreme Court case on magistrate judge authority to conduct juror voir dire. Gonzalez v. United States extends Peretz, holding that the consent to have a magistrate judge preside over jury selection required by Peretz may be given by counsel and not directly by the defendant. 83 Gonzalez v. United States, 553 U.S. 242, 253 (2008) (“[A] magistrate judge may preside over jury examination and jury selection only if the parties, or the attorneys for the parties, consent. Consent from an attorney will suffice.”). The Court did not address whether an attorney’s consent may override timely objections by the defendant, but some language in the opinion may imply that this is the case. See id. (describing demand that Article III judge preside over jury selection as consideration “more significant to the realm of the attorney than to the accused”). Gonzalez does not address whether a defendant or defendant’s counsel may consent to have a magistrate preside over jury selection by implied consent, after holding in an earlier case that implied consent was sufficient for a magistrate judge to preside over a civil trial. See Roell v. Withrow, 538 U.S. 580, 590–91 (2003) (holding implied consent should suffice where “litigant or counsel was made aware of the need for consent and the right to refuse it, and still voluntarily appeared to try the case before the Magistrate Judge”). Gonzalez continues the trend from Gomez and Peretz of recategorizing the authority to preside over jury selection vis-à-vis authority to conduct mis­demeanor and civil trials. Gomez had viewed magistrate authority to preside over felony jury selection as implicating more serious Article III concerns than magistrates handling misdemeanor and civil trials and held that choosing a felony jury was more like presiding over a felony trial than presiding over a misdemeanor or civil trial. 84 See supra note 60 and accompanying text (discussing positioning of case types in Gomez). Peretz, by contrast, saw the Article III issues raised by a magistrate judge handling felony jury selection as roughly equivalent to those raised by a magistrate judge pre­siding over a misdemeanor or civil trial. 85 See supra note 73 and accompanying text (describing categorization of civil and misdemeanor trials and selection of felony jury in Peretz). Gonzalez goes further, finding that presiding over a misdemeanor or civil trial with consent of the litigants raises Article III concerns above and beyond those present in the context of presiding over felony jury selection. 86 See Gonzalez, 553 U.S. at 252 (“[I]t is not obvious that Congress would have thought these matters required the same form of consent.”). The Court reaches this conclusion out of necessity: The petitioner argued that, because magistrates must receive express personal consent by the parties before presiding over civil and misdemeanor trials, this requirement should extend to the more or equally important authority to select a felony jury. Id.

In addition to reorienting the civil and misdemeanor trial authority relative to felony jury selection, Gonzalez touched on the third factor identified in Gomez and Peretz: Whether the magistrate is the final arbiter of the issue. Because the district court would have to consider all object­tions raised during jury selection and because it would not be “difficult or disruptive for a district judge to review” such objections, the Gonzalez Court determined the final decisional authority ultimately remained with the district judge. 87 Id. at 251–52. This position does not consider the question of whether district courts must review objections not raised before the magistrate judge, an issue that had led to a split in the circuits at the time Gonzalez was decided. See Kevin Koller, Note, Deciphering De Novo Determinations: Must District Courts Review Objections Not Raised Before a Magistrate Judge?, 111 Colum. L. Rev. 1557, 1573–91 (2011) (describing split). If district judges were required to hear objections not raised before the magistrate judge, this would strengthen the Court’s position on this point because additional final decisionmaking authority would fall to the district judge.

The issue of magistrate judge authority to preside over felony jury selection demonstrates how the Court tends to conflate magistrate auth­ority over civil and criminal matters. The Court chooses to eschew the formalist distinction between the section of the Federal Magistrates Act that confers power to magistrate judges to enter sentences for mis­demeanors 88 28 U.S.C. § 636(a)(5) (2012) (giving magistrate judges power to “enter a sentence for a class A misdemeanor in a case in which the parties have consented”). and the section authorizing them to preside over civil trials 89 Id. § 636(c)(1) (granting power to conduct civil trials on consent of parties). in favor of a functionalist identification of common issues between the two grants of authority.

C. Magistrate Judge Authority Today

Notwithstanding the Court’s jurisprudence on magistrate judge authority to preside over felony jury selection, the broad outline of magistrate judge duties has remained largely consistent since 1979. 90 See supra notes 38–41 and accompanying text (discussing 1979 Federal Magistrates Act). Then and now, magistrate judge duties fit into four categories: “(1) initial proceedings in criminal cases; (2) criminal misdemeanors; (3) pretrial matters and other proceedings in civil and criminal cases; and (4) civil cases on consent of the parties.” 91 McCabe, Guide, supra note 2, at 23. As discussed above, the civil consent provision was added by statute in 1979, 92 28 U.S.C. § 636(c). and a magistrate judge’s power to try and enter sentences for petty offenses and misdemeanors is also cited elsewhere in § 636. 93 See id. § 636(a)(4)–(5) (conveying to magistrate judges “power to enter a sentence for a petty offense” and “power to enter a sentence for a class A misdemeanor in a case in which the parties have consented”). The additional-duties clause 94 28 U.S.C. § 636(b)(3). authorizes nearly all the magistrate judge work in the third category, pretrial matters, and much of the first category of magistrate judge duties in initial proceedings in criminal cases. Hence, this Note focuses on the role of magistrate judges in handling initial proceedings in criminal cases and pretrial matters in all cases.

There are three main categories of initial proceedings in criminal cases that constitute the bulk of magistrate judges’ work in that area: “(1) issuance of criminal process; (2) conduct of bail and detention pro­ceedings; [and] (3) conduct of arraignments” and pleas. 95 McCabe, Guide, supra note 2, at 26. In addition to these responsibilities, magistrate judges also handle various other preliminary criminal matters, including mental competency and extradition proceedings and international prisoner transfers. Id. at 32–34. Most of these responsibilities are assigned to magistrate judges by the Federal Rules of Criminal Procedure and so tend to be consistent across districts. 96 See, e.g., Fed. R. Crim. P. 3 (“The complaint . . . must be made under oath before a magistrate judge . . . .”); Fed. R. Crim. P. 4(b)(1)(C) (“A warrant must . . . command that the defendant be arrested and brought without unnecessary delay before a magistrate judge . . . .”); Fed. R. Crim. P. 4(c)(4)(A) (“At the request of an attorney for the government, an unexecuted warrant must be brought back to and canceled by a magistrate judge . . . .”); Fed. R. Crim. P. 4.1(a) (“A magistrate judge may consider information communicated by telephone or other reliable electronic means when reviewing a complaint or deciding whether to issue a warrant or summons.”); Fed. R. Crim. P. 5(a)(1)(A) (“A person making an arrest within the United States must take the defendant without unnecessary delay before a magistrate judge . . . .”); Fed. R. Crim. P. 6(f) (“The grand jury . . . must return the indictment to a magistrate judge in open court.”); Fed. R. Crim. P. 41(b)(1) (“At the request of a federal law enforcement officer or an attorney for the government . . . a magistrate judge with authority in the district . . . has authority to issue a warrant to search for and seize a person or property located within the district . . . .”); Fed. R. Crim. P. 41(e)(1) (“The magistrate judge . . . must issue the warrant to an officer authorized to execute it.”); Fed. R. Crim. P. 41(f)(1)(D) (“The officer executing the warrant must promptly return it . . . to the magistrate judge designated on the warrant.”); Fed. R. Crim. P. 58(b)(2) (“At the defendant’s initial appearance on a . . . misdemeanor charge, the magistrate judge must inform the defendant of . . . the charge, . . . the right to retain counsel[,] . . . the . . . right not to make a statement, . . . the right to trial, judgment, and sentencing[,] . . . [and] the right to a jury trial . . . .”). Guilty plea allocution and acceptance in felony cases is one exception to this general consistency in the criminal caseload of magistrate judges and is discussed in detail in Part II. 97 See infra section II.B.2 (discussing split among circuits on whether magistrate judges may accept felony guilty pleas).

The scope of magistrate judge responsibilities in the pretrial context is immense and varies considerably from district to district. 98 See infra Appendix A (summarizing interviews with magistrate judges describing differences across circuits). Most dis­tricts cast magistrate judges as active case managers in both the civil and criminal contexts 99 See McCabe, Guide, supra note 2, at 43–45 (describing case management role of magistrate judges). and task them with handling civil settlement con­ferences. 100 Id. at 45–46. The Federal Magistrates Act makes a distinction between dispositive motions and nondispositive motions, with magistrate judges allowed to handle both but only permitted to enter orders with absolute finality on the latter. 101 See 28 U.S.C. § 636(b)(1)(A) (2012) (distinguishing dispositive from nondispositive motions). When a magistrate judge handles a dispositive motion, she must issue a report and recommendation to the assigned district judge, who is required to make a “de novo determination” on the motion when a party objects, though the district judge is not required to reconduct the motions practice. 102 Id. § 636(b)(1). Pursuant to the additional-duties clause, magistrate judges have historically handled a wide variety of other matters, including special master proceedings, social security appeals, state and federal habeas corpus cases, 103 See infra section II.B.1 (identifying circuit split on authority of magistrate judges to enter final orders in federal habeas cases). voir dire, 104 See supra section I.B (chronicling history of magistrate judge power to supervise voir dire in felony cases). post-trial motions, modification or revocation of supervised release, and naturalization proceedings. 105 See McCabe, Guide, supra note 2, at 49–58 (noting these additional duties).

Regardless of the source of authority for a magistrate judge to engage in some exercise of her authority, the primary objection from critics is rooted in Article III. The next section summarizes common arguments against the exercise of powers typically vested in Article III judges by either administrative courts or magistrate judges and other adjuncts to Article III judges.

D. Magistrate Judges and Article III

Before turning to the recent developments in magistrate judge juris­diction and the permissible powers of non-Article III courts at the focus of this Note, some background is needed on the general limits of non-Article III courts and objections to the exercise of certain powers traditionally reserved for Article III judges by magistrate judges. 106 This Note uses “non-Article III court” and “Article I court” interchangeably even though this is not strictly correct, as Article IV territorial courts are also non-Article III courts. As this Note does not discuss any issues specific to Article IV courts, that term does not appear outside of this explanatory footnote. As used in this Note, “Article I courts” is shorthand for courts or tribunals Congress creates pursuant to its Article I authority. This is in contrast to Article III courts, meaning primarily, though not exclusively, the U.S. district courts and circuit courts of appeals. The plainest statement of why certain exercises of authority by magistrate judges raise Article III concerns is that magistrate judges are not Article III judges. While they are employees of the federal judiciary, typically have chambers and judicial clerks, and carry the title “Judge,” 107 See supra notes 42–43 and accompanying text (discussing retitling of “federal magistrates” to “federal magistrate judges”). magis­trate judges do not possess the two defining characteristics of Article III judges: life tenure and irreducible salary. 108 See supra note 16 (describing Article III protections and lack thereof for magistrate judges).

Article III concerns about magistrate judges should not be described in a vacuum. The legislative court doctrine, which endeavors to cabin Congress’s power to create non-Article III courts generally, 109 Richard B. Saphire & Michael E. Solimine, Shoring Up Article III: Legislative Court Doctrine in the Post CFTC v. Schor Era, 68 B.U. L. Rev. 85, 87 (1988). can shed light on how courts of appeals limit the power of districts to assign responsibilities to magistrate judges. Though the scope of Congress’s power to empower judicial officers without providing them with Article III protections is unclear, 110 See James E. Pfander, Article I Tribunals, Article III Courts, and the Judicial Power of the United States, 118 Harv. L. Rev. 643, 647 (2004) (“Scholars have searched, with mixed success, for an organizing and limiting principle in the somewhat muddled jurisprudence that governs the relationship between Article III courts and Article I tribunals.”). the Supreme Court has developed the legislative court doctrine “to preserve the integrity of [A]rticle III while accommodating Congress’s need for flexibility in the exercise of its enumerated powers.” 111 Saphire & Solimine, supra note 109, at 91. The doctrine developed through three Supreme Court cases in the 1980s, two concerning the authority of bankruptcy judges and one the powers of an administrative law judge. While none of these cases emerged from an exercise of authority by a magistrate judge, the general principles of the cases remain relevant. For purposes of this Note, it is important only to understand that the first Supreme Court case on the legislative court doctrine, Northern Pipeline Construction Co. v. Marathon Pipe Line Co., established a methodological framework for assessing whether a challenged power of a non-Article III court was constitutional and in so doing, severely curtailed Congress’s powers to create Article I courts. 112 458 U.S. 50 (1982); see also Saphire & Solimine, supra note 109, at 99 (discussing Justice Brennan’s framework from Northern Pipeline). Justice Brennan, writing for a plurality of the Court, concluded that non-Article III courts could only decide three categories of cases: “cases adjudicated by military courts, cases adjudi­cated by territorial courts, and those cases which fell within the public rights doctrine.” 113 Saphire & Solimine, supra note 109, at 96. The public-rights doctrine, or public-rights exception, covers cases between the government and citizens that arise out of consti­tutional exercises of power by the executive or legislative branch. 114 See infra note 141–142 and accompanying text (summarizing public rights exception as discussed in Stern v. Marshall).

The plurality opinion held that these categories represent formal distinctions that draw the line between permissible and impermissible delegation of judicial decisions to courts and tribunals deriving their authority from Article I, with no exceptions. 115 See N. Pipeline, 458 U.S. at 63–64 (interpreting prior cases to authorize jurisdiction for non-Article III courts in only “three narrow situations”). It should be obvious that many matters that come before magistrate judges fall outside of those three categories and that any effort to apply this formalist framework to magistrate judges would significantly restrict their authority.

The framework from Northern Pipeline was first eroded 116 Thomas v. Union Carbide Agric. Prods. Co., 473 U.S. 568 (1985), eschewed the formalism of the categories expressed in Northern Pipeline in favor of a pragmatic focus. See Saphire & Solimine, supra note 109, at 102 (noting “avowedly pragmatic inquiry” of majority in Thomas, in contrast to formalism of Northern Pipeline). and finally eliminated in CFTC v. Schor, 117 478 U.S. 833 (1986). whichabandoned what was left of the Northern Pipeline rule” 118 Saphire & Solimine, supra note 109, at 111. by establishing a balancing test in which courts are tasked with “test[ing] the degree to which a grant of judicial power to a non-[A]rticle III court actually impinged upon [A]rticle III values.” 119 Id. at 107. According to some commentators, the modern doctrine provides Congress with the practical authority to replace the entire Article III judiciary with non-Article III tribunals so long as they can articulate “‘evidence of valid and specific legislative necessities.’” 120 Id. at 121 (quoting Schor, 478 U.S. at 855). The vagaries of the legislative court doctrine’s limits on Congress’s ability to create Article I courts may be part of the reason that the majority opinions in Gomez, Peretz, and Gonzalez never explicitly invoke the doctrine. 121 See Peretz v. United States, 501 U.S. 923, 936 (1991) (citing Schor for holding “litigants may waive their personal right to have an Article III judge preside over a civil trial”); Gomez v. United States, 490 U.S. 858, 864 (1989) (citing Schor for use of constitutional avoidance canon). Justice Marshall’s dissent in Peretz devotes several pages to a discussion of Schor and relies on its “institutional interests” language to support his Article III concerns about magistrate authority to lead jury selection. See Peretz, 501 U.S. at 949–52 (Marshall, J., dissenting) (distinguishing Schor from instant case due to unavailability of “de novo judicial review in upholding the performance of core Article III powers by an Article I tribunal” (quoting Schor, 478 U.S. at 851)); supra note 75 and accompanying text (discussing Justice Marshall’s Peretz dissent).

Just before both Gomez and Peretz were decided, Richard B. Saphire and Michael E. Solimine published an analysis of the legislative court doctrine in light of Schor. 122 Saphire & Solimine, supra note 109. Saphire and Solimine view the development of the doctrine as a swing from an “extremely formalistic” conception that left “no congressional power to create legislative courts” to a “prag­matic” approach that gave Congress “(almost) unlimited power” in this field. 123 Id. at 144. Their article argues for a reinterpretation of the legislative court doctrine that focuses on the presence or lack of Article III review as the solution to the overly formalist and overly pragmatist past approaches to defining Congress’s power to create legislative courts. 124 See id. at 144–45 (arguing “[t]he mere threat of [A]rticle III review . . . may deter non-[A]rticle III tribunals from succumbing to political pressure,” “requiring . . . [A]rticle III review would impose costs on the choice to create legislative courts,” but noting this requirement is not sufficient). As discussed above, the Court and Justice Marshall disputed the importance of a magis­trate judge being the final arbiter of a given matter in Peretz. 125 See supra notes 79–81 and accompanying text (noting Peretz majority felt magistrates were always sufficiently controlled by Article III judges while Justice Marshall disagreed in context of jury selection).

Ultimately, this Note will use this factor discussed by Saphire and Solimine, as well as Justice Marshall, and apply it to several modern disputes over magistrate judge authority, developing a new interpretation of the constitutionally permissible scope of magistrate judge duties. 126 See infra notes 245–260 and accompanying text (proposing reinterpretation). Saphire and Solimine’s “[A]rticle III review” theory has been criticized for, among other defects, failing to account properly for the importance that Raddatz placed on as-of-right de novo review of a magistrate’s findings in an evidentiary hearing by the district court. Pfander, supra note 110, at 765–66. Pfander argues that it is “the magistrate’s status as an adjunct to the exercise of Article III judicial power” that has protected most exercises of magistrate authority from the Court’s invalidation. Id. at 766. Because he identifies Saphire and Solimine’s theory as one of “appellate review,” he argues that it fails to account for the singular importance of district, as opposed to appellate, review of magistrate decisions. Id. This is an uncharitable reading of Saphire and Solimine, who use the phrase “appellate review” only when referring to past cases and term their own theory one of “[A]rticle III review.” Saphire & Solimine, supra note 109, at 144–45. Even assuming that this critique of their theory is generally correct, it does not impact the theory as it applies to this Note, which will explicitly extend Saphire and Solimine’s analysis to magistrate judges.
Pfander’s account of the limit of congressional power to create Article I courts, which he calls the “inferior tribunals account,” is significantly less useful as a tool to analyze magistrate judge authority than Saphire and Solimine’s account. Pfander effectively excludes magistrate judges from consideration, noting simply that when an Article I tribunal hears a dispute that “fits squarely within traditional conceptions of the judicial power . . . the Court has required the Article I tribunal or agency to do its work as an adjunct to an Article III court.” Pfander, supra note 110, at 747–48. As applied to magistrates, this is a truism: Magistrate courts are by definition courts adjunct to a particular Article III court. Saphire and Solimine’s “[A]rticle III review” approach provides guidance on which exercises of magistrate authority should be invalidated by the Court and which should not, while Pfander’s “inferior tribunals account” identifies magistrate judges as outside his categorization scheme.
It may seem strange to use a doctrine that more typically applies to executive agencies and their Administrative Law Judges (ALJs) to magis­trate judges, but when a magistrate judge presides over jury selection, she is doing so pursuant to the same congressional power that allows ALJs to adjudicate common law counterclaims that come before them. 127 See CFTC v. Schor, 478 U.S. 833, 844 (1986) (finding CFTC’s position “that it has the power to take jurisdiction over counterclaims” to be “eminently reasonable and well within the scope of its delegated authority”). This formalist similarity between magistrate judges and ALJs notwithstanding, later on this Note argues that magistrate judges’ relative closeness in working relationship with and proximity to 128 Magistrate judges are often located in the same courthouse as district judges. See infra Appendix C (describing relationship between district and magistrate judges). In the Southern District of New York, magistrate judges are the only nondistrict or circuit judges in the same courthouse, with bankruptcy judges located in a separate “bankruptcy” courthouse. See id. district judges significantly diminishes the Article III problems attendant in delegating adjudicatory authority to magistrate judges.

II. Stern v. Marshall and Contested Exercises of Magistrate Judge Authority

This Part examines recent developments in the area of Article I judicial authority in general and magistrate judges in particular. Spe­cifically, it discusses Stern v. Marshall, 129 131 S. Ct. 2594 (2011). a 2011 Supreme Court decision on the scope of bankruptcy judge authority; Stern’s implications for magistrate judge authority; and the relation of two recent circuit splits on magistrate judges to Stern. Section II.A covers Stern, and section II.B introduces the two contested exercises of magistrate judge authority that are discussed in the bulk of the Note: rendering final decisions in state habeas cases and accepting felony guilty pleas. Section II.B also connects the recent cases condemning these exercises of authority to Stern, finding that they share the same Article III concerns about excessive magistrate judge authority.

A. Stern v. Marshall

Two bankruptcy attorneys recently argued that “[t]here is a striking contrast between the jurisdictional harmony that animates the federal magistrate system and the disagreements about the scope of federal bankruptcy jurisdiction.” 130 Jonathan W. Young & Dana G. Hefter, Creating a Defensive Border: Reasonable Limits on Post-Stern Bankruptcy Jurisdiction, Am. Bankr. Inst. J., July 2013, at 28, 28. While, as detailed earlier, there have been periods of time when there was significant dispute over the authority of federal magistrate judges, 131 See supra notes 33–37 and accompanying text (summarizing jurisdictional disputes arising out of original 1968 Federal Magistrates Act); supra notes 53–87 and accompanying text (discussing conflicting jurisprudence on magistrate judge authority to preside over felony jury selection). today there is little dispute about the juris­diction of federal magistrate judges, largely because the flexibility of the system leads to few opportunities for conflict. 132 See supra note 11 and accompanying text (noting value of flexibility in federal magistrate system).

Those attorneys were writing in the wake of Stern v. Marshall, a 2011 Supreme Court decision that one commentator describes as having rendered “the bankruptcy bench and bar . . . topsy-turvy.” 133 Geoffrey K. McDonald, The Question of Consent in Executive Benefits: Can Bankruptcy Courts Exercise the Judicial Power of the United States Under Article III Based on Litigant Consent Alone?, 87 Am. Bankr. L.J. 271, 304 (2013). Another writer has described the case as an “earthquake” that has “caused aftershocks.” Jolene Tanner, Comment, Stern v. Marshall: The Earthquake that Hit the Bankruptcy Courts and the Aftershocks that Followed, 45 Loy. L.A. L. Rev. 587, 612 (2012). Stern’s central holding is specific to the authority of bankruptcy judges, but like the legislative court doctrine cases that arose from exercises of authority by a bankruptcy judge, Stern has important implications for non-Article III judges generally. 134 Northern Pipeline also came out of bankruptcy proceedings. N. Pipeline Constr. Co. v. Marathon Pipe Line Co., 458 U.S. 50, 56 (1982).

Stern had sensational facts, 135 The facts are sensational enough to deserve inclusion in a footnote. Stern is the last chapter in the lengthy fight over the estate of oil magnate J. Howard Marshall II (“Marshall”), who died in 1995, leaving nothing for his wife, Vickie Lynn Marshall (known to the public as Anna Nicole Smith, described henceforth as “Vickie”). See Marshall v. Marshall, 547 U.S. 293, 300 (2006) (describing factual background). One of Marshall’s sons, E. Pierce Marshall (“Pierce”), was the beneficiary of Marshall’s estate, and several parties initiated claims against the estate in both state and federal courts. Id. While probate proceedings were ongoing in Texas state court, Vickie filed for Chapter 11 bankruptcy in California. Id. In the (federal) bankruptcy proceeding, Pierce filed a proof of claim, a written statement indicating that the writer is a creditor to someone going through the bankruptcy process and expressly setting forth a claim or claims, Fed. R. Bankr. P. 3001(a), alleging that Vickie defamed him. Marshall v. Marshall, 547 U.S. at 300. In answering the defamation claim, Vickie asserted a state-law counterclaim (still in bankruptcy court) of tortious interference with a gift. Id. at 301. It is important that the claim was based in state common law because this made the action over the claim “independent of the federal bankruptcy law and not necessarily resolvable by a ruling on the creditor’s proof of claim in bankruptcy.” Stern v. Marshall, 131 S. Ct. 2594, 2611 (2011). The bankruptcy court eventually ruled for Vickie on both claims. Marshall v. Marshall, 547 U.S. at 301. Importantly, that court found that the state counterclaims were “core proceedings” to the bankruptcy and thus within its jurisdiction. Id.
After unsuccessfully fighting these holdings under lack of subject matter jurisdiction, Pierce sought district court review of the bankruptcy court rulings. Id. at 301–02. The district court held, inter alia, that Vickie’s state tortious interference claim did not qualify as a “core proceeding,” so the bankruptcy court could only “issue proposed findings of fact and conclusions of law . . . reviewed de novo by the district court.” Id. at 302–03. The district court performed this de novo review and affirmed the bankruptcy court’s decision, id. at 304, and the Ninth Circuit reversed. Id. at 304–05. The Supreme Court reversed this decision in Marshall v. Marshall on grounds unimportant for this Note. Id. The case was sent back to the lower courts, both original parties died, and the case wound its way back to the Supreme Court in Stern v. Marshall with different issues, reality television star and attorney Howard K. Stern a party as executor of Vickie’s estate, and Pierce’s widow Elaine T. Marshall the other party as executor of Pierce’s estate. 131 S. Ct. at 2594.
The case grew so complicated during its two trips to the Court that Chief Justice Roberts opened his opinion for the Court in Stern with a lengthy quote comparing the matter before him to Charles Dickens’s Bleak House, in part revealing his opinions about the case: “A ‘long procession of [judges] has come in and gone out’ during [the proceeding], and still the suit ‘drags its weary length before the Court.’” Stern, 131 S. Ct. at 2600 (quoting Charles Dickens, Bleak House, in 1 Works of Charles Dickens 4–5 (1891)) (first alteration in Stern).
but for purposes of this Note, it is only important to know that the Court was presented with two questions: Whether a particular state-law counterclaim was a “core” claim in the bankruptcy proceeding and whether the statute that authorized the bankruptcy courts to hear this counterclaim was constitutional. 136 See Marshall v. Marshall, 547 U.S. at 315 (noting as “open for consideration on remand” whether “Vickie’s claim was ‘core’”). On the first part of the question, the Court concluded that the plain text of a section of the Bankruptcy Code, which includes as a “core proceeding” before a bankruptcy judge any “counterclaims by the estate against persons filing claims against the estate,” 137 28 U.S.C. § 157(b)(2)(C) (2012). unambiguously authorized a bankruptcy court to hear and enter final judgment on the counterclaim at issue. 138 Stern, 131 S. Ct. at 2604 (“That provision specifies that core proceedings include ‘counterclaims by the estate against persons filing claims against the estate.’” (quoting 28 U.S.C. § 157(b)(2)(C))). However, the Court held this grant of authority to bankruptcy courts by the Bankruptcy Act of 1984 unconstitutional, concluding that, while bank­ruptcy courts could hear these counterclaims under the statute, they lacked the constitutional authority to enter final judgment on a counter­claim that is not “resolved in the process of ruling on a creditor’s proof of claim.” 139 Id. at 2620.

The Court based this decision in large part on a formalist con­ception of the legislative court doctrine. 140 See supra notes 113–114 and accompanying text (discussing categories of Northern Pipeline). The Court found that adjudi­­cating a state-law counterclaim did not fall within the so-called “public rights exception,” which allows non-Article III judges to handle cases “arising ‘between the Government and persons subject to its authority in connection with the performance of the constitutional functions of the executive or legislative departments.’” 141 Stern, 131 S. Ct. at 2612 (quoting Crowell v. Benson, 285 U.S. 22, 50–51 (1932)). In entering final judgment on a claim arising under state common law, the Court concluded, the bankruptcy court unconstitutionally “exercised the ‘judicial Power of the United States’” because it strayed outside the public-rights exception. 142 Id. at 2611 (quoting U.S. Const. art. III, § 1).

In concurrence, Justice Scalia drew attention to the muddled legislative court doctrine without invoking that doctrine by name: “The sheer surfeit of factors that the Court was required to consider in this case should arouse the suspicion that something is seriously amiss with our jurisprudence in this area.” 143 Id. at 2621 (Scalia, J., concurring). Justice Scalia was concerned not just with the “sheer numerosity” of reasons the Court gives for its opinion but also because “they have nothing to do with the text or tradition of Article III.” Id. Justice Scalia argued for a rule that would require “an Article III judge . . . in all federal adjudications, unless there is a firmly established historical practice to the contrary.” Id. His Stern concurrence was the first time Justice Scalia presented this formulation, despite writing separately in Gonzalez v. United States, 553 U.S. 242, 254–58 (2008) (Scalia, J., concurring in the judgment), and writing a separate dissent in Peretz v. United States, 501 U.S. 923, 952–56 (1991) (Scalia, J., dissenting). Two of the factors Justice Scalia col­lected 144 Justice Scalia identified seven factors in his concurrence. Stern, 131 S. Ct. at 2621 (Scalia, J., concurring). are worth evaluating in more detail because they relate to reasons that motivated the Court’s Gomez and Peretz decisions. 145 See supra text accompanying notes 82–83 (laying out “three factors” important to Court in Gomez, Peretz, and Gonzalez and to Justice Marshall dissenting in Peretz). The Court in Stern expressed two concerns to which Justice Scalia attached particular emphasis. First, a party filing a proof of claim in a bankruptcy proceeding cannot meaningfully consent to bankruptcy court juris­diction, 146 See Stern, 131 S. Ct. at 2614–15, 2614 n.8 (majority opinion) (“Pierce did not truly consent to resolution of Vickie’s claim in the bankruptcy court proceedings . . . [because] [c]reditors who possess claims that do not satisfy the requirements for nondischargeability . . . have no choice but to file their claims in bankruptcy proceedings if they want to pursue the claims at all.”). and second, a bankruptcy court has authority to issue final orders on a counterclaim under 28 U.S.C. § 157(b)(2)(C) subject to review only if a party chooses to appeal the ruling to a district court. 147 See id. at 2619 (“[A] bankruptcy court resolving a counterclaim under 28 U.S.C. § 157(b)(2)(C) has the power to enter ‘appropriate orders and judgments’—including final judgments—subject to review only if a party chooses to appeal . . . .” (quoting 28 U.S.C. §§ 157(b)(1), 158(a)–(b) (2012))).

The Stern majority’s invocation of the public rights exception recalls Northern Pipeline, which included it as the most important category of cases that non-Article III judges could decide. 148 See supra notes 112–117 and accompanying text (discussing Northern Pipeline factors). Under this reasoning, it could be argued that magistrate judges sometimes (unconstitutionally) exercise the “judicial Power of the United States” because they often decide matters outside the public-rights exception. The application of Stern to magistrate judges is the focus of the remainder of this Note.

Most commentary on Stern focuses on the issue of litigant consent, ignoring the latter point about Article III court review of the non-Article III court, in part because the most significant post-Stern ambiguity con­cerned litigant consent. 149 The Sixth and Ninth Circuits reached opposing conclusions on how a bankruptcy court should proceed when a judge identifies that a “Stern claim” has appeared in the matter. See McDonald, supra note 133, at 291–99 (describing relevant opinions of Sixth and Ninth Circuits). McDonald argues that the Sixth Circuit decision, Waldman v. Stone, 698 F.3d 910 (6th Cir. 2012), cert. denied, 133 S. Ct. 1604 (2013), is “simply wrong” because of an either “deliberate or inadvertent” failure to cite to the directly on-point 28 U.S.C. § 157(c)(2), which unambiguously allows a bankruptcy judge to enter final judgment in noncore proceedings with the consent of the litigants, an omission “which rendered [the section] unconstitutional sub silentio.” McDonald, supra note 133, at 293, 297. The Ninth Circuit, by contrast, relied on § 157(c)(2) in Exec. Benefits Ins. Agency v. Arkison (In re Bellingham Ins. Agency, Inc.), 702 F.3d 553 (9th Cir. 2012), and affirmed a bankruptcy court’s use of the statute to decide a core matter with the consent of the litigants. See id. at 567 (“If consent permits a non-Article III judge to decide finally a non-core proceeding, then it surely permits the same judge to decide a core proceeding . . . .”). The Ninth Circuit required implied consent, relying on a comparison with the Federal Magistrates Act as discussed in Roell. Id. at 569 (referencing Roell v. Withrow, 538 U.S. 580, 586 (2003)); see also supra note 83 (discussing holding of Roell).
The Supreme Court sided firmly with the Ninth Circuit, identifying no constitutional problem with a bankruptcy court proceeding under § 157(c) to dispose of noncore claims with the consent of the parties. See Exec. Benefits Ins. Agency v. Arkison, 134 S. Ct. 2165, 2173 (2014) (“If the claim satisfies the criteria of § 157(c)(1) [as non-core], the bankruptcy court simply treats the claims as non-core . . . .”). Thus, courts can continue to handle truly noncore matters with the consent of the parties. Courts may also adjudicate “Stern claims,” proceedings that meet the statutory definitions for a “core” proceeding but may not, after Stern, be treated as such by a bankruptcy court; instead courts will treat them as non-core under § 157(c). See id. (holding a “Stern claim . . . may be adjudicated as a non-core claim”). This ruling helped “to mitigate the procedural disarray in which the Court threw the bankruptcy process” in Stern. Ronald Mann, Opinion Analysis: Bankruptcy Judges, Attorneys Breathe Sigh of Relief as Court Affirms Bankruptcy Court Authority, SCOTUSblog (June 10, 2014, 10:07 am), http://www.scotusblog.com/2014/06/opinion-analysis-bankruptcy-judges-attorneys-breathe-sigh-of-relief-as-court-affirms-bankruptcy-court-authority/[http://perma.cc/E8RL-V827].
Additionally, the only scholarship on the impact Stern could have on magistrate judges focused exclusively on litigant consent. 150 See generally Lori Yount, Comment, Litigant Consent as a Constitutional Threat: Reconsidering the Jurisdiction of Magistrate Courts After Stern v. Marshall, 55 S. Tex. L. Rev. 197 (2013) (describing background of litigant consent to magistrate and bankruptcy judge authority and suggesting “solutions to the constitutional problem posed by Stern”). There has been no effort as of yet to examine the other six “factors” Justice Scalia identified in his concurring opinion in Stern in any detail, let alone as applied to magistrate judges. Below, this Note will argue that the persistent availability of review by an Article III court, one of Justice Scalia’s factors and the crux of the legislative court doctrine as articulated by Saphire and Solimine, 151 See supra notes 122–125 and accompanying text (describing legislative court doctrine as articulated by Saphire & Solimine). is the crucial factor for determining if an exercise of authority by a magistrate judge is proper.

B. Stern and Recent Challenges to Magistrate Judge Authority

In 2014, the scope of magistrate judge authority was the central concern of circuit courts in assessing two particular exercises of district court adjudicatory delegation to magistrate judges. 152 United States v. Harden, 758 F.3d 886 (7th Cir. 2014); Brown v. United States, 748 F.3d 1045 (11th Cir. 2014). The first case, from the Eleventh Circuit, raised the issue of magistrate judge authority to enter final orders in actions brought by federal prisoners attacking their sentences under 28 U.S.C. § 2255. 153 28 U.S.C. § 2255 (2012); Brown, 748 F.3d at 1047.

1. Magistrate Judges and § 2255 Proceedings. — 28 U.S.C. § 2255 authorizes suits brought by federal prisoners attacking their sentences. 154 28 U.S.C. § 2255; cf. id. § 2254 (outlining federal writ of habeas corpus for state prisoners). Procedurally, § 2255 proceedings must be brought in “the court which imposed the sentence” and the filing consists of a motion to “vacate, set aside or correct the sentence.” 155 Id. § 2255 (a). Motions under § 2255 differ from habeas corpus actions filed under 28 U.S.C. § 2254 to contest a state sen­tence in many ways, including that while habeas petitions are “separate civil action[s]” for all purposes, 156 Id. advisory committee’s note to rule 1. different courts have cate­gorized a motion filed under § 2255 as both civil and criminal. Reflecting the lack of clarity on this question, there are three different approaches to cate­gorizing § 2255 proceedings as either civil or criminal. First, “[t]he Supreme Court and lower courts have repeatedly characterized [the § 2255 proceeding] as civil in nature.” 157 3 Charles Alan Wright & Sarah N. Welling, Federal Practice and Procedure: Federal Rules of Criminal Procedure § 622 (4th ed. 2014). Second, “[a]t least three Courts of Appeals have suggested that [the] proceedings . . . are criminal, not civil.” 158 Id. Supporting the notion that § 2255 proceedings are criminal and not civil is the advisory committee note on § 2255, which says such a motion “is a further step in the movant’s criminal case and not a separate civil action.” 28 U.S.C. § 2255 advisory com­mittee’s note to Rule 1. This language from the note is far from conclusive, however, as evidenced by the disparity in approaches among the various circuit courts, the Supreme Court’s occasional rejection of this approach, and the fact that this conclusion in the note is based on a single comment from the legislative history from the 1948 statute that created § 2255. See Wright & Welling, supra note 157, § 622 (“This is based on a single paragraph from the legislative history of the 1948 statute and without any reference to the cases that have reached a contrary conclusion.”). Third, “[s]everal Courts of Appeals have characterized [§ 2255] habeas actions as ‘hybrid’ in nature.” 159 Wright & Welling, supra note 157, § 622.

Whether a § 2255 proceeding is “criminal” or “civil” rarely has practical significance, but an important exception is in the area of magis­trate judge authority to handle § 2255 proceedings because of the dif­ferent requirements and rules for magistrate judge handling of civil and criminal matters. 160 See supra note 91 and accompanying text (discussing magistrate judge authority in civil and criminal cases). If a § 2255 proceeding is purely civil, a magis­trate judge may, as a statutory matter, handle the proceeding alone and enter a final order with the consent of the litigants, under a magis­trate judge’s civil consent power. 161 See supra notes 38–41 (discussing civil consent authority). On the other hand, if a court agrees with the Advisory Committee that the § 2255 proceeding is an extension of the movant’s criminal case, the civil consent authority would not allow a magistrate judge to handle the matter. Either it would fall into the additional-duties bucket or a magistrate judge could not handle the matter. Adopting the view that § 2255 proceedings are extensions of the underlying criminal case would lead to the conclusion that a magistrate judge could not handle any part of a § 2255 proceeding resulting from a felony conviction because the proceeding is just an extension of the original felony trial, which is definitively outside the magistrate judge’s scope of authority.

In Brown v. United States, the Eleventh Circuit recently held that a § 2255 proceeding is not a “civil matter” for purposes of the Federal Magistrates Act, despite concluding that the statute “could plausibly be read to authorize a magistrate judge to enter final judgment in a § 2255 proceeding.” 162 Brown v. United States, 748 F.3d 1045, 1072 (11th Cir. 2014). The court chose to adopt this particular reading largely because it found that in light of Stern v. Marshall, “allowing a magistrate judge to enter final judgment on a federal prisoner’s § 2255 motion raises serious constitutional concerns.” 163 Id. The court stated that it “har­bor[s] serious concerns as to the facial constitutionality of” the section of the Federal Magistrates Act authorizing magistrate judges to preside over civil trials with the consent of the parties. 164 Id. at 1068. These concerns are directly connected to Stern: “Congress’s conclusion that magistrate judges are adjuncts of the district courts cannot be deemed correct.” 165 Id. at 1069. Instead, quoting Stern, the court concluded that “magistrate judges exercise the ‘judicial Power of the United States,’ despite the fact that they lack Article III protections.” 166 Id. (quoting Stern v. Marshall, 131 S. Ct. 2594, 2601 (2011)). Brown is the first example of a circuit court reconsidering the constitutionality of an established magistrate judge practice in light of Stern v. Marshall.

In addition to its concerns about the facial constitutionality of the Federal Magistrates Act, the Eleventh Circuit identified two unique constitutional problems with granting a magistrate judge authority to rule on § 2255 motions. First, having a magistrate judge in such a posi­tion “would create an ‘ironic situation whereby non-Article III magistrate judges review and reconsider the propriety of rulings by Article III district judges, but do not themselves have to worry about review’ by the district court.” 167 Id. at 1070 (quoting United States v. Johnston, 258 F.3d 361, 371 (5th Cir. 2001)). Second, and as discussed above, 168 See supra note 126 (discussing importance of “as-of-right” de novo district court review to magistrate authority). “the authority of a district court to review the magistrate judge’s decision, even if neither party invokes such authority, is essential to ensuring that Article III values are protected.” 169 Brown, 748 F.3d at 1071. To the court, litigant consent “does not . . . obviate the Article III concerns.” 170 Id. at 1070. The Eleventh Circuit in Brown, then, highlighted that it is the lack of district court review, and not litigant consent, that raises the most troubling Article III problems. 171 See supra notes 124–126 and accompanying text (discussing “Article III review”); infra notes 233–241 and accompanying text (applying Article III review theory to magis­trate judge authority generally). One magistrate judge interviewed for this Note presented a different argument for allowing only district judges to handle § 2255 motions. Rather than express a concern rooted in Article III, the judge raised an efficiency problem with having magistrate judges handle § 2255 proceedings. Infra Appendix E. This magis­trate judge thought it a waste of a district’s time to have a magistrate judge handle a case with which the district judge is already familiar, at the very least because district judges handle sentencing of felony offenders. Infra Appendix E.

There is no dispute between the circuits as to whether magistrate judges can enter final orders on a § 2255 motion: Only one other circuit has squarely addressed this question, and that court also found that such a delegation violated Article III. 172 See Johnston, 258 F.3d at 372 (“By allowing consensual delegation of § 2255 proceedings to magistrate judges, we exact a deadly blow to the vitality and strength of a [sic] independent judiciary.”). The Eleventh Circuit’s decision effectively dodged the constitutional question by finding that a § 2255 proceeding is not a “civil matter,” obviating the need to decide whether the handling of this particular matter by a magis­trate judge was a violation of Article III. See Brown, 748 F.3d at 1047 (“[W]e need not decide whether that delegation would violate Article III because we hold that a § 255 proceeding is not a ‘civil matter’ for purposes of [the relevant section of the Federal Magistrates Act].”). Though consistent with the outcome of the Fifth Circuit’s Johnston case, the Eleventh Circuit’s reasoning in Brown stands in contrast to the holding of the Fifth Circuit and other courts that a § 2255 proceeding is, in fact, a “civil action.” See, e.g., Johnston, 258 F.3d at 366 (holding “§ 2255 proceeding is a civil matter over which Congress intended magistrate judges to exercise jurisdiction upon consent of the parties”). In fact, Brown overruled a Fifth Circuit precedent in the Eleventh Circuit to this effect. See United States v. Williamson, 255 F.2d 512, 515 (5th Cir. 1958) (per curiam) (denying petition for rehearing) (noting “long-established fact” that proceeding under § 2255 is “a civil matter”). Williamson was binding Eleventh Circuit precedent under Bonner v. City of Prichard, which held that the decisions of the Fifth Circuit prior to September 30, 1981 (when the Fifth Circuit was split, creating the Eleventh Circuit) were binding precedent in the Eleventh Circuit. Bonner v. City of Prichard, 661 F.2d 1206, 1210 (11th Cir. 1981). However, the Eleventh Circuit’s opinion in Brown expressly raised the related question of the constitu­tionality of a district judge delegating a state habeas action brought under § 2254 to a magistrate judge. For several reasons, allowing magistrate judges to handle § 2254 state habeas petitions on their own but forbidding them from entering final orders in § 2255 proceedings for federal prisons makes some sense. While there is some dispute over the extent to which a § 2255 proceeding is civil or criminal, there is no doubt that a habeas corpus petition is a civil action, entirely separate from the state criminal proceeding that led the prisoner to be incarcerated. 173 See supra notes 155–161 and accompanying text (discussing differences between § 2254 and § 2255 proceedings). The Fifth Circuit in Johnston also focused on some of the unique peculiarities of having a magistrate judge enter final orders in a § 2255 pro­ceeding. See 258 F.3d at 368–71 (arguing “§ 2255 motion directly questions the validity of a prior federal court ruling,” “may unwittingly embroil a magistrate judge in the uncons­titutional conduct of a felony trial,” and presents “severe” “reviewability problems”). Further, in Brown, one particular reason the Eleventh Circuit furnished against allowing magistrate judges to enter final orders in § 2255 proceedings does not apply in the context of state habeas actions: A magistrate judge handling a habeas corpus petition would not “recon­sider the propriety” of a federal district court judge’s decision because the original criminal matter was handled by state courts. 174 See supra note 167 and accompanying text (noting this concern).

Until Brown, no federal circuit court had questioned the authority of magistrate judges to handle state habeas actions under § 2254. The Fifth, Sixth, Seventh, and Eighth Circuits have all held that a magistrate judge may reach a final disposition on a § 2254 habeas corpus petition. 175 White v. Thaler, 610 F.3d 890, 897–98 (5th Cir. 2010); Farmer v. Litscher, 303 F.3d 840, 845 (7th Cir. 2002); Norris v. Schotten, 146 F.3d 314, 326 (6th Cir. 1998); Orsini v. Wallace, 913 F.2d 474, 476 (8th Cir. 1990). In its decision upholding this authority of magistrate judges, the Seventh Circuit expressly disavowed any comparison between motions brought under § 2254 and § 2255. 176 See Farmer, 303 F.3d at 845 (declining to extend analysis of § 2254 habeas petitions to § 2255 motions). In addition to these four circuits, the Eleventh Circuit has spoken previously on this question, holding in Sinclair v. Wainwright that magistrate judges may enter final judgment on § 2254 habeas petitions, in line with the holdings of the other circuits. 177 814 F.2d 1516, 1519 (11th Cir. 1987).

In Brown, the Eleventh Circuit had no cause to overrule Sinclair, but found that “[i]n light of Stern, our holding in Sinclair . . . has certainly been called into question.” 178 Brown v. United States, 748 F.3d 1045, 1069 (11th Cir. 2014). Of particular concern to the court was that in handling either § 2254 or § 2255 matters, a magistrate judge exercises the “judicial Power of the United States,” which was the problem with bankruptcy judges disposing of state-law counterclaims in Stern v. Marshall. 179 See supra note 166 and accompanying text (discussing this language from Stern as applied to Brown and § 2255). It is thus conceivable that the Eleventh Circuit could rule in a future case that delegating final disposition in a § 2254 habeas case to a magistrate violates Article III, a ruling that would create a split with the aforementioned circuits.

While Brown is the first example of a circuit court reconsidering the constitutionality of an established exercise of magistrate judge authority (handling § 2254 motions) in light of Stern v. Marshall, it is unlikely to be the last. 180 See, e.g., Yount, supra note 150, at 198 (“[T]he outcome in Stern will implicate magistrate courts and force them to reexamine whether the entering of final judgments on state-law claims by magistrate judges is constitutional regardless of litigant consent.”). The following subsection provides an example of a court of appeals questioning the constitutionality of a magistrate judge’s addi­tional duty. While, unlike Brown, the example that follows does not expli­citly invoke Stern v. Marshall, the decision does reflect some of the same underlying concerns about Article III.

2. Magistrate Judges and Felony Guilty Pleas. — Before accepting a guilty plea, a federal court must conduct a “plea colloquy,” addressing the defendant in open court. 181 Fed. R. Crim. P. 11(b)(1). During the colloquy, the court “must inform the defendant of, and determine that the defendant under­stands,” fifteen enumerated aspects of the plea and must also ensure that the plea is voluntary and determine its factual basis. 182 Id.

In 1994, the Magistrate Judges Committee of the Administrative Office of the United States Courts endorsed a pilot program to authorize magistrate judges to accept guilty pleas with the consent of the parties. 183 See Inventory, supra note 10, § 10, at 14 (describing Magistrate Judges Committee supplement to Long Range Plan for Magistrate Judges System). Today, all circuits allow magistrate judges to conduct the plea colloquy required by Federal Rule of Criminal Procedure 11 and make a recom­mendation on accepting the plea to the district judge, and a few circuits further allow magistrate judges to accept the plea without any involve­ment by a district judge. 184 See id. (“A large number of courts have authorized magistrate judges to conduct allocution proceedings to accept felony guilty pleas . . . as an additional duty . . . . Some courts have further held that a magistrate judge may accept the defendant’s guilty plea in a felony case . . . .”).

Before 2014, four circuits (the First, Second, Eighth, and Ninth Circuits) held that magistrate judges can perform the plea colloquy but did not decide whether magistrate judges may accept the plea because in each case the magistrate judge conducting the colloquy merely made a recommendation to the district judge to accept the plea. 185 See United States v. Reyna-Tapia, 328 F.3d 1114, 1116 (9th Cir. 2003) (holding magistrate judge may administer Rule 11 colloquy and submit recommendation to district judge regarding acceptance of plea); United States v. Torres, 258 F.3d 791, 796 (8th Cir. 2001) (same); United States v. Williams, 23 F.3d 629, 634 (2d Cir. 1994) (same); see also United States v. Vega-Martinez, 425 F.3d 15, 17–19 (1st Cir. 2005) (finding magistrate judge had adequately determined plea was voluntarily given when conducting colloquy and making recommendation). Four circuits (the Fourth, Fifth, Tenth, and Eleventh Circuits) have gone further, finding that magistrate judges may directly accept the plea. 186 United States v. Benton, 523 F.3d 424, 431–32 (4th Cir. 2008); United States v. Woodard, 387 F.3d 1329, 1330 (11th Cir. 2004); United States v. Dees, 125 F.3d 261, 269 (5th Cir. 1997); United States v. Ciapponi, 77 F.3d 1247, 1251–52 (10th Cir. 1996). The Fourth Circuit found that allowing a magistrate judge to accept a felony guilty plea is “merely the natural culmination of a plea colloquy.” 187 Benton, 523 F.3d at 431.

Over a decade ago, one sitting magistrate judge predicted that the circuits would come into conflict on the question of whether a district judge can accept a felony plea upon a magistrate judge’s recommen­dation. 188 See Durwood Edwards, Can a U.S. District Judge Accept a Felony Plea with a Magistrate Judge’s Recommendation?, 46 S. Tex. L. Rev. 99, 108 (2004) (“It is very likely that the appellate courts will disagree as to how to resolve this. If so, it will then be left up to the Supreme Court to do so.”). Conflict on this issue occurred in 2014, when the alignment of the circuits on this question changed in two ways. First, in dicta in Brown, the Eleventh Circuit narrowed the holding of a 2004 Eleventh Circuit decision allowing magistrate judges in that circuit to accept guilty pleas without district judge involvement. 189 Brown v. United States, 748 F.3d 1045, 1071 n.53 (2014) (reinterpreting United States v. Woodard such that “magistrate judge’s action in such proceedings are [sic] akin to a report and recommendation rather than a final adjudication of guilt”). Second, the Seventh Circuit, which previously had no precedent on this issue, 190 See Inventory, supra note 10, § 7, at 15 (listing no Seventh Circuit cases concerning acceptance of guilty pleas by magistrate judges). became the first circuit to affirmatively rule that magistrate judges may not accept guilty pleas, limiting their authority to conducting the colloquy and issuing a report and recommendation to the district judge. 191 United States v. Harden, 758 F.3d 886, 891 (7th Cir. 2014). United States v. Harden is the first decision by a circuit court to invalidate a district’s practice of having a magistrate judge accept a felony guilty plea.

With this decision, the Seventh Circuit now requires that a magis­trate judge conduct the plea colloquy and issue a report and recommen­dation to the district judge, who will decide whether or not to accept the plea. 192 Id. (noting this became voluntary practice of district from which this case ori­ginated after this plea was accepted). While it seems that many districts across the country operate under this system as a matter of course, 193 See infra Appendix E (noting none of judges interviewed for this Note had ever signed felony guilty plea). only the Seventh Circuit forbids any district from allowing a magistrate judge to accept a plea bargain.

The specific holding in Harden is that there is no statutory authori­zation for this exercise of magistrate judge authority, but the decision is extraordinary in its commitment to the formal structural requirements of Article III. 194 See Harden, 758 F.3d at 891 (“A felony guilty plea is equal in importance to a felony trial leading to a verdict of guilty. And without explicit authorization from Congress, the district court cannot delegate this vital task.”). The opinion details the many important determinations that must be made before a plea can be accepted, including “whether the defendant is competent” and making a voluntary choice to forego their right to a trial and “whether there is a legal and factual basis for the guilty plea.” 195 Id. at 889. These are important protections for criminal defendants, but under Harden, they remain determinations made on the basis of the defendant’s appearance before a magistrate judge at the plea colloquy rather than directly before the district court judge, who need only review and approve or conduct a de novo review of the magistrate judge’s recommendation. 196 A district judge considering objections made during a plea process before a magistrate judge could arguably use the Supreme Court’s latest decision on magistrate judge authority, Gonzalez v. United States, 553 U.S. 242 (2008), to handle felony jury selection to reach the opposite conclusion of the Seventh Circuit. The Court found that it would not be “difficult or disruptive” to review objections raised during jury selection on an appeal before a district judge, objections that are arguably analogous to those made during a plea process. Id. at 252; see also supra note 87 and accompanying text (discussing review issue in Gonzalez).

In contrast to Brown, Harden makes no reference to Stern v. Marshall, or to Gonzalez, the most recent Supreme Court decision on magistrate judge authority, but its reasoning is rooted in the same formalist strand of the legislative court doctrine. 197 See supra notes 83–87 and accompanying text (discussing Gonzalez). The Seventh Circuit concludes “the acceptance of the guilty plea is quite similar in importance to the con­ducting of a felony trial” because both result “in a final and con­sequential shift in the defendant’s status.” 198 Harden, 758 F.3d at 889. Using a familiar canon of statutory construction and Gomez v. United States, 199 The court relies on the canon of expressio unius est exclusio alteriusmeaning “the expression of one thing . . . implies the exclusion of other things of the same sort.” Caleb Nelson, Statutory Interpretation 89 (2011). In this context, the Seventh Circuit used the canon to conclude that the Federal Magistrates Act’s grant of authority to magistrate judges to preside over civil matters and minor criminal cases constitutes an “‘implicit with­holding of the authority to preside at a felony trial.’” Harden, 758 F.3d at 889 (quoting Gomez v. United States, 490 U.S. 858, 872 (1989)). the Court con­cludes that Congress did not intend to authorize magistrate judges to accept felony guilty pleas under the additional-duties clause. 200 See Harden, 758 F.3d at 889 (“[T]he acceptance of a guilty plea in a felony case, a task no less important [than conducting a felony trial], is also not authorized by the statute.”).

As with the Eleventh Circuit’s decision in Brown and the Supreme Court’s concerns in Stern, litigant consent is irrelevant in Harden, 201 See id. at 890 (quoting recent Supreme Court decision that “stated . . . reversal would have been proper ‘[e]ven if the parties had expressly stipulated to the participation of a non-Article III judge in the consideration of their appeals, no matter how distin­guished and well qualified the judge might be’” (quoting Nguyen v. United States, 539 U.S. 69, 80–81 (2003))). with the court instead focusing on the importance of having the “ultimate” decision made by an Article III judge. 202 See supra text accompanying note 198 (noting Harden court’s concern with “final and consequential shift in the defendant’s status” caused by accepting felony guilty plea). The following subpart will examine the acceptance of plea bargains and issuance of final orders on § 2254 habeas petitions and § 2255 postconviction review proceedings, drawing similarities and contrasts between these various exercises of magistrate authority.

C. The Legislative Court Doctrine and Stern, Harden, and Brown

The Eleventh Circuit’s decision in Brown v. United States and the Seventh Circuit’s decision in Harden v. United States bear both substantive and thematic similarities. While Brown is a case about § 2255 postconviction proceedings that bases its restriction on magistrate judge authority on Article III, it also limits a prior Eleventh Circuit decision on the power of magistrate judges to accept felony guilty pleas. 203 See supra note 189 and accompanying text (discussing Brown’s limiting of United States v. Woodard, 387 F.3d 1329 (11th Cir. 2004)). While Harden found that magistrate judges could not accept felony guilty pleas on the basis of a statutory argument, Article III concerns featured prominently in the decision. 204 See supra notes 194–198 and accompanying text (discussing Article III concerns expressed in Harden).

Both decisions are examples of formalist thinking about Article III and in particular, a brand of formalism heavily influenced by Stern v. Marshall. 205 131 S. Ct. 2594 (2011). Looking back at the three Supreme Court decisions on magistrate judge authority to preside over felony jury selection—Gomez, 206 490 U.S. 858 (1989). Peretz, 207 501 U.S. 923 (1991). and Gonzalez 208 553 U.S. 242 (2008). —each used three factors to assess whether a particular exercise of judicial authority by a non-Article III judge was constitutional: (1) whether litigants consent to magistrate auth­ority, (2) whether the magistrate judge is the “final arbiter” of the matter before her, and (3) whether the matter at issue is more analogous to presiding over civil and misdemeanor trials with litigant consent or conducting a felony trial. 209 See supra text accompanying notes 82–83 (summarizing these factors).

Most of the modern jurisprudence on the scope of magistrate judge authority under the additional-duties clause has developed from Peretz and specifically from its focus on litigant consent as the overriding concern. In the Inventory of United States Magistrate Judge Duties, the Administrative Office of the U.S. Courts notes that “[i]n light of the Peretz decision, many courts have focused on whether the litigant consented . . . when deciding whether the magistrate judge’s exercise of authority was proper.” 210 Inventory, supra note 10, § 7, at 3. The other two factors, the finality of the magistrate judge’s decision and its similarity to well-defined magistrate responsibilities and restrictions, have been given significantly less attention. That should change.

The Eleventh Circuit properly recognized that Stern v. Marshall requires a focus on something other than litigant consent in assessing whether a particular exercise of magistrate judge authority is consti­tutional. 211 See supra text accompanying note 170 (noting Eleventh Circuit’s disregard of litigant consent factor). Stern specifically represents a return to the legislative court doctrine, with the Court heavily weighing a formalist conception of the limits on Congress’s powers to create legislative court and leaving no room for litigant consent to cure structural Article III problems. 212 See supra notes 113–114 and accompanying text (discussing three categories from Northern Pipeline). Both the Seventh Circuit in Harden and the Eleventh Circuit in Brown are con­cerned about magistrate judges making a “final adjudication” regard­less of litigant consent, 213 See Brown v. United States, 748 F.3d 1045, 1071 n.53 (11th Cir. 2014). linking both of those decisions to a long strand of Article III formalism.

However, the early formalism of the legislative court doctrine, from which Stern derives much of its argumentation, is difficult to apply in the context of magistrate judges, who almost exclusively handle matters traditionally tasked to district judges. 214 Magistrate judges also perform the duties of the United States commissioners under 28 U.S.C. § 636(a)(1) (2012) (“Each United States magistrate judge . . . shall have . . . all powers and duties conferred or imposed upon United States commissioners . . . .”). The great bulk of duties exercised by magistrate judges falls outside all three formal categories of permissible non-Article III judicial power expressed in the earlier legis­lative court doctrine cases. 215 Magistrate judges are not exclusively military or territorial judges and do not handle only cases that fall within the public-rights exception. See supra notes 113–114 and accompanying text (discussing three categories from Northern Pipeline). Hence, while the Eleventh Circuit may be correctly interpreting the Supreme Court’s expressed interest in Stern to rely on something other than litigant consent in assessing the constitu­tionality of non-Article III courts, the court is placing undue influence on the reasoning of Stern because the formalist expression of the legis­lative court doctrine cannot apply to magistrate judges.

Instead of looking to Stern and in turn, its reliance on strict legislative court formalism, courts confronted with difficult or ambiguous appli­cations of magistrate judge authority should consider the matter from a functionalist perspective. The next Part describes a func­tionalist rule for assessing exercises of authority by magistrate judges that still respects the Article III supremacy of district judges and applies this rule to the issues in Harden and Brown.

III. Reading Stern Orthogonally: Magistrate Judge Functionalism

Federal magistrate judges derive their authority from the same place as do administrative law judges and bankruptcy judges: Article I. 216 See supra note 127 and accompanying text (comparing magistrate judge and CFTC ALJ). This fact, however, does not necessitate extending the formalist Article III approach Stern applied to bankruptcy judges to magistrate judges, as the Eleventh Circuit explicitly did in Brown. Rather, Stern should be read orthogonally to magistrate judges. Stern applies a formalist under­standing of the legislative court doctrine to bankruptcy judges, but ela­borating on the differences between magistrate judges and bank­ruptcy judges that carry weight in an Article III analysis demonstrates that Stern itself provides support for applying a functionalist legislative court doc­trine to magistrate judges. The overriding concern in the later legis­lative court cases and Stern is that the assignment of judicial authority to judges without Article III protections should not “actually impinge[] upon [A]rticle III values.” 217 Saphire & Solimine, supra note 109, at 107. While one way to ensure that Article III values are protected is through the proliferation of formal rules, the effort to secure Article III values in the context of magistrates must be more nuanced. 218 See supra note 215 (noting magistrate judges and their duties do not fit into Northern Pipeline’s formal categories).

The differences between magistrate judges and bankruptcy judges for purposes of Article III have been underexplored in the federal courts literature. Stern and its possible implications for magistrate judges pre­sent the perfect backdrop for engaging in such an analysis. This Part evaluates the differences between magistrate judges and bank­ruptcy judges and why they are meaningful under Article III, synthe­sizing argu­ments raised throughout this Note. Section III.A covers these differences, building to a theory of magistrate judge auth­ority that focuses on the supremacy of district judges and the district judge–magistrate judge rela­tionship. Section III.B applies that theory to the examples of Harden and Brown.

A. Distinguishing Magistrate Judges from Other Non-Article III Decisionmakers to Create a Theory of Permissible Magistrate Judge Authority

Magistrate judges require a unique Article III analysis for a few reasons. First, the system was designed to be flexible and provide the maximum latitude for each district to set magistrate duties for their court. 219 See supra notes 7–11 and accompanying text (discussing district court flexibility in setting magistrate judge responsibilities and benefits of this flexibility). Second, magistrate judges are connected to district judges to a degree far beyond any other non-Article III judge because of some particular features of the position. 220 The scope of this statement runs the gamut from the obvious—e.g., district judges are more connected to magistrate judges than they are to, say, a Securities and Exchange Commission ALJ—to the nuanced: District judges are generally more closely tied to magistrate judges than they are to bankruptcy judges, despite the fact that both magistrate and bankruptcy judges are associated with individual district courts. See supra note 128 (discussing physical and relational proximity between magistrate and district judges). The Judicial Conference of the United States, not Congress, authorizes new magistrate judge posi­tions, so the very existence of new magistrate positions is entirely within the control of Article III judges. 221 See McCabe, Guide, supra note 2, at 18 (noting Judicial Conference has authorized new positions “very deliberatively”). New magistrate judges, whe­ther they are filling a new position or replacing a departing judge, are selected entirely by the district court, where they hope to sit through a “merit-selection process.” 222 Id. at 13. Under the merit-selection process, districts are required to provide public notice of all magistrate judge vacancies and appoint a citizen selection panel, “composed of residents of the individual judicial districts, to assist the courts in identifying and recommending persons” to be magistrate judges. 28 U.S.C. § 631(b)(5) (2012).
While magistrate judges must be selected from the list of recommendations the committee of citizens gives the district judges, “district judges remain the focal point of the magistrate appointment process.” Christopher E. Smith, United States Magistrates in the Federal Courts: Subordinate Judges 32 (1990). Smith conducted a survey of district courts to determine how they selected merit committee members because the Judicial Conference has never specified how these committees were to be chosen. Id. at 34. While noting that his data are old, he found that roughly a quarter of districts employed a “Blue Ribbon” committee method, with the committee composed of “predominately white, male attorneys” who were “well-known to the district judges.” Id. at 35 tbl.1. Two-thirds had a “Representative” committee, with members of the committee representing various groups in the community. Id. Lastly, about eight percent of districts surveyed used a “Proxy” committee, wherein each district judge selected one or more committee members. Id. District judges both appoint members of the selection committee and choose from the recommendations provided. Id.
Magistrates are appointed for renewable terms of eight years 223 28 U.S.C. § 631(e). and can be removed by the district judges for cause. 224 See id. § 631(i) (“Removal . . . shall be only for incompetency, misconduct, neglect of duty, or physical or mental disability.”). Several specific exercises of magistrate judge authority require special desig­nation by a district judge, including handling pretrial matters, 225 Id. § 636(b)(1)(A). serving as a special master in a civil case, 226 Id. § 636(b)(2). and presiding over civil trials with consent of the litigants. 227 Id. § 636(c)(1). Under this rule, then, even if both parties consent, they will not be allowed to have the trial before the magistrate judge if the district judge objects.

Several of these factors led Justice Blackmun to conclude “the magistrate himself is subject to the Art[icle] III judge’s control.” 228 United States v. Raddatz, 447 U.S. 667, 685 (1980) (Blackmun, J., concurring). This was also the conclusion of the several circuit courts of appeals that have affirmed magistrate judge authority to accept felony guilty pleas and preside over state habeas actions under § 2254. The Fourth Circuit, holding that magistrate judges may accept felony guilty pleas after pre­siding over the plea colloquy, focused on “the role district courts play in protecting the structural integrity of Article III” when magistrates exer­cise this authority. 229 United States v. Benton, 523 F.3d 424, 432 (4th Cir. 2008). The Fifth Circuit similarly stressed the impor­tance of review by the district court as ensuring the constitutionality of a delegation to a magistrate judge to accept a guilty plea. 230 See United States v. Dees, 125 F.3d 261, 268–69 (5th Cir. 1997) (describing importance of review and concluding “plea proceedings conducted by magistrate judges are sufficiently reviewable so as not to threaten Article III’s structural guarantees”). In Stern, the Court found the possibility of review of a bankruptcy judge’s ruling on a state-law counterclaim by a district judge to be insufficient Article III review to save the constitutionality of such a ruling. 231 See Stern v. Marshall, 131 S. Ct. 2594, 2619 (2011) (concluding “‘[t]he authority—and the responsibility—to make an informed, final determination . . . remains with’ the bankruptcy judge, not the district court” (quoting N. Pipeline Constr. Co. v. Marathon Pipe Line Co., 458 U.S. 50, 81 (1982))). It is not enough that the district court acts merely as a stand-in for the court of appeals, reviewing the decision “only if a party chooses to appeal” because in that case, “a bankruptcy court can no more be deemed a mere ‘adjunct’ of the district court than a district court can be deemed such an ‘adjunct’ of the court of appeals.” 232 Id.

District court review of decisions by magistrate judges is not subject to this criticism the Court leveled against the review of decisions by bankruptcy judges. First, regardless of the quality of district court review, magistrate judges are definitively adjuncts of the district court in a way bankruptcy judges are not. 233 See supra note 126 (discussing adjunct nature of magistrate judge). While bankruptcy judges handle all bank­ruptcy matters that occur in a district as a matter of course, all but the most ministerial duties performed by magistrate judges must be either specifically delegated by an individual district judge or referred to magistrate judges through an established and deliberate practice of the district. 234 Bankruptcy cases begin with the debtor filing a petition with the bankruptcy court. Admin. Office of the U.S. Courts, Bankruptcy Cases, http://www.uscourts.gov/about-federal-courts/types-cases/bankruptcy-cases [http://perma.cc/JN3M-YYC4] (last visited Nov. 15, 2015). No specific case type or initial filing is ever made specifically to a magistrate judge; all matters that come before magistrate judges are filed in the district court generally, which assigns cases on the basis of district practice. See, e.g., infra Appendix A.2 (noting some districts automatically assign civil matters to magistrate judge and district judge, while others leave discretion to district judge whether magistrate judge is needed at all). Second, magistrate judges work with district judges every day, typically in the same courthouse. 235 See supra note 128 (noting in some large districts, magistrate and district judges share courthouse, with bankruptcy judges located elsewhere). Third, the assignment process in most districts means that district judges work with magistrate judges on the same cases (or at least, both district judges and magistrate judges work on different instances of the same type of cases). 236 See infra Appendix A (describing assignments of magistrate judges). As a result, the relationship between a district judge and the magistrate judges in the court is iterative: Over many professional interactions, a district judge comes to develop a pattern for working with magistrate judges and gets a better sense of how to use them, bolstering the oversight of Article III judges over magistrate judge authority. 237 One of the magistrate judges interviewed for this Note provided the example of a sitting court of appeals judge who, when he was a new district judge, delegated a sub­stantial portion of his criminal case load to a single magistrate judge, so he could learn how to handle a criminal case. Over time, this judge learned the ropes and cut down on his delegation of criminal cases. Infra Appendix B. Fourth, a previous survey has demonstrated that magistrate judges decide cases in the same way and reach the same results as district judges in the aggregate. 238 The only large-scale survey of the decisionmaking of magistrate judges and outcomes of cases brought before them was conducted by Bruce A. Carroll and published as a book. Bruce A. Carroll, The Role, Design, and Growing Importance of United States Magistrate Judges (2004). Carroll reached numerous conclusions about the approaches of magistrate judges, identifying regional, racial, and gender differences in decisionmaking. See generally id. at 55–71 (describing results of magistrate judge surveying). After comparing his data on magistrate judges to the decisions of district judges, Carroll concluded that “Magistrate and District Judges largely decide cases in the same manner.” Id. at 87. Carroll concluded that district judges decide cases in a very different manner than circuit judges. Id. Under a functionalist legislative court doctrine, courts are asked to consider if an exercise of authority by a non-Article III judge will “actually impinge[]” Article III values. 239 Saphire & Solimine, supra note 109, at 106–07. There is little to no indication that the autonomous choice of a district judge to delegate a task to a magistrate judge will “actually impinge” the values of Article III.

A focus on the relationship between magistrate judges and district judges as providing adequate Article III review is consistent with the theory of Article III review advanced by Saphire and Solimine. 240 See supra notes 122–125 and accompanying text (discussing Article III review generally). As discussed earlier, those authors argued that the “mere threat” of Article III review will help support Article III interests by diminishing the political influence on non-Article III judges. 241 See supra note 124 (describing this aspect of Saphire and Solimine’s theory). Undue political influence on non-Article III judges is severely curtailed in the case of federal magistrate judges because of the high degree of control district judges exercise over their duties. The Article III review district judges apply to magistrate judges is both a “threat” and a reality: District judges exercise control over magistrate judge authority both at the macro level, through the selection and retention process and through setting general guide­lines and rules for magistrate judges, and at the micro level, in reversing or vacating orders of magistrate judges.

Lastly, it is important to remember why the magistrate judge system exists. The complexity of the modern system and its attendant Article III issues can make it easy to forget that the Federal Magistrates Act pro­vided for delegation of additional duties to magistrate judges to allow district courts to “remain free to experiment” with the system. 242 S. Rep. No. 94-625, at 10–11 (1976). The Senate Report endorsing the additional-duties clause hoped to increase “time available to judges for the careful and unhurried performance of their vital and traditional adjudicatory duties.” 243 Id. The purpose of the statutory scheme and the system that has developed from it would not be served by tying the hands of district courts to treat magistrate judges like bankruptcy judges.

The unique features of the magistrate judge position and its exercise of authority detailed above may help explain the puzzlement the two bankruptcy law commentators on Stern expressed over the relative calm­ness of magistrate judge authority compared to bankruptcy judge auth­ority. 244 See supra note 130 and accompanying text (relaying these statements). These features may also explain why Gonzalez was an 8-1 decision and Stern was 5-4, despite both being decided by nearly identically composed Courts. See Stern v. Marshall, 131 S. Ct. 2594 (2011); Gonzalez v. United States, 553 U.S. 242 (2008). For these reasons, it would be a mistake to see Stern v. Marshall as portending a sea change in the authority of magistrate judges.

When evaluating a particular exercise of magistrate duties, a court should evaluate if the duty was properly delegated to a magistrate judge under the additional-duties clause or another part of the Federal Magistrates Act. If the reviewing court concludes that the delegation was proper under the Act, it should only be rejected as unconstitutional if the district court that delegated the duty does not retain meaningful review of the magistrate judge’s determination. This is a theory of “district court supremacy”: In all exercises of magistrate judge authority, the district judge must be supreme. To avoid Article III problems, the district judge must always be “waiting in the wings, fully able to correct errors.” 245 United States v. Raddatz, 447 U.S. 667, 686 (1980) (Blackmun, J., concurring). This quote was cited by the Court in both Peretz and Gonzalez. Gonzalez, 553 U.S. at 251; Peretz v. United States, 501 U.S. 923, 938 (1991). This is the standard circuit courts should use to determine whether a district court judge exercised sufficient oversight of a magistrate judge.

B. Applying “District Court Supremacy”

A theory of magistrate judge authority that focuses on the magistrate judge–district judge relationship instead of on litigant consent is expansive in its conception of what magistrate judges may constitutionally do, but their powers are not unlimited. Specifically, the notion of “district court supremacy” restricts magistrate judges from acting outside the explicit delegation of the district court or particularly enumerated powers in the Federal Magistrates Act. It also would not allow a district judge to delegate authority to a magistrate judge that the district judge could not later recant or overrule.

Under this interpretation of magistrate authority, the Eleventh Circuit’s holding in Brown v. United States was correct: Allowing a magistrate judge to enter a final order in a § 2255 postconviction review violates the supremacy of the district court because it requires the magis­trate judge to overrule the superior court. 246 See supra notes 167–171 and accompanying text (describing this concern in Brown). However, the Eleventh Circuit’s other arguments, expressing a concern about the facial consti­tutionality of several portions of the Federal Magistrates Act and doubt over the legitimacy of magistrate judges handling state habeas cases under § 2254, represent an overreaction to Stern v. Marshall and a failure to consider the unique features of magistrate judges as compared to bankruptcy judges. 247 See supra notes 163–166 and accompanying text (noting Stern-motivated constitutional concerns in Brown); supra notes 179–180 and accompanying text (providing Eleventh Circuit’s concerns about constitutionality of magistrate judge duties in state habeas actions). The same is true of the footnote in Brown that overrules the previous Eleventh Circuit holding that magistrate judges can accept felony guilty pleas. See supra note 189 and accompanying text (discussing this aspect of Brown). In handling a state habeas action after delegation from a district judge, a magistrate judge is still respecting the supremacy of the district court and the higher court will have an opportunity to review the magistrate judge’s handling of the habeas action. 248 One limit on the authority of magistrate judges to handle state habeas actions that is justified under this theory was explored in a recent Ninth Circuit decision. The court held that a magistrate judge generally may not hear and determine a motion to stay and abey a state habeas petition while state claims were exhausted. The court’s holding was based on the conclusion that denying such motion would be dispositive of the unex­hausted claims because it would preclude availability of a federal forum. Mitchell v. Valenzuela, 791 F.3d 1166, 1173 (9th Cir. 2015).

By contrast, the Seventh Circuit’s decision in Harden v. United States is not compatible with this theory of magistrate judge authority. If a district court judge delegates the taking of a guilty plea to a magistrate judge, the district court will still retain review of the proceeding via tran­script and will face the defendant at sentencing regardless. 249 See supra note 230 (discussing reviewability of plea proceedings by district judges). The magis­trate judge is not undermining the supremacy of the district court by accepting the plea and is not implicating Article III concerns under a functionalist theory because there is no meaningful difference between issuing a recommendation on accepting a plea to a district judge and actually accepting the plea. 250 See supra notes 194–195 (noting similarities between issuing report and accepting plea). For this reason, Harden’s argument that the Federal Magistrates Act does not permit magistrate judges to accept pleas as an additional duty does not hold up because it was based on the court’s argument that accepting a guilty plea and presiding over all stages of a felony trial are roughly equivalent for the purposes of Article III. 251 See supra notes 194–200 and accompanying text (summarizing statutory and constitutional arguments in Harden).

A recent student note argues that Harden does not go far enough in limiting magistrate judge authority to hear felony plea colloquies. 252 See Tomi Mendel, Efficiency Run Amok: Challenging the Authority of Magistrate Judges to Hear and Accept Felony Guilty Pleas, 68 Vand. L. Rev. 1795, 1829 (2015) (arguing magistrate judges’ findings should be treated “as non-final for all purposes” of Rule 11). The note attempts to resolve the uncertainty around magistrate judges’ roles in the felony process by making all magistrate judge decisions in the Rule 11 context nonfinal, thus allowing defendants to withdraw their guilty plea at any time and for any reason before the plea is accepted by a district judge. 253 Id. Though the note is correct in noting that “efficiency is not the sole concern of the Federal Magistrates Act,” 254 Id. at 1830 (emphasis added). as discussed above, the intense workload of district judges was the main driving factor in the creation of the magistrate system and in each subsequent expansion of magistrate judge authority. 255 See supra notes 24, 35–41 (discussing origin of magistrate system and subsequent expansion). As the Fourth Circuit concluded in Benton, making magistrate judge findings in the Rule 11 context always nonfinal would make a Rule 11 proceeding before a magistrate judge a “dress rehearsal” of the plea process, rendering the involvement of the magis­trate judge “meaningless.” 256 United States v. Benton, 523 F.3d 424, 432 (4th Cir. 2008). The Fourth Circuit speculated that the result would be that district courts “could stop delegating plea hearings to magistrates,” which would only “exacerbate the docket tensions.” 257 Id. at 433. The note’s motivation for taking magistrate judges out of the Rule 11 process is that the “most vulnerable” defendants, those who misunder­stand their rights and the consequences of their plea, are the ones most likely to withdraw a guilty plea. 258 Mendel, supra note 252, at 1830. But this is equally true of pleas before district judges. The note is not able to identify a connection between the protection of vulnerable defendants and the fact that magis­trate judges lack Article III protections because there is no such con­nection. As discussed above, the close connection between district judges and magis­trate judges and the ever-present possibility of review of a magistrate judge’s decision eliminate the possible harms of involving magistrate judges in accepting felony guilty pleas.

Consistent with this Note’s position that Harden and its effort to limit the role of magistrate judges in the context of felony guilty pleas is mis­taken, the only circuit court to take up magistrate judge-accepted felony guilty pleas since Harden was decided rejected its reasoning and conclusion. 259 Since Harden was decided, the Fourth Circuit has thrice reaffirmed its decision in Benton that “the Magistrates Act authorizes magistrate judges to accept a guilty plea and find a defendant guilty” when the parties consent. United States v. Shropshire, 608 Fed. Appx. 143, 144 (4th Cir. 2015) (per curiam); see also United States v. Ross, 602 Fed. Appx. 113, 115 (4th Cir. 2015) (per curiam) (“Benton rejected the precise argument that Ross now makes.”), cert. denied, 136 S. Ct. 794 (2016) (mem.); United States v. Farmer, 599 Fed. Appx. 525, 526 (4th Cir. 2015) (per curiam) (“Regardless of the Seventh Circuit’s contrary decision in Harden, we are bound by Benton.”), cert. denied, 136 S. Ct. 794 (2016) (mem.).
In addition, an Eleventh Circuit district judge (adopting a magistrate judge’s report and recommendation) has reaffirmed that circuit’s precedent in Woodard on magistrate-accepted guilty pleas and reached the same holding. See Finley v. United States, No. 3:13-cv-565-WHA, 2015 WL 4066895, at *9 (M.D. Ala. June 30, 2015) (rejecting Harden and concluding Brown “did not undermine the [Eleventh Circuit’s] holding in Woodward [sic] that magistrate judges have statutory and constitutional authority to accept felony guilty pleas”).

The persistent district judge review of magistrate judge decisions forestalls the Article III problems of allowing Article I judges to exercise the “judicial Power of the United States.” 260 U.S. Const. art. III, § 1. Magistrate judges exercise their authority through controlled and well-reasoned delegation by dis­trict judges, providing invaluable assistance to the Article III courts. Over-reading Stern to conflict with the long-established practice of district courts would be a disservice to the federal judiciary.

Conclusion

“A particular genius of the Federal Magistrates Act is that it does not mandate the assignment of particular duties to Magistrate Judges.” 261 McCabe, Guide, supra note 2, at 23. Another particular genius of the Act is that its restrictions on the assign­ment of duties to magistrate judges that district judges would want to delegate are limited and generally unobtrusive. 262 See infra Appendix D (suggesting magistrate judges generally do not have concerns about scope of their authority on matters they handle). The limits on magis­trate authority are generally left up to district courts, and that is the system’s great virtue.

Magistrate judges are unique in their relationship to Article III judges and in their importance to the operation of the federal judicial system. Using a theory of “district supremacy” to limit the authority of bankruptcy judges is the mechanism most compatible with the avowed purpose of the magistrate judge system. It would be a mistake to reduce the flexibility of district courts in assigning duties to magistrate judges in response to Stern v. Marshall. Appellate judges should not compromise the value of flexibility in the magistrate system in the name of Article III formalism when the actual control of magistrate judges by district judges is strong. Districts deserve deference.

Appendix

This Appendix includes more detailed conclusions and summaries of the interviews with magistrate judges conducted for this Note. References to interviews in the body of the Note are cited to sections of this Appendix. The sources for the Appendix itself are interview notes written during the interviews.

A. Magistrate Judge Powers

The mix of powers exercised by magistrate judges varies from district to district. In general, exercises of authority under the additional-duties clause (in either the criminal or civil context) were described by one magistrate judge as “whatever came along” from a district judge.

1. Criminal Authority and Practice. — Generally, criminal authority is more structured and routine than work on the civil side, despite only making up ten to fifteen percent of the total amount of work, according to one magistrate judge’s approximation. In the Southern District of New York (S.D.N.Y.), magistrate judges do five criminal duties a year in what the district calls “Magistrates Court.” Magistrate judges in the S.D.N.Y. sign all arrest warrants and are the first judicial officers that an arrested person sees; a magistrate judge advises an arrested person of their rights and determines bail. At that point, a felony case is transferred to the assigned district judge (a magistrate judge and a district judge are assigned randomly to every case that is docketed in the S.D.N.Y., civil or criminal, but magistrate judges handle misdemeanors on their own), but many district judges refer felony cases back to a magistrate judge for plea negotiations. When handling a plea, the magistrate judge will make a finding of competence and order submission of the plea to the district judge, with a recommendation that the district judge accept the plea. In the S.D.N.Y., magistrate judges do not accept pleas.

Other districts structure the criminal authority of magistrate judges differently. The Western District of Missouri gives all preliminary criminal matters to magistrate judges, up to and including jury selection. In that district, magistrates’ caseload is heavier on the criminal side than in other districts. In the S.D.N.Y., magistrate judge involvement in a criminal case typically ends with indictment or taking the guilty plea. Magistrate judges do not supervise criminal discovery or preside over jury selection in the S.D.N.Y.

2. Civil Authority and Practice. — The scope of civil authority varies dramatically from district to district, to a degree greater than in the criminal context. Some districts automatically assign all civil matters to magistrate judges (e.g., the Eastern District of New York (E.D.N.Y.)) while others leave discretion to district judges to decide what to assign to magis­trate judges and when (e.g., the S.D.N.Y.). In smaller districts with fewer judges (or smaller courthouses in districts with more than one court­house), magistrate judges are more likely to preside over civil trials with the consent of the litigants and spend a correspondingly greater per­centage of their time on the civil side than judges who preside over fewer trials. One magistrate judge in a relatively small district told me that he had presided over twenty-five jury trials during his four-year service as a magistrate judge.

B. Historical Evolution of Magistrate Judge Role

Without exception, every magistrate judge said that the reason that some particular area of magistrate judge authority operated in the way it did in their district was because of historical practice and tradition and not because of any deliberate choice to have magistrate judges be more or less involved than in another district. How these historical practices developed varies between districts. Again, the S.D.N.Y. provides an illus­trative example. There, one magistrate judge observes, “Everything is done by committee” because the district is so large.

The S.D.N.Y. was described as a very “decentralized” district in terms of what responsibilities district judges may delegate to magistrate judges. The decision whether to assign something to a magistrate judge is deter­mined individually by a district judge with no oversight or “district rules” besides fidelity to the Constitution and the Federal Magistrates Act. There are a small handful of district judges in the S.D.N.Y. that do not use magistrate judges at all or severely limit their duties. One judge relayed an experience with a new district judge whereby the new district judge assigned a far greater than typical portfolio of criminal matters to the magistrate judge early in the new district judge’s tenure so that he could learn how to handle criminal cases. Over time, the new district judge relied on the magistrate judge less.

C. Relationship with District Judges

In small districts where there is no formal system for assigning responsibilities to magistrate judges, one judge observed that the personal and professional relationship between a district judge and the magistrate judge assigned to the case matters a great deal in determining what responsibilities and duties the magistrate judge assumes. In larger districts, magistrate judges uniformly indicated that the personal relationship between a district judge and the assigned magistrate judge did not deter­mine what work was assigned to the magistrate judge on that particular case.

One judge discussed the differences in the relationships district judges have with magistrate judges and bankruptcy judges. In the S.D.N.Y. in Manhattan, bankruptcy judges are not in the same physical building as are district and magistrate judges and this affects the rela­tion­ship. There are also far fewer appeals of bankruptcy court decisions to district judges than there are referrals of matters from a magistrate judge to a district judge.

D. Uncertainties or Concerns

When faced with a potential exercise of authority with which they are unfamiliar, most magistrate judges first consult the Inventory of United States Magistrate Judge Duties. If the Inventory is silent on the question or indicates that the precedent is ambiguous or conflicting, magistrate judges consider if the motion presented is dispositive. If the motion is dispositive and in a “gray area,” magistrate judges will issue a report and recom­mendation to the district judge; otherwise, they will rule on the matter. As a result, the most consistent areas of uncertainty are where the dispositive/nondispositive distinction becomes cloudy. Several judges expressed uncertainty about whether they could grant or deny a motion to amend the complaint. The practice in the S.D.N.Y. is that magistrate judges can deny a motion to amend a complaint, but two magistrate judges I spoke to thought this was arguably a dispositive motion and thought they should only be allowed to issue a report and recommendation.

Two S.D.N.Y. magistrate judges expressed some concerns about the scope of their authority on sanctions because of a recent Second Circuit decision on this issue. 263 The magistrate judges were referring to Kiobel v. Millson, 592 F.3d 78 (2d Cir. 2010). One magistrate judge noted that the three judges on the panel “all wrote different things” in their respective opinions, leaving the law on magistrate judge authority over sanctions somewhat unclear in the Second Circuit.

Another area of uncertainty was how magistrate judges can handle default judgments. When there are two defendants and one appears before the magistrate judge and consents but the other does not appear, one magistrate judge stated that she would only issue a report and recommendation on the plaintiff’s motion for default judgment against the nonappearing defendant to the district judge, despite proceeding to final judgment (including possibly trial) on the matter as between the plaintiff and the appearing defendant. This judge stated that some magis­trate judges in her district would enter default judgment against a non­appearing defendant.

One issue specific to the S.D.N.Y. was related to the time gap between the completion of a plea allocution and a district judge’s accep­tance of the magistrate judge’s recommendation on accepting the plea. The government will often seek to remand the defendant without bail after the completion of the allocution, which could happen when a dis­trict judge accepts a plea on the spot after an allocution, but the district takes the position that the plea is not final until the recom­mendation is formally accepted by the district judge and that the magistrate judge can still grant bail for the time between the allocution and the acceptance of the plea by the district judge. This is still an area of contention between the U.S. Attorney and the district.

E. Discussion of Particular Powers

Because this Note concerns magistrate judge authority to accept felony guilty pleas and enter final orders in § 2255 proceedings, I asked every judge I spoke to about these particular issues. One magistrate judge noted that in his district, there has been some contention over whether magistrate judges should handle § 2255 proceedings as a matter of effi­ciency, not over Article III concerns. The argument against having magis­trate judges hear § 2255 is that the district judge has already heard the case at some point, possibly including preliminary matters (depending on the district) and certainly the trial (if one occurred) and sentencing, so it is a waste of time to have a magistrate judge retread over work that the district judge already completed.

One judge also argued that the issue of Article III and § 2255 proceedings depends on the nature of the district judge’s referral to the magistrate judge. This judge thought there was no Article III problem with a magistrate judge issuing a report and recommendation on a § 2255 to a district judge but that it was a closer question when the pro­ceeding is referred to the magistrate judge for entering a final order with consent of the parties. When asked to extend the thinking about § 2255 proceedings to § 2254 (as the Eleventh Circuit mentioned in a footnote), this judge reached the same conclusion: There may be some Article III issues with having a magistrate judge enter final orders in a state habeas case, but there are no problems with a judge issuing a report and recommendation.

I did not talk to any judges who had signed felony guilty pleas. All had only issued a report and recommendation to a district judge on whether they should accept a plea for which they conducted the plea colloquy. One S.D.N.Y. magistrate described the process of “recommending” that a plea be accepted as “an oddity.” The judge noted that the opportunity to make a finding of competence is the only interaction the district judge could have with an arrested person taking a plea outside of the sentencing pro­cess and argued that, by passing the finding of competence onto the magistrate judge, the district judge was “missing out on something impor­tant” in the criminal plea process.