Dubious Delegation: Article III Limits on
Mental Health Treatment Decisions
A common condition of supervised release requires a defendant, post-
incarceration, to participate in a mental health treatment program. Federal
district courts often order probation officers to make certain decisions an-
cillary to these programs. However, Article III delegation doctrine places
limits on such actions. This Note addresses the constitutionality of delegat-
ing the “treatment program” decision, in which a probation officer decides
which type of treatment the defendant must undergo; the choice is often be-
tween inpatient treatment and other less restrictive alternatives. The
resolution of this issue ultimately depends on whether this decision consti-
tutes a “judicial act.” Finding support in lower court case law, this Note
argues that a “judicial act” encompasses decisions affecting the defend-
ant’s significant liberty interests. The Supreme Court case law and the
mental health literature make clear that significant liberty interests are at
stake in these “treatment program” decisions. Thus, delegating the
“treatment program” decision to probation officers is unconstitutional un-
der Article III. The Note concludes by suggesting a constitutionally
permissible scheme whereby the judge orders a maximally intrusive treat-
ment while giving the probation officer the discretion to choose a less
Table of Contents
Introduction .................................................................................... 1554
I. Background on Conditions of Supervised Release
and Probation Officers ..................................................... 1557
A. Conditions of Supervised Release ................................... 1557
B. Probation Officers ........................................................... 1559
II. Supreme Court Delegation Framework ....................... 1561
A. Development of the Doctrine .......................................... 1561
B. Application of the Adjunct Theory ................................... 1565
1. Article III Oversight .................................................. 1566
2. Presence of a Judicial Act ......................................... 1568
III. A Liberty-Centered Framework...................................... 1568
A. Confusion in the Circuits ................................................. 1569
B. A Proposed Framework: Finding a Judicial Act
Where There Has Been a Significant Deprivation
of Liberty ......................................................................... 1572
1. Case Law on Delegation to Probation Officers ......... 1572
2. Case Law on Delegation to Federal Parole Boards ... 1574
Michigan Law Review
3. Other Potential Tests ................................................. 1575
IV. Liberty and Inpatient Mental Health Treatment ..... 1577
A. Application of the Test ..................................................... 1577
B. Possible Solutions ............................................................ 1580
Conclusion ....................................................................................... 1582
In the federal criminal justice system, individual liberty is placed in the
hands of an array of decisionmakers. There is the police officer who decides
whether to investigate a suspicious character; the magistrate judge who ap-
proves a warrant to search someone’s home; the jury which decides whether
or not to convict; the district judge who orders a sentence; and the probation
officer who oversees the conditions of supervised release. Sometimes, these
roles bleed together. For example, a police officer may conduct a warrant-
less search in certain situations,1 and a judge may override a jury conviction
not supported by the evidence.2 This Note addresses the overlapping roles of
a probation officer and a judge in determining the type of mental health
treatment a defendant must undergo pursuant to a condition of supervised
Imagine that you have been convicted of a federal drug crime and you
have a history of addiction and mental health issues. You are now in court to
be sentenced. Once you hear that you will be spending the next five years in
prison, you do not even give a second thought to everything else the judge is
saying. You do your time, you are released from prison, and you meet your
probation officer. At that point, the officer, not the judge, decides you need
to spend more time confined—this time, in a restrictive inpatient mental
health facility. In fact, the judge, five years ago at sentencing, ordered the
probation officer to make such a decision.
This situation has become surprisingly common: judges, in their orders
regarding conditions of supervised release, regularly delegate to probation
officers certain mental health treatment decisions. While potentially trou-
bling from a policy perspective, this issue also has a constitutional
dimension. Article III limits the exercise of the judicial power to federal
courts,3 and thus may impose constitutional restraints on this practice. This
* J.D., May 2012, University of Michigan Law School. I would like to thank my Note
Editors, Rob Boley and Ted Koehler, as well as the entire Volume 110 Notes Office for their
friendship and support. Thanks are also due to Professors Nina Mendelson and Margaret Jane
Radin, along with the student participants in the Fall 2011 Student Scholarship Workshop, for
their candid comments and criticisms. Most importantly, I would like to thank all of the
Teitelbaums, Martins, and Heidts for their constant love and support.
1. United States v. Robinson, 414 U.S. 218 (1973) (holding valid a warrantless search
made incident to a lawful arrest).
2. Fed. R. Crim. P. 29.
3. “The judicial Power of the United States, shall be vested in one supreme Court, and
in such inferior Courts as the Congress may from time to time ordain and establish.” U.S.
Const. art. III, § 1; see also N. Pipeline Constr. Co. v. Marathon Pipe Line Co., 458 U.S. 50,
Article III Limits on Mental Health Treatment 1555
Note examines the constitutionality of delegating the “treatment program”4
decision: when judges order mental health treatment upon a defendant’s
release but allow the probation officer to determine—at a later time—the
type of treatment program the defendant must undergo.
The pressure on district judges to delegate more duties to probation of-
ficers is not surprising. Undoubtedly, it is necessary that probation officers
assume certain tasks in order to “support judicial functions,”5 as probation
officers “play a vital role in effectuating the sentences imposed by district
courts.”6 District court judges face an “ever-increasing workload.7 For ex-
ample, in California from 1983 to 1992, judges’ caseloads grew by 73
percent.8 This puts pressure on all facets of the judicial system, and it sug-
gests that judges will be driven to delegate more duties to probation officers.
As a practical matter, “courts cannot be expected to map out every detail of
a defendant’s supervised release.”9
It is also intuitive to leave certain decisions to a probation officer, who is
“a liaison between the sentencing court, which has supervisory power over
the defendant’s term of supervised release, and the defendant, who must
comply with the conditions of his supervised release or run the risk of revo-
cation.”10 With regard to mental health decisions, probation officers are
closer to the situation and can monitor the defendant’s progress over time,
making adjustments when necessary.11 If these decisions were left to judges,
judges would be forced to choose between making a single determination at
the time of sentencing or delaying a hearing for a later date.12
Despite the administrability and efficiency justifications for delegating
decisions to probation officers, there are benefits to having an Article III
judge make this decision. The Constitution imparts to Article III judges cer-
tain important protections; it ensures that (1) they have life tenure, unless
59 (1982) (plurality opinion) (“The judicial power of the United States must be exercised by
courts having the attributes prescribed in Art. III.”).
4. This phrase was created for the purposes of this Note and does not appear in the
5. United States v. Nash, 438 F.3d 1302, 1305 (11th Cir. 2006) (per curiam) (quoting
United States v. Taylor, 338 F.3d 1280, 1284 (11th Cir. 2003) (per curiam)).
Taylor, 338 F.3d at 1284.
7. S. Scott MacDonald & Cynthia Baroody-Hart, Communication between Probation
Officers and Judges: An Innovative Model, 63 Fed. Probation, no. 1, 1999, at 42, 42.
9. United States v. Pruden, 398 F.3d 241, 250 (3d Cir. 2005). Further, “district courts,
to remain efficient, must be able to rely as extensively as possible on the support services of
probation officers.” United States v. Johnson, 48 F.3d 806, 809 (4th Cir. 1995).
10. United States v. Bernardine, 237 F.3d 1279, 1283 (11th Cir. 2001) (quoting United
States v. Davis, 151 F.3d 1304, 1306–07 (10th Cir. 1998)) (internal quotation marks omitted).
11. Heather Barklage et al., Probation Conditions Versus Probation Officer Directives:
Where the Twain Shall Meet, 70 Fed. Probation, no. 3, 2006, at 37, 37 (“Because every
change in circumstance cannot be anticipated at the time of sentencing, it is helpful if the
conditions of supervision can be adjusted and modified, sometimes on very short notice, in
order to meet a particular offender’s needs or answer a particular concern in the community.”).
12. MacDonald & Baroody-Hart, supra note 7, at 42.
Michigan Law Review
impeached, and that (2) their compensation can never be diminished.13 Both
of these protections were meant to ensure the independence of the judiciary
from other branches of government, and this independence must be “jeal-
ously guarded.”14 Probation officers, by contrast, are not insulated by these
constitutional protections—it would be expensive and impractical to give
these employees life tenure and salary protections. Further, a defendant may
feel that he is being treated more fairly when a judge makes this type of de-
cision after hearing and weighing the arguments through the typical judicial
process—as opposed to the probation officer making the decision on his
own whims, potentially biased by his closer relationship to the defendant.15
While these practical concerns provide perspective, this issue must ulti-
mately be resolved according to Article III delegation doctrine. This Note
argues that it is unconstitutional for a judge to delegate to a probation officer
the decision regarding the type of mental health treatment a defendant must
undergo following his or her release from jail. Part I discusses the duties and
training of probation officers, as well as the mechanics of conditions of su-
pervised release.16 Part II argues that this delegation is impermissible under
the Supreme Court’s delegation precedent if and only if a “judicial function”
has been delegated—an issue on which the Court has provided no guidance.
Part III argues that a “judicial function” has been delegated when the proba-
tion officer is allowed to make a decision depriving the defendant of a
significant liberty interest. This framework finds support in the plentiful
lower court case law17 dealing with delegations to probation officers and
parole boards. Finally, Part IV analyzes the liberty interests at hand by fo-
cusing on Supreme Court civil commitment cases and inpatient treatment
literature and ultimately determines that this delegation is impermissible.
The Note concludes by outlining a proposed solution whereby a judge or-
ders the most restrictive treatment at the time of sentencing while giving the
probation officer discretion to choose a less restrictive alternative at a later
13. U.S. Const. art. III, § 1; see also N. Pipeline Constr. Co. v. Marathon Pipe Line
Co., 458 U.S. 50, 59 (1982) (plurality opinion).
N. Pipeline, 458 U.S. at 59–60.
See infra Section II.B.1.
16. This Note uses the term “supervised release” instead of probation, as this is the
terminology used for the federal system. E.g., 18 U.S.C. § 3583 (2006 & Supp. IV 2010).
17. Because Article III only applies to federal courts, state court delegation will not be
analyzed in this Note. ASARCO Inc. v. Kadish, 490 U.S. 605, 617 (1989) (“[T]he constraints
of Article III do not apply to state courts . . . .”). This applies even when state courts are adju-
dicating federal issues. Id.; see also Lucinda M. Finley, Note, Article III Limits on Article I
Courts: The Constitutionality of the Bankruptcy Court and the 1979 Magistrate Act, 80 Col-
um. L. Rev. 560, 572 (1980) (“[State courts do] not have to conform to article III requirements
but could nonetheless exercise some of the subject matter jurisdiction described in article III
. . . .”).
Article III Limits on Mental Health Treatment 1557
I. Background on Conditions of Supervised
Release and Probation Officers
This Part provides background information on the statutory and practical
considerations involved in the implementation of post-release mental health
conditions. Section I.A focuses on conditions of supervised release, specifi-
cally those dealing with mental health treatment. Section I.B details the
appointment, duties, and training of probation officers.
A. Conditions of Supervised Release
Several important statutory provisions enable judges to fashion condi-
tions of supervised release and allow courts to supervise a defendant
following his release from prison. These statutes limit the types of condi-
tions that can be placed upon a defendant, and conditions of supervised
release must comply with these statutes. However, compliance with a statute
is not sufficient to dispose of the challenge to delegation. Article III is a con-
stitutional provision, and it places limits even on these federal laws.18
18 U.S.C. § 3583(c) references factors to be used in determining condi-
tions of supervised release. These factors are also used in determining the
length of a sentence:
(1) the nature and circumstances of the offense and the history and charac-
teristics of the defendant;
(2) the need for the sentence imposed—
. . . 19
(B) to afford adequate deterrence to criminal conduct;
(C) to protect the public from further crimes of the defendant; and
(D) to provide the defendant with needed educational or vocational
training, medical care, or other correctional treatment in the most
. . . 20
(4) the kinds of sentence and the sentencing range established . . . ;
(5) any pertinent policy statement . . . ;
(6) the need to avoid unwarranted sentence disparities among defendants
with similar records who have been found guilty of similar conduct; and
(7) the need to provide restitution to any victims of the offense.21
See Marbury v. Madison, 5 U.S. (1 Cranch) 137, 180 (1803) (“[A] law repugnant to
the constitution is void . . . .”).
19. 18 U.S.C. § 3553(a)(2)(A) is omitted because 18 U.S.C. § 3853(c) excludes this as
20. 18 U.S.C. § 3553(a)(3) is omitted because 18 U.S.C. § 3853(c) excludes this as a
21. 18 U.S.C. § 3553.
Michigan Law Review
The statute makes clear that a court may impose a condition of super-
vised release that is “reasonably related” to the enumerated factors and that
“involves no greater deprivation of liberty than is reasonably necessary” to
accomplish the above goals.22 While all of these factors are relevant, “it is
not necessary for all of the factors . . . to be present before a special condi-
tion of supervised release may be imposed.”23 The Sentencing Guidelines
specifically envisioned mental health treatment conditions: “If the court has
reason to believe that the defendant is in need of psychological or psychiat-
ric treatment—a condition requiring that the defendant participate in a
mental health program approved by the United States Probation Office [is
permitted].”24 The U.S. Probation and Pretrial Services System defines this
as a “special condition the court imposes to require an individual to undergo
evaluation and treatment for a mental disorder,” which “may include psychi-
atric, psychological, and sex offense-specific evaluations, inpatient or
outpatient counseling, and medication.”25
A condition with no basis in the record violates this statute, and there-
fore a district court must state its reasons on the record for imposing a
condition.26 Many delegation cases are disposed of on this basis because
courts seek to avoid constitutional questions where possible.27 For example,
in one case a condition requiring mental health treatment was struck down
because it was based on the groundless assumption that the defendant would
abuse his wife once released, even though he had not abused her or threat-
ened abuse for over a decade.28 On the other hand, in a case where the
defendant had a history of violent behavior and admitted to homicidal and
suicidal thoughts, such a condition was deemed warranted.29 It is in this se-
cond class of cases—where the condition is statutorily permissible—that
there are potential Article III concerns because even the impermissible dele-
gation of a permissible condition violates the Constitution. In other words,
the question of who decides has constitutional ramifications apart from the
content of the decision itself.
A condition of supervised release, once ordered by the court, is not au-
tomatically reviewed at any time. It is considered a final order. It may be
modified on rehearing,30 but this may occur only rarely for docket control
Id. § 3583(d).
23. United States v. Sicher, 239 F.3d 289, 291 (3d Cir. 2000).
24. U.S. Sentencing Guidelines Manual § 5D1.3(d)(5) (2011).
Probation and Pretrial Services: Commonly Used Terms, U.S. Courts, http://
www.uscourts.gov/federalcourts/ProbationPretrialServices/CommonlyUsedTerms.aspx (last visit-
ed Feb. 7, 2012) (emphasis added).
26. United States v. Loy, 191 F.3d 360, 371 (3d Cir. 1999).
See, e.g., Zobrest v. Catalina Foothills Sch. Dist., 509 U.S. 1, 7 (1993).
28. United States v. Kent, 209 F.3d 1073, 1077 (8th Cir. 2000); cf. United States v.
Pruden, 398 F.3d 241, 249 (3d Cir. 2005) (holding that the statute was violated where there
was no evidence of any mental health problems).
29. United States v. Wynn, 553 F.3d 1114, 1120 (8th Cir. 2009).
See, e.g., United States v. Heath, 419 F.3d 1312, 1314 (11th Cir. 2005) (per curiam).
Article III Limits on Mental Health Treatment 1559
reasons.31 While defendants have a right to appeal,32 conditions of super-
vised release are ordinarily reviewed under the deferential “abuse of
discretion” standard.33 However, when a timely objection is not made, courts
employ the even more deferential “plain error” standard.34 These considera-
tions will prove to be important factors in analyzing the constitutionality of
probation officer delegations.35
B. Probation Officers
Probation officers are considered both district court employees and
federal law enforcement officers.36 They are appointed by district courts
and can only be removed for cause.37 The probation officer’s tasks primari-
ly revolve around information. The officer investigates the defendant,
monitors the defendant’s actions, and relays observations to the court. Of-
ficer obligations are statutorily defined and include, among other express
duties,38 that the officer shall “perform any other duty that the court may
designate.”39 It is this provision that allows Article III judges to delegate
almost any duty they wish to probation officers, setting up the issue ad-
dressed in this Note.40
Probation officers are considered the “eyes and ears of the federal
courts.”41 They conduct thorough investigations, which include interviews
of the defendant and the defendant’s family and friends, and they review
criminal, medical, financial, and other relevant records.42 Probation officers
present this information to the courts in the form of a Presentence Report
(“PSR”). The PSR recommends a sentence, addresses any fines or restitu-
tion, and most importantly for this Note’s purposes, recommends conditions
of supervised release.43 When a mental health condition is recommended,
See supra notes 7–9 and accompanying text.
32. 18 U.S.C. § 3742 (2006) (providing the right to appeal one’s sentence); see also
United States v. Yellow Mule, 427 F. App’x 572, 573 (9th Cir. 2011) (“This [statute] unambig-
uously includes any right to appeal conditions of supervised release.”).
33. United States v. Allen, 312 F.3d 512, 514 (1st Cir. 2002).
See infra Section II.A.
Probation and Pretrial Services: Mission, U.S. Courts, http://www.uscourts.
gov/federalcourts/ProbationPretrialServices/Mission.aspx (last visited Jan. 9, 2012).
37. 18 U.S.C. § 3602.
38. For example, the officer must instruct the probationer of his conditions of super-
vised release, keep informed of the probationer’s conduct and report that conduct to the court,
and aid the probationer to improve his conduct. Id. § 3603(1)–(3).
Id. § 3603(10).
See United States v. Bernardine, 237 F.3d 1279, 1283 (11th Cir. 2001) (“However,
‘the catch-all provision’ is not all encompassing. A court may not delegate a judicial function
to a probation officer.”).
41. U.S. Courts, supra note 36 (internal quotation marks omitted).
Michigan Law Review
the PSR will often provide a detailed mental health history of the defend-
ant.44 While nothing in the PSR is binding, its influence should not be
underestimated. The probation officer is a primary source of facts for the
judge and is given a great deal of discretion. Some have even described the
position as a quasi-judicial role, like that of a federal magistrate.45 The
Supreme Court in discussing PSRs has stated that “[t]o deprive sentencing
judges of this kind of information would undermine modern penological
procedural policies that have been cautiously adopted throughout the na-
tion after careful consideration and experimentation.”46 The Court has
further emphasized the importance of this tool in holding that all PSRs are
Probation officers also have an important supervisory role, as they are
sometimes considered “social workers employed by the courts.”48 These
duties include implementing the court’s conditions of supervised release.
Probation officers ensure that defendants comply with the conditions and
direct them to certain services pursuant to orders by the court.49 Supervi-
sion is meant to satisfy three core criminal justice goals: (1) enforcing the
court’s order, (2) protecting the community, and (3) providing treatment
and assistance to the defendant.50 With regard to mental health treatment, a
probation officer is required to interview the family, review mental health
records, consult with medical health professionals, look for danger signs,
and refer the defendant to mental health treatment as ordered by the
court.51 There are clear challenges to this role. For example, defendants
with mental health issues are often less responsive to supervision.52 They
can also require more intensive supervision, and they may be violent or
It is important to note that these investigative and supervisory duties
do not alone raise any Article III concerns. As long as the Article III judge
makes the final decision, it makes no constitutional difference how he ob-
tains the information informing that decision. This Note deals instead with
a situation in which the final decision itself—not just the investigative
E.g., United States v. Vazquez, 381 F. App’x 168, 173 (3d Cir. 2010). Indeed, a
court may only order mental health treatment where there is evidence of a mental disorder.
United States v. Pruden, 398 F.3d 241, 249 (3d Cir. 2005).
45. Gary M. Maveal, Federal Presentence Reports: Multi-Tasking at Sentencing, 26
Seton Hall L. Rev. 544, 557–58, 596 (1996).
46. Williams v. New York, 337 U.S. 241, 249–50 (1949).
47. Gardner v. Florida, 430 U.S. 349, 359 (1977) (plurality opinion).
48. Maveal, supra note 45, at 552.
49. U.S. Courts, supra note 36.
Probation and Pretrial Services:
www.uscourts.gov/federalcourts/ProbationPretrialServices/Supervision.aspx (last visited Jan.
Supervision, U.S. Courts, http://
Article III Limits on Mental Health Treatment 1561
II. Supreme Court Delegation Framework
In determining the constitutionality of the “treatment program” deci-
sion, the first logical place to look is Supreme Court jurisprudence. As
Justice Rehnquist wryly noted, the Court’s Article III delegation cases “do
not admit of easy synthesis.”54 This was perhaps an understatement, as the
contours of the doctrine change with each subsequent Supreme Court
opinion on the subject. Section II.A summarizes the development of the
doctrine. Section II.B discusses the applicability of the “adjunct” theory
developed in those cases, ultimately concluding that the delegation of the
“treatment program” decision to a probation officer is impermissible if the
decision is considered a “judicial act.”
A. Development of the Doctrine
The first major Supreme Court case to address Article III delegation
came in 1932 with Crowell v. Benson,55 in which the Court considered the
constitutionality of delegating judicial functions to the United States Em-
ployees’ Compensation Commission, a non-Article III administrative
agency. The case arose out of the Longshore and Harbor Workers’ Com-
pensation Act, which created rights to compensation arising out of injuries
on navigable waters, while giving the Commission jurisdiction to hold
evidentiary hearings over some of those claims.56 The Court’s decision
turned on whether “essential attributes” of the judiciary remained in the
Article III court; if so, the delegation would be permissible.57 While the
Court noted the utility and policy reasons for desiring a separate fact-
finder to aid the Article III courts, it made clear that these reasons could
not drive the decision to delegate.58 The Court also rejected the contention
that this was a “public rights” case, which would constitute an exception
to Article III delegation concerns, because the case involved the liability of
one private party to another.59
Nonetheless, the Court found the delegation permissible. While the
agency was delegated the role of primary fact finder, the Court found determi-
native that all questions of law, constitutional fact, and jurisdictional fact were
subject to de novo review by an Article III court, and other questions of fact
54. N. Pipeline Constr. Co. v. Marathon Pipe Line Co., 458 U.S. 50, 91 (1982)
(Rehnquist, J., concurring in the judgment).
55. 285 U.S. 22.
56. Id. at 36–37.
Id. at 51.
Id. at 56–57. For a discussion of the policy benefits of this delegation, see supra
notes 5–12 and accompanying text.
Crowell, 285 U.S. at 50–51. These cases typically involve suits between a private
individual and the United States government. Id. at 50. This distinction sounds in separation-
of-powers considerations. When Congress creates the right, it is logical that Congress have
some leeway in defining the boundaries of those rights through adjudication. However, no
such justification exists for the encroachment on the judiciary with regard to constitutional
rights. N. Pipeline, 458 U.S. at 83–84 (plurality opinion).
Michigan Law Review
were subject to substantial evidence review.60 Further, authority for the
ultimate execution of the order was left to the district courts.61 Thus, the
essential attributes of the judicial power remained with the Article III
The Court refined its delegation doctrine again in 198062 in United
States v. Raddatz.63 The Court addressed the constitutionality of the Feder-
al Magistrates Act,64 which allowed magistrates to conduct evidentiary
hearings and propose findings of fact and recommendations for rulings on
certain motions.65 In deeming the delegation permissible, the Court noted
a key distinction from Crowell—in Raddatz, judicial tasks were delegated
to an officer appointed by the court, rather than to an administrative agency
as was the case in Crowell.66 Moreover, the Article III court retained plenary
discretion regarding whether to allow the magistrate to rule on these mo-
tions,67 magistrates were appointed and subject to removal by the district
courts,68 and less deference was given to magistrates than to the agencies in
Crowell.69 Finally, the magistrate’s findings resulted only in recommenda-
tions reviewed de novo; it was the Article III court that ultimately issued the
order.70 The Court thus concluded that it was not addressing a situation in
which the ultimate decision had been delegated.71
Crowell, 285 U.S. at 45–47, 54–57, 62–63.
Id. at 44.
62. In the meantime, the Court in 1973 held that non-Article III courts in the District of
Columbia that tried federal criminal cases were unconstitutional. Palmore v. United States,
411 U.S. 389 (1973). Some of the language in Palmore could be broadly read as requiring no
Article III limits on delegations involving criminal cases. Id. at 402 (“Nor, more particularly,
has the enforcement of federal criminal law been deemed the exclusive province of federal
Art. III courts.”). However, Palmore has since been read very narrowly, as the Court has sug-
gested that the reasoning only applies in situations where Congress has exercised its “ ‘plenary
authority’ under the District of Columbia Clause.” N. Pipeline, 458 U.S. at 76. The Court has
suggested that, in such a situation, there are lesser separation-of-powers concerns, as “Con-
gress’ power over the District of Columbia encompasses the full authority of government, and
thus, necessarily, the Executive and Judicial powers as well as the Legislative.” Id. Since
Northern Pipeline, the Court has not once relied on Palmore’s reasoning in any of its delega-
tion cases. The rejection of a broad reading of Palmore is further supported by the fact that
federal criminal cases are not deemed “public rights” cases. See infra note 109.
63. 447 U.S. 667, 669.
64. Pub. L. No. 90-578, 82 Stat. 1107 (1968) (codified as amended at 28 U.S.C.
§§ 631–639 (2006 & Supp. IV 2010)).
See generally J. Anthony Downs, Comment, The Boundaries of Article III: Delega-
tion of Final Decisionmaking Authority to Magistrates, 52 U. Chi. L. Rev. 1032 (1985);
Claudia Psome, Note, Magistrates: Constitutionality and Consent, 36 N.Y.L. Sch. L. Rev.
Raddatz, 447 U.S. at 682–83.
Id. at 681.
Id. at 685 (Blackmun, J., concurring).
See id. at 683 (majority opinion).
Id. at 682.
Id. at 681.
Article III Limits on Mental Health Treatment 1563
In 1982, the Court attempted to synthesize its previous delegation opin-
ions in Northern Pipeline Construction Co. v. Marathon Pipe Line Co.72 The
Court addressed the constitutionality of the Bankruptcy Act of 1978, which
created non-Article III bankruptcy courts and judges.73 The scheme allowed
the bankruptcy judges to hear all cases arising under or related to Title 11 of
the U.S. Bankruptcy Code.74 The case has no majority opinion, but Justice
Brennan’s plurality opinion has maintained the most influence. Justice
Brennan began by delineating three categorical historical exceptions to Arti-
cle III: (1) territorial courts,75 (2) military tribunals,76 and (3) public rights
cases.77 Most importantly for the purposes of this Note, the Court then went
on to discuss the “adjunct” theory. That theory provides that even if one of
the categorical exceptions does not apply, Article III still allows for certain
adjudicative functions to be performed by subordinates of the courts.78 Jus-
tice Brennan characterized the Crowell and Raddatz cases as prime
illustrations of the permissible use of adjuncts.79 Justice Brennan distilled
from these cases two key requirements for adjuncts. First, where the under-
lying substantive right is congressionally created (as in Crowell), more can
be delegated than where those rights are not congressionally created (as in
Raddatz).80 Second, adjuncts “must be limited in such a way that ‘the essen-
tial attributes’ of judicial power are retained in the Art. III court,”81 so that
“the ultimate decision is made by the district court.”82
Applying these principles to the case before it, the Court found the
delegation to bankruptcy courts impermissible.83 Justice Brennan noted that
state law, not Congress, created the rights at issue here.84 Accordingly, the
Court compared the scheme to that at issue in Crowell, finding that (1) the
subject matter heard by the bankruptcy courts was much broader than in
Crowell,85 (2) the bankruptcy court had total jurisdiction, not just
jurisdiction over fact finding functions,86 (3) the bankruptcy court had all
72. 458 U.S. 50 (1982) (plurality opinion).
N. Pipeline, 458 U.S. 50.
Id. at 54.
Id. at 65.
Id. at 66.
Id. at 67.
See id. at 76–77.
Id. at 78–79.
Id. at 80–81. Thus, Justice Brennan noted that the level of delegation to the magis-
trates in Raddatz was significantly more restricted than the delegation to the agency in
Crowell. Id. at 82–83.
Id. at 81.
Id. (quoting United States v. Raddatz, 447 U.S. 667, 683 (1980)).
Id. at 84–85.
Id. at 84.
Id. at 85.
Michigan Law Review
ordinary powers of the district courts, unlike the agency in Crowell,87 (4) the
orders in Crowell were set aside if “not supported by the evidence,” while
the standard for the bankruptcy courts was the more deferential “clearly
erroneous” standard,88 and (5) the bankruptcy courts, unlike the agency in
Crowell, could issue binding final judgments.89 These distinctions led the
plurality to conclude that the act impermissibly delegated to the bankruptcy
courts essential attributes of the judicial power.90
Four years later in Commodity Futures Trading Commission v. Schor,91
the Court took a more pragmatic approach to Article III delegation issues.
The Court addressed the constitutionality of the Commodity Futures Trading
Commission (“CFTC”), a non-Article III body that adjudicated claims under
the Commodity Exchange Act.92 Seemingly eschewing Brennan’s formalistic
Northern Pipeline categories,93 the Court adopted a multifactor balancing test
to determine whether Article III had been violated. The Court considered:
(1) “the extent to which the ‘essential attributes of judicial power’ are reserved
to Article III courts”; (2) “the extent to which the non-Article III forum exer-
cises the range of jurisdiction and powers normally vested only in Article III
courts”; (3) “the origins and importance of the right to be adjudicated”; and
(4) “the concerns that drove Congress to depart from the requirements of Arti-
cle III.”94 Noting that the CFTC was closer to the scheme in Crowell than the
scheme in Northern Pipeline,95 the Court also emphasized Congress’s practi-
cal justifications for the CFTC in determining its permissibility.96 Justice
O’Connor’s majority opinion concluded that there was only a de minimis
encroachment on the realm of the judicial branch.97 In his dissent, Justice
Brennan criticized this emphasis on practical considerations, warning that it
could lead to the “incremental erosion” of Article III powers.98
The Court seemed to return to a Northern Pipeline formalistic view of
Article III in its recent opinion in Stern v. Marshall.99 The Court considered
the constitutionality of the new bankruptcy court scheme, which replaced
the one invalidated in Northern Pipeline.100 The Court specifically consid-
Id. (noting the bankruptcy courts’ authority to preside over jury trials, issue declara-
tory judgments, and issue writs of habeas corpus).
Id. at 85–86.
Id. at 87.
91. 478 U.S. 833 (1986).
Schor, 478 U.S. at 835–36. Specifically at issue was the Commission’s adjudication
of state common-law counterclaims. Id. at 852.
Id. at 851 (“[T]he Court has declined to adopt formalistic and unbending rules.”).
Id. at 852–53.
Id. at 855–57.
Id. at 856.
See id. at 861–62 (Brennan, J., dissenting).
99. 131 S. Ct. 2594 (2011).
Stern, 131 S. Ct. at 2610.
Article III Limits on Mental Health Treatment 1565
ered the bankruptcy courts’ powers to decide a narrow category of claims
without de novo review by district judges.101 Writing for the majority, Chief
Justice Roberts first determined that the scheme did not fall within the “pub-
lic rights” exception to Article III.102 The Court next addressed the question
of whether the bankruptcy courts should be considered permissible adjuncts
to the district courts. In determining that the bankruptcy courts were not
adjuncts and that the Act had delegated “essential attributes” of the judici-
ary, the Court relied on the following considerations: (1) the bankruptcy
courts adjudicated issues along a broad range of subject matter;103 and
(2) the bankruptcy courts could issue final orders, reviewable only on ap-
peal.104 It made no difference to the Court that the bankruptcy judges were
appointed by the district courts rather than the president.105
In Stern, the Court notably disagreed with two elements of Justice
O’Connor’s opinion in Schor. First, echoing Justice Brennan’s dissent in
Schor, the Court rejected the concept of de minimis intrusions, stating that
“[a] statute may no more lawfully chip away at the authority of the Judicial
Branch than it may eliminate it entirely.”106 Second, the Court in Stern
seemed to dispel the role of practical considerations in impermissible-
delegation analysis. According to Chief Justice Roberts, “[T]he fact that a
given law or procedure is efficient, convenient, and useful in facilitating
functions of government, standing alone, will not save it if it is contrary to
B. Application of the Adjunct Theory
The delegation of tasks to a probation officer must be viewed under the
adjunct theory. None of the categorical Article III exceptions applies;108 the
probation officer is neither a territorial court, nor a military tribunal, nor is
the officer adjudicating public rights.109 Therefore, as in the bankruptcy and
magistrate cases, the only potential justification for this delegation is under
the adjunct theory.
Id. at 2620.
Id. at 2611.
Id. at 2618–19.
Id. at 2619.
Id. at 2620.
Id. at 2619 (quoting INS v. Chadha, 462 U.S. 919, 944 (1983)).
See N. Pipeline Constr. Co. v. Marathon Pipe Line Co., 458 U.S. 50, 65–67 (1982)
109. Public rights cases have been defined as those “between the Government and per-
sons subject to its authority.” Crowell v. Benson, 285 U.S. 22, 50 (1932). Although the United
States is a party in federal criminal cases, those types of cases do not fall within this excep-
tion. N. Pipeline, 458 U.S. at 70 n.24 (“Of course, the public-rights doctrine does not extend
to any criminal matters, although the Government is a proper party.”); see also Griffith v. Oles
(In re Hipp, Inc.), 895 F.2d 1503, 1511 (5th Cir. 1990) (“[The public rights] doctrine has never
encompassed criminal matters.”).
Michigan Law Review
The adjunct analysis can be broken down into two questions: (1) wheth-
er any judicial function is being delegated away from Article III courts; and
(2) if so, whether Article III oversight is sufficient, such that the “essential
attributes” of the judicial power remain with the Article III courts. This Note
considers these questions in reverse order. It first establishes that Article III
oversight of the “treatment program” decision is insufficient. The rest of the
Note then discusses whether the “treatment program” decision is a “judicial
function,” such that it cannot be delegated consistent with Article III.
1. Article III Oversight
Article III oversight of “treatment program” delegation is clearly insuffi-
cient for the judiciary to retain its “essential attributes.” The Court has found
the following factors relevant to whether “essential attributes” have been
delegated: (1) whether the delegee could issue final orders,110 (2) whether
the rights at issue were congressionally created,111 (3) whether the delegee’s
powers extend beyond fact finding,112 (4) whether the delegee can exercise
all ordinary powers of the district courts,113 (5) the breadth of the subject
matter delegated,114 (6) the standard of review for factual findings,115 and (7)
the standard of review for legal findings.116
All but two of these factors cut in the direction of an unconstitutional
delegation. First, the probation officer’s decision to require a certain type
of mental health treatment is, in essence, a “final order.” The officer is not
merely making a recommendation to the judge; rather, the judge has or-
dered that the probation officer has final authority to make the treatment
decision.117 Second, as in Raddatz, the officer is not deciding any congres-
sionally created rights.118 Third, this power is broader than fact finding
because the officer is making a final determination based on the facts.
Fourth, the standard of review of supervised release conditions is either
“abuse of discretion” or “plain error.”119 With regard to factual findings,
“abuse of discretion” is more deferential than the “clearly erroneous” stand-
See Stern, 131 S. Ct. at 2619; N. Pipeline, 458 U.S. at 85–86.
N. Pipeline, 458 U.S. at 81–82.
Id. at 81.
Stern, 131 S. Ct. at 2618–19.
N. Pipeline, 458 U.S. at 85. In one sense, the probation officer does not have the
power to enforce his own decision, as he cannot hold the defendant in contempt for disobeying
him. However, the court could hold the defendant in contempt for disobeying the probation
officer’s order—even without reviewing the decision itself. In contrast, in Crowell, the order
was not enforced until the matter was appealed and the district court had reviewed the deter-
mination. Crowell v. Benson, 285 U.S. 22, 44–45, 48 (1932).
See Crowell, 285 U.S. at 54–55, 62–63.
Contra United States v. Raddatz, 447 U.S. 667, 682 (1980).
See supra note 80 and accompanying text.
119. United States v. Allen, 312 F.3d 512, 514 (1st Cir. 2002). The “plain error” standard
applies if no timely objection is made at trial. Id.
Article III Limits on Mental Health Treatment 1567
ard120 found to weigh against constitutionality in Northern Pipeline.121 More
notably, this same standard of review applies to the officer’s legal determina-
tion of the appropriate level of treatment. However, the cases emphasize the
necessity of de novo review for legal determinations. In fact, in every
scheme the Court has deemed permissible, such review has been in place.122
The only factors on the other side of the scale are related: the probation
officer has not been delegated the ordinary powers of the district courts, and
the subject matter conferred is not especially broad. However, if these two
factors alone could tip the scale in favor of permissibility, a judge could reg-
ularly delegate narrow (but important) decisions to a variety of other
decisionmakers without salary and tenure protections. Article III could not
tolerate such an erosion of a judge’s judicial functions. Further, the vast ma-
jority of the factors weigh in favor of finding the delegation impermissible,
especially given the importance of the level of review for legal findings. Fi-
nally, even though there are legitimate policy reasons for desiring that
probation officers make this decision,123 the Court has rejected the im-
portance of functional concerns as a factor.124
Article III delegation, at its core, is concerned with the separation of
powers.125 One may argue that no significant separation-of-powers concerns
are at issue here, since probation officers are appointed and employed by the
district courts.126 Yet the Stern Court did not find this factor persuasive when
it considered delegation to bankruptcy court judges, who are also appointed
by the district courts.127 Even delegations that “may seem innocuous at first
blush” risk a “compromise [of] the integrity of the system of separated pow-
ers and the role of the Judiciary in that system.”128 In other words,
Congress’s mere creation of a scheme that tempts the judiciary to delegate
even minor tasks raises separation-of-powers issues. Each individual judge
will always desire to reduce his workload—a concrete, immediate benefit—
without properly considering the long-term, abstract costs of delegating
tasks to an individual not insulated by Article III protections.129
120. Amanda Peters, The Meaning, Measure, and Misuse of Standards of Review, 13
Lewis & Clark L. Rev. 233, 243–45 (2009).
121. N. Pipeline Constr. Co. v. Marathon Pipe Line Co., 458 U.S. 50, 85–86 (1982)
See, e.g., Commodity Futures Trading Comm’n v. Schor, 478 U.S. 833, 853 (1986);
Raddatz, 447 U.S. at 673–74; Crowell v. Benson, 285 U.S. 22, 65 (1932).
See supra notes 5–12 and accompanying text.
See supra note 107 and accompanying text.
N. Pipeline, 458 U.S. at 83 (discussing “the delicate accommodations required by
the principle of separation of powers reflected in Art. III”).
See U.S. Courts, supra note 36.
See supra note 105 and accompanying text.
128. Stern v. Marshall, 131 S. Ct. 2594, 2620 (2011).
See Geras v. Lafayette Display Fixtures, Inc., 742 F.2d 1037, 1052 (7th Cir. 1984)
(Posner, J., dissenting) (“The independence of the federal judiciary is as threatened by smoth-
ering Article III judges in non–Article III auxiliaries as by vesting the judicial power of the
Michigan Law Review
These delegations risk the “incremental erosion” of the judiciary.130
One need not be convinced by the concededly speculative assumption that
a probation officer may be biased by his close relationship to the defend-
ant. The assumption of bias is made in the Constitution itself: anything
less than Article III’s tenure and salary protections fails “to give [the deci-
sionmaker] maximum freedom from possible coercion or influence by the
executive or legislative branches of the Government.”131 When those Article
III protections are not in place, and the above oversight factors are left un-
fulfilled, a constitutional violation occurs.
2. Presence of a Judicial Act
The oversight factors appear unfulfilled when a probation officer makes
any decision regarding a condition of supervised release—even, for exam-
ple, choosing the day of the week for a program the judge ordered. But such
a conclusion would render almost every unreviewed action by a probation
Implicit in the Supreme Court case law, however, is a threshold require-
ment that some “judicial act” has been delegated. Only then does the extent of
Article III oversight become relevant. The Supreme Court delegation cases
each deal with clear judicial functions, such as fact finding or the issuance of
orders.132 The Court has therefore never needed to define a judicial act.
Choosing the day of a week for a program is clearly nonjudicial, but the
“treatment program” decision is more difficult. The remainder of this Note
attempts to answer this question: is the “treatment program” decision a
judicial act? If it is, then the Article III oversight is insufficient, and the
delegation is impermissible.
III. A Liberty-Centered Framework
This Part argues that a liberty-centered framework is appropriate for de-
termining whether the “treatment program” decision qualifies as a judicial
act. Section III.A summarizes the confusion in the lower courts surrounding
the “treatment program” issue, concluding that all of the circuit courts,
which have directly addressed the issue, have focused on the liberty interests
at stake in deeming the delegation impermissible. Section III.B proposes an
analytical framework, whereby the delegated decision constitutes a judicial
act if it allows for “a significant deprivation of liberty beyond that ordered
by the judge.” This formulation finds support from lower court case law ad-
United States directly in those auxiliaries, creating a corps of federal judges more dependent
on Congress’s good will than Article III judges are.”).
130. Commodity Futures Trading Comm’n v. Schor, 478 U.S. 833, 861 (1986) (Brennan,
131. U.S. ex rel. Toth v. Quarles, 350 U.S. 11, 16 (1955); see also Stern, 131 S. Ct. at
2609 (“Article III protects liberty not only through its role in implementing the separation of
powers, but also by specifying the defining characteristics of Article III judges.”).
See supra Section II.A.
Article III Limits on Mental Health Treatment 1569
dressing other types of delegations to probation officers and parole boards.
Section III.B also addresses and rejects other potential tests for resolving
A. Confusion in the Circuits
Although some have claimed that there is a circuit split on the “treat-
ment program” issue,133 all of the courts addressing the issue directly have
deemed the delegation impermissible by focusing on the liberty interests at
stake. Confusion persists, however, because so many courts fail to recognize
that the decision is being delegated. Consequently, these courts uphold or-
ders that can be read to allow probation officers to make the “treatment
For example, the Second Circuit defines the test for the permissibility of
a mental health condition delegation as follows:
If [the defendant] is required to participate in a mental health intervention
only if directed to do so by his probation officer, then this special condition
constitutes an impermissible delegation of judicial authority to the proba-
tion officer. On the other hand, if the District Court was intending nothing
more than to delegate to the probation officer details with respect to the se-
lection and schedule of the program, such delegation was proper.135
This test, followed by several circuits,136 can plausibly be read to allow
the delegation similar in type or nature to the mental health treatment deci-
sion because such a decision is inclusive of the “selection and schedule of
the program.”137 Similarly, an order of mental health treatment “[a]s deemed
necessary” by the probation officer is a common order held to be permissi-
ble.138 Under this language, a probation officer could presumably decide
which type of treatment is necessary.
Another circuit has permitted an order of treatment “under the supervi-
sion of” the probation officer.139 There, because treatment was not
E.g., United States v. Turpin, 393 F. App’x 172, 173 (5th Cir. 2010) (per curiam).
134. Few of these cases contain even a single citation to any of the Supreme Court dele-
gation cases discussed above. For example, the First Circuit upheld a delegation of the
“treatment program” decision without even recognizing that there was a constitutional issue.
United States v. Chan, 208 F. App’x 13, 16 (1st Cir. 2006) (per curiam).
135. United States v. Peterson, 248 F.3d 79, 85 (2d Cir. 2001) (per curiam) (citations
E.g., United States v. Vazquez, 348 F. App’x 168, 171 (3d Cir. 2010); United States
v. Heckman, 592 F.3d 400, 410 (3d Cir. 2010); United States v. Pruden, 398 F.3d 241, 250–51
(3d Cir. 2005); United States v. Allen, 312 F.3d 512, 516 (1st Cir. 2002).
Peterson, 248 F.3d at 85.
138. United States v. De Los Santos, 332 F. App’x 993, 993–94 (5th Cir. 2009) (per
curiam) (alteration in original) (quoting Judgment at 5, United States v. De Los Santos, No.
2:05-cr-00020 (S.D. Tex. filed Jan. 12, 2009), ECF No. 51); United States v. Grubert, 339 F.
App’x 406, 406–07 (5th Cir. 2009) (per curiam).
139. United States v. Wynn, 553 F.3d 1114, 1117, 1120 (8th Cir. 2009).
Michigan Law Review
specified,140 a probation officer would likely be the one deciding which type
of program the defendant undergoes. The most common language orders a
defendant to treatment “as directed by” the probation office or officer.141
Again, this order could be read to imply that the probation officer can direct
the defendant to that officer’s choice of mental health program.
Because circuit courts have upheld these vague orders, probation officers
can make “treatment program” decisions pursuant to their unclear terms.
But appellate court approval should not be construed to support a conclusion
that these circuit courts find “treatment program” delegation permissible. In
fact, these courts have failed to recognize that the decision has been
delegated in the first place. This is the result of several factors. First, this
issue is often one of many issues brought on appeal and one to which courts
rarely devote more than a paragraph or two.142 Another reason for the lack of
attention to the “treatment program” decision is that the debate has focused
primarily on whether the ultimate decision of “treatment versus no treatment”
has been delegated.143 It is plausible that these courts, after analyzing whether
the judge made the ultimate mental health treatment determination, have
failed to consider the constitutional implications of the remaining decisions
left to the probation officer through these broad orders.144 This lack of
thorough analysis has also resulted in lower courts conflating the
constitutional and statutory issues in these cases.145
Id. at 1117.
See e.g., Heckman, 592 F.3d at 403, 409; United States v. Miller, 341 F. App’x 931,
934 (4th Cir. 2009) (per curiam); United States v. Smart, 472 F.3d 556, 558 (8th Cir. 2006);
United States v. Stephens, 424 F.3d 876, 879, 882 (9th Cir. 2005); Allen, 312 F.3d at 515.
See, e.g., Wynn, 553 F.3d at 1120; De Los Santos, 332 F. App’x at 993; Allen, 312
F.3d at 516.
See, e.g., Wynn, 553 F.3d at 1120; Miller, 341 F. App’x at 932–33; United States v.
Nash, 438 F.3d 1302, 1306 (11th Cir. 2006); United States v. Pruden, 398 F.3d 241, 251 (3d
Cir. 2005); Stephens, 424 F.3d at 880; Allen, 312 F.3d at 516; United States v. Peterson, 248
F.3d 79, 85 (2d Cir. 2001).
See, e.g., Grubert, 339 F. App’x at 406 (“[Defendant shall] participate in a mental
health program as deemed necessary and approved by the probation officer.” (quoting Judg-
ment at 4, United States v. Grubert, No. 6:08-cr-00007 (S.D. Tex. filed Sept. 26, 2008), ECF
No. 38)); Wynn, 553 F.3d at 1117, 1120 (“[Defendant shall] participate in mental health coun-
seling under the supervision of the U.S. Probation Office . . . .” (quoting Judgment at 3, United
States v. Wynn, No. 4:07-cr-00011 (E.D. Ark. filed Aug. 27, 2007), ECF No. 24)); Allen, 312
F.3d at 515 (“[D]efendant shall participate in a program of mental health treatment, as directed
by the probation officer . . . .” (quoting United States v. Allen, No. 2:01-cr-00035 (D. Me. filed
Nov. 6, 2001)).
145. Several circuits, in determining whether a delegation was constitutionally permissi-
ble, have looked at whether there was factual support for conclusions on the defendant’s
mental health issues. E.g., United States v. Vazquez, 381 F. App’x 168, 173 (3d Cir. 2010);
Heckman, 592 F.3d at 410; Allen, 312 F.3d at 516. The First Circuit argued that “[t]he exten-
sive evidence of [the defendant’s] mental illness indicates that the court was imposing
mandatory counseling and delegating the administrative details to the probation officer, ac-
tions constituting a permissible delegation.” Allen, 312 F.3d at 516. That court appears to be
saying that the evidence of the defendant’s mental illness helps show that the judge intended
to make the treatment mandatory. This argument is a weak one, since it is difficult to imagine
this evidence being conclusive of a judge’s intent. The reasoning would be valid, however, if
Article III Limits on Mental Health Treatment 1571
The circuits that have expressly addressed the “treatment program” is-
sue—the Third, Ninth, and Tenth—have determined that this delegation is
impermissible.146 The Third Circuit, in United States v. Mangan, rejected as
unconstitutional an order that the defendant undergo mental health treatment
“as directed by the U.S. Probation Office.”147 This holding was made both
on the determination that the ultimate decision was delegated and because
the probation officer would have the authority to choose the treatment pro-
gram.148 In United States v. Esparza, the Ninth Circuit held unconstitutional
an order that the defendant participate in either an inpatient or outpatient
program, as determined by the probation officer.149 Most recently, in United
States v. Mike, a district court ordered that the defendant undergo mental
health treatment but left it to the probation officer to decide whether the
program would include residential placement.150 The Tenth Circuit found
this delegation by the district court impermissible as well.151
While none of these circuits conducted a rigorous constitutional analy-
sis, they each placed a strong emphasis on liberty interests. The Ninth
Circuit in Esparza noted the restrictiveness of inpatient psychiatric care in
terms of the “liberty interest at stake,” finding support in Congress’s protec-
tions applying to civil commitment in such facilities.152 The Third Circuit in
Mangan emphasized the “deprivation of liberty” resulting from the mental
health treatment condition,153 which the court defined as a “substantive as-
pect of Mangan’s sentence.”154 Finally, the Tenth Circuit in Mike laid out a
specific test, holding that “any condition that affects a significant liberty
interest, such as one requiring the defendant to participate in residential
the issue was not the delegation’s constitutionality but its permissibility under the United
States Sentencing Guidelines. Under the Guidelines, a condition of supervision requiring
mental health treatment is permitted “[i]f the court has reason to believe that the defendant is
in need of psychological or psychiatric treatment.” U.S. Sentencing Guidelines Manual
§ 5D1.3(d)(5) (2011). In United States v. Pruden, the Third Circuit properly determined that
the delegation was not allowable under the Guidelines because of the defendant’s lack of a
mental health history. See 398 F.3d 241, 250–51 (3d Cir. 2005). The court, for other reasons,
also determined that the delegation was unconstitutional. Id. at 251. Later Third Circuit cases,
in an attempt to distinguish Pruden, used the defendant’s mental history as support for the
delegation’s constitutionality. Vazquez, 381 F. App’x at 173; Heckman, 592 F.3d at 410.
146. United States v. Mike, 632 F.3d 686 (10th Cir. 2011); United States v. Mangan, 306
F. App’x 758 (3d Cir. 2009); United States v. Esparza, 552 F.3d 1088 (9th Cir. 2009) (per
147. 306 F. App’x at 761 (quoting Judgment at 3, United States v. Mangan, No. 2:07-cr-
00136 (D.N.J. filed Sept. 27, 2007), ECF No. 27).
Mangan, 306 F. App’x at 761 (“In our view, it is the District Court that must deter-
mine the type and duration of mental health treatment. These are substantive aspects of
Mangan’s sentence and not simply administrative details.”).
149. 552 F.3d at 1091.
150. 632 F.3d at 698–700.
Mike, 632 F.3d at 699.
Esparza, 552 F.3d at 1091; see also Mike, 632 F.3d 686.
Mangan, 306 F. App’x at 759.
Id. at 761.
Michigan Law Review
treatment . . . must be imposed by the district court.”155 This emphasis sug-
gests that any assessment of the constitutionality of “treatment program”
delegation must consider the significant liberty interests at stake.
B. A Proposed Framework: Finding a Judicial Act Where There
Has Been a Significant Deprivation of Liberty
This Note argues that a modified version of the Mike test, asking wheth-
er the delegated decision allows for “a significant deprivation of liberty
beyond that ordered by the judge,” provides the proper framework for re-
solving the “treatment program” issue. Answering the question affirmatively
is sufficient to conclude that a “judicial act” has been delegated.156 There is
nothing more inherent to the judicial power than the ability to restrict a per-
son’s liberty—a power most evident in the court’s authority to order
physical confinement. Additionally, decisions affecting liberty interests have
some of the most serious consequences, and therefore, benefit most from
Article III protections.157
This intuition finds support in two other sources of case law: other types
of delegation to probation officers, as discussed in Section III.B.1, and dele-
gation to federal parole boards, as discussed in Section III.B.2. Finally,
Section III.B.3 rejects other potential tests for defining a “judicial function.”
1. Case Law on Delegation to Probation Officers
This Section discusses delegation of decisions with regard to drug test-
ing, personal contact provisions, polygraph testing, and the issuance of a
summons and argues that the decisions in these cases are consistent with the
proposed liberty test.
Several circuits have encountered delegation issues regarding conditions
of probation that require drug testing. The issue arises when the court does
not specify the number of drug tests to be administered, leaving the decision
to the probation officer. The First and Seventh Circuits have determined that
Mike, 632 F.3d at 696 (citations omitted).
156. However, this Note does not argue that passing this test is necessary for an act to be
deemed judicial. For example, a majority of circuits have found the delegation of restitution
payment schedules impermissible. E.g., United States v. Overholt, 307 F.3d 1231, 1256 (10th
Cir. 2002); United States v. Merric, 166 F.3d 406, 409 (1st Cir. 1999); United States v. Gra-
ham, 72 F.3d 352, 357 (3d Cir. 1995); United States v. Johnson, 48 F.3d 806, 809 (4th Cir.
1995); United States v. Porter, 41 F.3d 68, 71 (2d Cir. 1994); United States v. Albro, 32 F.3d
173, 174 (5th Cir. 1994) (per curiam); United States v. Gio, 7 F.3d 1279, 1292 (7th Cir. 1993).
Thus, there may be situations where a decision significantly affecting proprietary or monetary
interests may also qualify as a judicial act, though this is beyond the scope of this Note.
This test, on its face, could imply that any significant liberty-depriving act of a private
party could be deemed judicial. This is obviously not the case. It would probably be more
accurate to say that the test only applies in a situation where the act is taken to decide a legal
controversy. See, e.g., John Locke, Second Treatise of Government § 131 (C.B. Mac-
pherson ed., Hackett Publ’g Co. 1980) (1690). However, it is not necessary to delve deeply
into the issue, as decisions by probation officers surely satisfy this prerequisite.
See supra notes 125–131 and accompanying text.
Article III Limits on Mental Health Treatment 1573
the delegation of this decision is unconstitutional.158 The Ninth Circuit held
that the probation officer could determine the number of tests, as long as the
court set a maximum number.159 No circuit that has addressed the issue has
found it entirely permissible to delegate this decision. Additionally, the
Third Circuit has held that it is not permissible to delegate to a probation
officer the decision as to whether the defendant can have contact with cer-
tain individuals.160 The Ninth Circuit, in its only case addressing the issue,
allowed such delegation because the court was only soliciting the probation
Yet there have been several types of delegation to probation officers that
courts have deemed acceptable. The First Circuit has determined that, once
a court orders that a defendant submit to polygraph testing, it is permissible
to allow the probation officer to determine the scope of the questions, the
frequency, and the duration of such tests.162 Also, the Eleventh Circuit ruled
that once a court decides to issue a summons, it can delegate to the proba-
tion officer the actual task of issuing the summons.163
Although the reasoning of these cases does not expressly turn on wheth-
er significant liberty interests are at stake, the outcomes suggest such a
distinction. Drug tests and no-contact provision decisions each implicate
significant liberty interests—in one’s own bodily fluids and in the freedom
to interact with others respectively. This informs why courts are hesitant to
allow their delegation. On the other hand, tasks that courts have found per-
missible, such as administering a polygraph test and issuing a summons, do
not implicate such serious liberty interests. While there may be other expla-
nations for these outcomes,164 it is significant that application of the Mike
test would not require overturning this substantial body of case law.
See, e.g., United States v. Tejeda, 476 F.3d 471, 473–74 (7th Cir. 2007); United
States v. Padilla, 415 F.3d 211, 224 (1st Cir. 2005) (en banc). However, both circuits, while
holding that the delegation did constitute error, held that no reversible plain error existed, as it
could not be shown that the defendants would be any worse off had the district court, rather
than the probation officer, determined the number of drug tests to be administered. Tejeda, 476
F.3d at 475; Padilla, 415 F.3d at 223. Again, there have been questions as to what type of
language constitutes delegation. See United States v. Figueroa-De-La-Cruz, 128 F. App’x 775,
777 (1st Cir. 2005) (per curiam).
159. United States v. Stephens, 424 F.3d 876, 883 (9th Cir. 2005). Interestingly, the court
narrowed this holding to situations in which drug tests were ordered specifically. If there was a
general treatment in which drug tests were a part of, the court is not required to set a maxi-
mum number. Id. This became a non-issue after district courts issued general orders capping
the maximum allowed of drug tests allowed. See, e.g., United States v. Toson, No. 94-1136GT,
2007 WL 1114052, at *2 (S.D. Cal. Apr. 4, 2007) (noting that the district court’s General
Order No. 547 “cured any possible Stephens problems”).
160. United States v. Voelker, 489 F.3d 139, 154 (3d Cir. 2007). The Fifth Circuit, citing
Voelker, has allowed the delegation where the condition was in place for only a period of sev-
eral years. United States v. Rodriguez, 558 F.3d 408, 416 (5th Cir. 2009).
161. United States v. Bowman, 175 F. App’x 834, 838 (9th Cir. 2006).
162. United States v. York, 357 F.3d 14, 21 (1st Cir. 2004).
163. United States v. Bernardine, 237 F.3d 1279, 1283 (11th Cir. 2001).
See infra Section III.B.3 (discussing and rejecting other potential tests).
Michigan Law Review
2. Case Law on Delegation to Federal Parole Boards
The United State Parole Commission (“USPC”)165 has also posed Article
III delegation issues. Although its constitutionality has been consistently
upheld, the reasoning in these decisions provides further support for the
“significant deprivation of liberty” test.
The USPC maintains broad authority to grant the release of incarcerated
individuals. The Parole Commission and Reorganization Act166 created the
USPC and empowered it to make parole release decisions for eligible feder-
al prisoners.167 According to the statutory framework, eligibility for release
depends on the sentence imposed by the Article III judge. If the judge orders
a definite sentence, the prisoner is usually ineligible for release until he has
served one-third of the sentence.168 However, if the judge sets only a mini-
mum term of incarceration, the prisoner is not eligible for parole until that
term ends.169 Finally, if the judge sets a maximum sentence only, the USPC
has authority to release the prisoner at any time.170 Once the prisoner is eli-
gible for release, the USPC follows a detailed set of statutory guidelines for
determining whether release should be granted.171 The constitutionality of
such actions has been “long-established” even in the face of Article III chal-
The ability of non-Article III actors to make such drastic incarceration
decisions is still consistent with the probation officer delegation precedent,
as discussed in Section III.B.2. The reasoning behind the USPC cases
turns on the fact that the judge retains control over the minimum and max-
imum deprivation of liberty:
The Commission may not require a prisoner to spend a single day in prison
longer than his judicial sentence dictates, regardless of whether it thinks
that the judge mischaracterized the nature of the offense or the offender’s
potential for committing further harm. In addition, the Commission may
not release a prisoner even one day earlier than his judicially set parole eli-
gibility date . . . .173
165. In 1984, Congress repealed the statute creating the USPC. Act of Oct. 12, 1984,
Pub. L. No. 98-473, § 218(a)(5), 98 Stat. 1837, 2027. Though the repeal was initially to take
effect five years after November 1, 1987, see id. § 235(b), 98 Stat. at 2032–33, the effective
date has been pushed back by a string of subsequent legislation, the most recent in 2011. See
United States Parole Commission Extension Act of 2011, Pub. L. 112–44, 125 Stat. 532 (ex-
tending the life of the USPC to November 2013).
166. Pub. L. No. 94-233, 90 Stat. 219 (1976) (repealed 1984).
167. 18 U.S.C. § 4203(b) (repealed 1984), reprinted in 18 U.S.C. ch. 311 note at 775
Id. § 4205(a), reprinted in 18 U.S.C. ch. 311 note at 776.
Id. § 4205(b)(1).
Id. § 4205(b)(2).
Id. § 4206(a), reprinted in 18 U.S.C. ch. 311 note at 777.
172. Hansen v. U.S. Parole Comm’n, 904 F.2d 306, 311 (5th Cir. 1990).
173. Geraghty v. U.S. Parole Comm’n, 719 F.2d 1199, 1212 (3d Cir. 1983).
Article III Limits on Mental Health Treatment 1575
Such a scheme is not much different than Congress’s ability to set terms
of punishment for specific crimes.174 Even if a judge may have had an ex-
pectation of the length of incarceration, the Supreme Court has deemed this
The USPC cases have found the parole scheme constitutional because
the USPC’s decision “does not enhance the sentence”176 or “extend [the
defendant]’s sentence.”177 In other words, any extension of a prison sen-
tence—any additional deprivation of liberty—beyond the sentence the
judge orders would be impermissible. This parallels the test proposed in
this Note, which would disallow “a significant deprivation of liberty be-
yond that ordered by the judge.”
Therefore, the USPC cases support this Note’s liberty-centered test,
subject to two important distinctions.178 First, in the “treatment program”
cases, the judge is not setting a maximum amount of deprivation (such as a
prison sentence), but rather is ordering participation in a generic mental
health treatment program.179 However, this is no less troublesome, since it is
impossible to know what the judge has in mind when making that order.
Unless the judge desired that the defendant undergo the most restrictive
program, there is always the possibility that the probation officer can choose
a program more restrictive—and therefore, more liberty depriving—than the
judge contemplated. Second, the Federal Parole Board scheme does not al-
low probation officers to release a prisoner below a minimum term.180
However, this is only a statutory requirement, and the case law provides no
reason to believe this plays a role in the constitutional analysis.
3. Other Potential Tests
Before proceeding, this Section will briefly address two other potential
frameworks for determining whether an act is a judicial function: the dis-
tinction between “ministerial tasks” and “ultimate responsibility,” and the
175. United States v. Addonizio, 442 U.S. 178, 190 (1979) (“To require the Parole
Commission to act in accordance with judicial expectations, and to use collateral attack as a
mechanism for ensuring that these expectations are carried out, would substantially undermine
the congressional decision to entrust release determinations to the Commission and not the
courts.”). Despite this language, the Supreme Court has never reached the Article III issue;
Addonizio was decided on statutory grounds. See id. at 186.
176. Page v. U.S. Parole Comm’n, 651 F.2d 1083, 1085 (5th Cir. Unit A July 1981) (per
177. Weppner v. U.S. Parole Comm’n, 95 F.3d 47, 47 (5th Cir. 1996) (per curiam).
178. A third distinction is that the USPC is an agency in the executive branch, see 18
U.S.C. §§ 4201–4218 (repealed 1984), reprinted at 18 U.S.C. ch. 311 note 774–81 (2006),
while probation officers are employees of the judiciary. 18 U.S.C. § 3602 (2006). However,
this is a distinction without a difference in the Article III delegation analysis. See supra notes
125-129 and accompanying text.
E.g., United States v. Stephens, 424 F.3d 876, 878–79, 882 (9th Cir. 2005).
180. Parole Commission and Reorganization Act § 4205(a).
Michigan Law Review
distinction between punishment- and rehabilitation-motivated decisions.
Neither distinction provides a workable test for permissible delegation.
A handful of courts have determined the constitutionality of delegations
to probation officers by distinguishing between “ultimate responsibility” and
“ministerial act[s].”181 But courts have not adequately fleshed out this test.182
They have not recognized any factors that courts should consider when
drawing this line, and the analysis includes little more than conclusory as-
sertions.183 Regardless of the merits of the test, the “treatment program”
decision cannot be deemed ministerial under any formulation. Because of
the open-ended nature of these orders, the probation officer has total discre-
tion in the choice of program; there are no guidelines to follow or factors to
consider. Thus, there is no support for the notion that probation officers can
make the “treatment program” decision “without regard to or exercise of
[the officers’] own judgment upon the propriety of the act being done.”184 In
addition, a decision with such significant implications for the lives of de-
fendants185 is not one “involving merely execution of a specific act arising
from fixed and designated facts.”186
At first glance, the punitive-versus-rehabilitative distinction has more
appeal. The First Circuit hinted at this test in United States v. Merric, when
it discussed the punitive aspects of restitution payments in determining that
a probation officer could not set the schedule for these payments.187 Howev-
er, while punishment is one judicial function, it is not necessarily the sole
judicial function.188 In deciding the constitutionality of a court setting aside
a sentence mandated by statute,189 the Supreme Court indeed stated that,
“under our constitutional system the right to try offences against the crimi-
nal laws and upon conviction to impose the punishment provided by law is
judicial.”190 Based on this language, it may very well be true that all punitive
E.g., United States v. Miller, 341 F. App’x 931, 932–33 (4th Cir. 2009) (per curiam)
(quoting United States v. Bernardine, 237 F.3d 1279, 1283 (11th Cir. 2001)) (internal quota-
tion marks omitted); see also United States v. Nash, 438 F.3d 1302, 1305 (11th Cir. 2006) (per
182. In fact, these courts supply little discussion at all as to why the delegation at issue
should fall into one category over another. See, e.g., Miller, 341 F. App’x at 932–33; Nash,
438 F.3d at 1305.
E.g., Nash, 438 F.3d at 1306 (“[T]he district court has delegated to the probation
officer the ‘ultimate responsibility’ of whether Nash must participate in mental health counsel-
ing instead of the ‘ministerial function’ of how, when, and where the defendant must
See Ballentine’s Law Dictionary 803 (3d ed. 1969) (defining “ministerial act”).
See infra Section IV.A.
See Ballentine’s Law Dictionary 803 (3d ed. 1969) (defining “ministerial du-
187. 166 F.3d 406, 409 (1st Cir. 1999).
188. Perhaps the same could be said about liberty—that not all judicial functions impli-
cate significant liberty interests. But this Note does not make that assertion; it only argues the
inverse, that all acts affecting significant liberty interests constitute judicial functions.
Ex parte United States, 242 U.S. 27 (1916).
Id. at 41.
Article III Limits on Mental Health Treatment 1577
actions must be left to the Article III judge. But this does not require that
punitive decisions be the only actions over which a judge has sole authority.
For example, the orders given by a judge in a typical bankruptcy action are
not punitive, yet the Supreme Court determined that they may be non-
delegable.191 Additionally, imposition of liability in civil cases is surely a
“judicial function,” though most civil remedies would not be considered
This distinction could also lead to an absurd result. The Supreme Court
has noted that the probation and parole system generally has a rehabilitative
rather than punitive focus,193 and the statutory scheme also reflects this.194
Therefore, under the punitive-versus-rehabilitative distinction, all decisions
regarding conditions of supervised release could be delegated to probation
officers. This would overturn whole bodies of case law. For example, such a
distinction would allow officers to make even the ultimate decision of treat-
ment versus no treatment—a decision that the circuits all agree is not
delegable,195 and which has serious implications.196
On the other hand, the liberty-focused test is grounded in case law, as
shown by its consistency across a range of delegation cases.197 Further,
courts commonly are adept at analyzing liberty interests, as shown by Due
Process Clause jurisprudence.198 Thus, the test proposed in this Note does
not face the obstacles burdening these alternatives.
IV. Liberty and Inpatient Mental Health Treatment
Part IV considers whether a significant liberty interest is at stake in this
“treatment program” decision. Section IV.A concludes, based on Supreme
Court language and inpatient mental health treatment literature, that such an
interest is implicated and the delegation is therefore unconstitutional. Section
IV.B briefly considers ways in which the current scheme can be amended to
comply with Article III.
A. Application of the Test
The Supreme Court, in dicta from a line of civil commitment cases,
has strongly suggested that the “treatment program” decision implicates a
See generally N. Pipeline Constr. Co. v. Marathon Pipe Line Co., 458 U.S. 50
192. Hence the distinction between compensatory and punitive damages.
193. Gagnon v. Scarpelli, 411 U.S. 778, 785 (1973).
Compare 18 U.S.C. § 3553 (2006 & Supp. IV 2010) (listing the need for providing
just punishment as a factor to be considered in sentencing), with id. § 3583 (incorporating
several of the sentencing factors to conditions of supervised release, but expressly omitting the
See supra note 143 and accompanying text.
See infra Section IV.A.
See supra Part III; infra Sections IV.A, IV.B.
See infra Section IV.A.
Michigan Law Review
significant liberty interest, and nonlegal mental health treatment literature
supports this conclusion. Because the delegation issue will be adjudicated
on appeal before the probation officer ever makes a decision, the analysis
must be facial rather than as applied. In other words, since it is impossible to
know ex ante what decision the probation officer would ultimately make, it
makes sense to look at the deprivation of liberty of the officer’s most restric-
tive option. Thus, the focus will be on inpatient mental health treatment.
The Supreme Court has mandated significant procedural protections for
involuntary civil commitment, primarily because inpatient treatment is ex-
tremely restrictive on liberty. The Court first addressed civil commitment in
O’Connor v. Donaldson,199 when it considered whether a state could confine
an individual solely because he or she is determined to have a mental ill-
ness.200 The Court acknowledged that civil confinement may in some cases
protect the individual from self-inflicted harm but rejected the notion that
the possible increased quality of life in an institution was sufficient to hold
him against his will.201 Because the individual was not dangerous, the state’s
confinement was deemed a violation of due process.202
Four years later, the Court addressed the proper burden of proof for civil
commitment decisions in Addington v. Texas.203 The Court noted the need to
balance the individual’s liberty interests against the state’s legitimate parens
patriae interests in caring for its citizens, as well as its police power inter-
ests in protecting its citizens from the dangerous mentally ill.204 In
conducting this balancing test, the Court determined that due process re-
quires at least a “clear . . . and convincing evidence” standard.205 At the core
of both of these decisions was the liberty interest being infringed by inpa-
tient commitment, and the Court used strong language to assert this. In
O’Connor, the Court spoke of the individual’s “constitutional right to free-
dom.”206 In Addington, the Court noted that “civil commitment . . .
constitutes a significant deprivation of liberty that requires due process pro-
tection.”207 Even where an individual has already been convicted of a crime
(and arguably has lesser liberty interests), the Court has deemed post-
incarceration civil commitment a “massive curtailment of liberty.”208 The
Court’s holding that the Due Process Clause requires certain procedures
199. 422 U.S. 563 (1975).
O’Connor, 422 U.S. at 575.
Id. at 576. The issue of due process is outside the scope of this Note.
203. 441 U.S. 418 (1979).
Addington, 441 U.S. at 426.
Id. at 433 (internal quotations marks omitted).
O’Connor, 422 U.S. at 576.
Addington, 441 U.S. at 425.
208. Humphrey v. Cady, 405 U.S. 504, 509 (1972).
Article III Limits on Mental Health Treatment 1579
before a defendant’s transfer from a prison to a mental health facility also
supports this proposition.209
Literature on inpatient treatment further confirms the Court’s conclusion
that a significant liberty interest is at stake in inpatient mental health treat-
ment decisions. The typical inpatient stay restricts liberty in a way mirroring
that of a prison sentence. For example, patients often have limited access to
cell phones or other forms of communicating with the outside world.210 The
hospital or treatment facility may also confiscate the patient’s belongings.211
Some patients are barred from leaving the facility under any circumstances
because the doors are locked twenty-four hours a day.212 At the very least,
most patients are restricted to the hospital unit’s boundaries for the first
twenty-four hours, with any subsequent increase in freedom left in the hands
of doctors and nurses.213
Inpatient treatment also entails a risk of physical harm. Even with in-
creasing quality of care,214 there are still significant dangers in inpatient
treatment. For example, it is estimated that 150 deaths are caused each
year in the United States from the use of seclusion and restraints in psy-
chiatric hospitals.215 Further, a random sampling of 31 psychiatric hospital
patients detected 2,194 medication errors during the patients’ collective
1,448 inpatient days.216 58 percent of those errors were judged to have the
potential to cause severe harm.217 More serious methods are used for dis-
ruptive patients: “Seclusion is the placement and retention of an inpatient
in a bare room for containing a clinical situation that may result in a state
of emergency. Restraint involves measures designed to confine a patient’s
bodily movements.”218 Patients may be subject to potentially more serious
See Vitek v. Jones, 445 U.S. 480, 494–96 (1980) (“[P]risoners facing involuntary
transfer to a mental hospital are threatened with immediate deprivation of liberty . . . .”).
210. Lauri Kuosmanen et al., Deprivation of Liberty in Psychiatric Hospital Care: The
Patient’s Perspective, 14 Nursing Ethics 597, 601 (2007). These results were based on inter-
views of fifty-one patients in two acute psychiatric inpatient wards. Id. at 597.
Id. at 601 (citing clothes, money, and other personal belongings).
Id. at 600.
213. Richard L. Munich & Pamela K. Greene, Psychosocial Approaches in Inpatient
Psychiatry, in Principles of Inpatient Psychiatry 17, 20–21 (Fred Ovsview & Richard L.
Munich eds., 2009).
214. Nancy P. Hanrahan & Linda H. Aiken, Psychiatric Nurse Reports on the Quality of
Psychiatric Care in General Hospitals, 17 Quality Mgmt. Health Care 210, 210–11 (2008),
available at http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2582015/pdf/nihms-75082.pdf. This
study sample consisted of 456 registered nurses permanently assigned to psychiatric units, as
well as a larger sample of 11,071 registered nurses who work permanently on medical, surgi-
cal, or medical–surgical units. Id.
Id. at 211.
218. Eila ES Sailas & Mark Fenton, Seclusion and Restraint for People with Serious
Mental Illnesses, Cochrane Database Systematic Revs. at 2 (2000), available at
D7F8DDD61E434CC071B2A425FC.d03t02 (“The rate of seclusion and restraint in the 23
Michigan Law Review
deprivations of liberty by being forced to take medications either orally or
by injection.219 These dangers have led to legitimate fears. In 2008, of those
with unmet mental health care needs, 9.6 percent did not seek treatment for
fear of being committed or being forced to take medicine.220 While such
serious liberty deprivations may not be the norm, these risks are probative of
the liberty interests at stake.
The literature and cases make clear that inpatient mental health treat-
ment can result in a significant curtailment of a defendant’s liberty interests.
Because imposing deprivations of liberty is a function of the judiciary, it
follows that the “treatment program” decision is an exercise of a judicial
function. And because the court delegates a judicial function and Article III
oversight is inadequate, delegating the “treatment program” decision to a
probation officer does not pass scrutiny under the “adjunct theory.” There-
fore, as interpreted by the Supreme Court, Article III requires that an Article
III judge make the “treatment program” decision.
B. Possible Solutions
While the main purpose of this Note is to establish the impermissibility
of delegation of the “treatment program” decision to a probation officer, the
discussion prompts the obvious question: how can the current scheme be
made constitutional? This Note outlines some potential solutions.
At one extreme, judges can simply decline to delegate this decision, al-
ways specifying in their orders the type of treatment program to be
administered. This would, of course, skirt the constitutional issue. It would
also result in the policy benefits of a having neutral, insulated decisionmaker
weighing all of the relevant considerations.221 The main objection to this
proposal is that probation officers can make more individually tailored deci-
sions, while making adjustments to the treatment over time.222 Judges, on
the other hand, have much less flexibility. While they technically could
modify their decision upon rehearing,223 docket control concerns would like-
ly preclude this.224 The practical result is that judges would be forced to
make an untailored treatment decision at the time of sentencing.
hospitals [studied] ranged from 0.4 to 9.4% of patients.”). These conclusions were drawn from
a survey of multiple studies. Id.; see also Munich & Greene, supra note 213, at 31 (“The pa-
tient is physically held and escorted with force or possibly put on the floor, and then carried by
staff to a seclusion area.”).
219. Kuosmanen et al., supra note 210, at 601.
220. Substance Abuse & Mental Health Servs. Admin., U.S. Dep’t of Health &
Human Servs., Results from the 2008 National Survey on Drug Use and Health:
National Findings 102 (2009), available at http://oas.samhsa.gov/nsduh/2k8nsduh/
See supra notes 13–15 and accompanying text.
See supra notes 11–12 and accompanying text.
See, e.g., United States v. Heath, 419 F.3d 1312, 1314 (11th Cir. 2005) (per curiam).
See supra notes 7–8 and accompanying text.
Article III Limits on Mental Health Treatment 1581
Another option is to create a framework with sufficient oversight by the
Article III judge so that delegation even of the judicial act can be permissi-
ble.225 Like the schemes in Crowell and Raddatz, the probation officers could
provide a proposal to be reviewed de novo by the district judge.226 Such a
scheme would be permissible, as it would avoid the main problem of having
all discretion delegated away in the judge’s final order. In fact, it would not
appear very different from the magistrate scheme approved in Raddatz.227
However, it is unclear whether this would be significantly less burdensome
than having a rehearing for the judge to modify the final decision. It still re-
quires further action taken by a judge and therefore raises the same efficiency
One proposal would likely avoid the constitutional issue while striking a
balance between the tailored decision by the probation officer and the neu-
tral decision-making of the Article III judge. The judge, at the sentencing
hearing, could set the maximum restrictiveness of the mental health program
while allowing the probation officer to choose a less restrictive program at
some point in the future. For example, the judge would order the defendant
to undergo “a health treatment program chosen by the probation officer, but
no more restrictive than an inpatient program [or program X].” This would
be consistent with the test outlined in this Note, as it would not allow the
probation officer to deprive the defendant of any additional liberty beyond
that which the judge ordered.228 The proposal would also allow probation
officers to tailor the decision over time, as long as they act within the con-
fines of the order. And here, as opposed to the impermissible wide-open
delegation, the judge must decide the maximum treatment after considering
its suitability for the particular defendant.
One may object that a judge can easily overcome this obstacle by rou-
tinely ordering inpatient treatment as the maximum, resulting in a delegation
of the entire decision to the probation officer. However, the same objection
could be made to other delegations the Supreme Court has deemed per-
missible. A judge could decide to always rubber stamp a magistrate’s
decision229 or the decision of an administrative agency.230 But we must
assume that the judge will follow the law—here, the law requires that the
judge give an independent judgment as to the propriety of the maximum
treatment. Further, as demonstrated above, a significant problem with this
See supra Section II.B.
See United States v. Raddatz, 447 U.S. 667, 681–82 (1980); Crowell v. Benson, 285
U.S. 22, 54 (1932).
See Raddatz, 447 U.S. at 682. Both magistrates and probation officers are appointed
by the courts. 18 U.S.C. § 3602 (2006); Raddatz, 447 U.S. at 685.
228. This is also consistent with the federal parole board case law, where judges were
permitted to set a maximum jail sentence and allow the parole board to shorten it. See supra
Section III.B. It is also consistent with the drug test cases, where the courts found it permissi-
ble for the judge to set a maximum number of drug tests in conditions of supervised release.
See supra notes 158–161 and accompanying text.
See Raddatz, 447 U.S. 667.
See Crowell, 285 U.S. 22.
Michigan Law Review
decision is that judges do not realize what it is they are delegating.231 Forc-
ing the judge to articulate the maximally intrusive treatment solves this
problem by bringing the delegation front and center. Under the proposed
solution, the judge would be required to contemplate the implications of the
treatment and its appropriateness for the particular defendant.
While “courts cannot be expected to map out every detail of a defend-
ant’s supervised release,”232 Article III places limits on which details must in
fact be mapped out by courts. The mental health “treatment program” deci-
sion is one such detail. Supreme Court precedent makes clear that the
scheme under which this decision is made does not provide sufficient over-
sight over judicial acts delegated from the Article III judge to the probation
officer. By applying the liberty-focused test proposed in this Note, the
“treatment program” decision—one with significant liberty ramifica-
tions—qualifies as a judicial act, and its delegation is therefore
impermissible. Instead of delegating this decision, judges should set a
maximum-restrictive treatment, allowing a probation officer to choose any
equally or less restrictive program.
See supra Section III.A.
232. United States v. Pruden, 398 F.3d 241, 250 (3d Cir. 2005).